Me CREA AE SF a a ert Se ye

i" ee ee iy CT el LY

apd Ret EME by 1

ere wa Pd ae dL oe

ee eee er ae

eC as yews

Wy fe FNM eerie:

Sw Sgt Atle i 1G

hel

GRO Rear Ue ee

SMe Me eet ts

Me Catan te le hate

La ate

Tethys beta ve

fetch Weems ott Ud OVI ENE otert SUp Oy Hiei ANE IRE Said aging

hig Ee ERS eR Marte a UE NT taint LNG facta eat tpt LN rene ith poral a pale A Seah YEE ANY EPO Obs LNG akeeiall,

the Ne heg eGin DAS We Po ae at a MCE Ge Petal ot WATS Sind etl BaD UAH MENS (VCE A ae Scpetal i tre UPN N pind dh TV PMEE SRL BAYOU EPIC ae SU UNdY Aa Faves Yet ate

Ey ate Woy amet

POW VOW CDs AU aE tees

wal Te

Sy yer aty

Pay sre ny tae flte MENERE CN elses el alte ead What e BY

Leg Mos yAieTra weber HAPS OT g by Ueda bee ate Serttg COn gn ty!

Cotas ee GAO et

TITY MPM ERS Eb Hoday at

wee Ae

eh Bs Be eget

Galan dy tte

eri prry

wrelids

ea Venn

Tee tt

Wy dathe by

ee adie fina we een ee KER CE UE NIE OM GOy 2 eLS INE BO SN te ot aaa Car we et SHC DA ERIVRL SOL Carty Ly Tipsisacn : Re NOE: SERENA hal yeaa i

NTE TN tee CeCe gH Td Bs Ct A EU te IIIS EM Dt Ute SSN Yd mathot Ano nran aTSeiD 3 sisip Ue E tages eke NETS

Set Cv gleams CA, AINS SY (488 ont Peat eenee eal SHE ey WEENIE UCL INENA CMB Te LONE EVES tie'y SPAean bar omic) sigtebig oe

mevede lass

: SE AE EM May Ge ay

Pert Oy fF VEE VAS MR aL ek Urpin ee gery oe fia ee wip WN witedion ot EROR Nays ges

SNE NUNN Osten MESS ANTS A ah E MOEN hw Saath

SEALY INK E Me Cie a eRe Nyy bye Oy

Sen gay

cert 7

meh Sek vgs

ON wae ae!

wd ghietal piel

pevmavas Vee Whee ethy

YA ln eae pee wy MR INOS yt

nea Te ee

fy he he enna ce ene ken oh eran

PVs paee ea tegey

<P gin by g o2 St

VS Path {Yiy SN tt int) stages

OU PLS ar why yo ty By!

Re Ty

oor

BDV Ry WORE Mbyte

as ( : a

Py batheseg «yeu es

SPT psa ag RENEHUES Oa ISE

tle Pe ee ee

ee het re

1 eee ty yh la inne

re em rerrs

aby Nga Ayli dara

EP eee rH MOF Ie WENGE Na AREY pire te eg Rat RN ee

ening s oi Anite Fe thm Man fle ey Teele Me rig tele iM

wo eta Sete Wy ob MMS Hal ee ie El a Dieta eT

bae4 ony

ee RUN REN BARI

SORT ct ane a

renerae

ee tata ens fe lty

Var he fave ta dis

CAYO i Oye

Vote ht etd

aS wae cin

Prete tnt

Ain Tm ie The OEE ne ea,

Petia rea ih atete het ety EYE

$2 GS vty

AGS MAb eV ab Na

ASM ate

Soe AE NE Nip eat Ig T yt aaa hes

Veep PN MAIS Ty

Move peat eis

See Uy ewaO

ca ae ae ee

Phar eet

watt vod yh

CLS ver

Gite! ala (yada ty Hedy alte AY ; ciudad ihe

= MaNeratiyr neha sng Cashed to eyeatwdtioe ast Ne we ict BN Meh te SEAT ET wth antisera Paci ahnad Sena MTaE Cher ai tent fie DUM gO ar cacyy a Bd?

MMT INET Aye sal tege dL A roth gem

Me Ee ey tate ete”

Hef

Reb ae AE oF TENT IROL Tt Mies athe Bee ah

re efi ep tea Pe Tin PHM im ht

Sr Cae Sater ty oF

YO EEN pee dye eh rye a Pe

SPM Obey en a

NaN Ge iba Rid ed! achs bY

Sr detiedie ny

ol Mam EY

Rete

Fs ede

DARI ea eR NANT WE AN Dl

re NTN SER LOE ey

Wyre saleby PUN Eye ae rire he

SPaveveUE em Ppa eno ; x4 PAWNS feds ONG AAP bythe ON eae Sg Dv ai ey Pe yl Ay vy

en ml eby ts abia BS Sanh Ciena BODE Sot Seip ald Mu vr 0 aie mi ey Ne DW PN bea IE IN bed aya whe Pate FeO a Sedalia eye “

aYensine

er enn Sere ra

Fatale Melted cinteret tot fen

ane

ferrin Og ates

% a ie

SSO rete cate faye pat ne he

Pew VTA WIT gd Sethe ny

shave

chase ane Menges

Soe eee en et

Veyeariee

Antal ie

Went w ern tant,

Ca gyts: Heeeriiceans

2 weeding hy Vgc

: WV oy PERE yb te vii a By five,

uey

s = Yet

hg bad stay se

NCES AEA a!

SS reared Slate

AEA et Su Mtn MUON ye TEE ESS rT eV a Rate Ste Tv ad we palale ce are SOAR WS Sota Sek K Ow RENES Nabe a Pada Maa a athy Peron hres

SEEN Sy 2 APD MCN AD opt OURS aealy ta UIE LER RES Lhe Beh CM wv WYP en eed ents, eNeietita Susatricpassa non, NA PN De ng tals Maas

ewcapasteitys pe gh NP da ie OR OEY Oe Capi ard a

PPILEN NEN Ge

yyy Mae tee

SoHE Te [Efe led rs Sty 8s

een Tear Liew rtng ene h

POPC Bly Sere se cog Sh bebe

Vendy ey

wehtes whines RAPT FAM bs ney Mem MAMLazE NG oTKWe owen ny, " PERMA Ue aeRO NE EY REIT Fg TTI TMotgtreenne eat a BLN Bau iN aS tha coy Dred ave

pane Wee al tah Sorted SSW TN ch gENDE omy roe nen Eos Pb Kent bbe Peteiagiag he SM cals Seni Ps VLA 9p LAE MARIAN pib Ro aytcgtatbcin Tv

a GT miley Me odie eB. Bete ryt ie LOW gig e felny : a, para oF bela VEN vet yer hv two we CR Ain Cah me PDN ET ON aes Va Ais

Be ett wae 5 Hind een en da 4 a a CeO IN ER ters Shen a WX ny Ra yy Sit» Med

PN fier bg tH0e

yobs rten VAN egN ESTE

yd iin eat

SPLEEN

bane

aretanid®

oe Peete

Ge aetna tiet er oe epi phn awe

wag Seb berg et WN Me Fieree eee

. By Marg tp baete te Tees

Ae eh ete CF PNAS tye

LM Coiled ee

Fee

CSET ate

nits RODS NG FAO bythe 0

aM ab cP HN acy Tet rental nt

Ren pe rey ere

FINE God NEE Ere edte oF

A a pte ae ot

ee Oe rents

rene a eres

iy aire Sethe NtteaMas yyy TY

shu wy ety

Se oa re ett ee Pa he

Atay ry tei?

banat

ont IN Ce ett

te Pusdig al =

Mading kes Fa ing 158 le

WO wD ae os Ste!

Naas

$y sue Ree

ays Teen ge

sh emathy:

UNISTS

GaSe Ferrey te ow!

Mt edie win tae! ger

Yer ey oy

tbe lig Bah

ria rlehatad they

Padget Ntie ate

; 57

: i

we |

1B}

ipl

ti io

} |

‘

{

‘

{

‘

{

‘

i f 2

. i}

d ie viel

1 h fr

yA ¥ y

)

ms! 7

N +P

VOLUME 84

Number 1

J Our nal of the March, 1996

WASHINGTON |

ACADEMY. SCIENCES

ISSN 0043-0439

Issued Quarterly

at Washington, D.C.

CONTENTS

Announcement:

MGC rMLOM Me EGIKOR (a suaiy alley ou ghard ie Vice Sn BO Uae 0

Articles:

GERALD J. SCHUELER, ‘‘The Unpredictability of Complex Systems’’ .....

WILLIAM H. ROGERS & KATHY H. ABBOTT, ‘‘Presenting Information

CUPRA EY (ELD Ere 2) LUV Aig eae a ake eat dO ee eee eee Aa et ear eR Un

C. R. SCHUMACHER, BARBARA HOWELL & H. UBERALL, ‘‘Plasmon

Exciter, Congucting SONGS i. iL A aN ic a sae) baie Baayen, Bade a adm

Washington Academy of Sciences

Founded in 1898

EXECUTIVE COMMITTEE

President

Rev. Frank R. Haig, S. J.

President-Elect

John S. Toll

Secretary

Neal F. Schmeidler

Treasurer

Grover C. Sherlin

Past President

John H. Proctor

Vice President, Membership Affairs

Cyrus R. Creveling

Vice President, Administrative Affairs

Stanley G. Leftwich

Vice President, Junior Academy Affairs

Marylin B. Krupsaw ~

Vice President, Affiliate Affairs

Terrell Hoffeld

Board of Managers

Marilyn S. Bogner

Norman Doctor

Herbert H. Fockler

George H. Hagn

James W. Harr

Nina M. Roscher

REPRESENTATIVES FROM

AFFILIATED SOCIETIES

Delegates are listed on inside rear cover

of each Journal.

ACADEMY OFFICE

2100 Foxhall Road, N.W.

Washington, D.C. 20007

Phone: (202) 337-2077

EDITORIAL BOARD

Editor:

Bruce F. Hill, Mount Vernon College

Associate Editors:

Milton P. Eisner, Mount Vernon Col-

' lege

Albert G. Gluckman, University of

Maryland

Marc Rothenberg, Smithsonian Institu-

tion

Marc M. Sebrechts, Catholic University

of America

Edward J. Wegman, George Mason

University

The Journal

This journal, the official organ of the Washing-

ton Academy of Sciences, publishes original

scientific research, critical reviews, historical

articles, proceedings of scholarly meetings of

its affiliated societies, reports of the Academy,

and other items of interest to Academy mem-

bers. The Journal appears four times a year

(March, June, September, and December). The

December issue contains a directory of the cur-

rent membership of the Academy.

Subscription Rates

Members, fellows, and life members in good

standing receive the Journal without charge.

Subscriptions are available on a calendar year

basis, payable in advance. Payment must be

made in U.S. currency at the following rates:

U.S: and Canada) 2s Ja... ee $25.00

Other Counties ».).:2.465 ose eee 30.00

Single copies, when available ...... 10.00

Claims for Missing Issues

Claims will not be allowed if received more

than 60 days after the day of mailing plus time

normally required for postal delivery and

claim. No claims will be allowed because of

failure to notify the Academy of a change of

address.

Notification of Change of Address

Address changes should be sent promptly to

the Academy Office. Such notification should

show both old and new addresses and zip

codes.

POSTMASTER: Send address changes to

Washington Academy of Sciences, 2100 Fox-

hall Road, N.W. Washington, DC 20007-1199.

Journal of the Washington Academy of Sciences (ISSN 0043-0439)

Published quarterly in March, June, September, and December of each year by the Washington

Academy of Sciences, 2100 Foxhall Road, N.W., Washington, DC, 20007-1199. Second

Class postage paid at Washington, DC and additional mailing offices.

Note from the Editor

To make the Journal of the Washington Academy of Sciences current, Volume

84 will include manuscripts received and accepted for the chronological years

1994-1996. Beginning in 1997 the journal will return to publishing manuscripts

received the chronological year of the volume’s publication in a single volume.

The Executive Committee listed on the back of the front cover is for the year of

service from May 1994-1995.

Journal of the Washington Academy of Sciences,

Volume 84, Number |, Pages 3—12, March 1996

The Unpredictability of Complex

Systems

Gerald J. Schueler

The Graduate School of America, Minneapolis, Minnesota 55402

Received December 1, 1994

ABSTRACT

One of the findings of modern chaos theory is that complex systems are predictable in the

short term, but not in the long term. There is an element of uncertainty in all complex

systems. Furthermore, this fact holds true across a very broad spectrum of scales and reference

frames. Does this finding hold true for human beings? When the human body and mind are

considered as complex systems we should expect to find that our futures are reasonably

predictable only in the short term, which does seem to be the case. Many of the findings of

chaos theory have been transposed to other disciplines, including biology. However, more

work needs to be done to transpose these important findings to psychology.

Introduction

Complex systems all have feedback mechanisms. These mechanisms provide

the means by which matter, energy, or information are fed back into the system.

Because material objects are not perfect, tiny errors will creep into the feedback

process. In the short term these errors can often be ignored. But in the long term

they can accumulate, through the repetitive feedback process of iteration, until

the error (noise) is as large as the system input or output itself, making any

predictions for the system impossible. Haken (1988) points out that deterministic

chaos theory has shown us that even in classical mechanics, predictability cannot

be guaranteed with absolute precision. However, according to Gordon (1991), in

certain circumstances it is possible to forecast values in a chaotic series over

limited ranges. Perhaps Kellert (1994) says it best when he states that complex

systems are associated with “‘predictive hopelessness’’ (p 33). He also points out

that if the ““time between noteworthy events’’ is short enough, a dynamic system

can have a “‘predictively worthwhile time,’’ which tells us the time required for

the situation to reach predictive hopelessness (p 34). In other words, complex

systems can be predicted in the short term, but not in the long term. The fact

4 SCHUELER

that deterministic systems can be unpredictable has required physicists to review

these two terms. Today, most physicists agree with chaos theory and eliminate the

requirement for predictability in their definition of determinism (Kellert, 1993).

The Chaos Theory of Unpredictability

The chaos theory of unpredictability is an extension of the well-known Hiesen-

burg uncertainty principle which states that

AM,Ax = h

where h is Plank’s constant. Essentially, this equation implies that either the

momentum of a particle, M,, can be known with certainty, or its position, x, can

be known with certainty, but not both together. This uncertainty is inherent in

how we measure things, and exists because every observer tends to influence, to

some degree, what is being observed. This basic principle of uncertainty at the

quantum level has been verified many times in the scientific community. Chaos

theory extends this uncertainty principle to the macroscopic level when we con-

sider complex systems which are sensitive to initial conditions. In chaos theory,

this principle is called Prigogine’s uncertainty after the 1977 Nobel Prize winner

in chemistry, Ilya Prigogine. Prigogine’s new uncertainty principle says that as

systems become more complex, a threshold of complexity will be reached such

that the system will begin functioning in unpredictable directions; such a system

will lose its initial conditions and these can never be reversed or recovered (Briggs

and Peat, 1989). Mathematically, this is

dS; = dS; + dSz_

where dS; is the total or net entropy, dS, is the internal entropy (this is the

traditional entropy of Clausius that can increase but not decrease for a closed

system), and dS; is the external entropy, or that which is exchanged by the system

with its environment. The last term, which can be positive or negative, extends

the traditional entropy definition to open systems (Cambel, 1993).

The future of any complex system is unpredictable. All that we can ever know

of the future is in terms of probabilities. The future of any complex system can

only be known totally (i.e., with certainty) by its moment-to-moment expression

in the present.

The chaos theory of unpredictability tells us that the future is a world of possibili-

ties. Every future event is associated with a probability of occurrence. We cannot

even say that the sun will rise tomorrow morning with 100% certainty. However,

chaos theory does provide techniques that make order out of apparent chaos. Probably

THE UNPREDICTABILITY OF COMPLEX SYSTEMS 5

position

A A

momentum

phase space

trajectory

Fig. 1. A Simple Pendulum Showing a Plot of position vs. Momentum.

the biggest area of concern today is in the stock market (Weiss, 1992). Using chaos

theory to predict the stock market has both adherents (Burke and Shirreff) and critics

(Hulbert). According to Stambler (1991), chaos theory is currently being investigated

for use in electric power research to predict energy usages.

Attractors in Phase Space

One dimension is a straight line. For example, if you are in one-dimensional

space, you can only go forward or backward. Two dimensions is a plane. In two

dimensions, you can go forward and backward as well left or right. In three

dimensions, you can also go up or down. But fractional (or fractal) dimensions

imply the ability to go between two and three dimensions.

Attractors are usually defined as a region in the state space of a system such

that all trajectories nearby converge to it (Kellert, 1993). Attractors are said to

reside in the state space of a system which is called phase space. Let’s take a

look at phase space and see if we can get a better grasp of just where it is located.

We will use the classical case of a swinging pendulum as an example of a simple

system (see Figure 1). Let’s envision a bob swinging from a string. The pendulum

has only two major characteristics,.the position of the bob in time, and its corre-

sponding momentum (mass times speed). We can let the lowest point of each

swing, where the bob will eventually come to rest in time, be position zero. At

the maximum height of each swing, just before the bob reverses its direction,

6 SCHUELER

position

momentum

Fig. 2. Phase Space of a Simple Pendulum.

the position of the bob is the greatest, and its corresponding momentum is zero.

We can make a simple plot of these two points on an xy-coordinate diagram

such as that shown in Figure 1. The x-axis measures the momentum of the bob.

The y-axis measures the corresponding position of the bob. We will mark these

two points as A and B. Since momentum is zero, they are both plotted along the

y-axis. For convenience, let’s let a swing to the right be plus along the x-axis,

and a left swing be negative. Point A is to the right of the swing and therefore

is plotted as positive along the y-axis. When the bob is at the lowest point of a

swing, midway between points A and B, its position is zero, and its momentum

is a maximum. When the bob swings from point A to point B it crosses this

lowest point, which is shown in Figure 1 as point C. When the bob swings back

from point B to point A it again crosses the lowest point, but in the reverse

direction (i.e., with a negative momentum). We call this point D on our plot. If

we were to plot a lot of these corresponding points on our diagram, a circle

would form around the center of the x and y axes as shown in Figure 1. A plot

such as that given in Figure 1 is called a phase space map. The circle itself is

called the phase space trajectory (because the trajectory here is a circle, it is

usually called an orbit) and represents the entire motion of the pendulum for one

cycle. Additional cycles of the pendulum will simply repeat the circle.

We all know that the pendulum, if left to itself, will not keep repeating this

cycle. Experience tells us that in time, it will slow down and will eventually

stop. This is shown in the phase space diagram in Figure 2. This attraction for

the same fixed point is true with all pendulums, no matter their size. The magnetic

THE UNPREDICTABILITY OF COMPLEX SYSTEMS 7

attraction of the orbit of the pendulum for a fixed point is caused by a special

attractor known as the fixed-point attractor. These are stable equilibrium points

where dynamic systems will tend to come to rest. According to Cambel (1993),

‘‘Dynamic systems are attracted to attractors the way fireflies are attracted to

ehtre (p59).

An easy way to think of the fixed-point attractor is to consider a ping-pong ball

and the surface of the sea. We can drop the ball over the sea where it will fall

until it contacts the surface. We can also hold the ball under the water and then

let go, where the ball will float upward toward the surface. Either way, the surface

of the sea acts as an attractor for the ball and no matter where the ball is released,

it will always wind up on the surface. But once the ball reaches the surface, it will

be buffeted by winds and currents in unpredictable ways because the dynamics of

the surface of the sea are very complex (Cohen and Stewart, 1994).

The fixed-point attractor is one of four known types. Other types include the

limit cycle, tori, and strange attractors. A fixed-point attractor draws a system

to a single point in phase space. A limit-cycle attractor attracts systems to a

cyclic path in phase space (a range of final resting points). The configuration of

attractors in phase space can help determine if a system is conservative (maintains

energy) or dissipative (energy must be supplied from outside the system) and

can also help determine if a system is chaotic. They serve as the geometric

counterpart to the thermodynamic entropy function (Cambel, 1993). Strange at-

tractors are found in conditions of turbulence. They attract complex systems from

order into disorder. Ruelle, who first called the attractor for turbulence ‘‘strange,”’

found that this attractor pulled complex systems into a space of fractional dimen-

sion, where they became caught somewhere between a two-dimensional plane

and a three-dimensional solid (Briggs and Peat, 1989, and Ruelle and Takens,

1971). To appreciate this situation, consider a piece of paper, which is essentially

a two-dimensional object. Crumple the paper. As it is compressed, the two-

dimensional sheet of paper will approach a three-dimensional object.

Reference Frames and Scales

A reference frame is an axis which we assume to be fixed, and by which we

can make relative measurements. According to Einstein (1961), every measure-

ment that we can make must be in relation to some fixed frame of reference.

Our everyday world is such a reference frame. However, modern relativity has

demonstrated that there is no such thing as a totally fixed reference frame. Every-

thing is moving or rotating. Our world hurtles through space around the Sun,

which itself is moving through space around the galaxy, and so on.

8 SCHUELER

Fig. 3. The Reduction of matter.

There are two main kinds of reference frames: inertial, which 1s relatively

fixed, and noninertial or accelerated, which is relatively in motion. Reference

frames can be constructed at different scales. .

This idea is graphically displayed in Figure 3 which depicts the current reduc-

tion of matter by modern physics. Take any physical object that you want. In

our normal everyday reference frame, your object will appear to be solid and

real. However, we now know that it is really made up of microscopic molecules.

The rest is simply empty space (for purposes of this argument, we will ignore

the theory of virtual particles which disprove the concept of empty space). When

we change our scale to the molecular level, we can have another reference frame.

If we look closer, we will see that each molecule is composed of atoms with

everything else being empty space. We can change our scale to the atomic level

and construct another reference frame. If we look deeply at each atom, we will

see that they are composed of subatomic particles such as electrons, protons, and

neutrons. The rest is empty space. We can change our scale to the subatomic

level and construct another reference frame. If we look closely at subatomic

particles called hadrons, we will find that they are composed of quarks (according

to Murray Gell-Mann’s quark theory). The rest is empty space. If we could

change our scale to the quark level, we could construct another reference frame.

This idea has also been proposed by quantum physicists. As we saw earlier,

THE UNPREDICTABILITY OF COMPLEX SYSTEMS 9

according to the Heisenberg uncertainty principle, we cannot measure both the

position and momentum of a subatomic particle with accuracy. This is because

when we measure the one, we simultaneously disturb the other thus making its

value uncertain. For this reason, subatomic particles are said to have a probabilis-

tic nature rather than determinant because we can only measure them in terms

of probabilities (Herbert, 1985).

An example, currently used by scientists to demonstrate the idea of reference

frames at various scales, is the question ““What is the dimension of a ball of yarn?’’

From a large distance, we could say that it is a point, with no dimensions. As we

get closer, we would see that it is a three-dimensional ball. When we get close

enough to see that it is a single piece of yarn wrapped around itself, we would

say that it is a one-dimensional object twisted into three dimensions. If we look

at the yarn very closely, we would see that the strands appear as three-dimensional

columns. From even closer, we would see individual one-dimensional fibers. This

example clearly shows that the number of dimensions that we assign to an object

depends on our relationship to that object, which in turn depends on the scale in

which we are making our observations (Gleick, 1987).

Linear and Nonlinear

Most of us like to think that our world is linear. We like to think that our lives

are continuous. In the world of mathematics, linear equations can be graphed as

a straight line. They can be easily solved. They can be taken apart and put back

together again.

On the other hand, nonlinear equations cannot be easily solved. They often

cannot be added back together. Engineers must use fudge factors or experiential

constants to arrive at meaningful solutions. We can say, then, that linear equations

reflect order while nonlinear equations reflect chaos.

Differential equations describe the way systems change continuously over time.

Differential equations also describe the way systems change erratically over time

(Gleick, 1987). Linear and nonlinear changes over time are both found in complex

systems. As much as we would all like to forget the nonlinearity of life, we must

face it squarely, and come to terms with it.

Let’s look at a simple pendulum, again. Aristotle taught that all things seek

their natural state-the state that they would find if left to themselves. The swinging

pendulum, according to Aristotle, was trying to get to the Earth, but was con-

strained by the string. Aristotle did not see forces, rather he saw changes, and

things that desired to change (Gleick, 1987).

When Galileo looked at a swinging pendulum, he saw a form of regularity or

10 SCHUELER

periodicity that could be measured. He believed that all things tended to keep

moving as they were, unless acted on by an external force. The pendulum, he

believed, would swing forever but for the forces of friction that slowly dampened

its movements (Gleick, 1987).

Galileo also believed that a pendulum, of a certain length, would keep the

Same time, no matter how wide or how narrow its swing. A wide-swinging bob

has farther to travel, but he believed that it would make up the difference by

moving faster. In other words, he taught that its period is independent of its

amplitude (Gleick, 1987).

Modern scientists have found that this is not true. It is only an approximation.

As the pendulum’s swing changes, a slight nonlinearity is induced. At small ampli-

tudes, the nonlinearity is so small that it is negligible. However, it 1s there, and

modern scientists have been able to measure it (Briggs and Peat, 1989).

The pendulum example shows us how easy it is to view the same thing in

many different ways. Although we may be unaware of them, almost everything

in our lives is associated with tiny nonlinearities. We usually dismiss these as

unimportant. However, these things are sometimes amplified by feedback loops

or iterative processes until they jump out at us when we least expect it, causing

all sorts of problems.

Looking at a somewhat bigger picture, Einstein’s general theory of relativity

is expressed in mathematical equations. These equations are essentially nonlinear.

In fact, the nonlinearity of his equations have led to the discovery of black

holes—tears in the fabric of spacetime, where the orderly laws of physics break

down and no longer apply (Briggs and Peat, 1989 and Hawking, 1988).

Weather Prediction

Our weather is an interrelated series of complex systems. In 1961, Edward

Lorentz discovered the butterfly effect. He was trying to forecast the weather. He

was running a long series of computations on a computer when he decided he

needed another run. Rather than do the entire run again, he decided to save some

time by typing in numbers from a previous run. Later, when he looked over the

printout, he found an entirely new set of results. The results should have been

the same as before. After thinking about this unexpected result, he discovered

that the numbers he had typed in had been slightly rounded off. In principle, this

tiny difference in initial conditions should not have made any difference in the

result. But it did. From this, Lorentz determined that long-distant weather fore-

casts are impossible. Tiny differences in weather conditions, on any one day,

will show dramatic difference after a few weeks, and these differences are entirely

THE UNPREDICTABILITY OF COMPLEX SYSTEMS 11

unpredictable. This phenomenon is called the Butterfly Effect because if a butter-

fly flaps its wings in one country, it can effect the weather in another. Technically,

it is called sensitive dependence on initial conditions. Although Lorentz’s discov-

ery was an accident, it planted the seed for what is now known as chaos theory

(Gleick, 1987).

Discussion

Strange attractors, associated with complex systems tending toward chaotic

conditions, are especially difficult to measure, although scientists have come a

long way in only the last few years. Our body is a complex open system. As we

go through life, we too experience the effects of attractors. We have single-point

attractors like college or retirement, limit-cycle attractors, such as our tendencies

toward habitual thinking, torus attractors such as our family and our attractions

to other people, and strange attractors, such as sickness and disease. However,

our ability to avoid many attractors in our daily lives constitutes the rationale

for the probabalistic nature of our future. For example, we may or may not marry.

We may or may not continue to think a certain way. In most cases, we may or

may not avoid an attractor. When encountered, they often will cause us to change

the direction of our life and thus send us on our way toward yet another attractor.

We can plan for tomorrow with a reasonable certainty. But our plans for ten

years from today will be very uncertain. Heisenberg’s uncertainty principle is

known to be true on the subatomic level. Chaos theory suggests that an element

of uncertainty is present in all scales and frames of reference, even in our everyday

lives. Our lives are inherently unpredictable, at least in their details. For example,

death itself is a certainty, a point of final rest and therefore a fixed-point attractor,

but how and when we reach this attractor are generally unpredictable factors.

According to Cambel (1993), ‘‘the concept of free will is related to chaos’”’

(p 193). Unpredictability ensures free will, and without it our futures would be

forever fixed. The choices that we make today can cause very different effects

in our future. Chaos theory suggests that our lives are not totally predetermined.

Instead, because the world includes both order and chaos, our lives tend to

oscillate between periods of free will and determinism, between sickness and

health, and between planned events and accidents.

Many important findings of chaos theory have already spilled over into other

disciplines. Hopefully, in the burgeoning spirit of interdisciplinary that we see

in Our universities today, some of these findings will be successfully transposed

to psychology; a task made possible by considering the human mind as a complex

system.

12 SCHUELER

References

Briggs, J., & Peat, F. D. (1989). The turbulent mirror: An illustrated guide to chaos theory and the science

of wholeness. New York: Harper & Row.

Burke, G. (1993). Measuring market choppiness with chaos (E. W. Dreiss’ Choppiness Index trends). Futures.

v22. 52-56.

Cambel, A. B. (1993). Applied chaos theory: A paradigm for complexity. Boston: Academic Press.

Cohen, J., & Stewart, I. (1994). The collapse of chaos: Discovering simplicity in a complex world. New

York: Viking.

Einstein, A., & Lawson, R. W. (trans.) (1961). Relativity, the special and the general theory: A popular

exposition by Albert Einstein. New York: Crown.

Gleick, J. (1987). Chaos: Making a new science. New York: Penguin Books.

Gordon, T. J. (1991). Notes on forecasting a chaotic series using regression. Technological Forecasting and

Social Change. v39. 337-348.

Haken, H. (1988). Information and self-organization: A macroscopic approach to complex systems. London:

Springer-Verlag.

Hawking, S. W. (1988). A brief history of time: From the big bang to black holes. Toronto: Bantam.

Herbert, N. (1985). Quantum reality: Beyond the new physics. Garden City, New York: Anchor Press/

Doubleday.

Hulbert, M. (1994). Rockfall? Or Avalanche? (Using the chaos theory to predict the stock market. Does it?).

Forbes. v154. n7. 211.

Kellert, S. H. (1993). In the wake of chaos: Unpredictable order in dynamical systems. Chicago: University

of Chicago Press.

Ruelle, D., & Takens, F. (1971). On the nature of turbulence. Commun. Math. Phys. 20: pp 167-192.

Shirreff, D. (1993). Efficient markets and the quants’ descent into chaos (portfolio theory). Uromoney. July.

60-64.

Stambler, I. (1991). Chaos creates a stir in energy-related R&D (chaotic dynamics). Research & Development.

v33. 16.

Weiss, G. (1992). Chaos hits Wall Street—the theory, that is. Business Week. Nov 2. 138-140.

Journal of the Washington Academy of Sciences,

Volume 84, Number |, Pages 13-38, March 1996

Presenting Information for Fault

Management

William H. Rogers

Bolt, Beranek & Newman, Inc.

Kathy H. Abbott

NASA Langley Research Center

Received December 5, 1994

ABSTRACT

While some issues of presentation of information are discussed here, the emphasis of this

paper is on information content, that is, what information should be presented for management

of aircraft systems faults, and additionally, when it should be presented. It is argued here

that information requirements, as traditionally defined, fall short of describing the larger set

of information that may be useful to the modern flight crew. We believe we can now

recast information requirements in terms of human information processing tasks, rather than

exclusively in terms of human actions and sensed data.

In the context of what information to present, this paper focuses on presenting information

to support fault management performed by the flight crews of commercial aircraft flight

decks. We discuss the characteristics of the application, and the similarities to and differences

from fault management in other application areas. We discuss human management of faults

using automation as an aid, and how this human-centered automation philosophy affects the

design of the decision aid and determination of information requirements. The results of

studies addressing the information needs and desires of flight crews are described, with a

particular emphasis on the manner in which the pilots process the information to support

fault management tasks. In addition to discussing presentation of information for managing

failures, we also discuss the need to present information to manage the fault management

aid. Lastly, we raise the issue of integrating the fault management aid with the remainder of

the flight deck systems.

Presenting Information for Fault Management

For most human factors and human-computer interaction researchers and prac-

titioners, discussion of presentation of information probably elicits concerns about

issues such as format, color, symbology, and display media. While some of those

issues are discussed here, the emphasis of this paper is on information content;

in particular, what information should be presented to flight crews for manage-

14 ROGERS AND ABBOTT

ment of aircraft systems faults, and additionally, when it should be presented.

‘“What’’ and ‘‘when’’ concerns are more typically described as information

content or requirements issues, and are usually resolved prior to addressing infor-

mation ‘‘form’’ issues.

It is argued here, however, that information requirements, as traditionally de-

fined, fall short of describing the larger set of information that may be useful to

crews of the modern flight deck. In the past, the pilot had a more physically-

active role, and available technology limited the set of information that could be

presented to a human operator to that which could be sensed, and a sensed value

was usually displayed on a dedicated gage or dial. Correspondingly, traditional

task analyses focused on observable tasks and information requirements were

identified down to the level of detail where an information element was a sensed

value. In these analyses, there was little application of research on human cogni-

tion by flight deck system designers. With the advent of digital computer and

display technology, great strides have been made both in processing and presenta-

tion capability. Additionally, the role of the pilot in current generation flight

decks accentuates cognitive abilities. Because of these developments, we now

should, and can, present information that directly supports human information

processing tasks, especially cognitive tasks, rather than just present sensed data

that supports observable human actions. To do so, we must understand the hu-

man’s cognitive activities, so that we can directly support those activities. For

fault management, this means understanding the human information processing

and decision-making processes involved. :

It is also argued here, as elsewhere (e.g., see Malin, Schreckenghost, Woods,

Potter, Johannesen, Holloway & Forbus, 1991), that information requirements

should include information needed, as a consequence of pilots working with

automated aids. These additional requirements include the need to monitor the

aid, which we discuss below. The more cognitive nature of pilot activities, the

new enabling processing and display technologies, and the new automation-

induced information requirements, point to the importance of reassessing what

information should be presented for fault management (with associated implica-

tions for when, how and where).

In this context, this paper focuses on presenting information to support fault

management performed by the flight crews of commercial aircraft flight decks.

We discuss the characteristics of the application, and the similarities to and

differences from fault management in other application areas. We discuss human

management of faults using automation as an aid, and how this human-centered

automation philosophy affects the design of the aid and determination of informa-

tion requirements. The results of studies addressing the fault management infor-

mation needs and desires of flight crews are described, with a particular emphasis

FAULT MANAGEMENT 15

on the manner in which pilots process information in performing fault manage-

ment tasks. In addition to discussing presentation of information for managing

failures, we also discuss the need to present information to help pilots manage

the fault management aid. Lastly, we raise issues about integrating fault manage-

ment aids with other flight deck systems.

Commercial Flight Deck Environment and Fault Management

The Domain

In modern commercial flight decks, two pilots work as a crew to manage the

flight. Managing in-flight subsystem failures is a task they are only occasionally

required to perform, because faults occur infrequently. As discussed below, even

monitoring for system faults and failures is becoming obsolete because of the

sophistication of automated alerting systems, the high reliability of systems, and

the redundancy of system functioning. When failures do occur, the flight crew’s

primary goal is to continue safe operation of the flight. This contrasts with

managing faults for the purpose of maintenance, for example. In maintenance,

the operator’s goal is to fix or replace the broken part of the equipment. Fixing

or replacing the equipment is rarely an option in in-flight fault management.

Because the tasks are different, the operator’s information requirements are corre-

spondingly different. Another complicating factor is that initial response to an

in-flight failure is often time pressured, so the pilot does not have extra time to

analyze the situation. This is why pilots are not trained to perform extensive

diagnosis of a failure once it is detected, but rather to respond rapidly in some

manner that hopefully will stabilize the situation.

In the commercial flight domain, since the flight crew is not performing fault

management as its primary function, the goal of fault management is to control

or compensate for the effects of the fault on the more important functions of

flight control and navigation. This coupled with the aforementioned infrequency

of failures means that pilots often do not have a great deal of practice managing

failures, other than in their training programs. Moreover, current training pro-

grams do not require the pilots to have as much detailed systems knowledge as

in the past. Systems management, including management of failures has become

more automated (‘‘Douglas new systems,’’ 1990), with the consequence that

pilots are not as involved with the systems as they once were.

Current Fault Management Philosophy

Up until the mid- to late-seventies, the number of unintegrated visual and aural

fault alerts on commercial flight decks was proliferating at an alarming rate, and

16 ROGERS AND ABBOTT

there was no well-thought-out, standardized information presentation philosophy

for fault management (Cooper, 1977). With the three-person flight crew, this was

tolerable because the second officer could spend much of his or her attention and

effort on systems management tasks, including fault monitoring and management.

Still, it was obvious that improvements were in order. The trend toward two-person

crews helped fuel the need for changes. In recognition of these issues, the Federal

Aviation Administration, Douglas, Boeing and Lockheed (e.g., Boucek, Berson, &

Summers, 1986) collaborated to begin development of guidelines for a standardized

and integrated caution and warning system for commercial flight decks. The notion

was to develop a standard set of visual and aural warnings and visual messages

which would attract the crew’s attention to the non-normal condition, indicate its

level of urgency (caution, warning or advisory), identify the problem, and provide

some feedback on the adequacy of the crew response (Boucek et al., 1986). The

resulting caution and warning concepts formed the basis for all modern commercial

aviation caution and warning systems.

The information presentation principles are elegantly simple: the urgency of

any condition is conveyed by the same information that is used to alert (e.g., a

master caution light and aural alert gets the crews’ attention and indicates to

them a ‘‘caution level’’ urgency requiring immediate crew awareness and future

corrective or compensatory crew action). Level of urgency is redundantly con-

veyed by the color and position of messages identifying the fault. Pilots are

trained to retrieve specific, prescribed procedures and checklists based on standard

phrases used in the identification messages, thus standardizing responses and

reducing the guesswork and the uncertainty (and the accompanying requirement

for difficult diagnostic and other decision making processes) related to fault

management. In summary, the current ‘‘information presentation’’ philosophy

for commercial flight decks is to alert and convey the urgency of a fault condition

quickly and precisely, and to identify the fault condition to a level of detail that

allows differentiation of prescribed procedures.

Fault Management as Information Processing

From an information processing perspective, fault management can be de-

scribed as a typical sequence of perceptual, cognitive, and response tasks, (e.g.,

see Wickens, 1984 for a standard information processing model). If the generic

perceptual, cognitive and response processes were translated into specific fault

management tasks, one might generate task labels such as fault monitoring, fault

detection, fault identification, fault diagnosis, fault localization, assessing the

effects of the fault, planning for operation with a faulted system, response plan-

ning and response execution to contain the fault, to correct the fault, to repair

the fault, and to compensate for the fault and the effects of the fault. This involves

FAULT MANAGEMENT 17

a considerable number of distinct types of information processing tasks. We will

not elaborate on the subtleties of these distinctions, but the point here is that

because of the nature of the commercial flight deck domain, fault management

by the flight crews of commercial aircraft normally entails only a subset of these

tasks: fault detection, identification and corrective and/or compensatory actions.

Hence the current information presentation supports these tasks to the exclusion

of others. As mentioned, much work has been done on consolidating and standard-

izing the presentation of information for these fault management tasks, particu-

larly in the alerting area.

Presentation of information for other fault management information processing

and response tasks such as fault monitoring and fault diagnosis, are de-emphasized

in the commercial flight deck environment because, for the most part with modern

technology, they are non-essential for responding adequately to the fault or failure

condition. ““For the most part’’ is not always, however. Further, ““non-essential’’

does not mean ‘‘not useful.’’ There are situations such as catastrophic faults,

multiple faults, extremely rare faults, or routine faults with exacerbating circum-

stances, in which these other fault management tasks come critically into play.

There are many documented cases where if the crew could have detected or

diagnosed the fault more quickly or accurately, or planned for or compensated

for the effects of the fault more effectively, lives would have been saved. Current

aircraft fault management systems do not offer pilots much, if any, support in

situations where these fault management tasks are essential. This is not a criticism

of current systems; not only are these rare events, thus making expensive informa-

tional aids hard to justify, but more importantly, technology to provide more

sophisticated information for tasks such as fault monitoring and fault diagnosis

has not existed. Advances in information processing and display technology are

beginning to change that.

As these technologies have progressed, more interest has been focused on

intelligent fault management aids (Abbott, 1991; Schutte, 1989). Concepts such

as Faultfinder (Schutte, Abbott, Palmer & Ricks, 1987) are intended to provide

an array of informational assistance to fault managers. This includes more sophis-

ticated information for tasks which are already supported, such as fault detection

and identification, but more importantly, new information for tasks for which

they have previously been left to their own devices. We are currently addressing

the effects of these advanced information aiding capabilities and concepts, the

accompanying human-automation integration issues, and the problems inherent

in presenting informational support for situations containing uncertainty (the very

reason they are traditionally human decision-making domains), on information

requirements and information presentation issues.

18 ROGERS AND ABBOTT

Human-Centered Automation Philosophy and Fault Management Aids

There has been a great deal of interest recently in human-centered automation,

or automation designed to support the human (Billings, 1991; Norman & Orlady,

1989). Billings (1991) defines human-centered automation as ‘‘automation de-

signed to work cooperatively with human operators in the pursuit of stated objec-

tives.’’ This contrasts with technology-centered automation, where a function or

task is allocated based on a comparison of human performance and machine

performance for each function. Inevitably, as technology progresses, this approach

results in more and more functions becoming automated. Unfortunately, this has

undesired ‘‘side effects’’ (Wiener, 1985). These include:

® the pilot is more remote from the primary flight functions (i.e., “‘out of the loop,”’

resulting in reduced situation awareness);

® pilots are forced into performing tasks for which they are not well suited, such as

monitoring; and

® new human error types and decrements in overall human/system performance can

occur that can nullify and even reverse the performance advantages of automating

individual functions. .

Additionally, when such automation performs reliably for some period of time,

pilots may become complacent or overly reliant on the automation. All of these

undesirable consequences have manifested themselves in incidents and accidents

that have been attributed, at least in major part, to the interaction between the

humans and the automated systems (e.g., NTSB, 1986).

Human-centered automation thus was born out of the need to keep pilots

involved in critical flight functions (both to maintain situation awareness and

because they occasionally need to perform as backups for automated functions)

and the need to consider the overall performance of the human-automation system

rather than system versus human performance on individual tasks and functions.

Because of the recognition that cooperation and complementariness of the human

and the automation are such important factors for the safe and efficient operation

of the flight, human-centered or user-centered automation has become the goal

of design of advanced systems or decision aids for any complex application

where the human is involved (Billings, 1991; Rouse, 1991).

For a resulting system or aid to be designed to support the human, it is

important to first identify the desired role of the human operator with respect to

the automation. One way to view the different possible roles that the human can

take on is shown in figure 1, which represents a continuum describing the alloca-

tion of processing and responsibility between the human and the system.

At one end of the continuum, the human receives only raw data and must do

all processing of that data and perform appropriate actions. At the other end, the

automation performs all processing and carries out the decision or answer to

FAULT MANAGEMENT 19

Human

Receives uman Receives

Raw Data Processed

To Information

Interpret

Fig. 1. Continuum describing possible human/automation roles.

whatever problem or task is being addressed. Between these two extremes, we

have identified two categories that represent qualitatively different roles for the

human/automation. In the leftmost of the two middle categories, the human

receives processed information, but in the form of situation or status information.

Much of the information provided by current caution and warning systems be-

longs to this category. The rightmost middle category is one where the automation

does the processing, and provides a command or response recommendation about

what to do, although the human actually carries out the action. An example of

a system that falls into this category is TCAS, or Traffic Alert and Collision

Avoidance System, a system which can give pilots a command, such as “climb

right,’’ to avoid other aircraft. The reason the two middle categories are separate

is because in the category where the automation provides a command or response

recommendation to the human, there is an opportunity for the human to follow

the advice and do what the automation says without fully understanding the

situation.

Clearly, this continuum represents only one of many dimensions along which

human/automation role could be described. Others include approaches taken by

Riley (1989) and Billings (1991). Nonetheless, it serves to support communication

of two important points: first, the desired role of the human should be selected

as part of an explicit design decision that is consistent with the automation

philosophy chosen, rather than have the automation be designed and have the

human’s role be to monitor the automation and do whatever the automation does

not do. Second, human-centered automation does not lie at a particular point

along the continuum. Rather, the appropriate role of the human differs depending

on the task. Many factors, such as time pressure, have a major effect on the

choice of appropriate role of the human (and hence that of the automation). Thus,

the pilot may have a different role for different tasks on the flight deck. We

would argue, however, that human-centered automation means that when the

total of all flight functions and tasks are considered, the human’s role, in the

aggregate, lies somewhere in the middle of the continuum.

Once the role of the human has been identified, the tasks to be performed

20 ROGERS AND ABBOTT

must be allocated to the human and the automation. As mentioned earlier, one

commonly-used approach is to identify what humans are good at and what ma-

chines are good at, and allocate tasks accordingly (the ‘‘Fitts list’’ approach, e.g.,

see Fitts, 1951). As Woods (1989) points out, however, this human-automation

comparability philosophy results in technology-centered automation with its asso-

ciated drawbacks. Human-centered automation principles suggest that allocation

of functions should be based on human-automation cooperation and evaluation

of overall human/system performance. This idea is not new: Jordan (1963) wrote

that ‘‘men and machines are complementary, rather than comparable.’’ He went

on to suggest that allocation of tasks to humans or machines is meaningless, and

we should think about tasks as done by humans and machines.

Information presentation to the flight crew is complicated by human-centered

automation, however. When a task or subtask is shared between the flight crew

and the automation, the human operator must manage the automation. This may

include monitoring the automation, understanding what it is doing, evaluating

its output, and knowing how to interact with it. Because additional tasks are

required of the human operator, there are corresponding additional information

requirements.

We can see, then, that many factors affect the need for information: not only

the fault management task, especially in terms of information processing tasks,

but also design and automation philosophy, technology, and allocation of func-

tions. Figure 2 shows a graphic depiction of some of the many factors that can

affect information presentation to the flight crew and examples of affected content

and form variables (Rogers, 1991). It behooves us to analyze the effects of all

the factors on all the aspects of information presentation as directly as possible.

New principles defining the ‘‘what,’’ “‘when,’’ “‘how’’ and ‘‘where’’ (and *“*how

much’’) of fault management information presentation are needed. These informa-

tion presentation principles will emerge from an understanding of the underlying

factors. Here, those factors have been divided into those dealing with an understand-

ing of fault management and those dealing with an understanding of human-

automation integration. Hence the subsequent division of the next two sections of

this paper into sections titled ‘‘presenting information for managing the fault’’ and

‘‘presenting information for managing the fault management aid.’’ It should be

pointed out that as the technological ability to provide information becomes less

and less constrained, pilots (and other operators too) become information managers

as well as fault managers and automation managers. Although we do not address

information management requirements here, others (Malin et al., 1991; Rogers,

1991) discuss the issue of information management as another set of factors that

must be considered in defining information presentation principles.

FAULT MANAGEMENT 21

Cognitive

MISSION GOALS &

AUTOMATION FUNCTIONS

CAPABILITIES

limitations

AUTOMATION and biases

PHILOSOPHY FUNCTION

ALLOCATION Depth of

DESIGN -

hal al

FLIGHT CREW

TASKS

AIRCRAFT !

& KNOWLEDGE

processing

ENVIRONMENT

STATES

OPERATIONAL

CONTEXT

INFORMATION

Cognitive

abilities and

strengths

COGNITIVE

Cognitive

strategy

PROCESSING

COGNITIVE CHARACTERISTICS

FACTORS

COGNITIVE TASKS

REPRESENTATION

OF DOMAIN

base

Perceived

task structure

eon

= FORM

Soy Me

LEVEL OF

DETAIL ; ORGANIZATION

La i

lei Ll

Fig. 2. Factors influencing the content and form of information presentation.

The following sections will raise some of the questions and provide some

recommendations associated with this expanded perspective of information pre-

sentation for fault management.

Presenting Information for Managing the Fault

As mentioned, there are many information processing tasks involved in fault

management. This section discusses cognitive task analysis as a way to gain

insight into some of these tasks and as a way of identifying information require-

ments for fault management in general. Then, the Faultfinder concept will be

discussed as an example of an aid for fault management, and information require-

ments for the specific tasks of fault monitoring, fault diagnosis, and determining

appropriate responses will be addressed.

Cognitive Task Analysis of Fault Management

Cognitive task analysis provides a means of capturing cognitive activities

associated with various operational tasks. As Roth and Woods (1990) point out,

‘there has been increasing recognition of the importance of performing a cogni-

22 ROGERS AND ABBOTT

tive task analysis as a basis for defining requirements . . . for decision aiding.’’

We have used cognitive task analysis as a way to try to understand fault manage-

ment tasks, and in particular, what information is needed to support them.

Cognitive task analysis can be thought of as a set of analyses aimed at under-

standing the cognitive processes involved in the conduct of a particular task or

function. It is not a single analysis technique. In fact, loosely speaking, it is

probably fair to say that the aggregate of experimental tools and techniques aimed

at understanding human cognition comprise cognitive task analysis. Many studies

have addressed human cognitive and information processes related to fault man-

agement. In several operational domains, Woods and his colleagues (e.g., see

Woods, Roth & Pople, 1990), have investigated cognitive activities in regard

to problem solving and emergency situations. Others (e.g., Rasmussen, 1982;

Johanssen, 1983) have analyzed human fault management in nuclear power plants

in terms of an information processing model, and described how individuals

differ in the specific processing stages they use, depending on whether they adopt

a knowledge-based, rule-based, or skill-based mode of behavior.

One way to get at the information requirements for fault management on

commercial flight decks, especially in regard to information processing tasks, is to

try to capture how pilots organize and prioritize various types of fault management

information. If pilots organize information around different information pro-

cessing tasks, as we hypothesize, then we can gain insight not only into informa-

tion organization and prioritization, but also into the pilots’ ‘‘mental model’’ of

fault management; that is, in a generic sense, how they organize and prioritize

fault management functions and tasks.

Such a study was conducted by the first author (Rogers, 1993). The objective of

this study was to provide recommendations for presentation of fault management

information based on how pilots mentally represent the fault management domain.

This study was conducted in three steps: (1) fault management information types

were identified, (2) pilots were asked to categorize and prioritize the information

types in terms of a “‘generic’’ fault management situation, and (3) results were

analyzed using scaling and clustering analyses to extract organizational schemes

pilots use to group and rank information.

The information types were determined by conducting a series of structured

pilot interviews, and by reviewing manuals and documentation of current and

planned commercial aircraft as well as advanced fault management concepts such

as Faultfinder (Schutte et al., 1987). These data were used to generate 26 generic

types of fault management information. Examples of information types are:

® an alert that there is a fault

® a deviation of an actual parameter value from the expected value; and

@ the cause or location of a fault or condition.

FAULT MANAGEMENT 23

A description and example of each type of information was typed on an index

card. A group of 26 pilots were asked to sort the cards into piles based on the

similarity of the information and to rank order the cards based on general impor-

tance. The data were submitted to multidimensional scaling analyses and cluster-

ing analyses to determine how the pilots organized the information.

The results of the scaling analysis of the sorting data indicated that pilots

generally think about fault management information along two dimensions,

interpreted by the author as priority (or sequence of use) and stimulus-response.

The priority dimension indicated that pilots think about information in terms

of its priority or urgency. The scaling analysis of the ranking data did not reveal

multiple dimensions underlying prioritization of information. The stimulus-

response dimension suggested that pilots think about information in terms of

whether it supports assessing the problem and situation, or determining conse-

quences and actions. In fact, cluster analysis revealed that the stimulus-response

dimension likely represents the natural sequence of information processing

tasks: six primary clusters were found, and they were interpreted as information

supporting the tasks of monitoring, detection and identification, diagnosis, de-

tailed system status assessment, effects and consequences determination, and

response.

The priority dimension of the scaling solution and the rank-ordering data both

suggested that pilots want monitoring, detection and identification, and response

information first. This corroborates the rationale of the current flight deck fault

management information presentation philosophy. In terms of information pro-

cessing, pilots are probably operating in a skill-based mode, and short-cutting

deeper processing by associating shallow situation descriptions with pre-defined

response requirements (e.g., see Pew, Miller & Feehrer, 1981). Interestingly,

pilots indicated they would like the lower priority, ‘‘deeper’’ information such

as detailed status, diagnosis and consequences later, primarily to confirm their

understanding of the situation and the appropriateness of their responses. How-

ever, they want this information after they have stabilized the aircraft, compen-

sated for the fault, and when there is less time pressure.

This suggests that pilots want information (and probably perform information

processing tasks) consistent with the stages proposed by information processing

models, but not in the typical order: the order of processing of information and

the order of use appears to be different for fault management on commercial

flight decks (Figure 3). It’s not simply a matter of eliminating some stages

by ‘short-cutting’; rather, after an information processing short-cut to response

information, ‘backtracking’ occurs in which the skipped stages of processing are

performed after stabilizing responses are executed. These results have implica-

tions for what and when information needs to be presented, and which may best

24 ROGERS AND ABBOTT

Determine

consequences

Respond

Processing Order- The order in which information processing tasks are typically

performed without short cutting

Utilization Order- The order pilots would like to have information available based on results

_of study (reflects short cutting and back tracking)

Fig. 3. The order in which information is typically processed versus the order in which it is typically used

by pilots.

be presented automatically, which can be accessed at the pilot’s discretion, and

which may not be useful at all.

One caveat in interpreting the results of this study is important to note: since

the “*stimuli’’ grouped and prioritized by the subjects were themselves already

categories (1.e., they were ‘types’ of information rather than individual instances

or tokens of information), the possible ways that they could be further grouped

by the subjects were necessarily dependent on the pre-groupings. Another pre-

grouping strategy may have led to different results.

Studies using measures such as those just described, as well as verbal protocols,

eye-tracking data, etc., help identify the information requirements associated with

cognitive task performance, and hence are important elements of cognitive task

analyses. These analyses help identify aspects of information presentation not

addressed by traditional information requirements analyses.

Faultfinder

Faultfinder is an advanced concept for aiding flight crews in management of

systems faults and failures. While it should provide improvements in the kinds

of information that are presented by caution and warning systems on today’s

aircraft, it is aimed primarily at presenting new kinds of information to aid in

fault management tasks that are de-emphasized in modern aircraft (at least from

FAULT MANAGEMENT 25

the informational aid standpoint), such as fault monitoring, fault diagnosis, and

response generation. Faultfinder was designed to generate information to support

management of all faults, but particularly those which are difficult for pilots to

manage, such as novel, complex, or multiple faults (Schutte et al., 1987; Schutte,

1989). It represents a substantial advance over current caution and warning sys-

tems with respect to the computational capability to generate types of information

about fault situations, but comes with a corresponding increase in concerns about

information presentation.

The information presentation philosophy of the Faultfinder concept is currently

being developed, but simply, it is to present information to the flight crew that is

consistent with their informational needs as they ‘‘walk’’ through the information

processing stages associated with fault management, within the constraints of their

environment. Additionally, since it is conceived as a human-centered automation

concept, it brings with it requirements for additional information presentation

associated with managing automation. And, although not covered here, because

aiding concepts like Faultfinder have the potential for greatly increasing the

overall information load on the pilot, information management issues need to be

part of the information presentation philosophy as well.

Monitoring. ‘‘Information processing in most systems begins with the detection

of some environmental event.’’ (Wickens, 1984). However, flight crew monitor-

ing for systems faults on commercial flight decks has not been a major concern

in recent years due to the effectiveness of automated alerting systems in detecting

faults and alerting the flight crew. Engine faults and failures have been an excep-

tion because it has proved difficult to develop engine sensors that could reliably

distinguish faults from non-faults. However, for most systems, continuous param-

eter values are monitored by the automation and a discrete alert is activated based

on a pre-determined threshold value. A system that just provides an alert to the

flight crew may withhold accumulating evidence (e.g., actual parameter values

or rates of change) that might be useful to the crew before the alert threshold is

reached. After an alert has been activated, good parameter information can pro-

vide a needed source of confirming or refuting evidence of the problem. A premise

of the Faultfinder concept is that better fault monitoring information may allow

pilots to respond better, or better anticipate responses, relative to presentation of

alerting information only.

Secondly, thresholds for alerts are currently set as hard limits; when a particular

absolute value is exceeded, an alert occurs. In the case of engine instruments,

this may simply be a color change on the parameter display. For many parameters

whose normal values vary with the situation, absolute numbers may not be as

important for monitoring the health or status of the system as the actual parameter

value’s relation to what is normal for that set of conditions. For many cases,

26 ROGERS AND ABBOTT

then, it is hypothesized that displays that depict actual parameter values relative

to expected values should be more useful for fault monitoring than ones which

depict only absolute values.

This is the basis for monitoring concepts such as E-MACS (Abbott, 1989) and

the Monitaur module of Faultfinder (Schutte, 1989). Presentation of information

for these concepts was based on analysis of the information processing required

to perform the task, and in this case, monitoring of relative values rather than

absolute values is a key aspect of the information processing associated with the

fault detection task. Pilot performance of a monitoring task using the E-MACS

display presenting relative parameter information was shown to be much 1m-

proved over the traditional round-dial presentation of absolute parameter informa-

tion (Abbott, 1990).

If the system can provide the expected ‘‘normal’’ value as well as the actual

value, then the cognitive processing can be reduced and the task becomes essen-

tially a perceptual matching one, which humans generally perform quickly and

accurately. Of course, this depends on the relative values being displayed in a

manner that is perceptually ‘“‘salient,’’ that is, that can be quickly and reliably

noticed and decoded by the human monitor. A recent NASA study (Palmer &

Abbott, 1994) showed that when a number of relative parameter values are to

be monitored at once, a column deviation display (which shows the amount each

parameter deviates from the expected value as an upward or downward bar

extending from a centerline which represents no deviation), was more effective

than a round dial that contained both actual and expected values. In a similar

vein, object displays and other types of integrated displays use deviations from

default object shapes to depict deviations in one or more parameters (e.g., see

Carsell & Wickens, 1987; Buttigieg & Sanderson, 1991). These types of displays

can exploit both the task-oriented notion of supporting monitoring tasks with

relative values, and the human perceptual strength of being sensitive to salient,

global changes in visual stimuli.

A key to all these relative value display concepts is the capability of the

automation to compute nominally ‘‘normal’’ values. An obvious question that

arises with this capability is, why not have the automation alert the pilot based

on relative values. In other words, why not alert the pilot when a parameter value

deviates from the expected value by some pre-set amount? While providing data

and formats that enhance human detection of problems is desirable, certainly

there is no question automation excels at monitoring and alerting functions. We

are currently including the combination of better monitoring information and

additional alerting information in up-coming simulation tests of Faultfinder.

Figure 4 shows an example of an engine parameter display we are experiment-

ing with for one of these upcoming studies, in which both actual and expected

FAULT MANAGEMENT 27

N1 N1

LENG N1 ABOVE EXPECTED

(b)

LENG N1 IN WARNING BAND

LENG N1 ABOVE EXPECTED

- green

(MMM = - white

Fig. 4. Examples of engine N1 displays which depict actual values, expected values, exceedence bands, and

appropriate alerting information for conditions in which parameter values are normal (a), deviate from expected

(b), exceed the caution threshold (c), and exceed the warning threshold and deviate from expected (d).

values are shown, and alerts occur both for a deviation from expected value, and

for exceeding absolute thresholds. The actual value is depicted by the digital

readout and the fill of the thermometer. The expected value is depicted by the

green band on the left side of the thermometer. Four different conditions are

shown. When the actual value is within the expected range, and not exceeding

the hard limits (condition a), the fill remains green. When the actual value goes

out of the expected range (but does not exceed a hard limit), the fill changes to

white and a white ‘‘deviation’’ message occurs (condition b). The notion is that

a deviation from expected would be advisory level or informational, simply

28 ROGERS AND ABBOTT

informing the pilot to be more aware of that parameter or component because

there may be a problem developing.

The hard limits are depicted by the yellow and red bands on the right side of

the thermometer, and when the actual value reaches these values, the fill and

digital readout change to yellow (condition c) or red (condition d). Values that

exceed hard limits are accompanied by a corresponding yellow or red message.

This would indicate a caution or warning condition which requires immediate

crew awareness and immediate or imminent crew response (this is experimental;

as mentioned earlier, the engine parameters on today’s aircraft are not currently

tied into the caution and warning system because of the difficulty in reliably

predicting engine health from parameter values). If a hard limit is exceeded and

a deviation exists, as in condition d, the color coding of the fill and digital readout

would be driven by the value exceeding the hard limit, and the deviation would

be apparent from comparing the expected value band with the actual value, and

a “‘deviation’’ message would be presented just as it would if there was only a

deviation condition. It is hoped that a concept such as this will offer the advantages

of both automatic alerting for detection, and display of parameter data in reason-

able formats to support the pilot’s ability to use features of “‘raw’’ data, such as

rates of change and relative values, particularly in relation to their superior

knowledge of variables which may influence these.

As mentioned earlier, another possible advantage of presenting parameter data

to pilots in addition to automated alerts, is that they can detect trends in parameter

values and may be able to anticipate faults and failures before they occur, thus

allowing them to get a “‘headstart’’ in managing the faults. A NASA study (Trujillo,

1994) evaluated different formats for depicting trend information. It assessed sym-

bology that was hypothesized to enhance the ability to predict when an alert thresh-

old would be exceeded or when a fault would occur from observing the rate of

change of a parameter value for five or ten seconds. Although the experimental

symbology did not appear to offer any advantage over standard symbology, it was

noteworthy that for both an easy (constant rate of change) and moderate (decreasing

deceleration) trend condition, pilots could fairly accurately (less than 20% error)

predict the onset of an alert that occurred 20 to 80 seconds later.

Diagnosis. Once the existence of an abnormality is detected, another cognitive

task is to diagnose the cause and the effects of the fault. In many applications,

it is sufficient to identify the cause or source of the fault, so that the part to fix

or replace can be identified. For in-flight diagnosis, however, it is also important

to identify aircraft subsystems that are affected by the fault, but not necessarily

broken. For example, an engine failure may cause an electrical generator to drop

off line. The generator itself may be working properly, but because of the effect

of the failure, the flight crew must take a corrective action. First, however, the

FAULT MANAGEMENT 29

pilots must know that there is an action required. Additionally, since the aircraft

systems are continuing to operate, the propagation of the fault over time must

be understood to identify the changing system status.

For generating these types of diagnostic information, the diagnostic module

of the Faultfinder concept, called DRAPhys (Abbott, 1991), was designed to

diagnose these subsystem failures, especially faults that pilots have difficulty

understanding or responding to, such as multiple or novel faults. To generate the

diagnostic information about such faults, it uses qualitative models of the aircraft

subsystems to identify the cause of the fault, its propagation behavior, and the

resulting system status.

Deciding how to present this information to pilots needs to account for the

fact that they are only trained to perform rudimentary diagnosis, and evidence

indicates that they do not perform particularly well at this task. An aid might

provide information to help pilots in areas where they have particular difficulty

in performing diagnosis. In fact, the example of the engine failure is one in which

pilots on current flight decks may receive an alert related to the electrical system,

but may not get any alert on an engine failure, even though the engine was the

cause of the electrical system problem. Under some circumstances, such as idle

descent, once the pilots’ attention is directed to the electrical system, they may

fixate on it as the source of the problem, and only discover the engine failure

when obvious cues (such as lack of thrust when the throttles are moved forward)

become available.

Accurate diagnosis of system faults may be most important in unusual or complex

failure situations. These types of situations may ‘trap’ pilots into mappropriately

taking information processing short-cuts to quickly get to required responses. Not

fully understanding the problem may lead them to err in deciding what to do. Yoon

and Hammer (1988) addressed human performance in diagnosing novel faults, or

faults the human had not seen or been trained to manage. In this study, they found

that subjects provided with information about the discrepancies between observed

and normal system behavior performed better in the diagnostic task. Note that this

assumes a particular role for the human, with the aid supporting the human informa-

tion processing tasks without actually solving the problem for him or her.

Given that diagnosis is not done very often and that they are not extensively

trained for it, how do we support pilots performing diagnosis? It is generally

agreed that they need to know the status of the system with which they are

concerned, whether it is calculated for them or they figure it out themselves. If

we show them status information, there is a wealth of literature on pictorial and

graphical formats in which to show this (e.g., Summers & Erickson, 1984).

Graphical or schematic presentations seem to be consistent with the way humans

mentally model physical systems, although research by Kieras (1992) revealed

30 ROGERS AND ABBOTT

that it is useful to integrate the status of parameter information into the system

status display, so that related parameters and system components were close to

each other on the displays.

It is not so clear how to present other diagnostic information that we can now

generate, such as fault propagation, nor is there a hard and fast guideline on when

to present it. A recent study (Rogers, 1993) suggests diagnostic information should

usually be presented after the situation has been stabilized and the flight crew has

time to confirm its actions. Yet if the diagnosis is critical to choice of the correct

response, which is exactly the case in which pilot errors might occur, diagnostic

information should be presented immediately, as part of the ‘identification’ infor-

mation which helps the pilot choose the appropriate actions. Many open issues

remain, both in information content and form for diagnostic information.

Response Generation. In determining information requirements for the flight

crew to decide how to respond to a failure, there are several factors to consider.

One is that routine faults differ from novel faults, and the generation of a response

is correspondingly different. Another is that responses must be made relative to

the mission, goals and functions being performed by the flight crew.

Currently, flight crews have standardized procedures in the form of checklists

to respond to routine faults. Until recently, these checklists have been in paper

form, and the pilots retrieve them as appropriate based on the identification of

the fault. In some of the new generation aircraft, these checklists are implemented

electronically, so that they can be called up on a digital display when needed, or

the display may be presented automatically. Electronic checklists seem like an

obvious advance in the quest towards the “‘paperless cockpit,’ but undesired

effects such as pilot complacency and relinquishment of perceived authority may

accompany the electronic version. Palmer and Degani (1991) examined different

levels of automation in electronic checklists and found that pilots did not monitor

the electronic checklist at the same level of performance as with the paper check-

list, and that electronic checklist designs encouraged flight crews to not conduct

their own checks. This appears to be an example of behavioral phenomena that

Wiener and Curry (1980) referred to as ‘*Primary/Secondary Task Inversion,’’

where operators begin to rely on a backup system as a primary system.

Another concern is that checklists don’t exist for every fault situation. For

novel faults or multiple faults, for example, the pilot may have to figure out how

to modify an existing checklist, merge two checklists, or create an entirely new

set of actions appropriate for the situation. Automated aid in these situations

could be extremely important, but the technology has not been available to provide

it. Concepts such as Faultfinder are beginning to look at the issues involved in

presenting such information.

Checklists identify actions that must be taken to directly deal with the aircraft

FAULT MANAGEMENT 31

subsystem to correct or compensate for a fault; for example, shutting down an

engine. However, responses must also compensate for the effect of the fault on

other functions, such as flight control, planning and navigation. For example,

how can the aircraft be controlled when hydraulics fail? Can the destination

airport still be reached, or should the flight crew divert to an alternate airport?

Additionally, a fault may cause a constraint on the operational limitations of the

aircraft that will not affect the flight until much later. For example, a landing

gear problem detected in cruise will not affect the situation until landing, but the

pilot must remember that information for when it is needed.

For many of these control and navigation responses, pilots currently rely on

their experience to deal with the fault. An example of such a situation occurred

in the crash at Sioux City, Iowa (NTSB, 1990). In this accident, the pilots

used the wing engines for pitch control, because the elevators were damaged.

Information on this unusual use of the engines for controlling the aircraft is not

provided as part of the on-board fault management system. However, when novel

faults occur, just such an action may be appropriate. The RECORS portion of the

Faultfinder concept was designed to provide aid for just these types of situations

(Hudlicka, Corker, Schudy, & Baron, 1990).

All the issues we have raised in this section are related to aiding the pilot in

performing fault management tasks. Clearly there are many associated open

research issues. There are also research issues in the information requirements

associated with working with an automated aid.

Presenting Information for Managing the Fault Management Aid

The Society for Automotive Engineers G-10 Committee on automation has

listed nine categories of pressing automation concerns. Situation awareness, de-

sign of the crew interface, and the role of the pilot in automated aircraft are three

of those concerns. We posit that these three areas are highly related: when casting

the flight crew in the role of fault manager and the automation in the role of fault

management aid, a key component of required situation awareness is automation

awareness, and that awareness must be supported by the crew interface to the

automated aid itself: it must provide information to help the pilot manage it.

In a workshop on flight deck automation promises and realities (Norman &

Orlady, 1989), Wiener described automation that exists today as ‘‘clumsy’’ auto-

mation. When asked if this referred to the automation or to the interface to the

automation, he replied that he did not make a distinction. The problem refers to the

combination (Wiener, 1989). Norman (1991) has referred to current automation as

being at ‘‘an intermediate level of intelligence that tends to maximize difficult-

32 ROGERS AND ABBOTT

ies.’ Underlying these descriptions is a dual problem-one of processing clumsi-

ness (1.e., automation can’t be intelligent in all situations until it can perform

sophisticated situation assessment and pilot intent inferencing), and one of clumsy

or inadequate information presentation. We would argue that automation is

clumsy in its presentation of information and providing feedback about itself,

not so much because of technological limitations, but more because we do not

have well-defined information requirements for the effective use of intelligent

automation by the flight crew. There is a substantial body of literature that could

be translated into information requirements, but it has not been fully exploited

by those designing the automation.

What are the information requirements in terms of the pilot’s need for informa-

tion to manage automated aids? An exhaustive listing is beyond the scope of this

paper, but we will try to touch on some that are known, but not necessarily

manifested in today’s flight deck automation. These requirements apply to auto-

mated aids in general, but are relevant to the fault management aids of concern

here. We will first discuss information needed to support human-computer interac-

tion via the interface to the automation (interface information requirements), and

then information the flight crew needs to be aware of what the automation is

doing, how it is doing it, how well it is doing it, how reliable its output is, etc.

We call this automation awareness information requirements.

Interface Information Requirements

Much work has been done on tailoring the interface for interactive systems

for users with different levels of experience. A system that may be used infre-

quently or by novice users should provide information that assists the user in

interacting with the aid. The problem is that there is a time penalty in interacting

with an information system, and that penalty is increased by providing information

that helps the user step through the interaction. In time-critical fault situations

this is unacceptable. It is likely that for the time-critical information and phases

of aircraft system faults, there will be no or little interactions required with the aid;

certain information may automatically be presented. Thus the issue of presenting

information to assist the user in inputting and accessing information is only

pertinent to later phases of a fault management situation when more time is

available and the flight crew may be looking for more in-depth information.

Nonetheless, such information is critical to extracting the full potential of an

informational aid.

Unless a fault management aid provides this kind of supporting information

about how to interact with it, it is not much of an aid, especially to a novice user.

As one becomes more experienced, much of this kind of information becomes part

of one’s knowledge base and the need for the aid to provide this supporting

FAULT MANAGEMENT 33

information decreases. In the context of decision aiding on the commercial flight

deck, however, it should be kept in mind that situations where an automated

decision aid such as Faultfinder might be used in real-time may be rare. Therefore,

unless the flight crew can get sufficient off-line familiarity with the tool in a

training or practice setting, they should probably be treated as novice users; at

least the design should account for this possibility, which means the burden of

extra “‘interface’’ information such as the type described above is required.

Automation Awareness Information Requirements

Norman (1991) contends that automation does not provide adequate feedback

about its operation to the pilot. Feedback includes mode annunciation, that is, an

indication of what it is doing or how it is performing a given function. As

a recommendation for human-computer interaction design involving intelligent

automation, Malin et al. (1991) state that the active operating mode should be

clearly distinguished by the user interface. While a very basic requirement, mode

confusions still occur with highly automated systems. There is speculation, for

example, that the recent Air Intel Airbus A-320 crash involved confusion between

the pilots over what flight mode was selected (Lenorovitz, 1992). While the

involved system did provide mode annunciation, the question will arise as to

whether it was ‘clearly distinguished’. On the MD-11, systems management, in

terms of reconfiguring or compensating for systems faults, can be performed

automatically or in a ‘‘manual’’ mode; it is imperative that the pilots are aware

of which mode is active. Feedback about the mode of operation may not be as

important for informational systems as for control systems, but one certainly

needs basic information about the operating status (whether it is fully functional,

partially functional, non-functional), whether the aid is waiting on pilot input,

searching for data, computing or processing data, or simply cannot provide a

given answer or piece of information.

In addition to mode annunciation, Norman (1991) argues that one should have

continuous information by which pilots can compare the current operation of the

automation to their expectations about what the operation of the automation

should be. Often this kind of feedback is available from natural cues. For example,

the functioning of the autothrottle system is normally reinforced with continuous

visual feedback from throttle position. If that cue is eliminated, then it is impera-

tive that some information be presented by the automation showing what the

autothrottles are actually doing moment to moment. As Norman (1991) points

out, several accidents and incidents, such as the 1985 China Airlines 747 loss of

engine power (NTSB, 1986), were contributed to by insufficient feedback from

the automation. These incidents often occur even when natural feedback cues are

available (such as the position of the wheel and column as a cue to the inputs the

34 ROGERS AND ABBOTT

autopilot is making to the control surfaces). This points to an area of information

presentation where advances in technology can perhaps augment rather than

replace other feedback cues.

The above example deals with controlling automation, but the same concerns

hold true for informational automation or automation that provides computational

or planning support. Sarter and Woods (1992) describe similar problems for pilots

effectively interacting with flight management systems (FMS). They conclude

from their studies that a majority of problems in using the FMS result from

‘*system opaqueness.’’ The FMS does not provide pilots ““with adequate feedback

on the past, present or future system state and behavior.’’ At best, this results in

mistrust and mystique (e.g., ‘“why did it do that?’’; “‘it’s doing its magic again’’),

and worst, the pilots may not use the aid, may not use it properly, or may accept

the aid’s output without a level of understanding of its processing that is required

to assess its accuracy. Making the operation of automation ‘‘transparent’’ cer-

tainly applies to fault management aids.

The ‘‘transparency of operation’’ requirement becomes even more important

with ‘‘intelligent’’ automation which may be reasoning and hypothesizing, and

supplying probabilistic or evidential information, which the human operators

must then evaluate. Malin et al. (1991) discuss problems such as “providing

visibility into intelligent system reasoning,’ ‘‘distinguishing hypotheses from

facts,’ and “‘understanding intelligent system reasoning strategy.’’ Information

about reasoning processes and strategies, how the information was computed,

what data it was based on, etc., is important for the pilot to have to perform the

necessary evaluation of the system’s output.

This is related to the second part of the clumsy automation problem alluded

to above. That is, the sophistication of today’s generation of intelligent decision

aids is such that the aids can not be expected to perform requisite processing

perfectly or provide definitive answers for all circumstances and situations. This

characteristic again points to the flight crew requirement for information about

the accuracy of the automation’s output under different sets of circumstances.

Wheeler, Bolton & Sanquist (1991), for example, suggest that one of the most

demanding and difficult aspects of emergency management is evaluating the

validity, timeliness, completeness and factuality of information. Pilots thus need

information on an aid’s overall reliability, its capabilities and limitations, when

those capabilities are most useful, and when the limitations are most likely to

manifest themselves. If, for example, pilots know that the TCAS has a limitation

in that it cannot account for the intention of the target aircraft (e.g., whether it

plans to level off or continue to descend), they may be better able to evaluate

whether the TCAS advisory is a ‘‘false alarm’’ because they will be more likely

to have acquired additional information about the intentions of the target aircraft.

FAULT MANAGEMENT 35

In this vein, a fault management aid which assists with detection, diagnosis,

and compensatory actions runs the risks of providing misinformation. In signal

detection terms, the aid can have misses or false alarms. Misses are when the

aid does not detect a fault or failure, or does not diagnose the correct cause.

False alarms are said to occur when the aid indicates something is wrong when

it is not. If the flight crew can evaluate the output of the aid based on an

understanding of how it operates, its strengths and limitations, as well as on the

basis of other, uncorrelated sources of information, then the fault management

aid’s misses and false alarms are less likely to become the pilot’s misses and

false alarms. The flight crew will not be as prone to extremes in reliance, either

ignoring or turning the system off or over-relying on the system. The output can

be evaluated and combined with other evidence relevant to the fault situation.

This seems to typify the appropriate human-automation relationship for a human-

centered automation approach, and the success of this relationship hinges on the

operator having information by which he or she can evaluate the automation.

Integration of the Fault Management Aid Into the Flight Deck

As our final major point, note that the preceding discussion of information

requirements for fault management aiding did not consider the rest of the flight

deck systems of which this aid must be a part. A major concern in the design of

any new automation must be the integration of the concept into the overall system

of which it will be a part. For a flight deck aid, integration into both the flight

deck as a whole, and the airspace system where Air Traffic Control interacts with

the aircraft, must be considered. Without explicit consideration of integration, a

concept might be internally consistent, but might be inconsistent when considered

in an overall system context. That is, the whole might be less than the sum of

the parts.

This should be of particular concern for a fault management aid, since there

are Other alerting systems on the flight deck now (such as, TCAS), and more are

expected in the future (such as, wind shear alerting systems). We believe it is

important, for example, to have a consistent alerting philosophy across all alerting

systems in the flight deck, so that the flight crew does not have to interpret an

alert differently depending on what system it is.

Summary

This paper emphasized information content, that is, what information should

be presented for management of aircraft systems faults, and additionally, when

36 ROGERS AND ABBOTT

it should be presented. In that context, this paper focused on information to

support fault management performed by the flight crews of commercial aircraft

flight decks. We discussed the characteristics of the application, and the similari-

ties to and differences from fault management in other application areas. We

discussed human management of faults using automation as an aid, and how this

human-centered automation philosophy affects the design of the decision aid.

We discussed the importance of determining the appropriate role of the human,

and the determination of information requirements. The results of studies ad-

dressing the information needs and desires of flight crews were described, with

a particular emphasis on the manner in which the pilots process the information

to support fault management tasks. In addition to discussing presentation of

information for managing failures, we also described the need to present informa-

tion to manage the fault management aid. Lastly, we raised the issue of integrating

the fault management aid with the remainder of the flight deck systems.

Acknowledgments

We wish to acknowledge our colleagues who are also working on fault manage-

ment at NASA Langley Research Center, especially Paul Schutte, Anna Trujillo,

and Michael Palmer, for useful discussions that contributed to the content of this

paper. We would also like to thank Paul Schutte for his review of this paper.

References

Abbott, K. H. (1991). Robust fault diagnosis of physical systems in operation (NASA TM 102767). Hampton,

VA: NASA Langley Research Center.

Abbott, T. S. (1989). Task-oriented display design: Concept and example (NASA TM-101685). Hampton,

VA: NASA Langley Research Center.

Abbott, T. S. (1990). A simulation evaluation of the engine monitoring and control system display (NASA

Technical Paper 2960). Hampton, VA: NASA Langley Research Center.

Billings, C. (1991). Human-centered aircraft automation: A concept and guidelines (NASA Technical Memo-

randum 103885). Moffett Field, CA: NASA Ames Research Center.

Boucek, G. P., Berson, B. L., & Summers, L. G. (1986). Flight status monitor system-Operational evaluation.

Presented at the 7th IEEE/AIAA Digital Avionics Systems Conference. Forth Worth, TX: IEEE.

Buttigieg, M. A., & Sanderson, P. M. (1991). Emergent features in visual display design for two types of

failure detection tasks. Human Factors, 33:631-—651.

Carsell, C. M., & Wickens, C. D. (1987). Information integration and the object display. Ergonomics, 30:511-

D2:

Cooper, G. E. (1977). A survey of the status of and philosophies relating to cockpit warning systems (NASA

CR-152071). Moffett Field, CA: NASA Ames Research Center.

Douglas new systems automation policy to ensure minimal MD-11 pilot workload (1990, June). Aviation Week

and Space Technology, p. 20.

Fitts, P. M. (Ed.) (1951). Human engineering for an effective air navigation and traffic control system.

Washington, D.C.: National Research Council.

Hudlicka, E., Corker, K., Schudy, R., & Baron, S. (1989). Flight crew aiding for recovery from subsystem

failures. (NASA-CR-181905). Cambridge, MA: Bolt, Beranek & Newman, Inc.

Johannsen, G. (1983). Categories of human operator behavior in fault management situations. Proceedings

of the International Conference on Systems, Man and Cybernetics, Vol. II (pp: 884—889). Bombay and New

Delhi, India: IEEE.

FAULT MANAGEMENT 37

Jordan, N. (1963). Allocation of functions between man and machines in automated systems. Journal of

Applied Psychology, 47:161-—165.

Kieras, D. (1992). Diagrammatic displays for engineered systems: Effects on human performance in interacting

with malfunctioning systems. Jnternational Journal of Man-Machine Studies, 36:861—895.

Lenorovitz, J. M. (1992). Confusion over flight mode may have role in A320 crash. Aviation Week and Space

Technology, 136(5):29—30.

Malin, J. T., Schreckenghost, D. L., Woods, D. D., Potter, S. S., Johannesen, L., Holloway, M., &

Forbus, K. D. (1991). Making intelligent systems team players: Case studies and design issues, volume

I: Human-computer interaction design (NASA TM 104738). Houston, TX: NASA Lyndon B. Johnson

Space Center.

National Transportation Safety Board (1986). China Airlines Boeing 747-SP, N4522V, 300 Nautical

Miles Northwest of San Francisco, California, February 19, 1985 (NTSB/AAR-86-03). Washington, DC:

Author.

National Transportation Safety Board (1990). United Airlines Flight 232, McDonnell Douglas DC-10-

10, Sioux Gateway Airport, Sioux City, Iowa, July 19, 1989 (NTSB/AAR-90-06). Washington, DC:

Author.

Norman, D. A. (1991). Cognitive science in the cockpit. CSERIAC Gateway, 2:1-6.

Norman, S. D., & Orlady, H. W. (Eds.). (1989). Flight deck automation: Promises and realities (NASA CP

10036). Moffett Field, CA: NASA Ames Research Center.

Palmer, E., & Degani, A. (1991). Electronic checklists: Evaluation of two levels of automation. In R. S.

Jensen (Ed.), Proceedings of the 6th International Symposium on Aviation Psychology (pp. 178-183).

Columbus, OH: The Ohio State University.

Palmer, M. T., & Abbott, K. H. (1994). Effects of expected-value information and display format on recogni-

tion of aircraft subsystem abnormalities (NASA Technical Paper 3395). Hampton, VA: NASA Langley

Research Center.

Pew, R. W., Miller, D. C., & Feehrer, C. E. (1981). Evaluating nuclear control room improvements through

analysis of critical operator decisions. In Proceedings of the Human Factors Society 25th Annual Meeting

(100-104). Santa Monica, CA: The Human Factors Society.

Rasmussen, J. (1982). Human errors. A taxonomy for describing human malfunction in industrial installations.

Journal of Occupational Accidents, 4:311—333.

Riley, V. (1989). A general model of mixed-initiative human-machine systems. In Proceedings of the Human

Factors Society 33rd Annual Meeting (pp. 124—128). Santa Monica, CA: The Human Factors Society.

Rogers, W. H. (1991). Information management: Assessing the demand for information. In R. S. Jensen (Ed.)

Proceedings of the 6th International Symposium on Aviation Psychology (pp. 66—71). Columbus, OH: The

Ohio State University.

Rogers, W. H. (1993). Managing systems faults on the commercial flight deck: analysis of pilots’ organization

and prioritization of fault management information. In R. S. Jensen & D. Neumeister, (Eds.) Proceedings

of the 7th International Symposium on Aviation Psychology (pp. 42—48) Columbus, OH: The Ohio State

University.

Roth, E. M., & Woods, D. D. (1990). Analyzing the cognitive demands of problem-solving environments:

An approach to cognitive task analysis. In Proceedings of the Human Factors Society 34rth Annual Meeting

(pp. 1314-1317). Santa Monica, CA: The Human Factors Society.

Rouse, W. B. (1991). Design for success: A human-centered approach to designing successful products and

systems. New York: Wiley.

Sarter, N. and Woods, D. (1992). Pilot interaction with cockpit automation I: Operational experiences with

the flight management system. International Journal of Aviation Psychology, 2(4):303-—321.

Schutte, P. C. (1989). Real-time fault monitoring for aircraft applications using quantitative sensor simulation

and expert systems. In AJAA Computers in Aerospace VII Conference Part 2 (pp. 876—885). Monterey, CA:

AIAA.

Schutte, P. C., Abbott, K. H., Palmer, M. T., & Ricks, W. R. (1987). An evaluation of a real-time fault

diagnosis expert system for aircraft applications. In Proceedings of the 26th IEEE Conference on Decision

and Control (pp. 1941—1947). New York: IEEE Control Systems Society.

Summers, L. G. & Erickson, J. B. (1984). System status display information (NASA CR 172347). Long

Beach, CA: McDonnell Douglas Airplane Co.

Trujillo, A. C. (1994). Effects of historical and predictive information on ability of transport pilot to predict

an alert (NASA TM 4547). Hampton, VA: NASA Langley Research Center.

Wheeler, W. A., Bolton, P. A., & Sanquist, T. F. (1991). Decision making in an emergency: When information

is not enough. In Proceedings of the Human Factors Society 34rth Annual Meeting (pp. 1137-1141). Santa

Monica, CA: The Human Factors Society.

Wickens, C. D. (1984). Engineering psychology and human performance. Glenview, Illinois: Scott, Foresman

and Co.

Wiener, E. L. (1985). Beyond the sterile cockpit. Human Factors, 27:75—90.

38 ROGERS AND ABBOTT

Wiener, E. L. (1989). Field studies in automation. In S. D. Norman & H. W. Orlady (Eds.), Flight deck

automation: Promises and realities (NASA CP-10036). Moffett Field, CA: NASA Ames Research Center.

Wiener, E. & Curry, R. (1980). Flight-deck automation: promises and problems. Ergonomics, 1980, 23:995—

1011.

Woods, D. D. (1989). The effects of automation on the human’s role: Experience from non-aviation industries.

In S. D. Norman & H. W. Orlady (Eds.), Flight deck automation: Promises and realities (NASA CP-10036).

Moffett Field, CA: NASA Ames Research Center.

Woods, D. D., Roth, E. M., & Pople, H. E. (1990). Modeling operator performance in emergencies. In

Proceedings of the Human Factors Society 34rth Annual Meeting (pp. 1132-1136). Santa Monica, CA: The

Human Factors Society.

Yoon, W. C. & Hammer, J. M. (1988). Aiding the operator during novel fault diagnosis. Transactions on

Systems, Man, and Cybernetics, 18:142—147.

Journal of the Washington Academy of Sciences,

Volume 84, Number |, Pages 39-52, March 1996

Plasmon Excitation in Conducting Solids

C. R. Schumacher’ and Barbara Howell

Naval Surface Warfare Center, Carderock Division, Annapolis Detachment,

Annapolis, MD 21402

H. Uberall”

Naval Warfare Center, Carderock Division, Annapolis Detachment,

Annapolis, MD 21402 and Department of Physics, Catholic

University of America, Washington, DC 20064

Received June 29, 1995

ABSTRACT

Drude model fits to the dielectric functions of 15 metals are available, largely based on

the experimental work of Ordal and collaborators. The resulting values of plasma and damping

frequencies provide predictions for the DC conductivity in reasonable agreement with mea-

sured data, and they also permit (approximate) predictions for the dielectric constant e, [the

w — OQ limit of the real part of the DC dielectric function e(w)] of these metals, which is

found to be large and negative. More accurate values for this and the plasma frequencies can

be based on empirical tabulations, and with the usual interpretation of the crossover of €,(w)

from negative to positive values near the optical region as corresponding to plasmon excita-

tion, these values allow the calculation of electron loss factors which are sharply peaked

close to the plasma frequency. A trend of the DC conductivity to rise with increasing plasma

frequency is pointed out, and this trend can also be followed through in the direction of

decreasing conductivity, as is shown here for several poor conductors (carbon and conducting

polymers).

Introduction

In a series of studies, Ordal et al. (1985; 1987; 1988) have carried out electro-

magnetic power reflectance measurements on various metals in the 1/A = 180 @

50,000 cm™' range, as well as absorptance measurements using a nonresonant

cavity in the 30 ~ 300 cm‘ range. Combining these data in the far-IR, IR and

visible region with those found by others, they were able to obtain, using Kramers-

' Deceased.

> US Navy—ASEE Distinguished Summer Faculty Fellow at NSWC Annapolis.

40 SCHUMACHER, HOWELL, AND UBERALL

Kronig analysis of their surface-impedance data, the real and imaginary parts of

the dielectric function,

E(W) = &\(W) + €2(w), (1)

as functions of frequency in the mentioned frequency range. Subjecting these

results to a Drude model (see, e.g., Bohren & Huffman, 1983) two-parameter fit

(parameters being the plasma frequency w, and the damping frequency w,, reason-

able fits were obtained which can be used to extrapolate e(w) down to its DC

value «(Q). In this way, Ordal et al. (1985) found DC conductivity values o(Q)

which generally compared well with conventionally-measured and tabulated

(Babiskin & Anderson, 1972) conductivities.

By combining the various Ordal data (Ordal et al., 1985; 1987; 1988) and the ©

Drude model fits, we tabulate the DC conductivities and also the previously

believed (Slater & Frank, 1947) unmeasurable DC dielectric constants for fif-

teenmetals, by extrapolation of the Drude model fits to low frequencies. We note

that DC conductivities exhibit a rising trend as the plasma frequency increases,

and we show that such a trend also holds for a few poor conductors considered

here (carbon and conducting polymers).

As a further step, we tabulate the crossover-frequencies w, of these metals at

which €,(w) changes sign; at this place, the electromagnetic field is longitudinal

(Bohren & Huffman, 1983) corresponding to a collective longitudinal oscillation

of the free electrons known as a plasma oscillation (in quantized form, a plasmon).

For the metals considered here, the values of w, lie in the visible (1/A ~ 1.4 X

10* — 2.5 x 10°* cm‘') or ultraviolet region. It should be emphasized that the

Drude model equations given below only contain the effects of the free electron

gas, neglecting the effects of bound electrons and/or of ionic vibrations on the

dielectric function. These resonance effects have been discussed at length (see,

e.g., Bohren & Huffman, 1983) and are illustrated there e.g. in Fig. 9.11, they

appear close to the visible region and can be noticed in Ordal’s data which are

presented on logarithmic plots. The resonance effects differ for individual materi-

als, so that no general statements can be made about them. As to our use of the

Drude fits to predict plasmon excitation frequencies, it could be argued that the

free-electron plasmon is screened by the dielectric constant at optical frequencies,

causing shifts in w.

Tabulations of earlier reflectivity and absorptivity data (Weaver et al., 1981)

do include these effects, as well as the above mentioned resonance effects, on

the dielectric function in an empirical fashion, and alternate (empirical) values

of w, and e,(0) can be obtained from Weaver et al. (1981), indeed leading

often to differences with the Drude values. For purposes of determining plasmon

frequencies, the empirical values should be preferable. Likewise, loss factors for

PLASMON EXCITATION 41

the excitation of plasmons by electron scattering, sharply peaked about the cross-

over frequencies, are obtained from the empirical data (Weaver et al., 1981)

together with their widths.

Drude Model Fits to the Optical Properties of Metals

The Drude model was proposed in 1904 and is the earliest realistic model for

a metal (see, e.g., Pines, 1964; Peierls, 1956). It has often been quoted as offering

a surprisingly good description of the optical properties of metals, considering

the crudity of its assumptions, and even the introduction of quantum theory with

its Fermi-gas picture of conduction electrons (Sommerfeld & Bethe, 1933), did

not change its results except explain the origin of its frictional resistive force

(Slater & Frank, 1947; Peierls, 1956). This frictional force introduces a mean

collision time 7 for each electron, and the electron current density becomes in a

harmonically time-varying field E « exp(iwt):

j = o(w\(E — T6E/6t) (2)

Where the component of j proportional to E is the conduction current, with the

coefficient of proportionality being the effective conductivity at frequency w:

a(w) = oJ + w’r’) (3)

where a, = ne’T/m* is the DC conductivity (n = electron density, m* = electron

effective mass). The component of j proportional to the time rate of change of

E, and hence cut of phase with it, is physically equivalent to the displacement

current arising from the dieloectric constanct.

In place of o and &,, the complex dielectric function e(w) of Eq. (1) may be

employed where €, contains the effects of conductivity. The Drude model gives

for it (Ordal et al., 1985) the expressions

= De wll@ + wr), (4)

Ey = ww I[w(w + w,’)). (5)

The two parameter contained therein are, in corresponding units, the plasma

frequency

W, = (4rne*/m*)'/(27c) (6a)

and the damping frequency

wo. = IO ne 7). (6b)

Using Eq. (5), the DC value of the conductivity may be obtained in terms of w,

42 SCHUMACHER, HOWELL, AND UBERALL

and w,, and shall thus be referred to (Ordal et al., 1985) as the ‘‘optical frequency’’

conductivity (units cm™'),

Gon = whamus.), (7)

to distinguish it from the standard conductivity o, as measured in DC experiments.

The assumed validity of the Drude model should then lead to an agreement of

Oop With a, which will be seen below to be generally the case, except for some

individual deviations.

Table I collects results of Ordal et al. (1985; 1987; 1988), listing the Drude

fit parameters w, and w, as well as the empirically determined values (Weaver

et al., 1981) for w,, and empirical values for w, to be discussed below. The table

also lists the DC resistivities p, and p,,, (based on the Drude model), the latter

being related to the conductivities by

Foop(cm™') = 9*10"'/(2rc[poop(Qem))]). (8)

The ratio p,/Po 18 seen to be generally close to unity, but it shows individual

differences as mentioned. We complete the table by also listing the DC dielectric

constant ¢,(0), given in the Drude model by

6(0) = 1 — w,/w,’, (9)

thus providing the values of the dielectric constant ¢, for metals. These should

be considered as approximations only, and in fact we also tabulate empirical

values of ¢,(0) following from Weaver et al. (1981) by extrapolation to w — 0.

Such an extrapolation is fraught with great uncertainties, and the large differences

with the Drude values could be due to uncertainties both in the Drude fits or in

this extrapolation. The ‘‘empirical’’ values for w, are obtained from those of

€,(O) via Eq. (9).

As Eq. (4) shows, the function ¢,(w) increases monotonically from the large

negative DC values displayed in Table I to the value of unity at infinite frequency,

passing through zero in the vicinity of the optical range. On the other hand, Eqs.

(3) and (7) show that o(w) decreases monotonically with increasing frequency.

This behavior is shown in Fig. 1 and is characteristic of plasma or collective

oscillations in a solid, resulting from the high density of electrons present in a

metal and the fact that they can act cooperatively due to the Coulomb interaction

among them.

The energy of these oscillations is large compared to the single-particle energy

because a plasma oscillation at long wavelengths involves the correlated motion

of a very large number of electrons. No single electron is greatly perturbed but,

because a large number of electrons are moving together in a coherent fashion,

the resultant energy of the collective mode is quite substantial.

PLASMON EXCITATION

(1861) ‘Te 10 JOAROM ,

(ese apni oJ Te) LpL ‘d (L861) ‘Te 19 [epso »

LOTI-SO7I “dd *(gg61) ‘Te 19 Tepio ,

porr d (C861) ‘Te 19 [eplo ,

‘Ob—-6 ‘6€-6 ‘dd “(ZL61) UOosIspuy 2 UTysIgeg ,

086'0 08°C vLT 000°09T cOe 779'0S VCV OOr IZI 00r'S6 AV

1160 9/01 08°6 00S°8 LC O9L'SE 9¢I 00S°S8 00S°67 Pea

€c0'l 88°9 vO'L 000°LT O8S 066 87 8LI 008°SZ OOr'6E pIN

oa vol OL'I 000°ST Os 006°799 eL 006 +9 009°6S qh)

9e£0 6c LI 08'S 000°6 89 OLL II S6C 006'r9 000°ZE gO

SPL? vse cc Ol 0000S 6L7 O16 S7I vl O0E 79 000'rr qPd

ISO'l 86 61 OIC = = Coot CLIT <3 00r'6S pid

pLit c6 91 6 61 000'8 009 9ETL 687 OOL'€eS 009° Ir aA

9€S°0 vr 6l cr ol 000°L e8¢ oss BSS 008'8r 00S Ir ad

7060 eve GE 000°SZ OLI OS9'rIT CIC 009°9r 008°CL qQh¥

8260 gol LOT 000°0L vil O8e 1Sz crl 007 0€ OOL'CL aoV

6Se'| 796 et 000°C OI 1788 vOL 00r'9T 007°99 eL

1820 c8°9 ces 000'Sr gs OSE IZ CIV OOL TT 007°09 VW

88r'0 £6 01 ces 000 12 VL Le L8v OOL'0I OOL'TS gM

SLL'‘0 c9'SS Ver 000'€ 99 ECS G C8E 009°€ 00E 02 qhL

deg 1° oe Sele , lida opniqg , 1idwia opniq [R10

(0) (to) 4 (Q) (ura) “ (ura) &

S[VJOUI ST JO SITIPIANSISAI pue sjUB}sUOD [edo AOJ S}Yy [edIAIdwIa pue japoul spnig—'] 2g],

4d SCHUMACHER, HOWELL, AND UBERALL

0.6um 3000A 2000A 1500A 1200A 1000A 857A _ _—s750A

25K

20K

15K

(S/m)

10K

DSS ss ——— i

0 500 1000 1500 2000 2500 3000 3500 4000

FREQUENCY (THz)

Fig. 1. Drude-fit values of and ¢ for Fe as functions of \ and frequency (THz).

Electrons, Plasmons, and Photons in Solids

Drude completely neglected the ionic cores present in solids and the fact that

electrons obey Fermi statistics. Pines provides a much more complete treatment,

including the effects of the periodic lattice, photons, electron-electron and elec-

tron-phonon interactions, and plasmons (Pines, 1964). The plasmon is the quan-

tum of plasma oscillation energy. Electron interactions in solids differ from those

in a free-electron gas, because changes in screening behavior due to the periodic

ion potential influence the plasmon spectra. Plasmons are bosons, and have a

distribution function of the characteristic boson form at finite temperatures.

Under certain conditions plasma oscillations represent normal modes of the

entire system, which means that once such oscillations are excited, they do not

decay in time. Since they maintain themselves, we would expect to find an

internal electric field E(w,) in the solid at the plasma frequency w,, due to the

oscillations in the electron density, without the presence of an external field

D(w,). By the constitutive relation, since

D(w,) = e(w,)E(w,) = 0 (10)

and E(w,) # 0, the condition for plasma oscillations is therefore

E(W,) = €\(W,) + 1€(W,) = 0. (11)

PLASMON EXCITATION 45

Table II.—Imaginary part of the dielectric function €, loss factor and half width Aw, at the (empiri-

cal) plasmon resonance frequency w,. Also, DC conductivities (Drude model)

Wo=Wp its hw, Loss Aw, om Toon

Metal (cm ') (THz) (eV) en Factor (cm7') (10®S/m) (10°8S/m)

Ti 3600 108 0.45 Aas 0.046 66 0.023 0.018

WwW 10700 321 1.35 19.80 0.051 74 0.188 0.091

Mo 11700 351 1.45 22°35 0.045 55 0.188 0.145

Ta 16400 492 2.03 7.46 0.134 110 0.076 0.104

Ag 30200 906 3.75 0.44 D215 114 0.621 0.606

Au 46600 1,400 5.78 2.90 0.345 170 0.455 0.412

Pt 48800 1,460 6.05 3.82 0.262 583 0.096 0.051

V 53700 1,610 6.66 1.87 0.535 600 0.050 0.059

Pb 59400 1,780 Tee 0.02 50.54 1175 0.048 0.050

Pd 62300 1,870 TG 1.32 0.758 279 0.095 0.260

Co 64900 1,950 8.05 2.06 0.485 684 0.172 0.058

Cu 64900 1,950 8.05 2.10 0.476 530 0.588 0.808

Ni 75800 2,270 9.40 1.83 0.546 580 0.142 0.145

Fe 85500 2,560 10.60 1.54 0.649 927 0.102 0.093

Al 121400 3,640 15.06 0.04 26.32 303 0.365 0.357

Thus, strictly speaking, plasma oscillations should exist as normal modes of the

system only if both real and imaginary parts of the dielectric constant vanish at

w,. In practice, however, plasma oscillations will exist if €.(w,) < 1 when €,(w,)

= Q. Since €, represents the damping of the plasma resonance, the condition

€>(w,) < | implies damping should be small.

More precisely, when plasma oscillations are damped, the frequency at which

€,(w) = O does not quite provide us with plasmon energy. By Eq. (4), the real

part of the dielectric function vanishes at the frequency w,, where

Because w,/w, < 1, w, is very nearly equal to w,. Values of w, in cm™', simply

approximated by the (empirical) values of w, of Table I, are given in Table II,

along with the equivalent frequency f, in THz and the corresponding photon

energy in eV for 15 metals. Table II also gives €.) = &,(w,) the value of the

imaginary part of the dielectric function at w, from the empirical data of Weaver

et al. (1981).

The principal experimental evidence for the existence of plasmons as a well-

defined excitation mode of the valence electrons in solids comes from characteris-

tic energy-loss experiments, where one observes the energy spectrum of kilovolt

electrons, either as they emerge from a thin solid film or after they are reflected

from.a solid surface (Peierls, 1956). In fact, the most familiar method of determin-

ing w, utilizes measurement of these characteristic electron energy losses, which

are proportional to the “‘loss factor’? — Im(1/e(w)), a function of width Aw, =

w, that is very sharply peaked about w, when the conditions for plasma oscilla-

46 SCHUMACHER, HOWELL, AND UBERALL

tions are fulfilled (Daniels et al., 1970). When one measures the angular distribu-

tion of the inelastically scattered electrons, one measures Im(1/e(w)) directly for

the electrons in the solid.

Whereas fast electron scattering is a longitudinal probe of the solid, in which

the electron gas responds to a time-varying longitudinal field, measurements of

the optical reflectivity of a solid constitute a transverse probe of the solid, because

the electromagnetic wave couples directly to the transverse current-density fluc-

tuations of the electrons. Therefore, the dielectric constant is a tensorial quantity

because, just as the response of the electron gas to a time-varying longitudinal

field defines a longitudinal dieletric constant, the system response to an external

electromagnetic field defines a transverse dielectric constant.

Pines (1964) presents a comparison of the optical values of plasmon energies

for the alkali metals, based on reflection experiments, with the values measured

in electron energy-loss experiments and remarks that agreement between the two

methods is quite good. Experiments have been made to compare the electron

energy loss function with that calculated from known optical constants. General

agreement has been found in the detailed structure of the two loss functions

(Daniels et al., 1970), so no experimental evidence exists to date for a difference

between the two dielectric functions.

Values of the loss factor, —Im(1/e,(w)), from the empirical tabulation of

Weaver et al. (1981) are given in Table II and exhibited for 14 metals (except

for Pb where only Drude values exist) as a function of f, and \ in Fig. 2. Also

entered in Table II, for the purpose of later discussion, are the values of o, and

Oop In units of 10° S/m (1 Siemens = S = (1"'), obtained from the Drude values

of Table I via Eq. (8).

Scattering of Electromagnetic Waves

The behavior of the dielectric constant in a region of appreciable conductivity

has a profound effect on the scattering of electromagnetic waves from metals. If

one considers scattering from a simple model of a solid with « = 1 and o equal

to its constant DC value, the reflection coefficient gradually decreases from unity

with increasing frequency, as shown in Fig. 3, in accordance with the familiar

Hagen-Rubens relation discussed by Born and Wolf (1975). Schumacher (1987)

has provided an explanation of the scaling laws that require o to increase with

increasing frequency for the reflection coefficient to remain constant.

Even though the Drude o decreases with increasing frequency in accordance

with Fig. 1, which by itself would lead to a reflection coefficient decreasing from

unity with increasing frequency more rapidly than the simple solid model shown

PLASMON EXCITATION 47

0.6um 3000A 2000A 1500A 1200A 1000A 850A

Loss Factor -Ilm(1/e)

on!

0 500 1000 1500 2000 2500 3000 3500

FREQUENCY (THz)

Fig. 2. Values of the loss factor = —Im(1/e) for 14 metals as a function of f, and X.

in Fig. 3, the strongly negative Drude (or empirical) ¢, causes a large impedance

discontinuity, which keeps the reflection coefficient very nearly equal to unity

until it drops very rapidly at the plasma resonance frequency to values far below

those of the simple solid model.

This finding has very significant practical consequences. Usually infinite o is

assumed for the metals used in radar calculations, even though one expects

eventual decreases in the reflection coefficient of physical scale models caused

by the polarization not being able to keep up with exciting electromagnetic fields

as the measurement frequency increases. In fact, scaling laws require that o

48 SCHUMACHER, HOWELL, AND UBERALL

30cm 3cm 3mm 300um = 30um 3m 3000A 300A 30A

1.0

0.8

0.6

0.4

0.2

0 S

1GHz 10GHz 100GHz 1THz 10THz 100THz Uiriae 1OPHz 100PHz

FREQUENCY

Fig. 3. Values of R, the reflection coefficient for power, as a function of frequency and 2 for the Drude

theory and for a simple solid modele with the DC value of o and ¢ = | for the case of Fe.

increase to very large unphysical values in the IR range for the reflection coeffi-

cient to remain constant (Schumacher, 1987).

However, our theoretical study of plasmon excitation in good conductors yields

the same reflection effect as a o actually increasing with frequency in the physical

scale modeling range, but finally decreasing to zero as photon energies reach the

electron volt range, in accordance with Fig. 3. Thus, not only is the assumption

of infinite o fully justified in physical scale model measurements, but also there

iS a new understanding of the reflection properties of metallic scale models.

The above results also have implications for radar absorbing materials. An

ideal absorber is one that allows incident electromagnetic energy to enter without

reflection and then to rapidly attenuate over a short distance. However, it does

not appear possible to satisfy these two conditions simultaneously. Undoubtedly,

attenuation is rapid if both electric and magnetic losses are high. However,

these losses are related to imaginary components of the dielectric and magnetic

polarization, both of which result in reflection.

DC Conductivity vs. Resonance Frequency: Good and Poor Conductors

In Fig. 4, we present a graph of o, (crosses) and o,,, (circles), from Table II,

plotted vs. the resonance frequencies of the 15 metals considered here. A trend

PLASMON EXCITATION 49

0.6um 3000A 2000A 1500A 1200A 1000A

6 (108 S/m)

0 500 1000 1500 2000 2500 3000 3500

FREQUENCY (THz)

Fig. 4. Values of DC conductivities o, and o,,, for 15 metals from the Drude model, at the emperical

resonancs frequencies.

of increasing DC conductivity with rising frequency is readily apparent here.

Such a trend can already be discerned from Eqs. (7) and (12), and the deviations

of 0,,, from a smooth curve are caused by the scattering of the w, values as seen

in Table I.

Here, the opportunity arises of extending our present study from solely good

conductors to also including some poor conductors, such as carbon (graphite)

and two conducting polymers, for which the resonance frequencies are believed

to be known. In Table III, we list the values of f,(THz) for the conducting polymer

poly[bis-(3-ethynylaniline)-N,N’-(1,4-dimethylydene)] pyrolyzed at 500° C, the

same sample pyrolyzed at 700° C, and for graphite. These resonance frequencies

were determined by use of a Hewlett-Packard 4191A impedance analyzer. The

DC conductivities were measured for the polymer by use of a Simpson 260

multimeter and for carbon, o, was taken from Babiskin & Anderson (1972). Also

included in Table III are the first metal of Table II (Ti), the last metal (Al), and

50 SCHUMACHER, HOWELL, AND UBERALL

that with the highest conductivity (Cu), for purposes of comparison. The same

data are shown graphically in Fig. 5, on a double-logarithmic plot. It is seen here

that by including poor conductors, the (semi-empirical) downward trend of DC

conductivity with decreasing resonance frequency f,, noted before for metals in

Fig. 4, has now been (empirically) extended further for poor conductors by many

decades. It would, of course, be possible to try describing these poor conductors

by the Drude model also. In that case, one would find from Eq. (7) the values

of the damping frequency w, = 6.94 cm™! (C), 7.4 X 107’ cm™! (Pyr. 700°), and —

6.7 X 10°* (Pyr. 500°); but this would (except for C) probably amount to pushing

the Drude model too far, in view of the large jumps between Pyr. 500° and Pyr.

700°. Nevertheless, the continuing trend of DC conductivities over many decades

for good and poor conductors remains an interesting feature.

Conclusions

Approximate numbers for the static values of the dielctric constant, ¢, = ¢,(0),

of 15 metals have been obtained from experimental data, both via Drude fits or

purely empirically. All these DC values are shown to be negative and very large.

Higher frequency values of €, increase monotonically with increasing frequency

passing through zero in the optical range between wavelengths of 1000 and 6000

Angstréms. This frequency dependence of ¢ is obtained from values of plasma

and damping frequencies reported by Ordal et al. (1985; 1987; 1988), which were

derived from Drude model fits to the Kramers-Kronig analysis of a combination of |

their new infrared measurements of the optical constants of 15 metals and other

previous measurements. It is also obtained empirically (Weaver et al., 1981) from

earlier data.

Ordal’s Drude-model fits predict values of the conductivity, a, that decrease

monotonically with increasing frequency but are in good agreement with mea-

sured static values. This behavior is characteristic of plasma or collective oscilla-

tions in a solid and is interpreted as plasmon excitation. Very sharply peaked loss

factors are tabulated for 14 metals. This behavior of € causes a large impedance

Table I1I.—Resonance frequencies of several poor and good conductors

Substance f, (THz) a, (10° S/m) Ton (10° S/m)

Pyr.500°C 6X 107° bo 10m =

Pyr.700°C PX 10% P< 107 ~

G 16.5 O727 <x 10s: -

Ti 108.0 0.0232 0.018

Cu 1,950.0 0.588 0.808

Al 3,640.0 0.365 0.357

PLASMON EXCITATION 51

Graphite

Pyr. 700°

DC conductivity (S/m)

Pyr. 500°

1 10 100 1000 1E4 1E5 166 1E7 1E8 1E9 1E10 1E11

(GHz) (THz) (1000 THz)

fy (MHz)

Fig. 5. Values of DC conductivities o, for three poor conductors and three metals.

discontinuity, which keeps the reflection coefficient very nearly equal to unity

until it drops very rapidly at the plasma resonance frequency to values ap-

proaching zero as photon energies reach the electron volt range. This behavior

is compared to scattering from a simple model of a solid with « = 1 and o equal

to its constant DC value, which yields a reflection coefficient decreasing more

slowly with increasing frequency, in accordance with the familiar Hagen-Rubens

relation.

A trend of the DC conductivity to increase with rising plasma resonance

frequency is noted for the 15 metals under consideration. This trend, which is

52 SCHUMACHER, HOWELL, AND UBERALL

semiempirical on the basis of the Drude model, is empirically extrapolated down-

wards by including a few poor conductors (carbon and conducting polymers),

and is seen to persist here over many decades.

Acknowledgement

The first author (C.R.S.) is thankful to Drs. W. E. Lukens, B. E. Douglas,

E. C. Fisher and to G. A. Wacker and J. R. Crisci for support and encouragement.

In addition he thanks Drs. P. O. Cervenka, A. J. Stoyanov, and Y. J. Stoyanov

for helpful discussions. The last author (H. U.) wishes to thank the American

Society for Engineering Education and the David Taylor Research Center Annap-

olis (now NSWC, Annapolis Detachment) for their award of a Distinguished

Summer Faculty Fellowship, under the auspices of which part of his contribution

to this work was made. He also acknowledges the hospitality of Prof. Staffan

Strom, Head of the Department of Electromagnetic Theory, Royal Institute of

Technology, Stockholm, Sweden, where part of his work on the present paper

was carried out.

References

Babiskin, J., and Anderson, J. R., Eds. (1972). American Institute of Physics Handbook. McGraw Hill: New

York, NY.

Bohren, C. F., and Huffman, D. R. (1983). Absorption and Scattering of Light by Small Particles. Wiley:

New York, NY.

Born, M., and Wolf, E. (1975). Principles of Optics, 5th ed., Chapter XIII. Pergamon Press: Oxford, England.

Daniels, J. V., Festenberg, C., Raether, H., and Zeppenfeld, K. (1970). Optical Constants of Solids by

Electron Spectroscopy, in G. Hohler (Ed.) Springer Tracts in Modern Physics 54:77—135. Springer, Berlin,

Germany.

Ordal, M. A., Bell, R. J., Alexander, R. W., Jr., Long, L. L., and Querry, M. R. (1985). Optical properties

of fourteen metals in the infrared and far infrared: Al, Co, Cu, Au, Fe, Pb, Mo, Ni, Pd, Pt, Ag, Ti, V, and

W. Appl. Opt., 24:4493-—4499.

Ordal, M. A., Bell, R. J., Alexander, R. W., Jr., Long, L. L., and Querry, M. R. (1987). Optical Properties

of Au, Ni, and Pb at submillimeter wavelengths. Appl. Opt., 26:744—752.

Ordal, M. A., Bell, R. J., Alexander, R. W., Jr., Newquist, L. A., and Querry, M. R. (1988). Optical

properties of Al, Fe, Ti, Ta, W and Mo at submillimeter wavelengths. Appl. Opt., 27:1203—1209.

Peierls, R. E. (1956). Quantum Theory of Solids. Oxford University Press: London, England.

Pines, D. (1964). Elementary Expectations in Solids. W. A. Benjamin: New York, NY.

Schumacher, C. R. (1987). Electrodynamic similitude and physical scale modeling of nondispersive targets.

J. Appl. Phys., 62:2616—2625.

Slater, J. C., and Frank, N. H. (1947). Electromagnetism. McGraw Hill: New York, NY.

Sommerfeld, A., and Bethe, H. A. (1933). Elektronentheorie der Metalle, in Handbuch der Physik, 2. Auflage,

Band 24, Teil 2. Springer: Berlin, Germany.

Weaver, J. H., Krafka, C., Lynch, D. W., and Koch, E. E. (1981). Physics Data: Optical Properties of

Metals, Part I and Part II. Fachinformationszentrum Karlsruhe, Germany.

DELEGATES TO THE WASHINGTON ACADEMY OF SCIENCES,

REPRESENTING THE LOCAL AFFILIATED SOCIETIES

Pilocopuical Society OF WasiinetOM, 62... 0 5.056 odes eee kee ee eek oie Thomas R. Lettieri

Paimemomcical SOclety OF WaSMNSION . 0.5.5.2. fi eg nk eae es dew lee eels Jean K. Boek

Pemacical SOCKIV OL Washimeton: 220.2. 626 ye ee ee eee ce eee ee ale Kristian Fauchald

eased SOCICty, GE VV ASENECON 0 cis he 2 eae e oe eee alee le ee cele oie bln oe ee ee Elise A. B. Brown

Paromiolacical Society of Washingion ....... 26. ..6 6606. eee eee eee F. Christian Thompson

MMA MEUPOPnAMC SOCILY 20.6302 2 6. os cae eagle eee dened anes Stanley G. Leftwich

PE ECCS OL. WASMMNGION. 2945. fot be be bese ee bee ee eee des beh VACANT

ime aecicty Or the Distticiot Columbia 3.2)... 5.0). ade be kee John P. Utz

emueesncicty OF Washineton, DC . 2... yoke eee ee eb ee eee dae VACANT

PE GCICTy, OF WASHINGTON . . (02). .2 oe hese ee hele ln Le ee be le Muriel Poston

eee ot American Foresters, Washington Section ..................6:-2-- Eldon W. Ross

DsenEMESOCIcIy OL EMCINCCES 2 24.50 ee ee ew es ee breed eee adele bale ¥ Alvin Reiner

Institute of Electrical and Electronics Engineers, Washington Section ........ George Abraham

American Society of Mechanical Engineers, Washington Section .......... Daniel J. Vavrick

inmmmnarical socicty Of Washington, .....-.. 2-2. .5.- 2-6. c eee eee eee nee VACANT

Amenacan Society for Microbiology, Washington Branch ..........:.....-...--.... Ben Tall

Society of American Military Engineers, Washington Post ............... William A. Stanley

American Society of Civil Engineers, National Capital Section ................... VACANT

Society for Experimental Biology and Medicine, DC Section ............. Cyrus R. Creveling

eameenaional, Washington Chapter .....-....-.-.-----26---- ene eee eee Richard Ricker

American Association of Dental Research, Washington Section ............ J. Terrell Hoffeld

_ American Institute of Aeronautics and Astronautics, National Capital

as Serge De ee hala s ite ele Shed so Po Bie Shin dyer baka al aoe Reginald C. Smith

mimemesa) Meteorological Society, DC Chapter ..............-.6.-++..-.-- A. James Wagner

memeeiemee Society Of Washington ........2-2.-2225-00 45 e eee ee eee eee To be determined

Acoustical Society of America, Washington Chapter ...................... Richard K. Cook

Ppmiemean Nuclear Society, Washington Section .............-.-222++5++++2---- Kamal Araj

Institute of Food Technologists, Washington Section ...................2.0-- Roy E. Martin

American Ceramic Society, Baltimore-Washington Section ................. Curtis A. Martin

Se RENAL EAMES US OCACUY 8 ona ohecoolh | boi ic dg wisela ae onic Whe nd oe eee eeh elas ee ales Regis Conrad

Mianaeseniiistory Of Science Club «.............-6 2.020000 0be eee eee Albert G. Gluckman

American Association of Physics Teachers, Chesapeake Section ............ Robert A. Morse

Optical Society of America, National Capital Section .................... William R. Graver

American Society of Plant Physiologists, Washington Area Section ........... Steven J. Britz

Washington Operations Research/Management Science Council ............... John G. Honig

instaumicnt Soecicty of America, Washington Section ...:................2...---.- VACANT

American Institute of Mining, Metallurgical and Petroleum Engineers,

Seem PUE SCC. 0) els iht oo ise his gine Boca dda ede ete bes Anthany Commarota Jr.

Mae AGIA ASIZONOMICTS 2 (6.642220 052 022 ecu bee ee be seneees Robert H. McCracken

Mathematics Association of America, MD-DC-VA Section ................ Sharon K. Hauge

Pismicro, Columbia Institute of Chemists’ ».............0. 022.000 0600- William E. Hanford

District of Columbia Psychological Association ...................5.+.- Marilyn Sue Bogner

imran e ant Kechnology Group... 20. os ee ee es ce ek ee ee eee Lloyd M. Smith

American Phytopathological Society, Potomac Division ................... Kenneth L. Deahl

International Society for the System Science, Metropolitan Washington

LUTE ES Ie ROTO De) ODE MESS Se PI LNCAP, Ot David B. Keever

Huma naciors Society, Potomac Chapter .........0.. 062002020005 ce ee: Thomas B. Malone

fineiedn Fisheries Society, Potomac Chapter .. 22... 2.226 fee eae eee Dennis R. Lassuy

Association for Science, Technology and Innovation ................... Clifford E. Lanham

an i SOLIOIOPICAL SOCICUY 6.50! jails vo sole fea le ale bende se Ronald W. Manderscheid

Institute of Electrical and Electronics Engineers, Northern Virginia

Bic MNO DE Mae terete Note RAY hs ALENT aby, Lea ae Re A coh Td Ya Blanchard D. Smith

Association for Computing Machinery, Washington Chapter ............. Charles E. Youman

Pecan erm Si AtIStical SOEIeIy, ys) eI Sho eee eel David Crosby

Society of Manufacturing Engineers, Washington, DC Chapter .............. James E. Spates

Institute of Industrial Engineers, National Capital Chapter ................ Neal F. Schmeidler

Delegates continue to represent their societies until new appointments are made.

Washington Academy of Sciences 2nd Class Postage Paid

2100 Foxhall Road, NW at Washington, DC

Washington, DC 20007-1199 vis and additional mailing offices.

Return Postage Guaranteed

es (7

A tt VOLUME 84

Number 2

Jour nal of the June, 1996

WASHINGTON

ACADEMY .. SCIENCES

ISSN 0043-0439

Issued Quarterly

at Washington, D.C.

CONTENTS

Articles:

HAL W. HENDRICK, ‘‘Cognitive Complexity, Conceptual Systems, and

|e SUPE RVIDT EG alt asl SHAR EWA Aa AME ata eG UO AS RUA eNO Ban al EU Ea I

JEANNE L. WEAVER, CLINT A. BOWERS & BEN B. MORGAN, JR.,

“*The Effect of Individual Differences on Anxiety and Team Performance’”’ ..

MICHAEL G. RUMSEY, NORMAN G. PETERSON, SCOTT H. OPPLER &

JOHN P. CAMPBELL, ‘“What’s Happened since Project A: The Future

Garce mR GiCem Nhe igs Maia aoe rue te anee ume Be anor 40% lana Un eakamaee elk

JANET J. TURNAGE, ROBERT S. KENNEDY & NORMAN E. LANE,

““Gender Differences in Human Abilities: Can Practice Moderate Results?’ ..

111

Washington Academy of Sciences

Founded in 1898

EXECUTIVE COMMITTEE

President

John S. Toll

President-Elect

Rita Colwell

Secretary

Louis F. Libelo

Treasurer

John G. Honig

Past President

Rev. Frank R. Haig, S. J.

Vice President, Membership Affairs

Cyrus R. Creveling

Vice President, Administrative Affairs

Benjamin Alexander

Vice President, Junior Academy Affairs

Marylin B. Krupsaw

Vice President, Affiliate Affairs

Neil F. Schmeidler

Board of Managers

Marilyn S. Bogner

Elise A. B. Brown

Norman Doctor

Herbert H. Fockler

John H. Proctor

Nina M. Roscher

REPRESENTATIVES FROM

AFFILIATED SOCIETIES

Delegates are listed on inside rear cover

of each Journal.

ACADEMY OFFICE

2100 Foxhall Road, N.W.

Washington, D.C. 20007

Phone: (202) 337-2077

EDITORIAL BOARD

Editor:

Bruce F. Hill, Mount Vernon College

Associate Editors:

Milton P. Eisner, Mount Vernon Col-

lege

Albert G. Gluckman, University of

Maryland

Marc Rothenberg, Smithsonian Institu-

tion

Marc M. Sebrechts, Catholic University

of America

Edward J. Wegman, George Mason

University

The Journal

This journal, the official organ of the Washing-

ton Academy of Sciences, publishes original

scientific research, critical reviews, historical

articles, proceedings of scholarly meetings of

its affiliated societies, reports of the Academy,

and other items of interest to Academy mem-

bers. The Journal appears four times a year

(March, June, September, and December). The

December issue contains a directory of the cur-

rent membership of the Academy.

Subscription Rates

Members, fellows, and life members in good

standing receive the Journal without charge.

Subscriptions are available on a calendar year

basis, payable in advance. Payment must be

made in U.S. currency at the following rates:

US. and Canada’ ou..ce 522 See $25.00

Other counties”... 5.6%.) 5.65 ee 30.00:

Single copies, when available ...... 10.00

Claims for Missing Issues

Claims will not be allowed if received more

than 60 days after the day of mailing plus time

normally required for postal delivery and

claim. No claims will be allowed because of

failure to notify the Academy of a change of

address.

Notification of Change of Address

Address changes should be sent promptly to

the Academy Office. Such notification should

show both old and new addresses and zip

codes.

POSTMASTER: Send address changes to

Washington Academy of Sciences, 2100 Fox-

hall Road, N.W. Washington, DC 20007-1199.

Journal of the Washington Academy of Sciences (ISSN 0043-0439)

Published quarterly in March, June, September, and December of each year by the Washington

Academy of Sciences, 2100 Foxhall Road, N.W., Washington, DC, 20007-1199. Second

Class postage paid at Washington, DC and additional mailing offices.

Journal of the Washington Academy of Sciences,

Volume 84, Number 2, 53-67, June 1996

Cognitive Complexity, Conceptual

Systems, and Behavior

Hal W. Hendrick

Institute of Safety and Systems Management, University of Southern California,

& Error Analysis, Inc.

Received April 30, 1996

ABSTRACT

Historical findings concerning the nature of the higher-order structural personality dimen-

sion of cognitive complexity and related conceptual systems, including systematic effects on

moral reasoning and behavioral style, are summarized. The author’s own research on early

trainer and traumatic event effects on one’s complexity level; and the effect of individual

differences in cognitive complexity on creativity, leader behavior and influence, interpersonal

and self perception, group task performance, and matching individual and position complexity

are reviewed. The author’s use of the cognitive complexity dimension in organizational

assessment and design is described.

During the past three decades, there has been a growing body of research,

largely outside the fields of engineering psychology and ergonomics, which I

believe to be relevant to our understanding of individual differences in human

performance, including values, attitudes, motives, creativity, and stylistic leader-

ship and work behavior. This research concerns the higher-order structural person-

ality dimension of cognitive complexity, or concreteness-abstractness of thinking.

The fact that the high technology societies of the world appear to be undergoing

significant, yet understandable age-related demographic changes in this higher-

order personality dimension further enhances its value to us in understanding

human performance.

This area of structured personality research had it origins in the classical work

of Piaget (1948) on child development. Among others, two groups of researchers

of particular relevance to this paper have extended the study of concreteness-

abstractness into the adult range. These are O.J. Harvey, D. E. Hunt, and H.

Schroder (1961) in the area of conceptual functioning or conceptual systems and

behavior, and Lawrence Kohlberg and his colleagues at Harvard with respect to

the structure of moral reasoning (1969).

54 HENDRICK

Cognitive complexity, or concreteness-abstractness is reported to have two

major structural aspects, differentiation and integration (Harvey, et al., 1961).

Operationally, differentiation can be defined as the number of dimensions ex-

tracted from a set of data, and integration as the number of interconnections

between rules for combining structured data (Barrif and Lusk, 1971). A concrete

cognitive style is one in which relatively little differentiation is used in structuring

concepts. Experiential data are categorized by the individual into relatively few

conceptual dimensions, and within concepts, there exists relatively few categories

or shades of grey. In the extreme, a concept is divided into just two categories,

characteristic of either/or, black/white, absolutist thinking. In addition, concrete

thinkers are relatively poor at integrating conceptual data in assessing complex

problems and developing unique or creative, insightful solutions. In contrast,

cognitively complex persons tend to demonstrate high differentiation and effective

integration in their conceptualizing (Harvey, 1966; Harvey, et al., 1961).

Although all persons tend to become more abstract over time, our development

curve tends to flatten in early adulthood. At what degree of cognitive complexity

this plateauing occurs appears to depend primarily on two factors: How open

one is to learning from one’s experience, and how much exposure one has had

to diversity. We all start out in life with very limited exposure to diversity, and

thus have few conceptual categories in which to place experiential information,

and few rules and combinations of rules to integrate our experiential information.

AS we gain new experiences, and if we are open to learning from those experi-

ences, we develop new conceptual categories and more rules and rule combina-

tions for integrating our conceptual data (Harvey, 1966). How open we are to

learning from our experiences depends on the nature of one’s early training

environment at home and school (Blatt, 1971). In particular, the nature of the

*“trainer’’ role appears to be critical. In general, the more absolutist and authoritar-

ian the parent, teacher or other trainer, the greater is the likelihood that active

exposure was inhibited, and that the child will plateau at a relatively concrete

level of conceptual functioning (Harvey, et al., 1961). The more relativistic and

less authoritarian the training, the more the trainer encourages the child to think

things through and draw personal conclusions, and the more the trainer instills

a strong positive sense of self worth in the child, the more abstract or cognitively

complex the child will become in his or her conceptual functioning as an adult.

Cognitive Complexity and Moral Reasoning

When Lawrence Kohlberg set out to study moral reasoning, he was looking for

structures, forms and relationships that are common to all societies and languages

(Kohlberg, 1969). Over the years, he gradually elaborated a topological scheme

COGNITIVE COMPLEXITY 55

for describing the general structure and forms of moral reasoning that he and his

colleagues had found throughout the world. Of particular importance was the

finding that moral reasoning could be defined independently of the specific moral

content of moral decisions and actions. Kohlberg found that their were three

different developmental levels of moral reasoning that lie along the concreteness-

abstractness dimension. At any given point in one’s life, the rationale underlying

a large majority of one’s moral decisions will be at a single level. Each level of

moral reasoning appears to have two variations or stages (Kohlberg, 1969).

Level 1: Preconventional Moral Reasoning

This level is characteristic of highly concrete functioning. It is oriented around

concepts of good and bad which are interpreted in terms of physical concepts

(punishment and reward) or in terms of the physical power of those who make

the rules (i.e., might is right). Within the preconventional level are two discernable

stages. Stage 1 is an orientation toward punishment and unquestioning deference

to superior power. Here a good decision is one which leads to avoidance of

punishment. Stage 2 is an orientation toward personal need satisfaction and,

occasionally, the needs of others. Elements of fairness, sharing and reciprocity

are present, but it is a “‘you scratch my back and I’ll scratch yours’’ kind, rather

than a reciprocity based on loyalty or justice. This stage sometimes is referred

to as the morality of the marketplace.

Level 2: Conventional Moral Reasoning

This level can be described as conformist in the sense of maintaining the

expectations and rules of one’s family, group, culture or nation. The maintenance

of existing ways is perceived as a valuable end in and of itself. Stage 3 is referred

to as the good-boy, good girl orientation. Here, the goodness of an action is

based on whether it pleases or helps others and is approved by them. Stage 4 is

an orientation towards authority, fixed rules, regulations, and the maintenance of

the existing social order. The goodness or rightness of behavior is judged on the

extent to which a person is doing one’s duty, showing respect for authority, and

maintaining the existing order as an end in itself; it is a reliance on external

sources for ones moral decisions.

Level 3: Postconventional Moral Reasoning

The postconventional level reflects cognitively complex or abstract functioning.

It is characterized by autonomous, universal moral principles, which provide an

internalized and more principled moral basis for one’s decisions. Stage 5 is a

social contract orientation with legalistic overtones. The rightness of an action

tends to be evaluated in terms of respecting the general individual rights of

56 HENDRICK

persons, and the standards which have been critically examined and agreed upon

by the whole society. Stage five is the ‘‘official’’ morality of the American

government and the U.S. Constitution. Stage 6 is an orientation of universal

moral and ethical principles. Morality is not defined by a given society, but by

one’s conscience in accordance with self-determined moral principles. These are

not concrete rules like the Ten Commandments; rather, they are broad and abstract

and include universal principles of justice and the reciprocity and equality of

human rights.

Cognitive Complexity and Conceptual Systems

Based on their original research, Harvey, et al. (1961) concluded that there

appear to be at least four fundamentally different ways in which people organize

or structure and integrate their experiences of reality. Further, these four ways

appear to lie along an invariant developmental continuum, with the underlying

dimension being concreteness-abstractness or cognitive complexity. As persons

develop greater differentiation and integration in their conceptual functioning,

they may move on to a new, more cognitively complex way of viewing reality.

These four systematic ways or stages, in order of their developmental occurrence,

are labeled simply as conceptual systems 1, 2, 3, and 4. Because a given person’s

level of cognitive complexity can fall at any point along the continuum, it is

possible to function primarily out of one conceptual orientation but secondarily

out of another. While the four conceptual systems represent points along the

cognitive complexity continuum, the specific location of those points can vary

somewhat from person to person.

In their initial research, Harvey, et al.(1961) identified the following sets of

characteristics associated with each conceptual level of functioning. Subsequent

research, including that by the author, consistently has confirmed these findings.

System I Functioning: Conventional (Conformist) Thinking and Behavior

Regardless of culture or nationality, all persons start out with a highly concrete

System 1 orientation toward reality. As persons gain experience, they become

more abstract in their conceptual functioning, but still may maintain a System 1

perspective. In comparison with persons with other, more abstract, conceptual

orientations, System 1 persons are characterized by conventional thinking and

behavior. If highly concrete, the majority of their moral decisions are at Kohl-

berg’s Preconventional Level. If somewhat more abstract, their moral reasoning

tends to be at the Conventional Level; and they tend to rely on rules, regulations,

tradition, and other external sources of ‘‘authority’’ as the basis for their decisions.

COGNITIVE COMPLEXITY 57

In comparison with more abstract functioning individuals, the research consis-

tently has shown System 1 persons to have a relatively high need for (a) structure

and order and (b) simplicity and consistency, to be relatively (c) authoritarian,

(d) absolutist, (e) closed in their belief systems, (f) ethnocentric, (g) paternalistic

(h) personally rigid, (1) have a low tolerance for ambiguity, (j) highly accepting

of prevailing rules, norms, and social roles - and to see them as relatively static

and unchanging, and (k) to have a high belief in external fate control.

Research, to date, suggests that approximately 60% of the American adult

population is operating primarily from a System 1 perspective; but that there are

systematic differences by age group. Among those who are in their late 60’s or

older, approximately 80% may be operating at the System 1 level. Among those

in their 40’s and younger, less than half are operating from a System 1 orientation.

Those in their 50’s and early to mid 60’s represent the national average. Limited

research in other industrialized countries suggests a similar pattern.

Although age related, these differences are not age caused. Rather, other fac-

tors, such as generational differences in child rearing patterns, addressed later,

and amount of childhood and adolescence exposure to diversity through such

things as media, education, travel, etc. appear to be the most important factors.

In fact, because of experience, those in their late 60’s and older are more cogni-

tively abstract now than they were when they were young adults.

System 2 Functioning: General Negativism

As persons become more abstract, the System | conception of reality eventually

may “‘break down’’. These individuals react by becoming focused on, and sensi-

tized to what is ‘‘wrong’’ with the ““system’’ - its institutions and persons who

exercise authority and restraint over their lives. From a developmental standpoint,

the individual appears to learn more about oneself as distinct from the generalized

cultural standards which had been applied to both self and others during System

1 functioning (Hunt, 1966). In their moral decision-making, System 2 persons

often seem in a kind of psychological vacuum. They tend to see the external

norms, which heretofore they had relied upon, as having let them down and,

thus, no longer reliable. As yet, they have not replaced this external basis for their

decisions with an internalized, principled basis, characteristic of Postconventional

moral reasoning. About all System 2 persons can do is react in a distrustful,

negative manner; which Kohlberg (1969) describes as a kind of Stage 2 hedonistic

relativism which is confused with more principled moral reasoning.

Like System 1 individuals, but to a lesser extent, System 2 conceptualizers

tend to have a high need for structure and order and simplicity and consistency,

be absolutist, closed minded, not highly creative, and personally rigid. Unlike

System 1 persons, they tend not to be highly authoritarian and to reject the

58 HENDRICK

prevailing rules, norms, and social order, and to advocate change. Often, move-

ment into System 2 conceptualizing occurs during the latter high school and

college age years. Behaviorally, persons may have either an ‘‘approach’’ or

‘avoidance’ reaction: They may suddenly become campus left-wing activists

or, alternately, simply ‘‘drop out’’ from the society which has disappointed them.

In either case, it is likely to be an indiscriminate ‘‘throw the baby out with the

bathwater’’ reaction. If asked what they believe should replace the existing social

order, they often will advocate some kind of anarchy - such as a simplistic belief

of ‘‘let everyone do their own thing’’.

For most persons who enter the System 2 stage, it is a reactionary, transitory

one - perhaps a few months to several years in duration. As they gain further

experience and become more abstract, they move on to a System 3 orientation.

Although as many as 10% of the teen age population may be in this stage at any

given time, only several percent of adults plateau at this level. Those that do

provide a source for the leadership of radical left wing organizations - groups

that have high sounding causes but inhumane means of accomplishing them, such

as terrorism. |

System 3 Functioning: The World is People

The System 2 breaking away from the norm, and learning about how one is

distinctly oneself, provides the basis for eventually empathically understanding

and accepting differences - from both oneself and the norm - in others (Hunt,

1966). With development of this more abstract realization about others, the indi-

vidual moves into System 3 conceptualizing. As an overview, for System 3

functioning persons the world is people. Instead of seeing differences in values,

religion, race, lifestyles, beliefs, etc. as deviant or ‘‘less than’’, as do more

concrete functioning persons, System 3 individuals tend to value these differences

as enriching their personal lives and the human condition. With this increase in

empathy comes a shift to Postconventional moral reasoning.

In marked contrast to System 1 individuals, System 3 functioning persons

tend to demonstrate a low need for structure and order, and for simplicity and

consistency, and often will express a preference for complexity and change. They

tend to have a high tolerance for ambiguity, low absolutism, low ethnocentrism,

and an openness of beliefs - in fact, they expect their beliefs to change with

increased experience. System 3 persons tend to be moderately authoritarian, but

do not hold authority figures in awe, and expect to be questioned when they are

in positions of authority. Rules, regulations, and procedures are accepted as

useful, but also are seen as needing review and, sometimes, modification as things

change to ensure that they remain functional. They tend to have both a high

need for people and for helping others. They thus tend to be ‘‘joiners’’ and are

COGNITIVE COMPLEXITY 59

empathically concerned with the human condition and factors that affect the

quality of life.

In the United States, approximately 25% of the adult population operates from

a System 3 orientation; but again, their is a systematic age relationship. Only

10% to 15% of those persons in their late 60’s and older appear to be operating

from a true System 3 orientation. Among those in their mid 20’s to late 40’s,

approximately one-third appear to be functioning at a System 3 level. Again,

those in their 50’s and early to middle 60’s represent the average. Based on

limited research, this same pattern appears to characterize other industrialized

countries.

System 4 Functioning: Autonomous, Creative Behavior; Conceptual Maturity

The major developmental task at the fourth conceptual level is the integration

of standards which apply to both self and others. This integration enables the

individual to understand both self and others as occupying different positions on

the same transcendent dimension, rather than seeing self and others simply as

being on different standards. In accomplishing this integrating task, the individual

develops greater autonomy in thought and action (Hunt, 1966).

To an even greater extent than System 3 thinkers, System 4 conceptualizers

rely on Postconventional moral reasoning, have a low need for structure and

order and simplicity and consistency, are relativistic rather than absolutist, open

minded, creative, flexible, and have a high tolerance for ambiguity. Like System

3 persons, System 4 individuals are people-oriented, but are not highly people

dependent - they thus tend not to be joiners; and will be very open and direct in

expressing their views, even when they may be unpopular. Although others may

have a high perceived self-worth, it seems to be universal among System 4

persons - a precondition for being able to function at the System 4 level.

In terms of their empirically identified characteristics, System 4 individuals

appear to be the same persons which, through very different research approaches

and models, Maslow identified as true self actualizers, and Carl Rogers as fully

functioning persons. All three approaches have identified approximately 10% of

the adult population as falling into this group (Hendrick, 1981).

Characteristics Unrelated to Conceptual Systems

It should be noted that several important characteristics that might seem highly

related to cognitive complexity and conceptual stage are not. First, when educa-

tion level is held constant, only a weak correlation is found between abstractness

of functioning and general intelligence (Harvey, et. al, 1961). Some of the most

brilliant persons from all walks of life appear to have been, and are System 1

60 HENDRICK

functioning individuals. Secondly, conceptual functioning does not appear related

to generosity, friendliness, or numerous other valued personality characteristics.

Other Characteristics Related to Abstractness and Conceptual Systems

More cognitively complex functioning has been found to be related to (a)

completeness and effectiveness of cue utilization, (b) readiness and ability of

persons to relinquish previous assumptions or approaches and change their set

in order to complete tasks, (c) use of more novel, yet appropriate responses to

problems, and (d) value differences. With respect to values, System 1 persons

score higher than more abstract functioning individuals on Scott’s Scale of Values

for self-control, honesty, kindness, loyalty, religiousness, and the desire for power

and influence; System 2 persons value self-control, honesty, kindness, loyalty

and religiousness less than all others, and creativity and independence somewhat

higher than System | persons; System 3 individuals score relatively low on their

valuing of self-control and independence, and as high as System 1 persons on

kindness, and intermediate on the other values; System 4 conceptualizers value

creativity and independence highly, give low value to self-control and reli-

giousness, and are intermediate on the other four dimensions. (Davis, 1966).

Contributions to Our Understanding of Cognitive Complexity and Behavior by

the Author

Cogntive Complexity Level and Childrearing Patterns

Harvey, et al. (1961) proposed that reaching a plateau at a particular stage of

conceptual functioning is related to exposure to a particular dominant trainer

pattern during one’s childhood. The essential characteristics of the four trainer

patterns identified are as follows.

System I Trainer Pattern. Trainers of System 1 adults were hypothesized to

have been authoritarian, absolutist, ethnocentric, and closed minded; and to have

relied on external sources in their moral reasoning. By the trainer’s behavior,

conformity rather than creativity of thought and action was emphasized, and the

child was given little opportunity to explore values and power relationships.

System 2 Trainer Pattern. System 2 adults’ trainers were hypothesized to have

characteristics similar to those of System 1 trainers, but also to have been arbitrary

and inconsistent. Consequently, the child learned not to trust authority figures or

the institutions of social control that they represent.

System 2 Trainer Pattern. System 3 adults were hypothesized to have had

trainers who were permissive, overprotective, indulgent, and somewhat socially

dependent on the trainee. This enabled the child to take advantage of the depen-

COGNITIVE COMPLEXITY 61

dency relationship to develop skill at manipulating others and, through this, to

avoid facing the world alone. The permissive atmosphere allowed the child greater

freedom than the System 1 and 2 trainer patterns to explore ideas, values, and

relationships.

System 4 Trainer Pattern. System 4 adults were hypothesized to have had

trainers who themselves functioned in a highly abstract manner. They tended to

relate to the child as an older, experienced adult to a younger, developing adult.

The child’s behavior was shaped primarily by positive reinforcement, including

being rewarded for exploring and trying the different rather than for overt re-

sponses matched to narrowly prescribed standards of the trainer. The child was

intrinsically valued by the trainer as a person in his or her own right.

To test the above hypothesized relationships, I had 198 practicing managers

and engineers, enrolled in nine sections of my graduate organizational behavior

classes between 1976 and 1979, write a one page essay describing their childrear-

ing. As part of their essay, these adults were specifically asked to (a) state the

nature of their relationship with each parent, or surrogate parent, including the

extent to which each parant was authoritarian or permissive, and (b) Indicate

how each parant responded when they deviated from parental rules. 156, or 79%

of the responses fell clearly into one of four trainer patterns, highly similar to

those hypothesized by Harvey, et al.(1961). The others could not be classified

into a distinct pattern. All 198 participants were administered Harvey’s This I

Believe Test (TIB), a measure of both cognitive complexity and conceptual

systems (See Harvey, et. al 1961). For the 79% that could be categorized, the.

correlation between trainer pattern and Conceptual system was .62 (p < .001).

In general, the conceptual system level of the individuals matched the hypothe-

sized corresponding trainer pattern, but with one exception: The System 4 adults

(21 students) were fairly evenly split between having had either the System 3 or

System 4 trainer pattern. Most frequently, those persons who had experienced

the System 3 trainer pattern reported it for only one of the parents, usually the

mother, with the father’s role being either largely absent or less influential, and/

or that of disciplinarian.

Cognitive Complexity and Traumatic Event

During the 1976-1995 period, I interviewed over 50 persons who experienced

the System | trainer pattern while growing up, yet had made the transition to

System 3 or 4 functioning. The one common characteristic that these persons

seem to possess, and to which they attribute their breaking away from the System

1 mold, is having undergone a traumatic event in their adult lives (e.g., near

death, death of a loved one, divorce, combat in Viet Nam) which upset their

lives and caused them to seriously question their views of reality.

62 HENDRICK

While it appears traumatic events can lead to development of greater ab-

stractness, it often is not the case. Harvey (1966) and Hunt (1966) have empha-

sized that exposure to diversity can be superoptimal as well as suboptimal, and

thus not facilitate conceptual growth.

Cognitive Complexity and Creativity

In 1968, during a required undergraduate introductory psychology class at The

U. S. Air Force Academy, over 600 students were asked to write down as many

uses as they could think of for their uniform shoulder boards within a five

minute period. All students also were administered the AOS measure of cognitive

complexity. The cognitive complexity scores then were correlated with both

number of uses and instructor ratings, including my own, of originality of the

uses. For both number of uses and originality of uses, the cognitively complex

students scored significantly higher (p < .001) than the more cognitively concrete

students. Similar, but less striking results were found for uses of both a pencil

and paper clip.

In a related study, the students were asked to write down how they would take

their roommate’s girl friend out on a date and have him appreciate it. The replies

were rated for creativity by two instructors, with a third instructor rater being

used to decide the issue when ever there was a disagreement between the first

two. The abstract functioning students scored significantly higher on creativity

than did the cognitively concrete group (p < .001). Using the students’ Harvey

TIB measures of conceptual system, we found no significant difference between

System 3 and System 4 students’ replies. The results of these studies are consistent

with the cognitive complexity and conceptual systems literature, and lend con-

struct validity to the model.

Cognitive Complexity and Leadership Behavior and Influence

117 male upper class undergraduate students, enrolled in eight sections of an

advanced leadership class, were administered the Abstract Orientation Scale

(AOS), a measure of cognitive complexity that has shown good correlation with

Harvey’s TIB and various measures of related dimensions (see Hendrick, 1990

for a brief summary of AOS validation studies). During the course, the students

participated in group discussions of case studies and reading materials and took

part in various classroom exercises involving dimensions of leadership behavior.

These discussions and exercises provided opportunity for each class section mem-

ber to become aware of each other member, the resources he brought to the class

section, and his method and pattern of participation. At the end of the course,

each student ranked all of the students in his section, including himself, in terms

of the degree of influence exercised in the classroom. These rankings were

COGNITIVE COMPLEXITY 63

summed for each student to determine his composite score. The product moment

correlation between composite score and AOS score was .29 (p < .01), suggesting

that the more cognitively complex students tended to have somewhat greater

influence on the group. (Hendrick, 1990)

Observations by myself and the other instructor of the cognitively concrete and

cognitively abstract students’ behaviors during the class exercises were generally

consistent with previous findings for System 1 and Systems 3 & 4 functioning

persons, cited earlier. Cognitively concrete persons were more authoritarian, less

open to opinions of others, more absolutist in their views, and took longer to

change their opinions in the light of new information. In contrast, the more

abstract functioning students were more truly participative and empathic, open

minded, relativistic in their views, and more flexible in their opinions.

More recently, I have replicated the above study with 53 practicing managers

and engineers in three of my graduate organizational behavior courses, and ob-

tained similar results.

Cogntive Complexity and Interpersonal Perception

Fundamental to the interpersonal influence or leadership process appears to be

the ability to perceive the behavioral cues of others (Harvey, 1966). Harvey

(1966) has summarized a number of studies by himself and others demonstrating

that abstract persons have a greater sensitivity to minimal cues and a greater

ability to use them appropriately and completely. In order to determine if this

ability also applies to interpersonal perception, I had 117 senior and junior male

undergraduate students view the film, Twelve Angry Men, which frequently has

been used in interpersonal perception research. The first 38 minutes of the film

depicts the deliberations of the jury at the end of a murder trial. The film is rich

in its portrayal of group dynamics phenomena. Issues of leadership, conformity,

and deviation are highly visible in the emerging patterns of interpersonal relation-

ships of the jurors. Each juror exemplifies a distinct personality and his arguments

and nuances of behavior easily suggest a degree of attitudinal and behavioral

flexibility. The initial vote of the jury is 11 to 1 “‘guilty’’. The film was stopped

at the point where the jurors are about to take a second vote. The students were

informed that during the remainder of the film, the jurors would change their

vote, one by one, resulting in a final vote of 12 to O for not guilty. Each class

member then was handed a form, depicting the jury seating arrangement, and

asked to number the jurors in the order in which they would change their votes

to “‘not guilty’’. Each student’s ranking then was compared with the actual order

in which the jurors switched their vote, and a composite error score was computed.

These error scores were correlated with their scores on the AOS measure of

cognitive complexity. The resulting correlation was .44 (p < .001), suggesting

64 HENDRICK

that the cognitively complex students indeed did make better use of the available

cues. (Hendrick, 1990).

As part of the leadership influence study, cited earlier, the students’ own

rankings of their effectiveness, as compared with that of the other class members,

were compared with the composite group rankings and an error score for each

student was determined. These error scores were correlated with the students’

respective AOS scores. The resulting r was .39 (p < .001), suggesting that the

more abstract functioning students were somewhat more accurate in their self

perceptions, as well as in their perceptions of their classmates. (Hendrick, 1990)

Cognitive Complexity and Group Task Performance

One of the most dramatic differences between abstract and concrete functioning

persons that I have observed has been in two studies of group task behavior

(Hendrick, 1979). In these two studies, the cognitive complexity levels of 100

cadets at the U. S. Air Force Academy and 100 experienced managers enrolled

in a graduate management program at a large private university were assessed

using the AOS. The five highest and five lowest scorers in each of 20 class

sessions were assigned to abstract and concrete problem-solving groups, respec-

tively. The group problem solving task used was the Broken Squares exercise,

described by Pfeiffer and Jones (1969) in their handbook. The exercise consists

of five identically sized cardboard squares, consisting of three sections, thus

forming a puzzle, with no two puzzles being alike. At the beginning of the task,

each participant is given an envelope containing three puzzle pieces, each from

a different square. The participants are instructed to complete all five squares as

quickly as possible. They further are instructed that they are not to talk or gesture,

can only pass pieces to the person to their right or left around the table, and must

wait for pieces to be passed to them.The 20 concrete groups took almost twice

as long as the 20 abstract groups to complete the task (p < .001). Compared to

concrete groups, abstract group members interacted at a faster pace and demon-

strated better cue utilization (p < .001). No differences were found between the

undergraduate and graduate groups.

In addition to the systematic observations of group problem solving behavior,

several other behavioral characteristics on which their were striking differences

between the concrete and abstract groups were informally noted. These included

(a) a tendency of abstract groups to demonstrate greater flexibility of set by a

willingness to break up completed squares to try alternate combinations; (b) a

tendency of abstract groups to test the rules to determine their real limits, whereas

concrete groups did not; and (c) a tendency of concrete group members to focus

primarily on their own individual task, whereas abstract team members tended

also to focus on the work of other team members.

COGNITIVE COMPLEXITY 65

Cognitive Complexity and Stratified Systems Theory

Stratified systems theory holds that hierarchical differentiation of jobs in orga-

nizations differ systematically in their cognitive complexity requirements, and

that managers perform most effectively and are happiest when their own cognitive

complexity level matches that of their position (Stamp, 1981). Based on this

hypothesis, the cognitive complexity levels of 22 hotel General and Resident

Managers for a large hotel chain were assessed using both the AOS and a compos-

ite of four scales of the Guilford-Zimmerman Temperament Survey. All 22 man-

agers were assessed by their superiors as being successful in their present posi-

tions. The 22 managers each were evaluated by the Hotel Division’s Vice

President for Operations Support in terms of his potential for promotion to Area

Manager - the next hierarchical level and one in which the manager exercises

- Supervision indirectly over a group of hotels and their employees, thus making

it more cognitively complex. Of the nine managers scoring as cognitively com-

plex, seven were evaluated as having high potential for promotion to Area Man-

ager. Of the thirteen managers scoring as cognitively concrete, only four were

evaluated as having high potential for promotion (¢@ =.57, p < .01). These results

appear to offer partial support for the stratified systems hypothesis.

Cognitive Complexity and Interpersonal Communication

Implicit in the cognitive complexity literature is a message concerning commu-

nicating with persons of differing levels of complexity which is consistent with

my 30 years of experience as both an organizational consultant and teacher of

practicing managers: Namely, express your message in terms of the other person’s

conceptual reality - to use a trite but true expression, from where he or she is

*“‘coming from’’. With concrete functioning persons, it is important to express

ideas in specific, concrete terms. For example, if trying to persuade a concrete

functioning manager to approve some organizational change intervention, it is

important to describe the intervention in a clear, step-by-step fashion and, espe-

cially, the rationale for the intervention in terms of how it will improve the

manager’s *‘bottom line’’. In contrast, more abstract functioning managers are

likely to respond positively to a description of the approach in terms of its

underlying rationale and its less tangible benefits, such as improving employee

job satisfaction and commitment, reducing stress, and being “‘the right thing to

do’’ from a human consideration point of view.

Use of the Cognitive Complexity Dimension in Organizational Design

Organizational structures generally are acknowledged to have three major di-

mensions: Complexity, formalization, and centralization (Bedeian and Zammuto,

66 HENDRICK

1991; Robbins, 1983; Stevenson, 1993). Like cognitive complexity, organiza-

tional complexity also has the two major components of differentiation and inte-

gration. Organizational differentiation refers to (a) the number of hierarchical

levels or vertical differentiation, (b) the degree of departmentalization and special-

ization, or horizontal differentiation, and (c) the geographical dispersion of organi-

zational units and employees, or geographical differentiation. Increasing any of

these increases the organization’s complexity. Organizational integration refers

to the mechanisms that are used to coordinate and control the differentiated

elements. These include such things as standard operating procedures, commit-

tees, task teams, information systems, integrating offices and vertical hierarchy

(e.g., One boss supervises two or more subordinate units, thus serving as the

integrator). Formalization refers to the extent to which operations rely on formal-

ized procedures, standardized communications and detailed job descriptions,

rather on employee expertise and decision-making. Centralization refers to the

extent to which decisions are made by managers, higher up in the organization,

versus being delegated to lower employee levels.

Among other factors, the optimal degrees of complexity, formalization and

centralization to incorporate into an organization’s design depend on the characteris-

tics of the work force. In particular, the (a) degree of education and training or

professionalism, and (b) the psycho-social characteristics of the employees (Rob-

bins, 1983). With respect to the psycho-social characteristics, I have found one of

the most useful integrating factors to consider is that of cognitive complexity.

Given the characteristics, already described, it is not surprising that, in my 30

years of consulting, I have found concrete functioning managers and employees

alike to prefer the clear, unambiguous structure and formalization of bureaucratic

organizational designs. In contrast, abstract functioning managers and employees

usually prefer low formalization, decentralized decision-making; and are very

comfortable in more complex or less structured, more ambiguous organizations

(e.g., professionalized, matrix, and continuously changing or free-form designs).

I also have found more abstract managers to be comfortable with the use

of employee participation and decentralized decision making, whereas concrete

functioning managers often are resistant to these practices, and prefer to use a

more authoritarian and controlling management style.

Conclusion

From the above historical review of cognitive complexity and its relation to moral

reasoning, conceptual systems, and behavior, and of my research contributions to

our knowledge of cognitive complexity and performance, it should be apparent that |

COGNITIVE COMPLEXITY 67

the cognitive complexity dimension can be of considerable benefit to engineering

psychologists and ergonomists. From my experience, I believe that many of the

inconsistencies in human performance, often found between persons in the same

or similar situations, can be explained by a knowledge of the complexity levels of

the individuals involved. Similarly, knowledge of the complexity levels of persons

in a given environment can better enable us to structure that work situation to

better enhance worker performance and satisfaction.

The cognitive complexity literature also suggests that knowing the complexity

level of individuals may better enable us to design both management information

systems and training programs for them. For example, more structured, step-by-

step linear presentation of data and procedures is likely to better serve cognitively

concrete persons, with their greater need for structure, order and formalization.

In contrast, systems which focus more on communicating or teaching general

principles and cognitive maps are likely to be more intrinsically interesting and

satisfying to cognitively complex persons. These are but two hypotheses of poten-

tially useful applications of cognitive complexity that remain to be validated.

References

Barrif, M. L., & Lusk, E. J. (1977). Cognitive and personality tests for the design of management information

systems. Management Science, 23, 820-837.

Blatt, M. (1971). The effects of classroom discussion upon children’s moral judgement. Jn: L. Kohlberg E. Turiel

(Ed.), Moral research: The cognitive-developmental approach. Holt, Rinehart and Winston, New York.

Bedeian, A. G., & Zammuto, R. F. (1991). Organizations: Theory and Design. Dryden Press, Chicago

Davis, K. (1966). Some correlates of responses to the ‘‘This I Believe Test.’’ In: O.J. Harvey (Ed.), Experience,

structure, and adaptability. Springer, New York.

Harvey, O. J. (1966). Experience, structure, flexibility, and creativity. In: O. J. Harvey (Ed.), Experience,

structure, and adaptability. Springer, New York.

Harvey, O. J., Hunt, D. E., & Shroder, H. M. (1961). Conceptual systems and personality organization.

Wiley, New York.

Hendrick, H. W. (1979). Differences in group problem-solving behavior and effectiveness as a function of

abstractness. Journal of Applied Psychology, 64:518—525.

Hendrick, H. W. (1981). Abstractness, conceptual systems, and the functioning of complex organizations. In:

G. England, A. Negandhi, & B. Wilpert (Eds.), The functioning of complex organizations (pp. 25-50).

Oelgeschalger, Gunn & Hain, Cambridge, MA:.

Hendrick, H. W. (1986). Matching individual and job complexity: Validation of stratified systems theory. In:

Proceedings of the Human Factors Society 30th Annual Meeting, (pp. 999-1001) Human Factors Society,

Santa Monica, CA: .

Hendrick, H. W. (1990). Perceptual accuracy of self and others and leadership status as functions of cognitive

complexity. Jn: K. E. Clark & M. B. Clark (Eds.), Measures of leadership, (pp. 511-520). Leadership

Library of America. West Orange, NJ

Hunt, D. E. (1966). A conceptual systems change model and its application to education. Jn: O. J. Harvey

(Ed), Experience, structure and adaptability. Springer. New York:

Kohlberg, L. (1969). The child as a moral philosopher. Jn: B. Henker (Ed.), Readings in psychology today

(pp. 180—186). CRM Books, Del Mar, CA:

Pfeiffer, J. W., & Jones, J. E. (1969). A handbook of structured experiments for human relations training

(Vol. 1). University Associates Press, Iowa City, IA.

Piaget, J. (1948). The moral judgement of the child. Free Press, Glencoe, IL.

Robbins, S. R. (1983). Organization theory: The structure and design of organizations Prentice-Hall, Engle

wood Cliffs, NJ.

Stamp, G. (1981). Levels and types of managerial capability. Journal of Mana gement Studies, 18:27 7-297.

Stevenson, W. B. (1993). Organizational design. Jn: R. T. Golembiewski (Ed.), Handbook of organizational

behavior. Marcel Dekker, New York.

Journal of the Washington Academy of Sciences,

Volume 84, Number 2, 68—93, June 1996

The Effect of Individual Differences on

Anxiety and Team Performance’

Jeanne L. Weaver, Clint A. Bowers and Ben B. Morgan, Jr.

Department of Psychology University of Central Florida

Received April 30, 1996

ABSTRACT

Because stressors impact individuals and teams in a variety of occupations and environ-

ments, research regarding occupational stress has become increasingly critical. A concurrent

trend is the increasing role of teams in a wide variety of environments. Because it is often

necessary for teams to perform under conditions characterized by stress, there is a clear need

for investigations of the factors which might impact the performance and subjective distress

of teams under stress. Although some past research has considered the impact of individual

differences on team performance under stress, the literature on individual differences and

stress has primarily been concerned with individual responses to stressors. The current study

investigated the relationships among individual difference characteristics, team performance,

individual distress, and coping in teams performing under conditions of stress related to the

task being performed and under conditions where the stress was unrelated to the task being

performed. Specific hypotheses were tested with regard to the impact of self-control and the

manipulation of perceived control with regard to individual anxiety, team performance, and

team member coping. Mixed support was obtained for these hypotheses. Possible explanations

are considered for these findings and directions for further research are discussed.

Stress is one of the most critical and enigmatic areas of psychological research.

Because stressors impact individuals and teams in a variety of occupations and

environments (e.g., fire fighting, the military, medical occupations) research re-

garding occupational stress has become increasingly critical. In fact, occupational

stress has been referred to as the threat to work (Cox, 1978). Thus, it is apparent

that stress is worthy of the attention of researchers interested in gaining an

understanding of the health, psychological, and performance effects of stress.

A concurrent trend is the increasing role of teams in a wide variety of environ-

ments. Although it has been noted that the quantity of team research conducted

regarding team processes and performance has not kept pace with the frequent use

' An earlier version of this work was presented at the Third Interdisciplinary Conference on Occupational

Stress and Health, Washington, D.C. (September, 1995). American Psychological Association.

INDIVIDUAL DIFFERENCES 69

of teams, it appears that this situation is beginning to change (Salas, Dickinson,

Converse, & Tannenbaum, 1992). Because it is often necessary for teams to

perform under conditions characterized by stress, there is a clear need for investi-

gations of the factors which might impact the performance and subjective distress

of teams under stress.

Because a team consists of at least two or more individuals working toward

a common goal in an interdependent fashion (Morgan, Glickman, Woodard,

Blaiwes, & Salas, 1986), team performance would appear to be particularly

susceptible to the effects of stress. That is, the requirement for teams to maintain

acceptable performance by interacting effectively with team members under stress

in addition to the need for individuals to maintain their own performance, places

an additional level of demand or ‘‘resource strain’’ on a team and its members.

Thus, there is a need for research regarding the relationship of stress to team

performance, and the investigation of factors which might impact a team’s perfor-

mance and the subjective distress of its members under stress.

The current effort addresses this need by selecting a well accepted, existing

model of stress deemed to be relevant for adaptation and application to the study

of team stress. The coping construct and the role of self-control and social support,

in this regard, are described in light of the model and a rationale is presented to

explain the stress manipulation employed in the current study. This rationale is

built around a discussion of past research conducted on the perceived control

construct. Finally, research hypotheses are delineated, and the method of study

is detailed.

Background

Stress in Teams

Relatively little systematic inquiry is available regarding stress in groups and

teams (Driskell & Salas, 1991). Morgan and Bowers (1995) note that this lack

of attention is becoming increasingly critical because many jobs in both the

military and civilian sectors require groups of individuals to work together effec-

tively in teams. Often these teams must make quick and effective team decisions

in complex, stressful occupational environments. Similarly, Driskell and Salas

(1991) note that the lack of knowledge regarding teams under stress is surprising

for several reasons: (a) because the complexity and breadth of many present-day

systems often require the combined efforts of a group, (b) in order to understand

outcomes associated with these tasks, an understanding of group processes must

be gained, and (c) because group interaction patterns will be impacted by stress

and other environmental factors.

70 WEAVER, BOWERS, AND MORGAN, JR

Because the nature of team tasks, such as tactical decision making (Cannon-

Bowers, Salas, & Grossman, 1991), aviation (Weitz, 1970), and fire fighting (Fuller-

ton, McCarroll, Ursano, & Wright, 1992), is inherently stressful, it is apparent that

these team situations should be investigated further. The impact of stress in a

variety of team environments and the criticality often associated with related out-

comes make the investigation of stress in teams an important undertaking.

Alluisi and his colleagues investigated the impact of various stressors (e.g.,

continuous work, illness, sleep loss) on team performance. Within the investiga-

tion of Morgan, Coates, Brown, and Alluisi (1974), two five-man crews performed

both team and individual synthetic tasks included in a Multiple Task Performance

Battery (Morgan & Alluisi, 1972). In their investigation of the effects of continu-

ous work and sleep loss, Morgan and his colleagues found that team performance

recovery was related to the length of the continuous work period. That is, a 44

hour period of continuous work produced an average decrement of 22%, while

a 36 hour period was associated with decrements between 14 and 18%.

Morgan and his colleagues’ study of team performance was complemented by

an investigation of the impact of individual differences on team and individual

performance under conditions of extended operations. Morgan, Winne, and Dugan

(1980) reexamined and reanalyzed data previously obtained in Morgan et al.

(1974) in order to investigate the range and consistency of individual differences

of members in the two five-man crews. Results indicated that individual differ-

ences in performance ranged from 0-40% after 32 hours of continuous work.

Their general findings indicated that individuals performed relatively consistently

throughout several exposures to such stress. These authors noted the proposition

of Wilkinson (1974) that individual differences in response to environmental

stimuli be investigated in order to gain an understanding of stressor effects in

team performance.

Although some past research has considered the impact of individual differ-

ences on team performance under stress, the literature on individual differences

and stress has primarily been concerned with the individual (Jex & Beehr, 1991).

Studies such as that conducted by Morgan and his colleagues are relatively rare.

Recently, however, interest has again been sparked in the role of individual

differences in relation to team performance under stress. The work of Weaver

and her colleagues (Weaver, Morgan, Adkins-Holmes, & Hall, 1992; Bowers,

Weaver, & Morgan, 1996) identified a number of variables which might serve

to moderate the effects of stress on team performance. These authors argue that

a clear understanding of stress and its effects can be gained only by identifying

and investigating factors which impact the relationship between stressors and

outcomes. An understanding of these so called moderator variables might enable

researchers to better predict the effects of stress, conduct research which would

INDIVIDUAL DIFFERENCES 71

allow an unobscured view of stress effects, and design effective interventions for

the control of stress (also see Jex & Beehr, 1991).

The literature regarding stress at the individual and team levels have often

differed in the outcome variables studied. That is, whereas the majority of the

individual stress literature often considers the ‘‘feelings’’ or subjective reactions

of people confronted with stressful circumstances (e.g., life events research), the

team stress literature reflects a preoccupation with gaining an understanding of

stressors in relation to performance outcomes. Perhaps this is to be expected

since ““team stress’’ has been investigated less, and it began as an outgrowth of

a need to understand the workings of teams in operational settings. However,

the current study attempts to merge these two approaches by simultaneously

considering subjective and performance outcomes. Furthermore, an understanding

of the relationship between these two outcomes and individual differences, within

the members of the team, might provide valuable insight into the stress phenome-

non. Thus, a primary purpose of this study was to investigate the effects of

ambient and performance contingent stress, in line with the suggestion of Driskell

and Salas (1991), with attention to individual differences and their impact on

member distress and team performance.

Stress and Coping

Lazarus and Folkman (1984) define stress as ‘‘a particular relationship between

the person and the environment that is appraised by the person as taxing or

exceeding his or her resources and endangering his or her well-being’’ (Lazarus &

Folkman, 1984, p. 19). The “‘resources’’ component of the definition refers to

what one draws upon in order to cope.

In addition to the inclusion of components of stress, appraisal, and coping

variables, person and environment antecedents of stress and coping, short- and

long-term adaptational outcomes are also included in Lazarus and Folkman’s

model of stress and coping. Person variables are defined as characteristics of

persons which influence their situational appraisal. The authors argue that person

variables influence appraisal ‘‘by (1) determining what is salient for well-being

in a given encounter; (2) shaping the person’s understanding of the event, and

in consequence his or her emotions and coping efforts; and (3) providing the

basis for evaluating outcomes’’ (Lazarus & Folkman, 1984, p. 55). Two such

person variables are commitments and beliefs. Lazarus and Folkman note that

person and situational variables are highly interdependent. That is, although many

situations might appear to be universally stressful, individual variations are great.

Therefore, it is important to consider situational and person variables simultane-

ously.

72 WEAVER, BOWERS, AND MORGAN, JR

The next component of the stress model is appraisal. Lazarus and Folkman

(1984) note that although certain situations produce stress in a large number of

persons, individual and group differences exist in the extent and type of reactions.

The authors define appraisal as ‘‘the process of categorizing an encounter, and its

various facets, with respect to its significance for well-being’’ (Lazarus & Folkman,

1984, p. 31). The process is argued to be evaluative, focused on significance, and

continuously occurring. In addition, there are two types of appraisal; primary and

secondary. Primary appraisal focuses on the implications of a stressor while second-

ary appraisal focuses on what can be done about the situation. Finally, Lazarus

and Folkman (1984) hypothesize that these appraisals are not always conscious

and can be determined by factors below an individual’s awareness.

Coping is defined as ‘‘cognitive and behavioral efforts to manage specific

external and/or internal demands that are appraised as taxing or exceeding the

resources of the person’’ (Lazarus & Folkman, 1984). Lazarus and Folkman

(1984) also argue that coping serves two functions: (1) problem-focused coping

manages or alters the problem within the environment which is causing distress,

(2) emotion-focused coping regulates emotional responses to problems. As a

method for investigating the coping process, Lazarus and his colleagues devel-

Oped an assessment instrument called the Ways of Coping Scale (Folkman &

Lazarus, 1980), which assesses both problem and emotion focused coping. Al-

though most stressors might trigger both types of coping, emotion focused coping

is most salient when people feel that nothing can be done to reduce or obviate

the stress, while problem focused coping becomes more salient when persons

feel that something can be done (Carver, Scheir, & Weintraub, 1989).

Lazarus and Folkman (1984) also describe the role that self-control, or self-

regulating cognitive skills, plays as a resource within the coping process. They

note that “‘the capacity for self-control’? (Rosenbaum, 1980b) is an important

predictor of problem solving ability in the context of coping. Thus, self-control

is considered within the context of individual coping because of its potential to

contribute resources in relation to problem and emotion focused coping.

Self-control or ‘‘learned resourcefulness’’ is defined as ‘‘an acquired repertoire

of behavioral and cognitive skills with which the person is able to regulate internal

events such as emotions and cognitions that might otherwise interfere with the

smooth execution of a target behavior’’ (Rosenbaum, 1983, p. 55). Rosenbaum

(1980a) argues that persons who use self-control methods to manage pain,

thoughts, and emotions are able to reduce the negative effects of maladaptive

behaviors. Therefore, the use of self-control techniques might prove particularly

effective as a method for coping with stressful situations.

Past research has supported the utility of the self-control or ‘‘learned resourceful-

ness’’ construct as discussed by Rosenbaum, for predicting effective performance

INDIVIDUAL DIFFERENCES 73

in aversive circumstances. For example, Gal-Or, Tenenbaum, Furst, and Shertzer

(1985) investigated the relationship of self-control and trait anxiety to performance

in a sample of 44 trainee parachutists. The results of the study indicated that best

performance was exhibited by parachutists with high self-control regardless of

their trait anxiety. Another study, conducted by Rosenbaum and Rolnick (1983)

investigated the relationship between seasickness and the use of self-control behav-

iors in eighty-nine sailors of the Israeli Navy. The results of this study indicate

that high self-control subjects who became seasick exhibited fewer performance

decrements than did those subjects who were low in self-control. Interestingly,

there was no difference between low and high self-control subjects in their suscepti-

bility to become seasick, but only a difference in their capabilities to perform when

seasickness was upon them. Thus, although self-control did not prevent the onset

of sickness, persons high in self-control were better able to cope with their illness

and maintain better performance under the stressor of being ill. Therefore, these

studies provide interesting data with which to speculate about the potential effects

of self-control on performance under stress. Investigations of self-control are also

interesting in order to provide support for (or against) the value of stress inoculation

training (SIT) for military team members as a method for coping and maintaining

effective performance under stress.

Team Coping

To date, coping has not been treated at the team level. However, similar

constructs have been investigated such as cohesion and social support. For the

purpose of the current study, only social support will be considered. Although

their reference was to individuals, Lazarus and Folkman (1984) have discussed

the role that social support can play in contributing resources within the coping

process. Social support has been defined as:

attachments among individuals or between individuals and groups that serve

to improve adaptive competence in dealing with short-term crises and life

transitions as well as long-term challenges, stresses, and privations through (a)

promoting emotional mastery, (b) offering guidance regarding the field of

relevant forces involved in expectable problems and methods of dealing with

them, and (c) providing feedback about an individual’s behavior that validates

his conception of his own identity and fosters improved performance based on

adequate self-evaluation (Caplan & Killilea, 1976, p. 41).

Social support is important in the context of coping as it ‘‘provides vital

resources from which the individual can and must draw upon to survive and

flourish’’ (Lazarus & Folkman, 1984, p. 243). Carver et al. (1989) also note that

seeking out social support acts as a coping response. People seek out social

74 WEAVER, BOWERS, AND MORGAN, JR

support for one of two reasons: social support for instrumental reasons and

social support for emotional reasons. Instrumental reasons can be seeking advice,

assistance, or information while emotional reasons are related to obtaining moral

support, sympathy, or understanding (Carver et al., 1989). The former is a type

of problem focused coping while the latter is a form of emotion focused coping.

Because social support, by definition, requires interactions with others, the con-

struct is particularly important for the investigation of team performance.

Ironically, relatively little attention has been paid to the role of social support

in research regarding teams under stress. One exception to this surprising trend

is a study conducted by Fullerton et al., (1992). These authors investigated the

psychological responses of fire-fighting teams. In particular, the authors were

interested in factors serving a moderating function of the extreme stress involved

in dealing with the traumatically injured (e.g., mass-casualty air disasters). Their

results indicated that all of the fire-fighters involved in the study noted the impor-

tance of social support from fellow workers. This was noted to be the case

particularly in regard to decision making, staying task focused, providing reassur-

ance, and maintaining a sense of humor (Fullerton et al., 1992).

For the purpose of consideration within the realm of team research, it is

suggested that the coping portion of the Lazarus model be considered in terms

of two separate resource components represented by self-control and social sup-

port. The first coping component represents the resource of self-control which

originates from within individual team members. Because a key component of

research regarding teams rests in the interdependent and interactive activity of

its members, the coping component is expanded to describe coping which is

facilitated or generated from the interactions of the team members. An example

of this type of coping is social support. Consistent with the paradigm established

by Lazarus and Folkman (1984), each of these components is then further divided

into two dimensions. These dimensions for both the individual and team resources

can be task (problem-focused) or non-task (emotion-focused) related. As noted

above, Lazarus and Folkman (1984) describe two sub-types of coping; namely,

emotion and problem focused coping. Problem focused coping is directed at

managing the problem itself, while emotion focused coping is directed at manage-

ment of the emotional response to the problem. One goal of the current study

was to test the extent to which social support and self-control are differentially

related to coping behaviors assessed by the Ways of Coping questionnaire, perfor-

mance, and state anxiety.

Perceived Control of Stress

Because of the belief that control perceptions are implicated in coping and

stress, the perceived control construct has provided a compelling topic of research

INDIVIDUAL DIFFERENCES 75

for psychologists. In fact, as early as 1966, it was argued by Lazarus that the

less control individuals possess in threatening situations, the more likely they are

to feel helpless and distressed by situations (Lazarus, 1966). Weaver et al. (1992)

argued in their review that perceived control might be the common mechanism

to explain the effectiveness of variables which moderate outcomes associated

with stress.

Perceived control has been defined as “‘the belief that one has at one’s disposal

a response that can influence the aversiveness of an event’? (Thompson, 1981,

p. 89). Thompson (1981) notes that this definition is advantageous in that it is

general enough to include all types of control while simultaneously recognizing

that control does not have to be exercised in order tc be effective. Furthermore,

the definition implies that control does not really even have to exist in order to

be effective as long as it is perceived. However, the effects of perceived control

are not as well understood as this definition would appear to imply. In fact,

reviews of the effects of perceived control have indicated that the belief in an

event’s controllability does not always lead to reduced stress while beliefs in the

uncontrollableness of a stress situation do not always lead to increased stress

(Folkman, 1984; Thompson, 1981; Averill, 1973). Thus, there is a need for further

research regarding the circumstances under which perceived control is associated

with positive performance and subjective outcomes.

The effects of perceived control have been investigated in relation to such

diverse stimuli as noise (Corah & Boffa, 1970; Glass & Singer, 1972; Cohen &

Weinstein, 1981), heat (Bell & Greene, 1982), and air pollution (Evans & Jacobs,

1982). One of the most often used manipulations in investigations of perceived

control is threat of shock (Haggard, 1943; Averill, O’Brien, & DeWitt, 1977;

Ball & Vogler, 1971; Staub, Tursky, & Schwartz, 1971). In addition, the effects

of perceived control have been investigated in the organizational literature in

relation to such factors as job satisfaction, commitment, and performance (see

Spector, 1986 for a meta-analytic review of this topic).

Behavioral control is most relevant for the manipulation of stress within this

investigation. That is, the current investigation exposes persons to stress in two

conditions: (a) one in which the stressor is ambiently derived and, thus, uncontrol-

lable, and (b) the other in which the stressor is task contingent and avoidable by

maintaining effective task performance. Results regarding the effect of behavioral

control on arousal during stressor impact are far from clear. While some studies

have found decreased arousal for subjects receiving shock (Geer, Davison, &

Gatchel, 1970), others have found no effect of behavioral control on arousal

during receipt of loud noise (Glass, Reim, & Singer, 1971). Still others have

found increased arousal at impact of loud noise (Gatchel & Proctor, 1976).

Similarly, studies of the effects of behavioral control on self-reported distress

76 WEAVER, BOWERS, AND MORGAN, JR

at stressor impact yield mixed results. For example, several studies (Averill &

Rosenn, 1972; Staub et al., 1971, experiment 1; Sherrod, Hage, Halpern, &

Moore, 1977) have found that behavioral control is not associated with a lesser

tendency to report subjective distress under shock or loud noise, while others

(Staub et al., 1971, experiment 2; Glass et al., 1971) have found that behavioral

control does have an impact on subjective distress under shock and noise stress

conditions.

Fortunately, the equivocality of the perceived control research has been ad-

dressed by Driskell, Mullen, Johnson, Hughes, and Batchelor (1991). These re-

searchers performed a meta-analysis of studies which assessed the effects of

uncontrollable shock on self-reported distress or performance accuracy. The au-

thors note that only shock studies were used due to the common nature of this

manipulation in the control literature. In addition to their investigation of the

effects of perceived control, the authors also sought to account for the effects of

the meaning of the delivery of shock. That is, the authors note that in some

studies shocks are actually delivered whereas other studies only threaten to deliver

shocks. Thus, Driskell et al. (1991) sought to determine the effects of this manipu-

lation as well.

Driskell et al. (1991) summarize their results in the following manner. Their

hypothesis that having control would be associated with less stress was supported

only when shocks were not actually delivered. Thus, in situations in which shock

was threatened but not delivered, less stress was reported when persons had

control. In their ““guidelines for manipulating uncontrollableness’’ Driskell et al.

(1991) state that ‘‘these analyses show that uncontrollable shock can be used

as a powerful laboratory stressor to elicit subjective stress and decrements in

performance accuracy. Greater effects are shown under conditions of anticipatory

stress, when the stress is perceived as:imminent, but never occurs’’ (Driskell et

al., 1991, pp. 105-106). This manipulation was used in the current study. That

is, subjects were threatened with shock, but they never actually received shocks

during task performance. A later section describes the method of past studies

which have used threat of shock, in order to establish the logistics of the manipula-

tion and the level of shock which can be used safely.

Rationale of the Current Study

One purpose of the current research was to explore the utility of the Lazarus

model for application to research regarding team stress. That is, it is suggested

that teams possess, to a greater or lesser extent, resources which contribute to their

ability to cope under stress. Consequently, outcomes (e.g., team and individual

INDIVIDUAL DIFFERENCES 77.

performance, and member distress) are impacted by this coping ability. Therefore,

the proposed definition of team stress is a direct outgrowth of the Lazarus defini-

tion of individual stress, that is, team stress is a particular relationship between

the team and their environment that is appraised by the team members as taxing

or exceeding their resources and endangering their well-being. Based upon the

theoretical underpinnings of Lazarus’ work, the current study sought to investigate

the relationship between two types of coping resources, social support and self-

control, and subjective and performance outcomes. Although the appraisal com-

ponent is obviously critical, this preliminary investigation did not attempt to

assess this component.

The current study investigated the relationships among individual difference

characteristics, team performance, individual distress, and coping in teams per-

forming under conditions of stress related to the task being performed and under

conditions where the stress was unrelated to the task being performed. Because

the coping variable of primary interest was self-control, and because social sup-

port has been demonstrated to have an impact under stress, the current study

assessed both. However, the study was designed so as to directly assess the

effects of self-control while controlling for the effects of social support. Because

relatively few studies have been devoted to the investigation of teams under

stress and the factors which influence related outcomes, it was anticipated that

this investigation would yield useful information regarding the processes and

performance of teams under stress. The following section details the hypotheses

tested within this study based upon the reviewed literature above.

It has been noted (Lazarus, 1993) that cognitive methods of coping (e.g.,

self-control behaviors) are relatively stable, while perceived social support is

cumulative and depends on prior interactions. Therefore, these variables were

differentially treated in the current investigation. That is, teams were formed by

selecting individuals for level of self-control. Social support was assessed from

these teams, and the data were analyzed on the basis of the social support ratings.

In order to assess the effects of social support, an existing measure for the

assessment of perceived social support (Inventory of Socially Supportive Behav-

iors) was adapted for use in the current study. The effects of each coping variable

were investigated in relation to performance and subjective stress outcomes.

Hypotheses

1) Team members high in self-control were expected to report less anxiety than low

self-control members.

2) Teams composed of high self-control members were expected to outperform teams

composed of low self-control members as assessed by team score, query time, penalty

points, and number of items queried.

78 WEAVER, BOWERS, AND MORGAN, JR

3) Teams composed of high self-control members were expected to report different

coping behaviors as assessed via the Ways of Coping Questionnaire than low self-

control member teams.

Preliminary investigations of team stress such as this one, should attempt to

ascertain the extent to which teams are differentially affected by stressors that

impact individual members more directly (e.g., more environmental types of

stressors) relative to stressors that are more relevant to the interactions within

the team (e.g., related to the task the team is attempting to perform). It has been

suggested that a new class of stressors be considered as relevant for investigation

in teams. These teamwork stressors, defined as ‘‘stimuli or conditions that (a)

directly impact the team’s ability to interact interdependently or (b) alter the

team’s interactive capacity for obtaining its desired objectives’ (Morgan & Bow-

ers, 1995), are argued to have a direct impact upon the team’s interaction and

coordination. It might be beneficial to divide this class of stressors further into

stressors which originate from an ambient vs. a performance contingent source.

Indeed, it has recently been suggested that there is a need for investigations of

ambient vs. performance contingent stress on the performance of teams

(Driskell & Salas, 1991). Although perceived control has been. relatively well

investigated in relation to individuals, there is a need for investigations of the

effects of perceived control on teams. Furthermore, there is a need for studies of

the extent to which teams are differentially impacted by stress originating as a

component of the task performed as compared to stress originating from other

sources. Such investigations will allow researchers to determine which source of

stress might be more deleterious for team performance. In turn, this will provide

direction for further research regarding team stress. Hypotheses regarding the

stress manipulation and the coping variables discussed above are listed in the

following section.

Hypotheses

1) Team performance in the ambient stress condition was expected to be inferior to that

of performance in the performance contingent condition, but performance in the

performance contingent condition was expected to be significantly worse than the

condition in which no stressor was presented.

2) A significant interaction (trial by self-control) was hypothesized such that high self-

control teams were expected to perform better in the ambient stress condition than

low self-control teams with a smaller difference between low and high self-control

team performance in the performance contingent condition.

3) State anxiety in the ambient stress condition was expected to be higher than in the

performance contingent condition, but distress in the performance contingent condi-

tion was expected to be significantly higher than the baseline condition.

INDIVIDUAL DIFFERENCES 79

Method

Participants

Sixty-four, University of Central Florida male undergraduate students were

solicited to comprise thirty-two, 2-member teams. Subjects were awarded re-

search credit for their participation.

Experimental Design

Independent Variables

Stressor type. Two types of stressor conditions were manipulated in the current

study. These conditions were threat of shock contingent upon poor task perfor-

mance and threat of shock presented as ambient or unrelated to task performance.

Both of these conditions were compared to a post-asymptotic baseline session

with no shock threat.

As previously noted, subjects were threatened with shock, but they did not

actually receive shocks during the experimental periods of task performance.

However, consistent with past shock threat research and in order to increase the

salience of the threat, subjects received a mild but uncomfortable shock just prior

to the two experimental periods. This is an adaptation of the method used in

other shock threat research. |

A common method of manipulating shock threat is to conduct a session prior

to the experimental period to establish ‘‘shock tolerance.’’ This procedure has

been utilized in a number of studies (Beck, Barlow, Sakheim, & Abrahamson,

1987; Beck & Barlow, 1986; Geer et al., 1970; Harris, 1981). The purpose of

this procedure is ‘‘to alleviate any ambiguity as to the nature of the shock

stimulus*‘ (Harris, 1981, p. 383). Thus, the salience of the threat manipulation

is insured. Another technique has been to include ’’dummy trials‘‘ on which

performance data are not collected (Averill et al., 1977). Thus, shock-threat

investigations which have exposed subjects to shock prior to, but not during the

experimental sessions, have taken steps to insure that the threat manipulation is

believable.

Researchers have noted that because the threat of shock may not always be

believable, it is possible for this manipulation to be ineffective with regard to

producing stress (Britt & Blumenthal, 1991). Therefore, it is necessary for re-

searchers to insure the effectiveness of the threat by including a manipulation

check or some manner of increasing the salience of the threat. In the current

study, the salience of the threat was increased by presenting subjects with a

80 WEAVER, BOWERS, AND MORGAN, JR

‘‘sample’’ shock prior to each experimental session. Overall, only two mild

shocks were given to each subject.

The research cited above has used shocks ranging from 1.0 to 10.0 mA for a

typical period of 0.5 second. Harris (1981) utilized a 60 Hz constant current

source with a shock intensity of 1 mA for a 0.5 sec duration. She notes that this

is consistent with that used by other investigators (Furedy & Doob, 1972), and

that pilot testing on volunteers indicated that they found this level painful but

not so much so that they would refuse to continue.

The current study employed the same level of shock as that used by Harris

(1981). This is the lowest amperage reported in prior studies to be effective in

its aversiveness. Subjects were informed prior to the beginning of the experiment

that an “‘uncomfortable but harmless shock’’ would be used and that they could

leave without penalty at that or any other time throughout the experiment. The

apparatus used to conduct the shock will be described in the method section of

this report.

Self-control. Two levels of self-control (low and high) were compared. Teams

of low self-control individuals and teams of high self-control individuals were

constructed by pairing individuals based on their scores on the Rosenbaum Self-

Control Schedule (SCS) (Rosenbaum, 1980b). High self-control teams were cre-

ated by pairing persons who scored in the upper quartile, and low self-control

teams were formed by pairing persons scoring in the bottom quartile.

The method of assessment used for determination of ievel of self-control is

the Self-Control Schedule (SCS) (Rosenbaum, 1980b). This 36-item self-report

instrument assesses the extent to which individuals utilize methods of self-control

in order to solve behavioral problems. The components of the scale are (a) use

of cognitions and self-instructions to cope with emotional and physiological

responses, (b) application of problem-solving strategies, (c) ability to delay imme-

diate gratification, and (d) belief in one’s ability to self-regulate internal events

(Rosenbaum, 1990). Rosenbaum (1990) also notes that past studies have shown

that there are relatively large individual differences even in rather homogenous

populations. Interestingly, responses on the Self-Control Schedule have been

investigated in relation to Lazarus and Folkman’s Ways of Coping Scale (1980).

Gintner, West, and Zarski (1989) studied responses of eighty graduate students

on the scales by comparing coping of high and low self-control students prior to

and following a midterm exam. High self-control persons used more problem

focused coping during the week of preparation while low self-control persons

resorted to more wishful thinking and distancing during preparation and more

wishful thinking and self blame during the waiting week. Overall, high self-

control persons reported fewer stress symptoms than low self-control persons.The

SCS has been demonstrated to possess adequate psychometric properties (Rosen-

INDIVIDUAL DIFFERENCES $1

baum, 1980b). Test-retest reliability of .86 has been computed over a four week

period and internal consistency coefficients have ranged from .78 to .84. A

measure of validity has been obtained by way of comparison with the Rotter

I-E Scale. It was hypothesized that individuals high in self-control would be

internal in their locus of control. A Pearson correlation of -.40 (p < .01) was

computed between the Self-Control Schedule and the Rotter scale. This coefficient

indicated that individuals who reported greater use of self-control methods be-

lieved less in the external control of their behavior.

Other indicators of the scale’s validity have also been found (Rosenbaum,

1990). As assessed by this instrument, high self-control has been found to be

associated with an individual’s ability to (a) tolerate clinical and laboratory in-

duced pain, (b) succeed in weight reduction, (c) cope with seasickness, (d) handle

helplessness manipulations, and (e) cope with demanding medical treatment regi-

mens (Rosenbaum, 1990).

Dependent Variables

Performance. Four performance measures were obtained from each of three

task sessions. These measures are explained in a later section which describes

the task.

State anxiety. The measure used to gauge subjective distress was the Spiel-

berger State-Trait Anxiety Inventory (STAI) (state portion) (Spielberger, Gor-

usch, Lushene, 1970). Subjects were asked to respond to this self-report measure

four times during their participation (i.e., when first entering the experimental

situation, immediately following baseline task performance, and immediately

following each stress condition).

The STAI has been effectively utilized as a measure of state anxiety and has

been shown to possess adequate psychometric properties. The instrument is a

two-part scale which assesses both state and trait anxiety. Each part contains 20

items that either indicate an absence of anxiety or describe anxiety symptoms.

The trait scale requires that individuals indicate on a four-point scale the frequency

of times that they experience anxiety symptoms, while the state scale calls for

ratings of intensity of anxiety symptoms at a particular time (Spielberger, Vagg,

Barker, Donham, & Westberry, 1980).

Test-retest reliabilities of the trait scale have ranged from .73 to .86, while the

reliabilities for the state scale have ranged from .16 to .54 (Anastasi, 1976). In

addition, Kuder-Richardson reliabilities for both scales have ranged from .83 to

.92. Validity has been established via factor analysis. Kendall, Finch, Auerbach,

Hooke, and Mikulka, (1976) factor analyzed the scale and distinguished one trait

factor and two state factors.

Coping. Coping actions were assessed through self-report using relevant items

82 WEAVER, BOWERS, AND MORGAN, JR

from the Ways of Coping Questionnaire (Lazarus & Folkman, 1984). Coping

was assessed following each of three task sessions.

Social support. A self-report measure of perceived social support was adapted

for this study from an existing measure of social support (Inventory of Socially

Supportive Behaviors; Barrera, Sandler, & Ramsay, 1981). This score was ob-

tained three times. Specifically, measures of social support were obtained follow-

ing each post-asymptotic performance session. The scores on the social support

measure were utilized as a covariate in analyses regarding performance and

anxiety.

Apparatus

Task

Two networked IBM compatible computers were utilized to present a pre-

viously developed team decision making simulation. The task used for this study

was the Tactical Naval Decision Making System (TANDEM) (Weaver, mee

Hall, & Compton, 1994).

The TANDEM is a networked radar simulation which requires team members

to query and integrate information in order to make accurate decisions regarding

the type, threat, and intent of incoming targets. Teams receive points based upon

the decisions made and consequent actions taken. In other words, teams are

required to both label targets based upon type, threat, and intent decisions and

engage targets accordingly. For a detailed description see Weaver et al. (1994).

The performance measures obtained were as follows. The first measure was

team score. This score is based on the number of points received for correct team

decisions minus the number of points lost for an incorrect team decision. In

addition, this score is also impacted by the number of penalty points assessed

for allowing incoming targets to get ‘‘too close.’’ Query time was the second

performance measure considered. Query time is the amount of time spent querying

target information. The number of information items queried was also measured.

This is the total number of items queried. The final performance measure consid-

ered was penalty points. This represents the number of points assessed for

allowing incoming targets to remain ‘‘too close’ as demarcated by the known

penalty circles. These four measures comprised the performance measures of

interest.

Shock Generator

In addition to the materials described above, a capacitor was used to present

two ’’test shocks** to both team members simultaneously. This apparatus gener-

INDIVIDUAL DIFFERENCES 83

ated 0.5-second shocks at a level of ImA and 70 volts. As previously described,

this is consistent with past research in that these shocks are at the low end of

the continuum of those used safely in the past.

Procedure

Pilot-Studies

Before the experimental studies were conducted, 20 pilot teams were tested in

order to determine the amount of time necessary to reach asymptotic performance.

These studies required subjects to perform eight, 20-minute sessions (a total of

2 hours and 40 minutes) for this purpose. This time period was based upon the

knowledge that previous three-person TANDEM studies required a period of 1

'/, hours to reach asymptote (Weaver, Bowers, & Morgan, 1994). Results of these

pilot studies indicated that there was no change in team performance for sessions

four through eight. Thus, asymptote was reached after the first hour of task

performance. |

In a second pilot-study, self-control levels of potential subjects were assessed

by administering Rosenbaum’s Self-Control Schedule questionnaires to 272

males. Their self-control scores were used in order to select sixteen teams com-

posed of high self-control members and sixteen teams composed of low self-

control members.

Experimental Study

A total of thirty-two, 2-person teams were run, 16 of which were composed

of high self-control members and 16 of low self-control members. The upper

and lower quartiles of the questionnaire data gathered in the pilot study were

used as cutoff scores. That is, low self-control was defined as a total self-control

score of 4.5 or less corresponding to the 25th percentile of the sample. High self-

control was defined as a total score of 36.5 or higher, corresponding to the top

25 percent of scores.

After obtaining informed consent, all teams were trained on a computer-based

simulation of the team task (TANDEM) and required to perform the task until

asymptotic performance was reached (three, 20-minute sessions, as determined

by the pilot studies conducted for this purpose). Subjects then performed four,

10-minute test sessions. First, subjects performed such a task session to adapt

them to the performance of 10-minute sessions (as opposed to 20-minute) without

visible score. The baseline task session to be used as a comparison condition

with subsequent stress sessions, was then performed. Subjects were then exposed

to the two stress sessions presented in a within-groups design, such that all

84 WEAVER, BOWERS, AND MORGAN, JR

subjects were exposed to both sessions. The order of the two stress sessions

was counterbalanced across teams. Approximately 10-minutes passed between

sessions as subjects completed questionnaires.

Stress was imposed in the form of the threat of shock. Just prior to the stress

conditions, subjects were informed that they had reached the portion of the study

in which shock would be introduced for the purpose of determining the impact

of the shock on their ability to maintain effective task performance. Each team

member was connected to the shock apparatus via an electrode on the index

finger of the non-dominant hand. Prior to each stress condition, team members

were instructed as to when they would be shocked. In the ambient stressor

condition, the threat was described as a ‘‘random occurrence unrelated to any

behavior on the subjects’ part’’. On the other hand, in the performance-contingent

stress condition, threat was described as ‘‘shock presented when the performance

of the team falls below their previous average performance.’’ It is important to

note that subjects received feedback in the form of a team score displayed as

part of the TANDEM task during skill acquisition. However, team scores were

not displayed during the baseline and stress conditions. That is, since subjects

were informed that shock was performance contingent during one of the stress

conditions, it was necessary to keep the availability of feedback regarding their

performance constant across the three task sessions of interest.

Immediately following task performance in each of the three conditions sub-

jects were required to respond to three questionnaires: the state portion of the

STAI, the Ways of Coping Questionnaire, and the social support questionnaire.

Subjective stress (as measured by the STAI) was conceptualized as the mean

reported anxiety score of teams’ individual members and performance stress

was considered as the team performance change from baseline under the stress

conditions. Measures of social support were averaged in the same manner as the

STAI. Following performance of the final session and completion of question-

naires subjects were debriefed and dismissed.

Results

Anxiety

In order to test hypotheses regarding anxiety differences as a function of self-

control and stressor condition, a 3 X 2 mixed model analysis of covariance was

computed with social support as the covariate. Results of this analysis indicated

that social support failed to act as a significant covariate for either the self-control

or stressor condition effects. Consequently, a 3 X 2 mixed model analysis of

variance was conducted. This analysis yielded a significant main effect of self-

INDIVIDUAL DIFFERENCES 85

control, such that low self-control teams experienced higher anxiety than did

teams composed of high self-control members F(1,30) = 9.26, p < .05. The eta

squared value computed for this effect was 0.14. The analysis also yielded a

significant main effect of trial, indicating that anxiety differed as a function of

stressor condition F(2,60) = 9.19, p < .05. Subsequent tests indicated that teams’

anxiety was higher in both stressor conditions than in the no-stress condition.

However, anxiety did not differ significantly between the two stressor conditions.

The eta squared value computed for the effect of stressor condition was 0.09.

There was no significant interaction.

Performance

A multivariate analysis of covariance (MANCOVA) was performed on the

four team performance measures in order to test hypotheses regarding team

performance differences as a function of self-control and stressor condition. This

analysis yielded no significant main effect of self-control or stressor condition,

nor was there any significant interaction.

Because apriori hypotheses were made regarding performance differences with

regard to self-control and stressor conditions, univariate analyses of covariance

were also computed with team score, query time, number of items queried, and

penalty points as dependent variables. Social support was not a significant covari-

ate in any of these analyses, therefore, univariate analyses of variance were

computed. Of these four analyses only a significant main effect of self-control

for penalty points was revealed such that low self-control teams scored signifi-

cantly more penalty points than did high self-control teams F(1,30) = 6.13, p <

.05. The computed eta squared value for this effect was 0.08. No other significant

main effect or interaction was found with regard to the performance variables.

Coping

Hypotheses regarding differences in subjective measures of coping as a func-

tion of self-control and stressor condition, were tested by computing mixed model

analyses of variance with team averages of the measures of problem focused

coping, and five measures of emotion focused (distancing, wishing, self isolation,

positive thinking, and social support) as dependent variables. These analyses

indicated a significant main effect of self-control F(1,30) = 17.37, p < .001 and

Stressor condition F(2,60) = 4.21, p < .05 for the problem focused coping

dimension. The analysis indicated that low self-control teams used more problem

focused coping than did high self-control teams. The eta squared value for this

86 WEAVER, BOWERS, AND MORGAN, JR

effect was 0.24. Subsequent tests regarding the significant main effect of stressor

condition indicated that more problem focused coping was used in the no stressor

and performance contingent condition than within the ambient stressor condition.

However, the performance contingent and no stressor conditions did not differ.

The eta squared value for this effect was 0.04.

No significant main effect of self-control or stressor condition was found for

any of the emotion focused dimensions. However, the analyses did indicate a

non-hypothesized interaction of self-control and stressor condition for the self-

isolation dimension of the coping inventory F(2,60) = 3.67, p < .05. Subsequent

tests revealed that low self-control teams used more self isolation as an emotion

focused coping strategy in the no stressor condition than high self-control teams,

with no further self-control differences in either of the stressor conditions. Further-

more, low self-control teams used self isolation more in the no stressor condition

than in the ambient stressor condition. However, there was no difference between

the no stressor and performance contingent conditions for low or high self-control

teams.

Manipulation Check

A paired samples t-test was computed with regard to subjects responses to two

questions. Each of these questions asked subjects to report their perceived control

in the ambient and performance contingent stressor conditions respectively. This

analysis indicated that subjects perceived significantly more control in the perfor-

mance contingent shock condition than within the ambient shock condition ¢ (63)

= 12.24, p <.05 (M = 75.0, 20.76).

Discussion

For the sake of clarity, this section will revisit and discuss the implications of

each hypothesis originally posited regarding the three dependent variables; anxi-

ety, coping, and performance.

Anxiety

It was originally hypothesized that high self-control teams would report less

anxiety than low-self control teams. Support for this hypothesis was found. This

finding is consistent with past research indicating that high self-control individuals

INDIVIDUAL DIFFERENCES 87

demonstrate greater tolerance of adverse conditions than low self-control individ-

uals (Rosenbaum, 1990).

It was also hypothesized that state anxiety in the ambient stressor condition

would be higher than in the performance contingent condition, but that state

anxiety in the performance contingent condition would be significantly higher

than in the baseline condition. This hypothesis received mixed support. Specifi-

cally, although anxiety was higher in both stressor conditions than within the no

stressor condition, no difference was obtained between stressor conditions.

The rationale for differences in anxiety between stressor conditions was ex-

plained previously with regard to the control perceptions of persons exposed to

either a controllable or uncontrollable stressor. That is, it was hypothesized that

anxiety would be higher in the ambient stressor condition because that stressor

was uncontrollable in comparison to the performance contingent stressor condi-

tion. In order to determine whether subjects perceived differences in control, a

manipulation check was utilized. Specifically, subjects were asked to report their

perceived control in each of the two stressor conditions. The analysis of these data

indicated that subjects did perceive greater control in the performance contingent

stressor condition than within the ambient stressor condition. Thus, the manipula-

tion was effective with regard to instilling these altered control perceptions.

However, there was no resulting difference in state anxiety.

Coping

It was hypothesized that teams composed of high self-control members would

report different coping behaviors (problem vs. emotion focused), as assessed with

the Ways of Coping Questionnaire, than low self-control teams. This hypothesis

was partially supported in that high self-control teams used more problem focused

coping than low self-control teams. This finding is consistent with past research

which indicates that high self-control individuals utilize more task relevant coping

than low self-control individuals (Rosenbaum, 1990).

Although problem focused coping differed as a function of self-control, there

was no meaningful difference in emotion focused coping. A possible explanation

for this finding relates to the salience of the task performance situation and the

salience of the stressor. That is, the salience of the requirement for task perfor-

mance in combination with the possibly low salience of the threat might have

induced subjects to feel less compelled to utilize emotion focused coping than

was previously hypothesized. .

The finding that social support failed to serve as a significant covariate in

relation to stressor condition was surprising given the literature indicating that

88 WEAVER, BOWERS, AND MORGAN, JR

social support is an effective means of ameliorating, to some extent, the effects

of stressors. Apparently, in the current laboratory context this measure did not

accurately capture resources contributed across team members and conditions.

Specifically, this might have been the case for two related reasons. First, the

manipulation of threat might not have been salient enough to cause team members

to feel the necessity to exchange resources. Thus, no social support differences

needed to emerge to ‘“‘buffer’’ these effects. Second, because team members were

males brought together in the context of a laboratory setting, they might have

been hesitant to offer and/or report the contribution of social support resources

as assessed by the instrument used within this study.

Anecdotally, the subjects often joked between themselves regarding the more

‘‘emotional’’ items included on the Ways of Coping and social support measures.

Perhaps these subjects were hesitant to respond truthfully regarding the more

‘‘emotional’’ items on the questionnaires. This might account, in part, for the

failure to find a significant difference regarding emotion focused coping and

social support.

Performance

Self-Control

With regard to team performance, it was hypothesized that teams composed

of high self-control members would outperform teams composed of low self-

control members as assessed by four TANDEM performance variables. Little

support was indicated for this hypothesis, with no difference evidenced for score,

query time or number. However, low self-control teams accumulated more penalty

points than high self-control teams. This indicates that high self-control teams

were more attentive to critical targets presented in the immediate vicinity of

‘‘ownship’’. A potential explanation for this difference in terms of self-control

might be related to the nature of this performance measure. That is, because

penalty points represent the loss of resources when targets are allowed to approach

too closely, it is possible that this measure represents the extent to which low

vs. high self-control teams respond to the demands of time-pressure. Specifically,

it appears that high self-control teams are better able to recognize the salience

of the targets approaching as threatening and to perform so as to minimize this

threat. Although for the purposes of the current study, team score was a primary

measure of interest, penalty points simulate a critical measure of performance in

the operational environment. Specifically, it is obviously critical for teams in

such settings as a command information center or fire-fighting to be able to

INDIVIDUAL DIFFERENCES 89

respond quickly, as well as correctly. It is possible that level of self-control is a

significant predictor of the ability to respond effectively to time-pressure.

Stress

It was also hypothesized that, 1) team performance in the ambient stress condi-

tion would be inferior to that of performance in the performance contingent

condition, which would be significantly worse than the baseline condition and,

2) that a significant interaction (condition by self-control) would be evidenced

such that high self-control teams would perform better in the ambient stress

condition than low self-control teams, with a smaller difference between low and

high self-control team performance in the performance contingent condition.

Neither of these hypotheses were supported.

Although it was expected that performance differences might be found within

the current study, past literature regarding individuals, and their performance

under stress has at times found evidence of performance degradation with stressor

exposure, and at other times evidence of performance facilitation. Thus, it is

reasonable to expect the relationship between stress and team performance to be

complex, given that both intra- and inter-individual processes and the complexity

of the task to be performed must be considered.

There is some literature to suggest that the performance of teams degrades

with stressor exposure (Driskell & Salas, 1991). However, in the case of the

study by Driskell and Salas, the stressor was highly threatening (1.e., possible

tear gas exposure) and the task performed was highly ambiguous. Other research

has been equivocal regarding teams and their performance while exposed to such

stressors as workload, ambiguity, and time pressure. For example, Urban, Bowers,

Monday, and Morgan, (1995) investigated the impact of workload on performance

and found no overall effects of workload. Other research (Weaver et al., 1994)

has investigated the effect of ambiguity and time-pressure. The study by Weaver

and her colleagues found no effects of ambiguity on performance. However,

time-pressure was associated with degraded performance. Conversely, Serfaty,

Entin, and Volpe (1993) who also investigated the effects of time-pressure and

ambiguity, found no significant performance differences for time-pressure, al-

though performance was degraded by increased ambiguity. Thus, prior research

regarding such task-related stressors has been equivocal. |

Within the current study, teams reported changes in anxiety and coping with

altered situational demands. However, no performance decrement was revealed.

One explanation is that the variability of team task performance was already so

great that the manipulation was ineffective in producing any significant change

within the stressor conditions. Another possible explanation for this finding is

related to the relatively low difficulty level of the task performed by teams in

90 WEAVER, BOWERS, AND MORGAN, JR

this study. It has been argued that the less attention a task demands, the less

vulnerable it is to the performance effects of stress (Hancock, 1981). Within the

context of the current study, the task procedure remained unchanged. That is,

although the presentation of targets was dynamic, the procedure for determining

the necessary outcome remained the same. Thus, this behavior became relatively

well practiced. In addition, the task was designed so that teams had the capability

to find ‘‘the right answer’’ if they followed the prescribed procedure. Thus, in

these respects, the task was relatively easy for a decision making task.

Another possible explanation for failure to find performance effects might lie

within the teams’ performance processes. In other words, it is possible that teams

were able to adapt their process in order to prevent the occurrence of performance

decrements. In fact, it has been argued that it is critical to consider stress at the

process level of teams in order to gain a thorough understanding of the impact

of stressors on team performance (Morgan & Bowers, 1995). Consequently, it

appears that there is a need for further research in order to attempt to explicate

these relationships between stressor exposure and team performance.

Directions for Further Research

In order to better understand the effects of stress on teams, team performance,

and team interactions, there appear to be several areas of research where valuable

knowledge is still needed. For example, research that utilizes highly salient stres-

sors, stressors that more closely approximate real world situations, would yield

valuable information. Future research needs to be conducted in settings which

allow the manipulation of such stressors.

Future research should also investigate the effects of stressors presented earlier

in training. Although there is clearly a need to gain an understanding of stress

in order to maximize performance in operational settings, perhaps an understand-

ing of the mechanisms by which stress functions would be best gained at this

juncture by investigating the effects of stressors and the processes involved (e.g.,

coping) with participants who are relatively inexperienced. There is some research

evidence to suggest that the presentation of stressors late in training will have a

less deleterious effect on performance than stressors introduced early in training

(Ryan, 1961). This research might prove valuable in the design and conduct of

research which would provide information regarding stressor effects in opera-

tional settings with experienced operators.

The current study represents a relatively novel approach to studying the effects

of stressors and the degree to which such exposures might be ameliorated by

coping resources originating intra- and inter-individually. It appears that the

INDIVIDUAL DIFFERENCES 91

modes of measurement suggested by Lazarus and Folkman’s treatment of the

stress phenomenon might be valuable in gaining further understanding of team

process and performance under stressful conditions. That is, it might be useful

to consider other individual and team coping resources (e.g., cohesion) within

the context of other types of team tasks. It may also prove useful to determine

the relative efficacy of such coping resources as social support and emotion

focused coping via another assessment approach. For example, behavioral ap-

proaches to measurement might be useful in order to minimize problems such

as shared variance and hesitancy of subjects to report particular types of feelings

and behaviors.

Given that past team research has suggested the criticality of team processes

as related to team performance, future research should include such team process

measures as communication. Because results of the current study indicated differ-

ences in terms of coping, it might prove beneficial to investigate the extent to

which further evidence of coping might be found in communication behaviors.

Subjective measures of coping as obtained within this study might yield less

information relevant to team performance than coping measures which capture

the dynamics of interpersonal interaction. That is, what team members report

doing individually in terms of coping might not be as relevant to team perfor-

mance as what team members are observed to do collectively. Perhaps the identi-

fication of such behaviors might be useful in the development of stress inoculation

training interventions relevant for operational teams. Further research regarding

teams and stress could attempt to address some of these issues.

References

Anastasi, A. (1976). Psychological Testing. Macmillan, New York.

Averill, J. R. (1973). Personal control over aversive stimuli and its relationship to stress. Psychological Bulletin,

80:286-303.

Averill, J. R., O’Brien, L., & deWitt, G. (1977). The influence of response effectiveness on the preference

for warning and on psychophysiological stress reactions. Journal of Personality, 45: 395-418.

Averill, J. R., & Rosenn, M. (1972). Vigilant and non-vigilant coping strategies and psychophysiological

stress reactions during anticipation of electric shock. Journal of Personality and Social Psychology,

23:128-141.

Ball, T. S., & Vogler, R. E. (1971). Uncertain pain and the pain of uncertainty. Perceptual and Motor Skills,

33:1195-1203.

Barrera, M., Jr., Sandler, I., & Ramsay, T. (1981). Preliminary development of a scale of social support:

Studies on college students. American Journal of Community Psychology, 9:435-447.

Beck, J. G., & Barlow, D. H. (1986). The effects of anxiety and attentional focus on sexual responding. II:

Cognitive and affective patterns in erectile dysfunction. Behavior Research and Therapy, 24:19-26.

Beck, J. G., Barlow, D. H., Sakheim, D. K., & Abrahamson, D. J. (1987). Shock threat and sexual arousal:

The role of selective attention, thought content, and affective states. Psychophysiology, 24:165-172.

Bowers, C. A., Weaver, J. L., & Morgan, B. B., Jr. (1996). Moderating the performance effects of stress.

In: J. E. Driskell & E. Salas (Eds.), Stress and human performance. Lawrence Erlbaum Associates, Inc.,

New York:.

Britt, T. W., & Blumenthal, T. D. (1991). On threats of shock as an elicitor of anxiety: A methodological

note. Psychological Reports, 69:649-650.

Cannon-Bowers, J. A., Salas, E., & Grossman, J. D. (1991, June). Improving tactical decision making under

92 WEAVER, BOWERS, AND MORGAN, JR

stress: Research directions and applied implications. Paper presented at the International Applied Military

Psychology Symposium, Stockholm, Sweden.

Caplan, G., & Killilea, M. (1976). Support systems and mutual help. Grune & Stratton,New York:

Carver, C. S., Scheier, , M. F., & Weintraub, J. K. (1989). Assessing coping strategies: A theoretically

based approach. Journal of Personality and Social Psychology, 56:267-283.

Cohen, S. & Weinstein, N. (1981). Non-auditory effects of noise on behavior and health. Journal of Social

Issues, 37:36-70.

Corah, N. L., & Boffa, J. (1970). Perceived control, self-observation, and response to aversive stimulation.

Journal of Personality and Social Psychology, 16:1-4.

Cox, T. (1978). Stress. MacMillan, New York:.

Driskell, J. E., Mullen, B., Johnson, C., Hughes, S., & Batchelor, C. (1991). Development of quantitative

specifications for stimulating the stress environment. (Report No. AL- 7R-1991-01-09) Wright-Patterson

AFB, OH: Armstrong Laboratory.

Driskell, J. E.. & Salas, E. (1991). Group decision making under stress. Journal of Applied Psychology,

76:473-478.

Folkman, S. (1984). Personal control and stress. and coping processes. Psychology and Aging, 2: 171-184.

Folkman, S., & Lazarus, R. S. (1980). An analysis of coping in a middle-aged community sample. Journal

of Health and Social Behavior, 21: 219-239.

Fullerton, C. S., McCarroll, J. E., Ursano, R. J.. & Wright, K. M. (1992). Psychological responses of

rescue workers: Fire fighters and trauma. American Journal of Orthopsychiatry, 62: 371-378.

Furedy, J. J., & Doob, A. N. (1972). Signaling unmodifiable shocks: Limits on human informational cognitive

control. Journal of Personality and Social Psychology, 21:11-115.

Gal-Or, Y., Tenebaum, G., Furst, D., & Shertzer, M. (1985). Effect of self-control and anxiety on training

performance in young and novice parachuters. Perceptual Motor Skills, 60:743-746.

Gatchel, R. J., & Proctor, J. D. (1976). Physiological correlates of learned helplessness in man. Journal of

Abnormal Psychology, 85:27-34.

Geer, J. H., Davison, G. C., & Gatchel, R. I. (1970). Reduction of stress in humans through nonveridical

perceived control of aversive stimulation. Journal of Personality and Social Psychology, 16:731-738.

Gintner, G. G., West, J. D., & Zarsky, J. J. (1989). Learned resourcefulness and situation-specific coping

with stress. Journal of Psychology, 123:295-304.

Glass, D. C., Reim, B., & Singer, J. E. (1971). Behavioral consequences of adaptation to controllable and

uncontrollable noise. Journal of Experimental and Social Psychology, 7:244- 257.

Glass, D. C., & Singer, J. E. (1972). Urban stress: Experiments on noise and social stressors. Academic

Press, New York:.

Haggard, E. (1943). Some conditions determining adjustment during and readjustment following experimen-

tally induced stress. In:S. Tomkins (Ed.), Contemporary Psychopathology. Harvard University Press, Cam-

bridge: .

Hancock, P. A. (1981). Heat stress impairment of mental performance: A revision of tolerance limits. Aviation,

Space, and Environmental Medicine, 52(3):177-180.

Harris, R. M. (1981). Conceptual complexity and preferred coping strategies in anticipation of temporally

predictable and unpredictable threat. Journal of Personality and Social Psychology, 41: 380-390.

Ivancevich, J. M., & Matteson, M. T. (1988). Application of the triangulation strategy to stress research. In:

J. J. Hurrell, Jr., L. R. Murphy, S. L. Sauter, & C. L. Cooper (Eds.), Occupational stress: Issues and

developments in research (pp.200-215), Taylor & Francis, Ltd., London.

Jex, S. M., & Beehr, T. A. (1991). Emerging theoretical and methodological issues in the study of work-

related stress. In: G. R. Ferris and K. M. Rowland (Eds.), Research in personnel and human resources

management: A research annual (Vol. 9, pp. 311-365), JAI Press, Greenwich, CT.

Kendall, P. C., Finch, A. J., Auerbach, S. M., Hooke, J. F., & Mikulka, P. J. (1976). The State-Trait

Anxiety Inventory: A systematic evaluation. Journal of Consulting and Clinical Psychology, 44:406-412.

Lazarus, R. S. (1966). Psychological stress and the coping process. McGraw-Hill, New York.

Lazarus, R. S. (1993). Coping theory and research: Past, present and future. Psychosomatic Medicine, 55:234-

247.

Lazarus, R. S., & Folkman, S. (1984). Stress appraisal and coping. Springer, New York.

Morgan, B. B., & Alluisi, E. A. (1972). Synthetic work: Methodology for the assessment of human perfor-

mance. Perceptual & Motor Skills, 35:835-845.

Morgan, B. B., Jr. & Bowers, C. A. (1995). Teamwork stress: Implications for team decision making. Jn:

Guzzo & Salas (Eds.) Team Effectiveness and Decision Making in Organizations. Jossey Bass Publishers,

San Francisco

Morgan, B. B., Coates, G. D., Brown, B. R., & Alluisi, E. A. (1974). Effects of continuous work and

sleep loss on the recovery of sustained performance. JSAS Catalogue of Selected Documents in Psychology,

4:104-105.

Morgan, B. B., Jr., Glickman, A. S., Woodard, E. A., Blaiwes, A. S., & Salas, E. (1986). Measurement

INDIVIDUAL DIFFERENCES 93

of team behaviors in a Navy environment (Tech. Rep. No. NTSC TR-86-014). Orlando, Fl: Naval Training

Systems Center.

Morgan, B. B., Jr., Winne, P. S., & Dugan, J. (1980). The range of consistency of individual differences in

continuous work. Human Factors, 22(3):331-340.

Rosenbaum, M. (1980). Individual differences in self-control] behaviors and tolerance of painful stimulation.

Journal of Abnormal Psychology, 89:581-590.

Rosenbaum, M. (1980). A schedule for assessing self-control behaviors: preliminary findings. Behavior

Therapy, 11:109-121.

Rosenbaum, M. (1983). Learned resourcefulness as a behavioral repertoire for the self-regulation of internal

events: Issues and speculations. Jn:M. Rosenbaum, C. M. Franks, & Y. Jaffe (Eds.), Perspectives on behavior

therapy in the eighties. (pp. 54-73), Springer, New York.

Rosenbaum, M. (Ed.) (1990). Learned resourcefulness: On coping skills, self-control and adaptive behavior.

Springer, New York

Rosenbaum, M., & Rolnick, A. (1983). Self-control behaviors and coping with sea-sickness. Cognitive

Therapy and Research, 7:93-98.

Ryan, D. E. (1961). Motor performance under stress as a function of the amount of practice. Perceptual and

Motor Skills, 13:103-106.

Salas, E., Dickinson, T. L., Converse, S. A., & Tannenbaum, S. I. (1992). Toward and understanding of

team performance and training. In: R. W., Swezey & E. Salas (Eds.), Teams: their training and performance.

Alex Publishing Corporation, Norwood.

Serfaty, D., Entin, E. E., & Volpe, C. (1993). Adaptation to stress in team decision making and coordination,

In: Proceedings of the 37th Annual Meeting of the Human Factors and Ergonomics Society, (pp. 1228-

1232).Human Factors and Ergonomics Society, Santa Monica, CA.

Sherrod, D. R., Hage, J., Halpern, P. L., & Moore, B. S. (1977). Effects of personal causation and perceived

control on responses to an aversive environment. Journal of Experimental Social Psychology, 13:14-27.

Spector, P. E. (1986). Perceived control by employees: A meta-analysis of studies concerning autonomy and

participation at work. Human Relations, 39:1005-1016.

Spielberger, C. D., Gorusch, R. L., & Lushene, R. E. (1970). Manual for the State-Trait Anxiety Question-

naire. Consulting Psychologists Press, Palo Alto, CA.

Spielberger, C. D., Vagg, P. R., Barker, L. R., Donham. G. W., & Westberry, L. G. (1980). The factor

structure of the State-Trait Anxiety Inventory. Jn: I. G. Sarason & C. D. Spielberger (Eds.), Stress and

anxiety. (Vol. 7.), Hemishere/Wiley. New York.

Staub, E., Tursky, B., & Schwartz, G. (1971). Self-control and predictability: The effects on reactions to

aversive stimulation. Journal of Personality and Social Psychology, 18:157-162.

Thompson, S. C. (1981). Will it hurt if I can control it? A complex answer to a simple question. Psychological

Bulletin, 90:89-101.

Urban, J. M., Bowers, C. A., Monday, S. D. & Morgan, B. B., Jr. (1995). Workload, team structure, and

communication in team performance. Military Psychology, 7:123-139.

Weaver , J. L., Bowers, C. A. & Morgan, B. B., Jr. (1994). TANDEM: An empirical test of ambiguity and

time-pressure as task parameters. In: Proceedings of the Applied Behavioral Sciences Symposium. Colorado

Springs:,CO.

Weaver, J. L., Morgan, B. B., Jr., Adkins-Holmes, C. & Hall, J. (1992). A review of potential moderating

factors in the stress-performance relationship. (Tech. Rep. No. NTSC TR-92-012). Orlando, Fl: Naval

Training Systems Center.

Weaver, J.L., Morgan, B.B., Jr., Hall, J.A. & Compton, D.A. (1994). Team decision making in the command

information center: Understanding and reducing the effects of teamwork stressors. Technical Report submit-

ted to Naval Training Systems Center, Orlando, FL.

Weitz, J. (1970). Psychological research needs on the problems of human stress. Jn: J. E. McGrath (Ed.),

Social and psychological factors in stress. Holt, Rinehart, and Winston, New York.

Wilkinson, R. T. (1974). Individual differences in response to the environment. Ergonomics, 17:745-756.

Journal of the Washington Academy of Sciences,

Volume 84, Number 2, 94-110, June 1996

What’s Happened Since Project A:

The Future Career Force’

Michael G. Rumsey

U.S. Army Research Institute for Behavioral and Social Science

Alexandria, VA 22333-5600

Norman G. Peterson and Scott H. Oppler

American Institutes for Research

John P. Campbell

Human Resources Research Organization

Received April 30, 1996

ABSTRACT

In 1982, a program of research was initiated to improve the Army’s selection, classification,

reenlistment and promotion systems. This program was conducted in two phases. The first

phase, known as Project A, was completed in 1989; the second, known as Building the Career

Force, in 1995. During Project A, a concurrent validation was completed which provided

information on the structure of initial entry performance and the relationship between a

variety of predictor constructs and performance dimensions. Also, during Project A, an effort

was initiated to link these predictors with both entry level and early supervisory performance

in a longitudinal fashion. This paper describes the results of this longitudinal validation,

completed under the Career Force project. It describes the generation of a model of early

supervisory performance. It shows how the powerful relationship between cognitive aptitude

and job proficiency demonstrated in Project A held across time, organizational levels and

cohorts, but that concurrent relationships between temperament and performance dimensions

of a ‘‘will do’’ nature declined over time. Relationships between early and later career

performance supported the notion that past performance predicts future performance on

comparable performance dimensions. Overall, Career Force has considerably extended our

knowledge of the nature of performance, of the relationship between individual differences

' The research reported here was funded by the U.S. Army Research for the Behavioral and Social Sciences,

Contract No. MDA903-89-C-0202. This paper was presented March 7, 1996 in Arlington, Virginia at the

meeting ‘‘Emerging Issues in Individual Differences,’’ sponsored by the American Psychological Association

Divisions 19 and 21 and the Potomac Chapter of the Human Factors and Ergonomics Society. All statements

expressed in this paper are those of the authors and do not necessarily reflect the official opinions or policies

of the Army Research Institute or the Department of the Army.

FUTURE CAREER FORCE 95

and job performance, and of the comparison of validity results obtained from a longitudinal

design with those obtained from a concurrent design.

In 1982, Project A was launched. It was an effort to improve the Army’s

selection and classification system, but that statement does not fully capture its

impact and scale. It was described by Hakel in the 1986 Annual Review of

Psychology as the ‘‘most significant effort in the measurement and interpretation

of human differences yet undertaken (p. 373).’’ Its major accomplishments were

summarized in a 1990 special issue of Personnel Psychology, an issue which

earned its contributors a scholarly achievement award from the Academy of

Management.

Now, six years after the publication of that special issue, it is time to reveal

what has happened since Project A. For those who thought the story ended with

the completion of Project A, it may come as a surprise to learn that this project

was just the first stage of a two-part Soldier Selection research program and that

the second stage, known as Building the Career Force, was just recently com-

pleted. In the course of this follow-up effort, a longitudinal design has been

completed which confirmed some of the Project A results and expanded upon

others. Some of the more interesting questions that could not be answered in

1990 can now be addressed. These questions include the following:

1. How do the components of entry level performance differ from those of first level

supervisory performance?

2. How well do measures of individual differences predict performance at different

career stages?

3. How do concurrent and longitudinal validation results based on essentially the same

predictor and criterion measures differ?

4. How well does early career performance predict later performance?

Before we explore the answers to these questions, we need to go back in time

to describe the circumstances that led to the Soldier Selection projects. At the

time of the initiation of Project A, there was considerable concern about the

qualifications of entering enlisted soldiers. Problems concerning the norming of

the Armed Services Vocational Aptitude Battery, or ASVAB, used to select and

classify all enlisted soldiers, had surfaced, indicating that those entering the Army

between 1976 and 1980 represented a lower segment of the population on this

measure than originally believed. Congress wanted to know what this meant in

terms of job performance (Shields & Hanser, 1990).

The Soldier Selection research program was launched as part of a Joint Service

effort to link enlistment standards to job performance. However, this program

went further than that. A variety of new equipment systems were soon to be

fielded, and there was a concern that soldiers be qualified to effectively use these

96 RUMSEY ET AL.

systems. There was interest not just in the verbal, quantitative, and information

types of measures represented in the ASVAB, but in spatial, psychomotor, tem-

perament, and interest measures as well.

Thus, the first objective of the research program was to link selection to job

performance by developing new entry tests and linking them and the ASVAB to

new performance measures. The second objective was to link selection, classifi- —

cation, and job performance to career progression in order to maximize perfor-

mance at higher levels as well as at initial entry.

Project A, which lasted from 1982 to 1989 and focused on initial entry perfor-

mance, was the first project in this research program. Project A looked at the

relationship between the ASVAB and job performance. Also in Project A, new

tests were developed and concurrently linked to performance.

Materials and Methods: Soldier Selection Research

Design: Overview

Two designs, as we will examine more closely later, were initiated in Project

A. One was concurrent and one was longitudinal. The results that have been

reported from Project A have all been from the concurrent design. When Project

A was completed, data from the longitudinal data collections had been partially

collected but not analyzed nor reported. Thus, for purposes of convenience, we

will associate all findings from the longitudinal sample with Career Force.

The concurrent design involved the concurrent collection of data from the new

predictors and performance measures on 9430 soldiers who had typically served

in the Army for a period of 12 to 24 months. A smaller sample of 4039 was

used in the validity analyses which will be discussed later. The amount of missing

data and the jobs from which the data were drawn were two factors determining

which cases were used in this smaller sample (Young, Houston, Harris, Hoff-

man, & Wise, 1990). In addition to the data that were collected in the field during

the course of this project, ASVAB data which had been collected at entry were

used in the analyses. Thus, although we will continue to refer to this as the

concurrent validation, it was more precisely a mixed design, with the ASVAB

linked to job performance in a longitudinal fashion (the overall design for the

Project A/Career Force projects is described in Campbell & Zook, 1994c).

Twenty-two jobs were examined in Project A and Career Force. However, for

a substantial subset of these jobs, only limited criterion information was collected.

This was done by design, so that it would be possible to compare analyses based

on a partial set of criterion measures with analyses based on a more comprehen-

sive set. For the most part, the analyses that will be examined in this paper were

FUTURE CAREER FORCE 97

based on nine jobs: infantryman, cannon crewman, tank crewman, radio teletype

operator, vehicle generator mechanic, administrative specialist, motor transport

operator, medical care specialist, and military police. These nine jobs represent

a reasonable cross-section of the full set of the hundreds of Army jobs. They

include combat jobs and combat support jobs. They include jobs which reasonably

represent African Americans and females, as well as those combat jobs which

females are by policy excluded from. They were by necessity jobs which were

heavily populated so that statistical analyses could reasonably be conducted (see

Campbell, 1987, for a discussion of the process of selecting jobs for this research

effort).

Predictor Measures

The purposes of this research dictated that the list of entry predictors examined

would include ASVAB. The ASVAB consists of ten subtests, which are combined

into ten different composites. Factor analyses have consistently identified four

factors: verbal, quantitative, technical, and speed (Kass, Mitchell, Grafton, &

Wing, 1983; Peterson, et al., 1990).

The approach to identifying additional predictors incorporated the following

elements. First, there was no attempt to replicate the content domain of the

ASVAB. The focus was on content categories which could best supplement

the ASVAB in predicting job performance. A comprehensive literature review,

followed by extensive pre-testing, resulted in the generation of 62 measures,

which were reduced to 27 composites in the following categories: spatial, com-

puter, temperament/biographical, job preferences, and career interests. Six spatial

tests, such as Map, Maze, and Assembling Objects, eventually reduced to a single

composite. Psychomotor and perceptual attributes were assessed through eight

computer test composites. A variety of temperament scales, such as conscientious-

ness and cooperativeness, were reduced to seven composites in an instrument

named the Assessment of Background and Life Experiences, or ABLE. Prefer-

ences for particular types of job outcomes, such as job security and autonomy,

were measured in an instrument known as the Job Orientation Blank (JOB).

Career interests were measured in the Army Vocational Interest Examination, or

AVOICE. This was based on an Air Force instrument known as the VOICE

(Peterson, et al., 1990).

Entry Level Performance Measures

One of the prominent features of Project A was its historically comprehensive

treatment of the criterion space. Performance measures included hands-on tests,

98 RUMSEY ET AL.

job knowledge tests, rating scales, and administrative measures. A series of

exploratory and quasi-confirmatory analyses resulted in the identification of five

substantive factors. The first two dimensions, Core Technical Proficiency and

General Soldiering Proficiency, have been described as ‘‘can do’’ dimensions,

and assess individual job proficiency. They were based on hands-on and job

knowledge tests. The last three, Effort and Leadership, Personal Discipline, and

Physical Fitness and Military Bearing, tended to be more ‘‘will do’’, or motiva-

tional in nature. They were based on ratings and self-report of information con-

tained in administrative records, such as awards and certificates and physical

readiness scores (Campbell, McHenry, & Wise, 1990).

Project A Results: Brief Summary

We will not attempt to comprehensively describe here the results from Project

A. These will be summarized later in this paper in the context of a comparison

of Project A and Career Force findings. However, a few points will be mentioned

here. First, ASVAB was found to have impressively high validity for predicting

the ‘‘can do’’ performance dimensions. The spatial tests added a small increment

to the prediction of these dimensions. The findings with respect to the “‘will do’’

dimensions were somewhat different. ASVAB was but a moderately effective

predictor of these dimensions. Adding composites from the ABLE, the tempera-

ment instrument, to the ASVAB enhanced the prediction of “‘will do’’ perfor-

mance substantially.

Career Force: Materials and Methods

As the title of this paper indicates, its principal focus is what has happened

since Project A. The follow-up effort, Career Force, lasted from 1989 to 1995

and focused on early supervisory performance. While Project A told us something

about both ASVAB and the new predictors, Career Force was designed to tell

us how to best combine ASVAB and the new predictors for selection and classifi-

cation. Also, by following soldiers into their second tour, it has given us a

basis for building an improved noncommissioned officer (NCO) corps through

selection, classification, reenlistment, and promotion decisions.

Subjects

Now that Career Force is over, we can complete the picture that was partially

painted in Project A. Results from four new data collection points have been

added. First, soldiers from the concurrent validation were followed into the second

FUTURE CAREER FORCE 99

tour. The remainder of the results, and those which we will concentrate on here,

are from the longitudinal validation. A total of 49397 soldiers were tested at

entry in 1986 and 1987 on the new predictors described earlier. Performance was

measured at three points—end of training, in the soldier’s first tour, and three

to six years after entry, which in many cases represented the soldier’s first supervi-

sory assignment, although a large number of this group still had not reached

NCO status. Training data were collected on 34305 soldiers (Campbell & Zook,

1991). Then 11266 were tested about 12-24 months after entry on first tour

performance in 1988 and 1989 (Campbell & Zook, 1990). The first tour perfor-

mance measures were those described earlier. Finally, in 1991 and 1992, 1595

soldiers were followed into the second tour, where performance measures were

again administered (Campbell & Zook, 1994b).

In a world ideally suited for this research project, each soldier would have

been tested at each stage—entry, end of training, first tour and second tour.

However, soldiers who had been tested at earlier stages were not always available

for testing at later stages, and there was a compelling need to have large sample

sizes for certain analyses which did not involve linking measures across all career

stages. Thus, at each stage beyond initial entry, the data collection strategy

involved a series of compromises between maximizing total sample size and

maximizing the number of soldiers who met all project criteria, and the number

of soldiers who were ultimately available for any particular analysis was a func-

tion of how this strategy was executed.

For the analyses reported here, sample sizes were constrained by two considera-

tions. First, analyses were limited to those jobs for which the full set of criterion

measures were administered. For entry level jobs and for end of training, these

were, for all practical purposes, the nine jobs listed above. For second tour analyses,

only seven jobs were included, with radio teletype operator and motor transport

operator excluded because of the small sample sizes available for these jobs.

The second constraint was that, for the end of training validity analyses, for

which 4039 cases were examined, and for entry level validation, where the total

sample size for analysis was 3090, soldiers were required to have complete

predictor and criterion data. Complete predictor data were not required for the

analyses linking entry predictors with first level supervisory performance mea-

sures. For those analyses, sample sizes varied from 810 to 1224 depending on

the predictor set being evaluated.

Training Performance Measures

Two types of measures of training performance were administered—knowl-

edge tests and rating scales. These measures yielded dimensions very similar to

those identified in first tour. Just as there was a core technical first tour dimension,

100 RUMSEY ET AL.

there was a technical school knowledge dimension based on written job specific

test items. School knowledge basic, based on written common task test items,

was comparable to general soldiering proficiency. The will do dimensions, based

on ratings, were comparable to the will do first tour dimensions, with the addition

of one new “‘will do’’ dimension in the training environment, “‘leadership poten-

tial (Campbell & Zook, 1990).”’

Entry Level Performance Measures

The entry level performance measures administered in the longitudinal valida-

tion (““Career Force’’ measures) were virtually the same as those administered

in the concurrent validation. The model of first tour performance generated in

the longitudinal validation replicated the concurrent validation model (Camp-

bell & Zook, 1994a).

Supervisory Level Performance Measures

As for the entry level, supervisory level measures included hands-on and job

knowledge tests, ratings, and administrative measures. In addition, they included

a written Situational Judgment Test requiring a response to a supervisory situation,

and special ratings provided in the context of training and counseling role play

exercises. These measures were grouped into six dimensions. Four dimensions,

core technical proficiency, general soldiering proficiency, personal discipline, and

physical fitness/military bearing, were strikingly similar to their identically-named

entry level counterparts. A fifth dimension, achievement and effort, was also

similar to a first tour dimension, effort and leadership. The principal difference

was the addition of a supervisory level sixth dimension, leadership, represented

by leadership performance ratings, performance on the role play exercises, and

scores on the Situational Judgment Test. The supervisory model presents, rather

than a can do and will do division, a can do, will do and leadership division

(Campbell & Zook, 1994b).

Analyses

Multiple correlations between each set of predictor scores and the substantive

factor scores were computed for training performance, entry level performance,

and first level supervisory performance. Results were (1) corrected for multivari-

ate range restriction (Lord & Novick, 1968) on the ASVAB subtests using the

intercorrelation matrix among the subtests in the 1980 Youth Population (Depart-

ment of Defense, 1982), and (2) adjusted for shrinkage using Rozeboom’s (1978)

Formula 8. Results were computed separately by job and then averaged.

Incremental validities for each additional predictor set over the ASVAB were

FUTURE CAREER FORCE 101

Table 1.—Average Multiple Rs and Increments in Multiple Rs over ASVAB for Predictor Sets Used

in End of Training Sample

Criterion ASVAB Spatial Computer JOB ABLE AVOICE

Tech Know. 76 63 (01) 61 (01) 41 33 44

Basic Know. 68 57 (01) 57 38 30 37

Eff/Tech Skill 41 35 (01) 36 (01) 24 19 (03) 22

Leadership 30 24 28 (01) 18 22 (05) 17

Discipline Zs 22 21 09 19 (09) 11

Fitness/Brg 14 05 11 (03) 05 (01) 29 (17) 07 (01)

Note: All values in Tables 1—5 should be interpreted as decimals but are presented as integers for ease of

reading (e.g., the first value in the ASVAB column is actually 0.76, but is presented as 76. Numbers in

parentheses represent incremental value for that predictor set over ASVAB. For example, the first value in the

Spatial column is 0.63. Spatial tests provide an increment of .01 over the ASVAB multiple R of 0.76. Thus,

when spatial tests are combined with the ASVAB, the multiple R increases to 0.77. Only incremental values

exceeding zero are shown. Data presented are from Building and Retaining the Career Force: New Procedures

for Accessing and Assigning Army Enlisted Personnel: Annual Report, 1991 Fiscal Year by J. P. Campbell

and L. M. Zook (Eds.), pages 41 and 49. Adapted with permission.

then computed, using the same range restriction and shrinkage adjustments, and

averaging across jobs as before.

Career Force Results

Now let us turn to the Career Force validity results, beginning with the predic-

tion of training performance. These results are summarized in Table 1. Multiple

regression coefficients are provided for each predictor composite linked with

each criterion dimension. There is a pattern that began to emerge in Project A

which, with some variations, we will find to pervade the Career Force results as

well. First, there was the impressive strength of the cognitive measures in pre-

dicting can do performance. This was most evident with respect to the ASVAB,

although the predictive validities of the spatial and computer composites were

nearly as high.

The cognitive validities dropped for the will do dimensions. The validities of

the noncognitive predictors, JOB, ABLE and AVOICE, also dropped from can

do to will do dimensions. ASVAB was the best predictor of all dimensions except

fitness and bearing, where ABLE emerged as the strongest predictor.

Table 1 also shows incremental validities for each predictor set over ASVAB.

While the ABLE validities tended to be modest across all dimensions, ABLE

composites collectively clearly emerged as the instrument providing the greatest

incremental validity over the ASVAB for all will do dimensions, particularly

fitness and bearing (Campbell & Zook, 1994a).

Now we look at the validities and incremental validities for entry level perfor-

102 RUMSEY ET AL.

Table 2.—Average Multiple Rs and Increments in Multiple Rs over ASVAB for Predictor Sets Used

for Entry Level Sample

Criterion ASVAB Spatial Computer JOB ABLE AVOICE

Core Tech 62 57 (01) 47 29 21 38

Gen. Soldier 66 64 (02) 55 29 23 37

Effort/Ldrship 37 3) 29 18 13 17

Discipline 17 14 10 06 14 (06) 05

Fitness/Brng 16 10 07 06 (01) 27 (14) 05

Note: Numbers in parentheses represent incremental value for that predictor set over ASVAB. Only incremen-

tal values exceeding zero are shown. Adapted from Building and Retaining the Career Force: New Procedures

for Accessing and Assigning Army Enlisted Personnel: Annual Report, 1991 Fiscal Year by J. P. Campbell

and L. M. Zook (Eds.), pages 164 and 171. Adapted with permission.

mance, shown in Table 2. The pattern was quite similar to that for training perfor-

mance, with the cognitive measures providing impressive prediction of can do

performance and all measures generally providing better prediction of can do than

will do performance. In general, the entry level validities tended to be somewhat

lower than the training performance validities. Note that the ABLE composites still

provided the most incremental validity over ASVAB, although the increments were

smaller than in the case of training performance (Campbell & Zook, 1994a).

At the completion of Career Force, we had the opportunity to compare concur-

rent and longitudinal validities for prediction of initial entry performance. In

order to make these comparisons, it was necessary to recompute the longitudinal

validities using the Claudy (1978) adjustment for shrinkage, as this was the

procedure used in Project A for computing the concurrent validities. Again, we

should note that the administration of the ASVAB was necessarily longitudinal

in both cases, while for all other measures the first administration was concurrent.

The comparisons are presented in Table 3. Because of the use of the Claudy

adjustment for these analyses, the validities shown for the longitudinal validation

are slightly different from those shown in Table 2. The most remarkable feature

of Table 3 is the consistency across both data collections for the cognitive tests.

The spatial and computer tests generally predicted as well, or almost as well, in the

longitudinal as in the concurrent administration. Curiously, all cognitive measures

predicted effort and leadership more effectively in the longitudinal than in the

concurrent administration.

Now consider the same comparison for the non-cognitive measures. These

measures tended to predict can do performance equally effectively in the concur-

rent and longitudinal validations. This was not the case, however, for prediction

of will do performance. Here there was a consistent decline, particularly for

the temperament instrument, ABLE. The incremental validities for ABLE over

ASVAB also show a consistent decline in the longitudinal relative to the concur-

FUTURE CAREER FORCE 103

Table 3.—Comparison of Entry Level Concurrent (Project A (Prj. A)) and Longitudinal (Career Force

(CF)) Validities

Cognitive Predictors

ASVAB Spatial Computer

Criterion Prj. A CF Prj. A CF Prj. A CF

Core Tech. 63 63 56 57 53 50

Gen. Soldier 65 67 63 64 57 57

Effort/Ldrship Si) 39 Zs a2 26 34

Discipline 16 22 12 14 12 15

Fitness/Brmg 20 21 10 10 1] 17

Non-Cognitive Predictors

JOB ABLE AVOICE

Criterion Prj. A CF Prj. A Cr Pry. A CF

Core Tech. 29 31 26 27 35 4]

Gen. Soldier 30 32 25 29 34 40

Effort/Ldrship 19 22. 33 20 24 25

Discipline 11 a! 32 22: 13 11

Fitness/Bmg 11 12 Sih 31 12 15

Note: Adapted from Building and Retaining the Career Force: New Procedures for Accessing and Assigning

Army Enlisted Personnel: Annual Report, 1991 Fiscal Year by J. P. Campbell and L. M. Zook (Eds.), page

179. Adapted with permission.

rent validation. This decline is particularly evident on the effort and leadership

dimension (Campbell & Zook, 1994a).

Finally, we look at the validity of entry-level measures for predicting perfor-

mance three to six years in the future. There is a fairly pervasive point of view

that the longer a person is on the job, the less important test scores administered

at the point of entry become. We now have results, shown in Table 4 (Campbell &

Zook, 1994c), that suggest that this point of view needs to be re-examined. All

predictors administered at entry predicted job specific technical proficiency in

the second tour as well or better than they predicted this measure in the first tour.

There was a decrement, but only a small one, with respect to general soldiering

proficiency. On the other hand, there were several instances where validities

for predicting will do performance declined, particularly in the prediction of

achievement and effort.

Another striking finding was the strength of all entry measures for predicting

the leadership dimension. The pattern of these relationships was very similar to the

pattern of validities with respect to can do performance. However, the measures of

leadership were very different from the measures of technical proficiency.

As Table 4 shows, once ASVAB was accounted for, the additional predictive

validity contributed by any entry predictor to second tour performance was very

104 RUMSEY ET AL.

Table 4.—Average Multiple Rs and Increments in Multiple Rs over ASVAB for First Level Supervisory

Sample

Criterion ASVAB Spatial Computer JOB ABLE AVOICE

Core Tech. 64 57 53 33 24 41

Gen. Soldier 63 58 (01) 48 28 19 29

Ach/Effort 29 27 (02) 09 07 13 09

Leadership 63 55 49 34 34 (01) 35

Discipline 15 15 12 03 06 06

Fitness/Brmg 16 13 03 07 17 (05) 09

Note: Numbers in parentheses represent incremental value for that predictor set over ASVAB. Only incremen-

tal values exceeding zero are shown. Adapted from Building and Retaining the Career Force: New Procedures

for Accessing and Assigning Army Enlisted Personnel: Annual Report, 1993 Fiscal Year, by J. P. Campbell

and L. M. Zook (Eds.), pages 132, 140, and 141. Adapted with permission.

limited. ABLE still provided some incremental validity to the prediction of fitness

and bearing, and added one point to leadership, but these increments were reduced

from those observed for first tour criteria.

Now we look at another question: How well does early career performance

predict later career performance? We had an opportunity to look at this question

twice, once in the concurrent validation, which we will refer to here as the

**Project A validation sample,’’ and once in the longitudinal validation, which

will be referred to as the ‘“Career Force validation sample.’’ Table 5 shows the

results from the Career Force sample, corrected for range restriction. The bivariate

correlational values along the diagonal, where similar first and second tour perfor-

mance dimensions were linked, were reasonably high, representing good conver-

gent validity. The off-diagonal values tended to be lower, often substantially

lower, representing good discriminant validity.

The results from Project A, where the samples were smaller (n = 102-121

Table 5.—Prediction of Performance from Prior Performance: Career Force Validation Sample

First Supervisory Level

Entry Level CT; GS? AES Disc* FB*

Core Tech 44, oul 28 —04 08

Gen. Soldier 41 57 34 04 —01

Effort/Ldrship 25 jp) 47 2 22

Discipline 08 09 29 26 14

Fitness/Bmg 02 —01 26 NG 46

Note: Sample sizes for bivariate correlations ranged from 322 to 412. Diagonal values are shown in boldface.

Off-diagonal values which are larger than the corresponding diagonal value are underlined. From Building and

Retaining the Career Force: New Procedures for Accessing and Assigning Army Enlisted Personnel: Annual

Report, 1993, by J. P. Campbell and L. M. Zook (Eds.), page 161. Adapted with permission.

* Core Technical Proficiency ° General Soldiering Profociency ‘ Achievement and Effort “ Personal Discipline

“ Fitness and Bearing

FUTURE CAREER FORCE 105

for the bivariate correlations), were similar. Here, again, a correction for range

restriction was used. Values along the diagonal ranged from 0.26 (discipline) to

0.48 (fitness and bearing). Two off-diagonal values involving comparison be-

tween general soldiering with core technical proficiency were particularly high

(entry level general soldiering with supervisory core technical proficiency r =

0.47; entry level core technical proficiency with supervisory general soldiering r

= 0.48), as well as one between entry level general soldiering and supervisory

effort and achievement (r = 0.36), but otherwise off-diagonal values were consis-

tently lower than diagonal values (Campbell & Zook, 1994c).

Discussion

Four Research Questions

Now let us return to the questions posed at the beginning of this paper. First,

the question of components of performance: How do the models of performance

compare at early and later stages in the soldier’s career? The results suggest that

the model for initial entry performance represents a foundation upon which the

soldier can build by adding new responsibilities at the first supervisory stage.

The old responsibilities do not disappear: the technical and motivational dimen-

sions remain important. However, a new leadership dimension appears, adding

to the scope of the soldier’s responsibilities.

Next is the question of how well individual difference measures predict perfor-

mance at different career stages. The most striking results were obtained with the

cognitive measures. We knew prior to Project A that cognitive measures predicted

soldier performance at the early, training stage from such research as that conducted

by Maier and Fuchs (1973), and we began to appreciate in the course of Project

A the strength of these measures for predicting job performance in the soldier’s

initial tour (McHenry, Hough, Toquam, Hanson, & Ashworth, 1990). We now

have information that strengthens this conclusion and which, additionally, shows

that these measures retain their predictive power three to six years out, when the

individual has generally assumed a higher level of responsibility.

The issues associated with non-cognitive measures are more complex. These

measures did not predict performance at a particularly high level, but the tempera-

ment measure ABLE more consistently provided incremental validity in combina-

tion with ASVAB than any other new predictor. Both the ABLE’s zero order

validities and incremental validities with respect to will do performance declined

over time, but this decline was not observed with respect to technical proficiency

criteria. We believe that temperament measures and biographical measures which

tap temperament constructs have promise for predicting performance, and Project

106 RUMSEY ET AL.

A and Career Force have stimulated follow-on research efforts by the Army

Research Institute (ARI) to explore this promise in greater depth.

The third question is: How do concurrent and longitudinal results differ? The

biggest surprise is how little they differed in this research, particularly when

cognitive predictors were at issue. The decline in predictive validity for the non-

cognitive predictors in the longitudinal validation may be a subset of the larger

issue of the decrement in the validity of such measures over time. Although many

explanations have been explored for this decline (Campbell & Zook, 1994a;

White & Moss, 1995), there does not appear to be a single one which totally

accounts for this phenomenon.

The fourth and final question is how well early career performance predicts

later performance. The generally high level of convergent validity is encouraging

and consistent with expectations. The consistency of performance over time has

potential payoffs in informing promotion decisions. The value of performance

information is likely to be greatest with respect to those will do dimensions that

are so difficult to predict by other means.

Additional Implications

A few other results from this research are worth noting. We have noted the

predictive power of cognitive measures in several contexts. There is a remarkable

consistency of the validity of cognitive measures in at least three respects: across

time, across organizational levels, and across cohorts. The results have supported

ASVAB as an initial entry selection tool and have provided evidence that it has

perhaps greater value as a predictor of leadership performance than previously

suspected.

Also, this research has shown that spatial tests can add small, pervasive incre-

ments for “‘can do’’ criteria. Despite the predictive power of the current ASVAB,

the Services have agreed to add a new spatial test from Project A, Assembling

Objects, to this joint service battery, in part because of evidence of its incremental

validity. Not since the ASVAB was introduced in 1976 has another test presenting

a new content category passed the many hurdles required for such a decision.

The Career Force results, in combination with the Project A results, have been

used by Campbell and others (Campbell, 1994; Campbell, McCloy, Oppler, &

Sager, 1992) to define a generalized multiple factor model of performance. The

Career Force results confirmed the factors identified in Project A and expanded

upon them to incorporate leadership.

The Career Force findings also provide guidance to those facing the challenge

of whether to be satisfied with a concurrent validation design or whether to invest

the additional time and resources required to conduct a longitudinal design. They

provide some support for using a concurrent design to link cognitive predictors

FUTURE CAREER FORCE 107

with proficiency criteria. As the predictors and criteria get softer, however, the

presumption that concurrent validities can substitute for longitudinal gets more

difficult to defend.

The Career Force results have tempered somewhat our enthusiasm for tempera-

ment measures, but they have by no means caused us to close the door on such

measures. At ARI, we have investigated temperament and biographical measures

in a variety of contexts, and have consistently found encouraging results (Mael &

White, 1994; White & Kilcullen, 1992). However, the results here suggest caution

in using temperament measures to make long-term predictions. We have begun

to conduct investigations examining results by item type. White and Moss (1995)

found that for items judged as likely to be influenced by one’s experiences in

the Army, concurrent validities were higher than predictive validities, but for

items classified as low on this dimension of organizational influence, concurrent

and predictive validities were comparable.

The encouraging concurrent validities for ABLE from Project A should not

be disregarded, as they may be particularly relevant when one is interested in

obtaining measures relating to one’s current job performance. The White and

Moss findings suggest that careful attention to the linkage between temperament

items and the organizational context in which such items are administered could

result in temperament instruments with enhanced long-term utility.

Some of the most interesting findings from this research have to do with the

prediction of junior level leadership. To fully appreciate these findings, we need

to first examine how leadership was measured in this project. Leadership was

identified as a separate factor, rather than encompassing all tasks the leader

performed. Multiple methods of measuring leadership were employed, including

ratings of job performance on leader dimensions, ratings of performance on role

play exercises, and written responses to hypothetical leadership situations. The

measures were based on extensive, exhaustive, and multi-method job analyses.

To say that they were carefully developed is an understatement.

The findings indicated that, to some extent, everything predicted leadership.

Both cognitive and non-cognitive measures did so, at levels from 0.34 to 0.63.

The most impressive predictor of leadership was a cognitive test battery, the

ASVAB. ASVAB not only predicted leadership well, it predicted it at virtually

the same level as it predicted job proficiency. It is probably unwise to draw too

many conclusions from a single finding, particularly when many soldiers in

the supervisory sample in this research were not formally required to exercise

supervisory responsibilities, but this finding does at least suggest that, at the

first supervisory level, leadership is perhaps more closely linked to conventional

measures of cognitive aptitude than might have been suspected.

The findings with respect to performance as a predictor are closely tied with

108 RUMSEY ET AL.

the performance modeling results. It is not enough to say that past performance

predicts future performance—rather past performance on a particular dimension

predicts future performance on that dimension, but not necessarily on other di-

mensions.

Finally, what are the implications for Army personnel selection and manage-

ment? First, we have re-affirmed that the ASVAB is a good test, in the sense

that it does what it is supposed to do—predict performance—and that the Army

should continue to use it. Project A generated this finding; Career Force confirmed

it. Second, we have generated further evidence of the incremental value of spatial

tests in predicting technical performance, a finding that complements research

showing that spatial and psychomotor tests are good predictors of such specialized

military tasks as tank and anti-tank gunnery (Busciglio, Silva, & C. Walker,

1990; Grafton, Czarnolewski, & Smith, 1988; Graham, 1988; Silva, 1989;

Smith & Graham, 1987; Smith & M. Walker, 1988). These results, in conjunction

with results from a joint service project (see, for example, Wolfe, Alderton,

Larson & Held, 1995 for a discussion of this project) and from a Marine Corps

project (Carey, 1994) helped stimulate the Department of Defense to begin limited

administration of Assembling Objects in an operational context, with plans for

expanded administration in the future.

Third, we have developed a better understanding of the potential value of

temperament measures, which we have begun at ARI to evaluate for use in

limited operational contexts and are exploring for more extended use. Finally,

we have gained a better understanding of which measures of present performance

are promising candidates for use in promotion decisions, an understanding we

plan to apply in future ARI research on the enlisted promotion system.

Acknowledgments

The authors would like to express their appreciation to the individuals and

organizations that made this research possible. We wish to acknowledge the

contributions of those scientists and support personnel from the Army Research

Institute, Human Resources Research Organization, American Institutes for Re-

search, and the Personnel Decisions Research Institute and the Army soldiers

and sponsors who made this research possible. We would also like to express

special thanks to the Career Force Scientific Advisory Group members who

provided invaluable technical guidance and to those whose contributions to Proj-

ect A laid the groundwork for this effort.

References

Busciglio, H. H., Silva, J., & Walker, C. (1990). The potential of new Army tests to improve job performance.

Paper presented at the Army Science Conference, Durham, NC.

FUTURE CAREER FORCE 109

Carey, N. B. (1994). Computer predictors of mechanical job performance: Marine Corps findings. Military

Psychology, 6:1-30.

Campbell, J. P. (Ed.) (1987). Improving the selection, classification, and utilization of Army enlisted personnel:

Annual report, 1986 fiscal year (Technical Report 813101). Alexandria, VA: U.S. Army Research Institute

for the Behavioral and Social Sciences.

Campbell, J. P. (1994). Alternative models of job performance and their implications for selection and

classification. In: M. G. Rumsey, C. B. Walker, & J. H. Harris (Eds.), Personnel selection and classification.

Erlbaum, Hillsdale, NJ.

Campbell, J. P., McCloy, R. A.,’Oppler, S. H., & Sager, C. E. (1992). A theory of performance. Jn:

N. Schmitt & W. Borman (Eds.), New developments in selection and placement. Jossey-Bass, San

Francisco, CA.

Campbell, J. P., McHenry, J. J., & Wise, L. L. (1990). Modeling job performance in a population of jobs.

Personnel Psychology, 43:313-333.

Campbell, J. P., & Zook, L. M. (Eds.). (1990). Building and retaining the career force: New procedures for

accessing and assigning Army enlisted personnel: Annual report, 1990 fiscal year (Technical Rep. No. 952).

Alexandria, VA: U.S. Army Research Institute for the Behavioral and Social Sciences.

Campbell, J. P., & Zook, L. M. (Eds.) (1991). Improving the selection, classification, and utilization of Army

enlisted personnel: Final report on Project A (Research Rep. No. 1597). Alexandria, VA:, U.S. Army

Research Institute for the Behavioral and Social Sciences.

Campbell, J. P., & Zook, L. M. (Eds.) (1994a). Building and retaining the career force: New procedures

for accessing and assigning Army enlisted personnel: Annual report, 1991 fiscal year (Research Note 94-

10). Alexandria, VA: U.S. Army Research Institute for the Behavioral and Social Sciences.

Campbell, J. P., & Zook, L. M. (Eds.). (1994b). Building and retaining the career force: New procedures

for accessing and assigning Army enlisted personnel: Annual report, 1992 fiscal year (Research Note No.

94-27). Alexandria, VA: U.S. Army Research Institute for the Behavioral and Social Sciences.

Campbell, J. P., & Zook, L. M. (Eds.). (1994c). Building and retaining the career force: New procedures for

accessing and assigning Army enlisted personnel: Annual report, 1993 fiscal year (Research Note No. 96-

45). Alexandria, VA: U.S. Army Research Institute for the Behavioral and Social Sciences.

Claudy, J. G. (1978). Multiple regression and validity estimation in one sample. Applied Psychological

Measurement, 2:595-601.

Department of Defense (1982). Profile of American youth: 1980 nationwide administration of the Armed

Services Vocational Aptitude Battery. Washington, DC: Office of the Assistant Secretary of Defense (Man-

power, Installations, and Logistics).

Graham, S. E. (1988). Selecting soldiers for the Excellence in Armor program. Paper presented at the annual

meeting of the Military Testing Association, Arlington, VA.

Grafton, F. C., Czarnolewski, M. Y., & Smith, E. P. (1988). Relationship between Project A psychomotor

and spatial tests and TOW gunnery performance (SCTA Working Paper RS-WP-87-10). Alexandria, VA:

U.S. Army Research Institute for the Behavioral and Social Sciences.

Hakel, M. D. (1986) Personnel selection and placement. Annual Review of Psychology, 37:351-380.

Kass, R. A., Mitchell, K. J., Grafton, F. C., & Wing, H. (1983). Factor structure of the Armed Services

Vocational Aptitude Battery, Forms 8, 9, and 10; 1981 Army applicant sample. Educational and Psychological

Measurement, 43:1077-1088.

Lord, F. M., & Novick, M. R. (1968). Statistical theories of mental test scores. Addison-Wesley, Reading,

MA.

Mael, F. A., & White, L. A. (1994). Motivated to lead: Dispositional and biographical antecedents of

leaders hip performance. Jn: H. O’Neil & M. Drillings (Eds), Motivation: Research and theory. Erlbaum,

Hillsdale, NJ.

Maier, M. H., & Fuchs, E. F. (1973). Effectiveness of selection and classification testing. (Research Report

No. 1179). Alexandria, VA: U.S. Army Research Institute for the Behavioral and Social Sciences.

McHenry, J. J., Hough, L. M., Toquam, J. L., Hanson, M. A., & Ashworth, S. (1990). Project A validity

results: The relationship between predictor and criterion domains. Personnel Psychology, 43:335-354.

Peterson, N. G., Hough, L. M., Dunnette, M. D., Rosse, R. L.., Houston, J. S., Toquam, J. L., & Wing,

H. (1990). Project A: Specification of the predictor domain and development of new selection/classification

tests. Personnel Psychology, 43:247-276.

Personnel Psychology (1990). Project A: The U.S. Army Selection and Classification Project (Special Issue),

43:2.

Rozeboom, W. W. (1978). Estimation of cross-validated multiple correlation: A clarification. Psychological

Bulletin, 85:1348-1351.

Shields, J. L., & Hanser, L. M. (1990). Designing, planning and selling Project A. Personnel Psychology,

43:241-245.

Smith, E. P., & Graham, S. E. (1987). Validation of psychomotor and perceptual predictors of Armor Officer

110 RUMSEY ET AL.

M-1 gunnery performance (Tech. Rep. No. 766). Alexandria, VA: U.S. Army Research Institute for the

Behavioral and Social Sciences.

Smith, E. P., & Walker, M. R. (1988). Testing psychomotor and spatial abilities to improve selection of

TOW gunners. Paper presented at the annual meeting of the Military Testing Association, Arlington, VA.

White, L. A., & Kilcullen, R. N. (1992). The validity of rational biodata scales. Paper presented at the

annual meeting of the American Psychological Association, Washington, DC.

White, L. A., & Moss, M. C. (1995). Factors influencing concurrent versus predictive validities of personality

constructs. Paper presented at the meeting of the Society of Industrial and Organizational Psychology,

Orlando.

Wolfe, J. H., Alderton, D. L., Larson, G. E., & Held, J. D. (1995). Incremental validity of Enhanced

Computer Administered Testing (ECAT) (TN-96-6). San Diego: Navy Personnel Research and Development

Center.

Young, W. Y., Houston, J. S., Harris, J. H., Hoffman, R. G., & Wise, L. L. (1990). Large-scale predictor

validation in Project A: Data collection procedures and data base preparation. Personnel Psychology,

43:301-311.

Journal of the Washington Academy of Sciences,

Volume 84, Number 2, Pages 111-129, June 1996

Gender Differences in Human Abilities:

Can Practice Moderate Results?

Janet J. Turnage

Star Mountain, Inc.

Robert S. Kennedy and Norman E. Lane

Essex Corporation, Inc.

ABSTRACT

The purpose of this review is threefold. First, we review the evidence regarding sex

differences in perceptual, cognitive, and motor skills in general. Then we focus on gender

differences in spatial abilities, an area where there are consistent and pervasive differences

between males and females. The nature of the differences are explored, including possible

reasons why these differences exist and whether differential performance on spatial tasks is

moderated by practice. Experimental evidence indicates some malleability of performance

through practice. These findings have implications for the future of women operating in an

increasingly technological world.

Cognitive Differences

Systematic attempts to draw conclusions about similarities and differences

between men and women achieved major prominence in the 1970’s with the

publication of Maccoby and Jacklin’s (1974) narrative account of past research,

‘The Psychology of Sex Differences’’, which maintained that the sexes differed

in several aspects of intellectual abilities (namely, verbal, quantitative, and spatial

abilities) and in aggression. After the mid 80’s, meta-analyses became the pre-

ferred method of research, providing a quantitative comparison between male

and female behavior across many studies in terms of effect size (or d) which

expresses the sex difference in units of the study’s standard deviation.

To understand whether sex-correlated differences are small, medium, or large,

it is helpful to keep in mind some general standards of comparison. Cohen’s

Corresponding Author: Janet J. Turnage, Star Mountain, Inc., 3601 Eisenhower Avenue, Suite 450, Alexan-

dria, VA 22304

111

112 TURNAGE, KENNEDY, AND LANE

Table 1.—Largest Sex Differences (Males Superior)

Throw Velocity 218

Throw Distance 1.98

Aggression 50

Mathematical Reasoning 44

Proportional Reasoning 44

Activity Level 48

Spatial ability 49

Spatial ability 45

Mental Rotation a3

Spatial visualization alle)

(1977) guidelines are expressed in terms of a difference, or d metric, where 0.20

is a small effect; 0.50 isa moderate effect; and 0.80 is a large effect. Medium

effect sizes correspond to group differences that would be apparent to the naked

eye, and large differences can be very readily perceived (Eagly, 1995). However,

it is important to note that even findings that are relatively large produce distribu-

tions that substantially overlap. Using the d metric, it is clear that some sex-

difference findings are quite large (about 0.80). These large effects occur with

respect to at least one test of cognitive abilities (a test of mental rotation) and

some physical abilities, such as throwing a ball (see Table 1). One researcher

(Deaux, 1984) suggested that 5% of explained variance in a specific variable is

an upper boundary for sex differences. Differences of this boundary magnitude

and their mean effect sizes include: spatial ability, .45 (Hyde, 1981); aggression,

0 (Hyde, 1984); mathematical reasoning, .44 (Rossi, 1983); proportional reason- _

ing, .48 (Meehan, 1984), and activity level, .49 (Eaton & Enns, 1986). Other

syntheses have shown male superiority in motor behaviors, such as reaction time,

flexibility, throw distance, and grip strength (Eaton & Enns, 1986; J. R. Thomas &

French, 1985.)

But are these differences ‘‘real?’’ Or are they diminishing? Considerable con-

troversy surrounds the proper interpretation of these syntheses, particularly in the

area of cognitive abilities (Eagly, 1995). There is a polarity between biologically

and socially oriented researchers. ‘‘Particularly now, when ‘‘political correct-

ness’’ has become a hot button, this area of research is something of a political

minefield’’ (Holden, 1991). For example, those at the biological end of the

spectrum assert that gender differences have as much to do with the biology of

the brain as the way we are raised (Gorman, 1992). Other more socially-oriented

researchers (e.g., Feingold, 1988; Hyde, 1981, 1990, 1994) argue that differences

in cognitive abilities are negligible in magnitude and decrease over time. For

example, using proportion of standard deviation measures where positive values

mean women scored higher and negative values mean men score higher, differ-

ences in scientific skill among 17-year olds went from —.50 in 1976 to —.28 in

GENDER DIFFERENCES IN HUMAN ABILITIES 113

1985; differences in math skills have gone from about —.2 in 1978 to about —.18

in 1986; differences in verbal skills declined from .23 prior to 1973 to .11 in

1988; and gender differences in spatial ability have declined from —.30 before

1974 to —.13 in 1986 (Adler, 1989). Others (e.g., Halpern, 1989, 1992), on the

other hand, maintain that the available findings show important cognitive gender

differences. Halpern (1989) points out that hundreds of studies have found consis-

tent gender differences on subtests in three cognitive area: female superiority in

verbal abilities and male superiority in visual-spatial and mathematical abilities.

According to Halpern (1989), there are three reasons for the unwarranted

conclusion that the gender differences for verbal and visual-spatial abilities are

shrinking: (a) Reliance on samples of high school students to address the issue

of cognitive gender differences has resulted in an underestimation of the female

superiority in verbal abilities; (b) the use of high school students has created

trends in the verbal and spatial tests that are artifacts caused by the changing

nature of the high school population; and (c) the questions asked on these tests

do not assess the areas in which the greatest gender differences are found.

Sex differences in central tendency, variability, and numbers of high scores

have been extensively studied, but research has not always been consistent. For

example, Jensen (1971) reviewed the literature on sex differences in intelligence

quotient (IQ) and concluded that the standard deviation of male IQ scores was

about 20% larger than that of females. These biases may be relatively small but

have effects that are not negligible when comparing real sex differences. Because

most studies have not used representative samples of national populations, there-

fore compounding biases due to selection and sampling, Hedges and Nowell

(1995) recently conducted an analysis of mental test scores from six major studies

that used national probability samples. Their study provided evidence that, al-

though average sex differences have been generally small and stable over time,

the test scores of males consistently have larger variance, as seen in Table 2.

The difference is typically 5% to 20%. Moreover, there is little indication that

variance ratios change over time. Except in tests of reading comprehension,

perceptual speed, and associative memory, males also typically outnumber fe-

males substantially among high-scoring individuals. In mathematics, science, and

social studies, more males than females were in the upper tails of the distribution

(ratios of 1.3 to 3.4 in the top 10%) and more females were in the lower tails.

The differences favoring males were more profound in the vocational aptitudes

scales, with 8 to 10 times as many males as females scoring in the top 10%. The

large sex differences in writing ability suggested by one of the data sets, the

National Assessment of Educational Progress (NAEP), are alarming, particularly

because the differences were found on assessments that used actual writing sam-

ples. The data show that males, on average, have a decided disadvantage in the

114 TURNAGE, KENNEDY, AND LANE

Table 2.—Average Difference In Means, Variance, And Numbers of Extreme Scores (adapted from

Hedges & Nowell, 1995)

Tail Region

Subject d VR <10% >90% >95%

Reading Comprehension 109 ial 2 1.44 .90 23)

Vocabulary .06 1.05 .98 1.10 1.14

Mathematics 16 JES) 1.05 1.62 1.94

Perceptual Speed = 28 1.06 1.66 liga) 76

Science O2 1725 ifs) PS 5.57,

Social Studies 18 1.20 1.06 1.94 2.62

Nonverbal Reasoning = iis) 1.09 £25 eye 84

Associative Memory =i25 oo 1.41 .60 56

Spatial Ability a9 127 81 1.88 2.36

Mechanical Reasoning 76 1.60 48 8.25 10.95

Electronic Information aT 2.14 58) 11.60 9.90

Auto & Shop Information an 1EO2 2.34 44 66.30 464.00

Note: d = effect size in standard deviation units; VR = variance ratio, male to female.

performance of this basic skill. Although these results do not shed light on the

origins of sex differences in either mean or variability, they do have significant "

implications for women in science. Hedges and Nowell note that small mean

differences combined with modest differences in variance can have a surprisingly

large effect on those that excel. ‘“The achievement of fair representation of

women in science will be much more difficult if there are only one-half to one-

seventh as many women as men who excel in the relevant abilities’ (p. 45).

Although we have seen that some researchers attribute male superiority in

some skills to greater score variability among males, there are some interesting

biological explanations that bear some relationship to behavior. For example,

Doreen Kimura, a researcher at the University of Western Ontario, suggests that

women with high testosterone levels and men with low testosterone levels have

higher spatial and mathematical abilities than low testosterone women and high

testosterone men. Relatedly, women respond more like men in tests during low-

estrogen phases of their menstrual cycle (Hampson & Kimura, 1988). A similar

enhancement of motor skill has been found during the estrogen phase of hormone-

replacement therapy (Kimura, 1989). Kimura studied the way men and women

think and found that men take more direct routes and make fewer errors, while

women rely more on landmarks and make more errors (Kimura, 1992). Evolution-

ary psychologists believe that men excel at thinking in three dimensions due to

ancient evolutionary pressures related to hunting, which requires orienting oneself

while pursuing prey. Such prehistoric pursuits may have conferred a comparable

advantage on women. In experiments in mock offices, women proved 70% better

than men at remembering the location of items found on a desktop—perhaps

GENDER DIFFERENCES IN HUMAN ABILITIES 115

reflecting evolutionary pressure on generations of women who foraged for their

food (Gorman, 1992).

These differences often show up during the acquisition of certain tasks (e.g.,

men tend to acquire some spatial, quantitative, and psychomotor skills easier

than women). Such differences have prompted instructional researchers (e.g.,

Regian & Shute, 1994) to observe a simple instructional intervention that signifi-

cantly reduces (and in some cases, eliminates) post-training gender differences

in a psychomotor/spatial task. Using the acquisition of a high-performance skill

on a video game-like task to study the relative contribution of nature (differences

in testosterone level) versus nurture (experiential differences), they found that,

if female subjects were required to participate in discussion groups with males

in which they talked about video game strategies, tactics, and motor skills, their

performance dramatically increased. Shute and her colleagues at Armstrong Labs

have just completed a comprehensive set of studies in which they explicitly tested

the relative contribution of instructional intervention and testosterone level across

a range of tasks varying in the importance of spatial skills. Regian and Shute

(1994) observe that *“The results from this type of research may have far-reaching

ramifications. In cases where experiential deficits account for gender differences

-in performance, the availability of appropriate instructional interventions may

double the potential pool of qualified trainees by allowing gender-blind assign-

ment of recruits to job skill categories. In cases where physiological differences

account for gender differences in performance, it may be better to select trainees

based on these physiological factors rather than gender’’ (p. 5). We will explore

later other studies that address whether equating for practice can erase gender

differences in the acquisition of a skill.

Visual-Spatial Abilities

The largest and most consistent differences between male and female perfor-

mance seem to be found on selected tests of visual-spatial abilities. Based on a

meta-analytic analysis of visual-spatial abilities, Linn and Petersen (1985; 1986)

concluded that reliable gender differences are found at around 7 or 8 years of

age (probably as early as they can be measured), increase at age 18, and continue

throughout the life span. However, it was determined that there are three distinct

categories of spatial tests: spatial perception, mental rotation and spatial visualiza-

tion. Tests of spatial perception, defined as the ability to determine spatial relations

despite distracting information, produced a mean effect size of 0.44. Tests of

mental rotation, defined as the ability to rotate quickly and accurately two- or

three-dimensional figures, in imagination, showed a mean effect size of 0.73.

116 TURNAGE, KENNEDY, AND LANE

Spatial visualization, defined as the ability to manipulate complex spatial informa-

tion when several stages are needed to produce a correct solution, yielded a mean

effect size of 0.13. In addition, these size effects increase with age (Voyer,

Voyer, & Bryden, 1995).

There are several theories as to why males are superior to females in spatial

abilities, most of them biological in nature, including genetic theories and brain

lateralization theories. Both theories are founded on the assumption that perfor-

mance on spatial tasks reflects an underlying, predetermined pattern of sex differ-

ences. However, such other factors as environmental pressures, verbal facility,

or learned problem-solving strategies may also affect hemispheric preference in

processing tasks. David Lohman of the University of Iowa (1987) hypothesizes

that the core difference has to do with what he calls the ‘‘visual-spatial

scratchpad,’’ the mental ability to retain and manipulate spatial and numerical

data that cannot be solved verbally. Several tests seem to rely on this ability.

One is a speed of closure task, which involves the identification of a distorted

or incomplete image. Another is a test of ““horizontality,’’ in which the subject

must draw a line to show the water level in a tilted glass. Males not only perform

these tasks better than females, they do them more quickly. When females get a

correct answer, they seem to get it by reasoning, but men just look at it and

know that’s the way itis. . . it’s almost as if they look at it without trying to

analyze and process it (Holden, 1992).

Gender differences in visual-spatial abilities are well- iocumened with males

performing more accurately on tasks involving such things as block design,

disembedding figures from background, and perceptual mazes (e.g., Harris, 1978;

Maccoby & Jacklin, 1974; Sherman, 1978). Sherman (1978) introduced the strat-

egy account which is opposite that of the genetic or biological theorists. According

to her ‘‘bent-twig’’ hypothesis, girls apply verbal solutions to visual-spatial prob-

lems due to early precocity, while boys find ‘‘mentally spatial’’ approaches more

successful. These styles are reinforced by social factors. Blough and Slavin (1987)

considered the hypothesis that men and women employ different problem-solving

strategies by examining studies that employ reaction time (RT). For example,

verbal solutions may be more time consuming than mentally spatial approaches.

Mental rotation, requiring kinetic mental imagery and included in many tests of

spatial ability, is closely related to RT. In previous studies of gender differences

in mental rotation (Kail, Carter, & Pellegrino, 1979; Tapley & Bryden, 1977),

females seemed to employ an analytic, feature by feature rotation strategy,

whereas males applied a holistic approach. Examining gender differences in

reaction time and accuracy on four visual tasks, Blough and Slavin found that

women were more accurate but slower on a standard visual choice task, which

demands minimal use of mental imagery. They had higher reaction times on the

GENDER DIFFERENCES IN HUMAN ABILITIES 117

two tasks which demand a great deal of mental imagery. These gender differences

interacted significantly with the degree of rotation and dissimilarity of the test

form, suggesting the presence of gender differences in visual-spatial strategies.

Women’s longer RTs overall conform to the verbal pattern identified by Cooper

(1976; 1982) as ‘‘analytic’’, while the spatial pattern adopted by men conforms

to a “‘holistic’’ type of strategy. Their longer RTs also support the notion that

women prefer accuracy to speed when confronted with mental rotation tasks

where there are speed-accuracy tradeoffs (e.g., Tapley & Bryden, 1977; Kail et

al., 1979; Lohman, 1986). Bryden, George, and Inch (1990) confirmed this find-

ing, but found that, although women take longer to manipulate three-dimensional

objects in space, they employ the same general strategy as men.

The suggestion that individuals may develop a learned strategy for problem-

- solving refutes the biological notion. For example, several studies (Connor,

Schackman, & Serbin, 1978; Goldstein & Chance, 1965; Johnson, 1976) demon-

strated that sex differences in performance on one spatial measure, the Embedded

Figures Test (EFT), could be eradicated through practice and through the training

of females. Tobin (1982) found that sex differences could be eradicated by a

single, brief practice session alone for adolescents between the ages of 13 and

16. This finding calls into question the validity of an interpretation based on the

hormonal differences because biological explanations would predict that a sex

difference in adolescence would be resistant to change.

Experimental Evidence for Gender Differences in Skill Acquisition

We now describe several somewhat similar experiments that examined gender

differences in the skill acquisition of complex tasks. The purpose is to determine

what influence basic skills have on performance and whether pre-experimental

levels of performance for males and females will be modified by practice or

specific training interventions.

A series of studies by McCloy and Koonce at the Air Force Academy (e.g.,

McCloy & Koonce, 1981) was designed to determine just how different males

and females are in cognitive style and psychomotor abilities, whether one could

expect a gender difference in the acquisition of skills, and whether or not predic-

tion equations developed to optimize the selection of males for a skill would be

adequate for the selection of females for the same task. They used a battery of

cognitive measures, consisting of tests of perceptual speed, visual memory, spatial

orientation, and spatial scanning from the Ekstrom, French, Harman, and Derman

(1976) kit of factor-referenced cognitive tests, the Embedded Figures Test (field

independence { Witkin, Oltman, Rasking, and Kark, 1971}), and three psychomo-

tor tests.

118 TURNAGE, KENNEDY, AND LANE

Table 3.—McCloy and Koonce (1981) Regression Results: Sex as a Moderator Variable for Skilled

Performance

Study 2

Overall Group Sex Within Group

0.593

CON 0.652 M 0.850

F 0.782

COG 0.783 M 0.733

F 0.871

COG + MOT 0.615 M 0.394

F 0.763

Study 3

Overall Group Sex Within Group

0.488

CON 0.589 M 0.980

F 0.690

COG 0.704 M 0.825

F 0.971

COG + MOT 0.684 M 0.889

F 0.805

In Study 1, 51 males and 52 female freshman cadets participated in three 50-

minute sessions over a two month period. Subjects were given the cognitive and

psychomotor tests in the first two sessions. In the third session each was instructed

how to perform four basic flight maneuvers (climb, cruise, descent, and turn)

using a desktop flight simulator. Each was given practice on the maneuvers, and

then tested on his or her performance while flying these maneuvers in simulated

smooth and turbulent air conditions. On all of the psychomotor tests and all of

the simulator flight maneuvers in rough and smooth air, the males were signifi-

cantly better than the females (p,0.001). Applying stepwise regression techniques

to predict flight performance, one of the perceptual motor coordination tests and

a perceptive speed test were the significant predictors for males, but for females

the pursuit rotor and a cube comparison test were the significant predictors.

The second study was conducted a year later on the same subjects. Based on

pre-test performances, they were divided into three equal groups of males and

females for control (CON), cognitive training (COG), and cognitive and motor

traning (COG + MOT). Following training, they were then tested on their

abilities to perform the four basic flight maneuvers. When multiple regression

equations were applied, they became stronger as the subjects overall were broken

down into their respective training conditions, and even more when divided into

sex within groups.

Results, as shown in Table 3, indicate that sex is indeed a moderator variable

GENDER DIFFERENCES IN HUMAN ABILITIES 119

in the prediction of basic flight performance. Also, novices that are not very

familiar with a skilled task to be performed will benefit more from both cognitive

and motor training than from cognitive training alone.

These results were replicated in a third study in which the same subjects were

then instructed on how to perform a chandelle, a more cognitively demanding

task than the four basic maneuvers previously learned. The results were similar

to those of the second study, continuing to support the idea of sex as a moderator

variable. The individual regression equations by sex also had different predictor

variables. Further studies by McCloy and Koonce actually “‘trained out’’ differ-

ences between males and females, 1.e., using comparable tasks, males and females

transferred their training equally from one task to another. These results seem to

suggest that there are different perceptual and mental approaches to attaining

- success on a task for males and females.

In recent research (Kennedy, Turnage, & Lane, 1996) again studying visual-

spatial cognitive predictors, we specifically examined changes in male and female

performance over extended practice. This is the type of study that can shed

some light on how skill acquisition curves differ between genders and whether

differences increase or decrease as a function of time on task. The purpose of

the study was the determination of predictive validity between multiple test

batteries and complex task performance on a high performance flight simulation

task. The Portable Inflight Landing Operations Trainer (PILOT) was developed

(Justiz, 1993) to assist NASA Shuttle Commanders and Pilots in maintaining the

highest possible level of shuttle approach and landing proficiency while on-orbit

for extended periods. The system is described elsewhere (Kennedy et al., 1996).

Thirty-one students (19 male and 12 female) participated in the study. Each

participant claimed to be experienced and familiar with computers and had played

video games to some degree.

The series of tests included video based tracking scenarios, postural stability

tests, cognitive performance tests, visual acuity/reflex tests, and the PILOT perfor-

mance task, all of which were microcomputer-based except for the postural stabil-

ity test. Each participant completed seven sessions within approximately a three-

week period. The Shuttle Landing task was given for eight trials per session,

with Sessions 1—6 presented under normal conditions and Session 7 presented

as a wind condition. All other tests were given once per session with the exception

of postural stability and dark focus which were administered before and after

each set of shuttle trials.

We found some consistent gender differences in most of the test scores. For

contrast sensitivity, a test of static visual acuity, there were small but consistent

gender differences at all frequencies, with means for females slightly below those

of the males. Differences ranged from 1.0 to 1.5 standard errors at all frequencies

120 TURNAGE, KENNEDY, AND LANE

other than 1.5 (no differences) and 18 cps (p,0.05). For the temporal factors

battery, which consists of seven tests of dynamic visual acuity, gender differences

were present, but modest and inconsistent. Males had a slight advantage on most

variables in early trials, with differences decreasing or disappearing in later

sessions. None of the differences was statistically significant taken across all

sessions.

For the cognitive tests, there were distinct gender effects, with means for

females below those for males for nearly every session on each variable (p <

0.05 or greater for all variables except Code Substitution and Tapping). This

result we think is in large part due to two factors. First, there were four males

in the present study who seemed to be unusually skilled on all tests, not just the

DELTA battery. Second, there were six subjects who evidently had consistent

difficulties with most of the tests, DELTA and others. They demonstrated ‘‘quit-

ting’’ behavior at one or more points during performance, perhaps due to lack

of motivation or boredom with repeated testing. All six were females; and three

were excluded from analyses. These motivational problems, as well as the gender

differences, are consistent with results reported by Ackerman, Kanfer, and Goff

(1995) that we will turn to next.

There were also strong gender differences on some of the video games. Differ-

ences were significant on Air Combat Maneuver (ACM) Score (p < 0.002),

Missed gates (p < 0.001), and Hill Time (p < 0.002). There were also Gender

X Practice interactions for ACM Score (p < 0.0001) and Hill Time (p < 0.001),

suggesting different shapes in the learning curves (very large initial differences,

decreasing with practice). A few females had considerable difficulty with the

tasks in early practice, and these negative experiences may have been reinforced

by the need to continue task performance over a large number of additional trials

to complete the experiment. Again, these problems of decreasing expectations

and their impact on performance are similar to those reported by Ackerman,

Kanfer, and Goff (1995). It is interesting to note that, for the Air Combat Maneu-

ver video game, there was a significant difference in the total number of shots

fired by females and males. Males shot as fast as possible, suggesting they were

using a speed strategy, while females who fire less rapidly seem to be employing

an accuracy Strategy.

Finally, with regard to the shuttle landing scores, seen in Figure 1 as JNM,

there were again large and consistent gender differences across the entire study.

These differences were highly significant (p < 0.0001) in a repeated measures

ANOVA. The expected decrement in performance due to wind effects was present

only for females; males actually tended to show a slight increase across those

blocks. This means that gender differences are unlikely to disappear without

considerable extended practice. When log regression curves were fitted to the data,

GENDER DIFFERENCES IN HUMAN ABILITIES 121

JNM

1 3 5 7 Sate as ets 6 17 TS e223 254027

BLOCKS OF TWO TRIALS

—ms— MALE (N=19) | — FEMALE (N=11)

BLOCKS 25 TO 28 ARE WIND CONDITION

Fig. 1. JNM by block by gender

they showed that, while the curves were converging, they remained considerably

discrepant even after the several hours of practice over the 56 trials.

Although not a skill acquisition study per se, a recently published study by ©

Tirre and Raouf (1994) from the Armstrong Labs at Brooks AFB offers some

insight into the nature of gender differences in perceptual-motor performance.

Tirre and Raouf asked ‘‘What roles do general cognitive ability and video game

experience have in determining perceptual-motor performance?’’ Perceptual-mo-

tor abilities include coordinated movements of two or more limbs, precisely

controlled movements in response to dynamic stimuli, speeded movements, and

steadiness of hand-and arm movements. They are measured by the same indices

of perceptual-matching/multilimb coordination that were used in the McCloy and

Koonce (1981) study previously described. Many tests that measure these abilities

include a substantial visual perception component. Based on the findings of Law,

Pellegrino, and Hunt (1993), Tirre and Raouf predicted that gender differences

would be independent of experience, using three perceptual-motor video game

tasks. Law et al. had found that dynamic spatial asks (e.g., predicting which of

two moving objects will reach a point first) reflected gender differences even

after video game experience was controlled. Three perceptual-motor components

and one spatial rotation component were used as dependent variables. Gender

122 TURNAGE, KENNEDY, AND LANE

differences favoring males were significant on two of the four components—a

perceptual matching with multilimb coordination factor and a speed of hit factor.

Video game experience was also correlated with performance on all dependent

variables. However, the gender X video game interaction was only significant

for one component, that of perceptual-matching with multilimb coordination. The

authors suggested that further investigation be devoted to the type of video

game experience women have had before concluding that video game experience

benefits only men. For example, they say ‘‘it is possible that women in our

sample who reported more video game experience might have engaged in this

activity mainly to please their boyfriends or might have been passive observers

of their boyfriends playing the games.’’ (p.A53). We suggest, on the other hand,

that the noncomparability with previous findings of Law et al. could be a reflection

of noncomparability between tasks ——Law and colleagues used a spatial reasoning

task while Tirre and Raouf used a perceptual-motor performance task.

Ackerman and Kanfer (e.g., Ackerman, 1992) are the only researchers that we

know of other than ourselves who are currently examining gender differences in

skill acquisition over time. Using a terminal radar approach controller (TRACON)

simulation common to air traffic controllers, they have validated and extended

Ackerman’s (1988) theory of cognitive ability determinants of individual differ-

ences in skill acquisition. That theory concerns ability-performance relationships

as a function of three task characteristics: (a) consistence of information pro-

cessing, (b) task complexity, and (c) degree of task practice. Consistent with

earlier findings (e.g., Maccoby & Jacklin, 1974), Ackerman (1992) found that

men, on average, tended to perform better than women on spatial ability tests;

women, on average, tended to perform better on the perceptual speed tests.

However, at the composite level, only the perceptual speed measure indicated a

significant difference in means, with higher scores by women. A general TRA-

CON task performance variable showed men performing better overall than

women (p < 0.01), and a significant interaction between sex and practice (p <

0.01), in addition to the main effect of practice. The interaction is indicated as

a divergence in performance between the two groups with increasing practice on

TRACON.

A more precise view of gender differences emerged in examination of arrivals,

departures, and overflights accepted. Arrivals and departures showed main effects

of sex (p < 0.01) and an interaction effect between sex and practice (p < 0.01),

with men having a greater advantage later in practice. Overflights did not indicate

main effects for sex or an interaction effect, perhaps due to the fact that overflights

seem to have a diminished demand for spatial abilities.

Ackerman suggested that the sex differences found in this experiment pose a

“vexing problem from both selection and training perspectives.’’ Although

GENDER DIFFERENCES IN HUMAN ABILITIES 123

women scored higher on perceptual speed composites than men, their perfor-

mance on the task component most dependent on perceptual speed ability was

only equivalent, and not superior, to that of men. For the task component that

featured arrivals, which would be most dependent on spatial abilities, women

performed significantly more poorly on average throughout the task. In addition,

there seemed to be a cumulative performance deficit, with increasing sex-related

differences with increasing task practice. Ackerman observed that many previous

studies regarding the effect of experience and practice on sex differences in spatial

ability have used tasks that overwhelmingly provide for controlled consistent

information processing. In these tasks, overall individual differences are attenu-

ated simply because the distribution of performance becomes leptokurtic. How-

ever, consistent with his theory of ability determinants of skill acquisition, he

_ observed that the current investigation focused on a task that required a substantial

amount of inconsistent information processing and, as such, precluded the attentu-

ation of individual differences in performance. Ackerman suggested that other

sources, perhaps noncognitive sex differences such as learning versus perfor-

mance orientation (e.g., Dweck, 1986) may be responsible for such effects.

Using a similar experimental paradigm, Ackerman, Kanfer, and Goff (1995)

recently followed up on the suggestion that noncognitive determinants are likely

operative in complex skill acquisition. The researchers examined a wide array

of predictors, including ability factors, personality factors, vocational interests,

self-estimates of ability, self-concept, motivational skills, and task-specific self-

efficacy. Ninety-three trainees were studied over the course of 15 hours (2 weeks)

of skill acquisition practice on the TRACON simulation task.

Similar to previous studies (e.g., Ackerman, 1992), Figure 2 shows that there

was a significant main effect for trainee gender on overall performance (p <

0.01), and an interaction between trainee gender and sessions of practice (p <

0.01). As before, the interaction is shown as a divergence in performance between

the two groups with increasing practice on TRACON. In agreement with previous

findings, as well, analyses of arrivals and overflights showed higher correlations

between arrival performance and math-spatial abilities than for overflights, and

the differences between gender groups showed greater advantages to men over

women in handling arrivals. For arrivals there were significant main and interac-

tion effects (p < 0.01), but overflights showed a smaller gender effect and a

nonsignificant interaction. Again, these results support the assertion that handling

arrivals requires more complex spatial information processing than does the han-

dling of overflights.

There were also interesting concomitant variables and consequences of prac-

tice. For example, women reported a significantly higher frequency of negative

motivational thoughts over practice (p < 0.01), and a significant interaction of

124 TURNAGE, KENNEDY, AND LANE

Overall Performance

—@— MALES

- # - FEMALES

Arrivals Handled

6 _a---4

4 Stall

pe —@— MALES

2 a---« - @ - FEMALES

‘ |

Overflights Handled

__-i---4

hea -

—@— MALES

- @ - FEMALES

Session of Practice

Fig. 2. Mean terminal approach controller task performance by gender over sessions of practice. From

with permission.)

Gender X Session (p < 0.01). This reflected a convergence of male and female

reporting of negative motivation thoughts over practice. In addition, for frequency

of planning thoughts, there was no main effect for gender, but significant increases

in planning over the course of practice and a significant interaction of Gender X

Practice. Examination of the means showed that the interaction was the result of

women reporting less planning than men early in practice but more planning later

GENDER DIFFERENCES IN HUMAN ABILITIES 125

in task practice. In terms of self-efficacy, four measures of self-efficacy for

TRACON performance were taken prior to each TRACON session. For each of

the four self-efficacy measures, there were substantial and significant effects of

gender on score (with women having lower mean self-efficacy scores for all

aspects of the TRACON task), and sessions of practice (with self-efficacy drop-

ping from Session 1 to Session 2 and then rising after Session 2). Significant

Gender X Practice interactions were found, SE(Handled) and SE (Arrivals), with

a divergence of scores for men and women. In hierarchical multiple regression

analyses, where gender was added after all the other predictor variables had been

taken into account, gender accounted for a significant amount of variance in the

arrival component of task performance but accounted for virtually no incremental

variance in the overflight component.

At this point, it is not clear whether the differences in performance between

men and women be attributed to sex differences in higher order spatial abilities

(such as visualization) or to differences in strategies for task engagement and

learning (although see Lohman, 1986, 1987, for a discussion of both sex differ-

ences in spatial information processing and on the structure of spatial abilities,

respectively, cited in Ackerman, 19972).

Implications

These cumulative results demonstrating that task performance that depends

particularly on spatial abilities diverges with practice for males and females has

distinctly negative implications for those who aspire to see women more equally

represented in scientific and technological disciplines. Newsweek (Kantrowitz,

1994) not too long ago ran a cover story titled “‘Men, Women, & Computers.”’

The article started by citing the story of a long-time “*Star-Trek’’ devotee who

happened to be a woman. After repeated tries to be part of a Trekkie discussion

group on the Internet, she was chased off the net by rabid hounds, or ‘‘fire

Phasers,’’ the male Trekkies who flooded her e-mail with nasty messages. So

she retreated into her own galaxy by starting the all-female Starfleet Ladies

Auxiliary and Embroidery/Baking Society. The article’s author observed, ‘‘From

the Internet to Silicon Valley to the PC in the family room, men and women

often seem to be like two chips that pass in the night’’ (p.50). In general, computer

culture is created, defined and controlled by men; women often feel as welcome

as a system crash. Anecdotal evidence seems to support the notion that men tend

to be seduced by the technology; they get into the faster-race car syndrome,

bragging about the speed of their microprocessors. Women, on the other hand,

are much more practical and much more interested in the machine’s utility; they

just want it to do the job.

126 TURNAGE, KENNEDY, AND LANE

Although boys and girls are equally interested in computers until about the

fifth grade, at that point boys’ use rises significantly and girls’ use drops (Kan-

trowitz, 1994). Many experts think that the increasingly violent nature of video

games— particularly against women—is playing a big role in creating a ‘‘gender

gap’’ in computer technology professions, where men vastly outnumber women.

Examples of the gender gap are evident around the country: Nationwide, less

than 28 percent of all computer science degrees go to women. Since the 1980s,

the number of women entering computer professions has dropped by 50 percent.

The number of women receiving bachelor’s degrees in computer sciences at the

University of Central Florida declined almost 70 percent from 1986 to 1995

(Burnett, 1996). Although game manufacturers are now trying to lure girls with

more appealing games, it will not be an easy task. However, Sherry Turkle, a

MIT professor who has studied gender differences in computer usage, thinks that

gender differences could actually help women. She says ‘‘We’re at a cultural

turning point; there’s an opportunity to remake the culture around the machine.”’

Practicality is now as valued as invention (Kantrowitz, 1994, p. 55).

These studies on gender differences, particularly in the realm of visual-spatial

abilities, need much further research. Several explanations have been offered

to account for the prevalence of sex differences in spatial abilities. For example,

variables such as choice of strategy (Bryden, 1980), rate of maturation (Sand-

ers & Soares, 1986; Waber, 1976), cerebral lateralization (Bryden, 1979; Levy,

1971), genetic complement (McGee, 1982), sex hormones (Imperatio-McGin-

ley, Pichardo, Gautier, Voyer, & Bryden, 1991); McGee, 1979), differential

experience and socialization (Baenninger & Newcombe, 1989), and sex role

identification (Nash, 1975; Signorella & Jamison, 1986) have all been proposed

as possible causes. As Voyer et al. have noted, because some tests consistently

show reliable and significant sex differences, it would seem important to under-

stand the processes underlying these tasks, including brain structure, hormonal

and experiential factors. It has also been shown that method of administration

also affects performance and may be responsible for the decreasing effect sizes

observed in recent years. This suggests that a detailed examination of the way

in which these tests are scored and administered may lead to valuable informa-

tion (Voyer et al., 1995).

To these potential explanations for gender differences, we would like to add

a few other observations and suggestions. The first observation has to do with

the choice of tasks in the skill acquisition studies that we have reviewed. In all

instances, the criterion task is a male-oriented task that features ability prerequi-

sites on which males are known to excel—where spatial abilities, psychomotor

behaviors, and quick reaction times equate to superior results. In our shuttle

landing study, we found that video games correlated so highly with shuttle landing

GENDER DIFFERENCES IN HUMAN ABILITIES 127

performance that game scores consistently dropped out from any prediction equa-

tions. But we have seen that video games and other perceptual-motor tasks do

not interest women. Both Ackerman and we found severe motivational problems

in keeping female participants focused on task performance in our studies. They

have negative thoughts and generally exhibit quitting behaviors when faced with

continued exposure to the task. Not surprisingly, then, their self-efficacy decreases

and performance differences between the genders become more pronounced with

increased exposure to the task. Would the same results occur if the task were

one which was not so male-oriented? What if the task were a problem-solving

task that was relatively more dependent on verbal abilities, associative memory,

or perceptual speed, on which women hold a performance advantage? What if

we looked at the bigger criterion picture and included tests for practical, tacit and

emotional intelligence (e.g., Sternberg, Wagner, Williams, & Horwath, 1995)?

We also saw that women profit more from verbal instructions than do men in

the McCloy and Koonce and the Regian and Shute studies. Could that advantage

be due to females’ greater verbal abilities? Likewise, they profited more from

training that featured explicit motor practice. They also use different combinations

of skills and abilities than men to achieve similar results. This conclusion is

supported by the very different prediction equations for males and females that

were found in the McCloy and Koonce studies. These observations suggest that

female-oriented interventions that capitalize on female interests and skills, such

as merely talking about strategies, can improve performance. In most of the

studies cited here, both predictor variables and criterion variables are those where

there are known male advantages, but it is highly likely that selecting female-

oriented predictors would shed a different perspective on skill attainment. We

know that males and females approach and solve problems differently, so let us

get on with the task of determining how best to use the best of nature and nurture

in both sexes to improve performance on a wide variety of activities that have

universal significance in our world of work and living.

Acknowledgments

This research was funded in part by NASA SBIR, Phase II, Contract no. NAS9-19107. Thanks go to Kelly

Gottfried, Jackie Depuy, and Pat Maddox for their technical assistance.

References

Ackerman, P. L. (1992). Predicting differences in complex skill acquisition: Dynamics of ability determinants.

Journal of Applied Psychology, 77(5), 598-614.

Ackerman, P. L., Kanfer, R., and Goff, M. (1995). Cognitive and noncognitive determinants and conse-

quences of complex skill acquisition. Journal of Experimental Psychology, Applied, 4, 270-304.

Ackerman, P. L. (1988). Determinants of individual difference during skill acquisition. Cognitive abilities and

information processing. Journal of Experimental Psychology: General, 117, 288-318.

Adler, J. (1989, March). Sex-based differences declining, study shows—APA Monitor, p. 6.

128 TURNAGE, KENNEDY, AND LANE

Baenninger, M., and Newcombe, N. (1989). The role of experience in spatial test performance: A meta-

analysis. Sex Roles, 20, 327-344.

Blough, P. M., & Slavin, L. K. (1987). Reaction time assessments of gender differences in visual-spatial

performance. Perception & Psychophysics, 41(3), 276-281.

Bryden, M. P. (1979). Evidence for sex-related differences in cerebral organization. In M. A. Wittig & A.C.

Petersen (Eds.). Sex related differences in cognitive functioning pp. 121—143. New York: Academic Press

Bryden, M. P. (1980). Sex differences in brain organization: Different brains or different strategies? Behavioral

and Brain Sciences, 3, 230—231.

Bryden, M. P., George, J., & Inch, R. (1990). Sex differences and the role of figural complexity in determining

the rate of mental rotation. Perceptual and Motor Skill, 70, 467-477.

Burnett, R. (1996, February 13). Games may discourage girls about computers. The Orlando Sentinel.

Cohen, J. (1977). Statistical power analysis for the behavioral sciences (Rev. ed.). San Diego, CA: Academic

Press.

Cooper, L. A. (1976). Mental transformations and visual comparison processes. Perception & Psychophysics,

19, 433-444.

Cooper, L. A. (1982). Strategies for visual comparison and representation: Individual differences. In R. J.

Sternberg (Ed.), Advances in the psychology of human intelligence (pp. 77—124). Hillsdale, NJ: Erlbaum.

Connor, J. M., Schackman, M., & Serbin, L. A. (1978). Sex-related differences in response to practice on

a visual-spatial test and generalized to a related test. Child Development, 49, 24—29.

Deaux, K. (1984). From individual differences to social categories: Analysis of a decade’s research on gender.

American Psychologist, 39, 105-116.

Dweck, C. S. (1986). Motivational processes affecting learning. American Psychologist, 41, 1040-1048.

Eagly, A. A. (1995). The science and politics of company women and men. American Psychologist, 50 (3),

145-158.

Eaton, W. O., & Enns, L. R. (1986). Sex differences in human motor activity level. Psychology Bulletin, 100,

19-28.

Ekstrom, R. B., French, J. W., Harman, H. H., and Derman, D. (1976). Manual for kit of factor-referenced

cognitive tests. Princeton, NJ: Educational Testing Service.

Feingold, A. (1988). Cognitive gender differences are disappearing. American Psychologist, 43, 95-103.

Goldstein, A. G., & Chance, J. E. (1965). Effects of practice on sex-related differences in performance of

embedded figures. Psychonomic Science, 3, 361-362.

Gorman, C. (January 20, 1992). Sizing up the sexes. Time. pp. 42—51.

Halpern, D. F. (1989). The disappearance of cognitive gender differences: What you see depends on where

you look. American Psychologist, 44, 1156-1158.

Halpern, D. F. (1992). Sex differences in cognitive abilities (2nd ed.). Hillsdale, NJ: Erlbaum.

Hampson, E., Kimura, D. (1988) Reciprocal effects of hormonal fluctuation on human and perceptual-spatial

studies. Behavioral Neuroscience, 102, 456—459.

Harris, L. (1978). Sex differences in spatial ability: Possible environmental, genetic, and neurological factors.

In M. Kinsbourne (Ed.), Asymmetrical function of the brain (pp. 405-521).

Hedges, L. V., & Nowell, A. (1995). Sex differences in mental test scores, variability, and numbers of high-

scoring individuals. Science, 269, 41—45. Cambridge: Cambridge University Press.

Holden, C. (1991). Is ‘‘gender gap’’ narrowing? Science, 253, 959-960.

Hyde, J. S. (1981). How large are cognitive gender differences? A meta-analysis using w~ and d. American

Psychologist, 36, 892—901.

Hyde, J. S. (1984). How large are gender differences in aggression? A developmental meta-analysis. Develop-

mental Psychology, 20, 722-736.

Hyde, J.S. (1990). Meta-analysis and the psychology of gender differences. Signs: Journal of Women in

Culture and Society, 16, 55-73.

Hyde, J. S. (1994). Can meta-analysis make feminist transformations in psychology? Psychology of Women

Quarterly, 18, 451-462.

Imperato-McGinley, J., Pichardo, M., Gautier, T., Voyer, D., & Bryden, M. P. (1991). Cognitive abilities ©

in androgen insensitive subjects—Comparison with control males and females from the same kindred.

Clinical Endocrinology, 34, 341-347.

Jensen, A. (1971). In R. Cancro (Ed.), Intelligence: Genetic and environmental influences. New York: Grune

and Stratton, pp. 107-161.

Johnson, S. (1976). Effect of practice and training in spatial skills on sex-related differences in performance

of embedded figures. Unpublished master’s thesis, George Mason University, Fairfax, VA.

Justiz, C. R. (1993). Brief overview of the PILOT self-correlation number. (Technical memorandum to PILOT

project participants). Houston, TX: NASA Johnston Space Center.

Kail, R., Carter, P., & Pellegrino, J. (1979). The locus of sex differences in spatial ability. Perception &

Psychophysics, 26, 182—186.

Kantrowitz, B. (1994, May 16). Men, women, & computers. Newsweek.

GENDER DIFFERENCES IN HUMAN ABILITIES 129

Kennedy, R. S., Turnage, J. J., & Lane, N. E. (1996). Development of a performance readiness evaluation

system. (Final phase II report, Contract NAS 9-19107). Houston, TX: Johnson Space Center.

Kimura, D. (1989, November). How sex hormones boost-or cut- intellectual abilities. Psychology Today, pp.

62-66.

Kimura, D. (1992). Sex differences in the brain. Scientific American, 267 (3), 118-125.

Law, D. L., Pellegrino, J. W., & Hunt, E. B. (1993). Comparing the tortoise and the hare: Gender differences

and experience in dynamic spatial reasoning tasks. Psychological Science, 4, 35—40.

Levy, J. (1971). Lateral specialization of the human brain: Behavioral manifestations and possible evolutionary

basis. In J. A. Kiger, Jr. (Ed.), The biology of behavior pp. 159-180. Corvallis: Oregon State University

Press.

Linn, M. C., & Petersen, A. C. (1986). A meta-analysis of gender differences in spatial ability: Implications

for mathematics and science achievement. In J.S. Hyde & M.C. Linn (Eds.), The psychology of gender.

Advances through meta-analysis pp. 67-101. Baltimore: Johns Hopkins University Press.

Linn, M. C., & Petersen, A. C. (1985). Emergence and characterization of sex differences in spatial ability:

a meta-analysis. Child Development, 56, 1479-1498.

Lohman, D. F. (1986). The effect of speed-accuracy tradeoff on sex differences in mental rotation. Perception &

Psychophysics, 39, 427-436.

Lohman, D. F. (1987). Spatial abilities as traits, processes, and knowledge. In R. J. Sternberg (Ed.), Advances

in the psychology of human intelligence (Vol.4, pp. 181—248). Hillsdale, NJ: Erlbaum.

Maccoby, E. E., & Jacklin, C. N. (1974). The psychology of sex differences. Stanford, CA: Stanford University

Press.

McCloy, T. M., & Koonce, J. M. (1981). Sex as a moderator variable in the selection and training of person

for a skilled task. Aviation, Space and Environmental Medicine, 53 (12), 1170-1173.

McGee, M. G. (1979). Human spatial abilities: Psychometric studies and environmental, genetic, hormonal,

and neurological influences. Psychological Bulletin, 86, 889-918.

Meehan, A. (1984). A meta-analysis of sex differences in formal operational thought. Child Development, 55,

pp. 1110-1124.

Nash, S. C. (1975). The relationship among sex-role stereotyping, sex-role preference, and the sex difference

in spatial visualization. Sex Roles, 1, 15-32.

Regian, J. W., & Shute, V. J. (1994, January). Understanding and reducing gender differences in task perfor-

mance. HFES Training Technical Group Newsletter, p. 5.

Ross, J. S. (1983). Ratios exaggerate gender differences in mathematical ability. American Psychologist, 38,

348.

Sanders, B., & Soares, M. P. (1986). Sexual maturation and spatial ability in college students. Developmental

Psychology, 22, 199-203.

Sherman, J. (1978). Sex-related cognitive differences: An essay on theory and evidence. Springfield, IL:

Thomas.

Signorella, M. L., & Jamison, W. (1986). Masculinity, femininity, androgyny, and cognitive performance: A

meta-analysis. Psychological Bulletin, 100, 207-228.

Sternberg, R. J., Wagner, R. F., Williams, W. M. & Horvath, J. A. (1995). Testing common sense. American

Psychologist, 50 (11), 912—927.

Tapley, S., & Bryden, M. (1977). An investigation of sex differences in spatial ability: Mental rotation of

three-dimensional objects. Canadian Journal of Psychology, 31, 123-130.

Thomas, J. R., & French, K. E. (1985). Gender differences across age in motor performance: A meta-analysis.

Psychological Bulletin, 98, 260-282.

Tirre, W. C., & Raouf, K. K. (1994). Gender differences in perceptual-motor performance. Aviation, Space,

and Environmental Medicine, A49—AS53.

Tobin, P. (1982). The effects of practice and training on sex differences in performance on a spatial task.

Unpublished master’s thesis; University of Toronto, Toronto, Canada.

Voyer, D., Voyer, S., & Bryden, M. P. (1995). Magnitude of sex differences on spatial abilities: A meta-

analysis and consideration of critical variables. Psychological Bulletin, 117 (2), 250—270.

Waber, D. P. (1976). Sex differences in cognition: A function of maturation rate? Science, 192, 572—574.

Witkin, H. A., Oltman, P. K., Raskin, E., and Karp, S. A. (1971). A manual for the embedded figures test.

Palo Alto, CA: Consulting Psychologists Press.

score ee r

oe vee

eae a eae a

mii:

ees

wee

eerie janet

Fat ea

Din a i

Sto . hig z

elites

ey. ry

ey ae ta ; ero; :

; j

* Ay i

; ii had,

ry wae ’ ;

AEP oR . ' ; t

i & }

‘ i; ¥ ia

‘ tf A

‘ j

a . ig

; \

aa Pe j

7 | 5

Pima thie, ary ee a, bi

Cte at if q LS 7, ¢ i f hy ! ;

. had r

iy y ve ri

‘ it, i Van

‘ F

~ 7 é i a Z i .

8 i

ra f bs ' -

; j wines

WADI PY / f : ne i ‘

aes eine = , M te

bea yea beat e, ae 1S Aes Bt

» = of

i " a4 4, ‘ : t

: 7 -. yee ee, ‘ ‘ i py 3 % ?

xt etapa * ‘

DELEGATES TO THE WASHINGTON ACADEMY OF SCIENCES,

REPRESENTING THE LOCAL AFFILIATED SOCIETIES

Philosophical Society of Washington ....:............6. 50 ee cece eee ees Thomas R. Lettieri

Pmntopolorical) Society Of Washington .... 2... 66... ese ee eee ce etc ee es Jean K. Boek

Pamacical Socicty Gl Washington 292.5. 6.5. owns eee ee eee ese cee Kristian Fauchald

macinical society Of WashiMetoOm .. 02)... 26 eee cee eee ee cee ees Elise A. B. Brown

Enromolocical Society of Washington ....:............2.06002 eee eas F. Christian Thompson

Pe MIBUICOPTAPING SOCICLY {260.056 a6 oe da ee eg ee ee lee oh eee Stanley G. Leftwich

ammo al SOCicty Of WaSMINPTON (2 ou. 060s ec kek ee eee Seles ee boa eee ee VACANT

miccical Society of the District of Columbia...................060- 2062 eec eee John P. Utz

ieee society of Washington, DC) 2... 24.0... ee ee ll ee eee es VACANT

Pecameam society Of Washinpton ...\.. 8... diss see ee ete ee eae bee enews ds Muriel Poston

Society of American Foresters, Washington Section .....................--. Eldon W. Ross

REM raneSOcicty OL EMPINEELS 22.6 o.oo. eee ee ee ee ede ee oes Alvin Reiner

Institute of Electrical and Electronics Engineers, Washington Section ........ George Abraham

American Society of Mechanical Engineers, Washington Section .......... Daniel J. Vavrick

Helmuimthological Society of Washington .................. 2... eee eee eee eee VACANT

American Society for Microbiology, Washington Branch .....................-.2.- Ben Tall

Society of American Military Engineers, Washington Post ............... William A. Stanley

American Society of Civil Engineers, National Capital Section ................... VACANT

Society for Experimental Biology and Medicine, DC Section ............. Cyrus R. Creveling

ASM intemational, Washington Chapter ..................2.. ccc e eee et eeee Richard Ricker

American Association of Dental Research, Washington Section ............ J. Terrell Hoffeld

-American Institute of Aeronautics and Astronautics, National Capital

eT eee.) ore ee shea SApeuens Hee ee dda We wea a Sade a Reginald C. Smith

amencan Meteorological Society, DC Chapter ..............5..-2-22---- A. James Wagner

Resmselemcee Society Of Washington .........-- 22.26.20. e bee ee eee eee To be determined

Acoustical Society of America, Washington Chapter .....................-. Richard K. Cook

American Nuclear Society, Washington Section ..................-0-02-e seen Kamal Araj

Institute of Food Technologists, Washington Section ....................205- Roy E. Martin

American Ceramic Society, Baltimore-Washington Section ................. Curtis A. Martin

Pee MEAG AINSOCICUY 2 2/c)..(4 0. ds 5 eb eam Vine eek d Slash Ls p lloadeciene bless ues Regis Conrad

MWasnimeton Eistory of Science Club ............-..2252-2--2 eee eee Albert G. Gluckman

American Association of Physics Teachers, Chesapeake Section ............ Robert A. Morse

Optical Society of America, National Capital Section .................... William R. Graver

American Society of Plant Physiologists, Washington Area Section ........... Steven J. Britz

Washington Operations Research/Management Science Council ............... John G. Honig

Instrument Society of America, Washington Section ..................02-0 eee eee VACANT

American Institute of Mining, Metallurgical and Petroleum Engineers,

Seem O AM SOCOM iarihs ea Wey eta See cy Se aya le Se dle be Se sale oss Anthany Commarota Jr.

Rirendbeapital AStFOMOMErS 6... 2.26. aie a ele nee ee eee cede wees Robert H. McCracken

Mathematics Association of America, MD-DC-VA Section ................ Sharon K. Hauge

Prcinctof Columbia Institute of Chemists ...............5./ se eeeeeae William E. Hanford

District of Columbia Psychological Association ....................0000- Marilyn Sue Bogner

Pashineton) Paint Technical Group .......... 62-605. bce ceeds eww nae Lloyd M. Smith

American Phytopathological Society, Potomac Division ................... Kenneth L. Deahl

International Society for the System Science, Metropolitan Washington

LoD STUB cg IT eS a SE UG Teresa eee ee David B. Keever

Human Factors Society, Potomac Chapter .........0.-....6-25.0.025582% Thomas B. Malone

American Fisheries Society, Potomac Chapter ................ 000.00 e eee Dennis R. Lassuy

Association for Science, Technology and Innovation ....... LS aa eet Clifford E. Lanham

EsteiiSOCtOlOPpiCal SOCIELY <2... cia s ceed se oaks eeek ebb debian Ronald W. Manderscheid

Institute of Electrical and Electronics Engineers, Northern Virginia

SCR RM itis A Poe Mechel oA Ee A Me tesla Ue ok Lia etn Blanchard D. Smith

Association for Computing Machinery, Washington Chapter ............. Charles E. Youman

Py asmineton Statistical SOCIELY 21...) koh oe eee chk heen va ede wiceeies David Crosby

Society of Manufacturing Engineers, Washington, DC Chapter .............. James E. Spates

Institute of Industrial Engineers, National Capital Chapter ................ Neal F. Schmeidler

Delegates continue to represent their societies until new appointments are made.

Washington Academy of Sciences Periodicals Postage Paid

2100 Foxhall Road, NW at Washington, DC

Washington, DC 20007-1199 and additional mailing offices.

Return Postage Guaranteed

SERIALS/ACQ. DEPT., LIBRARY

UNIVERSITY OF SASKATCHEWAN

3 CAMPUS DRIVE

SASKATOON SK S7N 5A4

CANADA

l {

wal 7

N tt VOLUME 84

Number 3

oJ Our nal of the September, 1996

WASHINGTON

ACADEMY .. SCIENCES

ISSN 0043-0439

Issued Quarterly

at Washington, D.C.

SMITHSON y

Rag a

Wat ERa qj oN

t j fe

Se , £

1998

LIBRARIES

CONTENTS

Articles:

THOMAS T. SAMARAS, ‘‘How Body Height and Weight Affect Our

Renionmance, ILONGeVIlys ANG SUEVIVAL (Goes) Gee eects cies il sid isabelle, 6 oye

JOSEPH DI RIENZI, ‘‘Locality, Realism, Lorentz Invariance and Quantum

Ty ES PTA CS Aa ct as ESR cay GLOW En OI J ga A 78 MU Oc eR Pa

H. S. EL KHADEM, ‘‘A Translation of a Zosimos’ Text in an Arabic

PRC HEIN Es OO Kern eee S On OURS UNA Je asa UNCLE) HMI Cheval S's) alla a lotsa Sialiait

Washington Academy of Sciences

Founded in 1898

EXECUTIVE COMMITTEE

President

Rita R. Colwell

President-Elect

Benjamin H. Alexander

Secretary

Louis F. Libelo

Treasurer

John G. Honig

Past President

John Toll

Vice President, Membership Affairs

Cyrus R. Creveling

Vice President, Administrative Affairs

George H. Hagn

Vice President, Junior Academy Affairs

W. Allen Barwick

Vice President, Affiliate Affairs

Lee Ohringer

Board of Managers

Marilyn S. Bogner

Elise A. B. Brown

Norman Doctor

Rex Klopfenstein

John H. Proctor

Grover Sherlin

Jammes Spate

REPRESENTATIVES FROM

AFFILIATED SOCIETIES

Delegates are listed on inside rear cover

of each Journal.

ACADEMY OFFICE

Washington Academy of Sciences

Room 811

1200 New York Ave. N.W.

Washington, DC 20005

EDITORIAL BOARD

Editor:

Bruce F. Hill, Mount Vernon College

Associate Editors:

Milton P. Eisner, Mount Vernon Col-

lege

Albert G. Gluckman, University of

Maryland

Marc Rothenberg, Smithsonian Institu-

tion

Marc M. Sebrechts, Catholic University

of America

Edward J. Wegman, George Mason

University

The Journal

This journal, the official organ of the Washing-

ton Academy of Sciences, publishes original

scientific research, critical reviews, historical

articles, proceedings of scholarly meetings of

its affiliated societies, reports of the Academy,

and other items of interest to Academy mem-

bers. The Journal appears four times a year

(March, June, September, and December). The

December issue contains a directory of the cur-

rent membership of the Academy.

Subscription Rates

Members, fellows, and life members in good

standing receive the Journal without charge.

Subscriptions are available on a calendar year

basis, payable in advance. Payment must be

made in U.S. currency at the following rates:

U.S. and Canada’) i230... 4 eee $25.00

Other countries, ...45:6.0:G4 eee 30.00

Single copies, when available ...... 10.00

Claims for Missing Issues

Claims will not be allowed if received more ~

than 60 days after the day of mailing plus time

normally required for postal delivery and

claim. No claims will be allowed because of

failure to notify the Academy of a change of

address.

Notification of Change of Address

Address changes should be sent promptly to

the Academy Office. Such notification should

show both old and new addresses and zip

codes.

POSTMASTER: Send address changes to

Washington Academy of Sciences, Room 811,

1200 New York Ave. N.W., Washington, DC

20005.

Journal of the Washington Academy of Sciences (ISSN 0043-0439)

Published quarterly in March, June, September, and December of each year by the Washington

Academy of Sciences, 2100 Foxhall Road, N.W., Washington, DC, 20007-1199. Periodicals

postage paid at Washington, DC and additional mailing offices.

Journal of the Washington Academy of Sciences,

Volume 84, Number 3, Pages 131—156, September 1996

How Body Height and Weight Affect

Our Performance, Longevity, and

Survival

Thomas T. Samaras

Reventropy Associates, 11487 Madera Rosa Way, San Diego, CA 92124-2877

ABSTRACT

Human body size is evaluated in terms of its impact on the earth and the future of

mankind. Key aspects of human existence are quantitatively evaluated, including performance,

longevity, resource consumption, pollution, and economics. Various U.S. and international

studies relating height, weight, and mortality are examined. Evidence is presented to show

that a population of smaller size people would not negatively impact human performance

and would ameliorate food, water, health, and environmental problems. A nutritional strategy

is presented for conserving resources and producing healthy and active people into old age.

Performance and survival are strongly affected by the average size of an animal

and its environment. Over 10000 years ago, North and South America saw the

extinction of over 135 species of animals. Virtually all of these species were

over 45 kg. The Dinosaurs also disappeared 65 million years ago but smaller

animals, such as turtles, crocodiles, snakes, and lizards survived to the present

time. It is obvious that larger animals, such as humans, need more resources to

survive and flourish. Each human being consumes enormous amounts of food,

water, and other resources. Over a lifetime, the average American requires 237

million liters of water, 4545 kg of meat, 4091 kg of grains, 12727 kg of milk

and cream, and 89040 liters of gasoline. When these resources are multiplied by

6 billion people, the scope of the problem is awesome. Yet, this area of study

has been neglected by virtually the entire scientific community. However, there

have been several scientists who have recognized the importance of human size

on our day-to-day lives and long-term survival. These have included Dr. Benjamin

H. Alexander, Dr. Ashley Montagu, Dr. John Yudkin, Dr. Lowell H. Storms, Dr.

Corresponding author: Thomas T. Samaras, 11487 Madera Rosa Way, San Diego, CA 92124-2877, Tele:

(619) 576-9283; Fax: (619) 292-1746

131

132 THOMAS T. SAMARAS

Harold Elrick, and Dr. Dennis D. Miller. Engineering professors Hansen and

Holley (1967) also recognized the importance of size many years ago when

they pointed out the impact of smaller human size on buildings, power sources,

transportation facilities, and food needs.

It is the purpose of this paper to explore and quantify the ramifications of

increasing human size on our performance, longevity, environment, and future.

However, a brief background review of some key areas will be given first.

Secular Growth

‘*Secular growth’’ refers to the progressive increase in the size of the average

person over the last 200 years. People in many countries experienced an increase

in stature after the initiation of the industrial revolution. This increase in Europe

has averaged about 2.54 cm per generation. However, growth has been uneven

over this period and some countries have lagged behind others. For example, the

Portuguese, Spanish, and Italians are a few inches shorter than the taller Northern-

ers, such as the Dutch and Norwegians. In 1760, the average Norwegian military

recruit was under 1.6 m. Today, he averages 1.8 m. More recently, the Japanese

and Chinese have had a remarkable growth rate of 2.54 cm per decade—a much

more rapid rate than Europeans have experienced. Since the end of WW II, the

Japanese have grown 7.6 to 12.7 cm.

Along with increasing height we have seen increases in weight. The average

American male during WW I was about 64 kg, during WW II 68 kg, and now

he exceeds 82 kg. Women have kept pace with men but they have generally

averaged 8 to 13 cm shorter. Their body weights have also been 15 to 20% less

than those of men. Overall proportions have remained fairly constant although

there have been some specific variations from symmetry in body structure

(Kroemer, 1990). American and Japanese studies have shown that the overall

body weight has increased as the cube of height increase for different generations.

From one generation to the other, weight has increased at a rate of 1 to 2 kg per

centimeter of height.

Most scientists believe that this growth process cannot continue indefinitely.

While we are still seeing increasing stature among the average American, the

rate of height growth among the upper middle class seems to be approaching a

plateau. However, scientists reported that growth had topped out over 30 years

ago. This has not yet occurred. With the advent of new growth promoting drugs,

we could see additional growth of our youth to satisfy what appears to be an

ideal goal of 1.83 m or more for men. (The average American is continuing to

put on weight at a rapid rate with no end in sight due to excess energy intake

and lack of exercise, and about 1 million of our youth are taking steroids to build

bigger bodies.)

HOW BODY HEIGHT AND WEIGHT AFFECT PERFORMANCE 133

Most experts in this area attribute secular growth to improved nutrition (Kunitz,

1987), sanitation, medical care, and standard of living. The Japanese growth spurt

began following WW II with increasing consumption of red meat, milk, eggs,

and other energy-dense foods. We also know that we can control the growth of

animals through nutritional restriction, especially protein and joules.

In general, most scientists have assumed that this growth is desirable and

reflects better health. However, this assumption has many pitfalls and requires a

more rigorous analysis. The following sections will describe some analyses and

research findings that indicate some dangers in feeding our children for maximum

growth and development. Before presenting these findings, a brief overview of

the concept of geometric similarity will be presented so that changes in body

characteristics with increasing stature will be readily understood.

Increasing Height and Geometric Similarity

Scientists who study animals are familiar with the disproportionate changes in

surface area and volume that occur when length increases. For geometrically

similar animals, this means that as an animal increases in length, its surface area

and volume increase at a more rapid pace. Thus, surface area increases as the

square of the increase in length, and body volume and mass increase as the cube

of the increase in length. Consequently, if the length of an animal is doubled, its

surface area increases by 4 times and its volume by 8 times (R. Alexander, 1989).

The same is true for human beings.

The consequences of these changes in body characteristics have been described

by various scientists, including the exercise physiologists, Astrand and Rodahl

(1986). Some of their key findings in relation to humans are summarized next.

Specific percentage changes are included to illustrate the principles involved.

These percentage changes are based on comparing two people of the same body

type or proportions but differing in height by 20%. A 1.83 m person weighing

86.4 kg is compared to a 1.52 m person weighing 50 kg [(1.829 m/1.524 m)° x

50 kg = 86.4 kg]. Thus, body height is the independent variable to which all

other body characteristics are related.

Body Dimensions: A 20% increase in height or length (L) results in a 20%

increase in body width and depth when geometric similarity is maintained.

Body Area. Body area is proportional to the square of the increase in height;

e.g. (1.83/1.52)* = 1.44. The 1.83 m person will have 44% more surface area

than the 1.52 m person or 2.06 m? vs 1.43 m’.

Body Volume. Body volume is proportional to the cube of the increase in body

height. The 1.83 m person will have a 1.73 times or 73% greater volume of 86.4

liters vs 50 liters for the 1.52 m person.

Body Weight. Body weight is proportional to the cube of the increase in body

134 THOMAS T. SAMARAS

height. The 1.83 m person will weigh 86.4 kg compared to 50 kg for the 1.52

m person. Thus, the taller person will weigh 73% more than the shorter person.

(Body density is about that of fresh water and is about the same for tall and short

people of the same body type.)

Strength. Strength increases as a function of muscle cross section. Since a

cross section is an area, maximum body strength increases as the square of the

increase in height, or by 44% for the 1.83 m person. The maximum strength of

tendons, ligaments, and bones also increase by 44% (or 1.44 times) with an

increase in height of 20%.

Strength to Body Weight. Since body weight increases at a faster rate than

muscle strength, the 1.83 m person will have a strength-to-weight ratio of only

83 (1.44/1.73 = .83). Thus, the shorter person will be 20% stronger in terms of

lifting his or her own body, as in chin-ups or climbing up a rope.

Acceleration. Acceleration is equal to force divided by mass. Since muscle

force increases at a slower rate than body mass, the 1.83 m person will have a

17% lower ability to accelerate his or her body compared to the 1.52 m person.

(acceleration = force/mass = 1.44/1.73 X 100 = 83%; thus, 100% — 83% =

17% less.)

Endurance. For short-term anaerobic activities, the 1.83 m person would be

able to produce energy in proportion to his or her weight. However, for aerobic

activities, the 1.83 m person’s 44% smaller lung surface area in proportion to

body weight will provide a lesser rate of oxygen intake and carbon dioxide

outtake. Similarly maximum cardiac output (1/min) and power output (kj/min) of

the taller person would be 44% greater for a 73% increase in weight or 17% less

on a kilogram-for-kilogram basis.

Brain size. The larger person’s brain would in theory increase by 73%. How-

ever, studies (Pilbeam & Gould, 1974) indicate it increases at a rate proportional

to M-°’ instead of M' (M = body mass). The relationship holds for animals

ranging widely in body size. (Women have 10% smaller brains but weigh 15 to

20% less than men.)

Metabolic rate. The taller person’s resting metabolic rate (RMR) will be lower

on a kg-for-kg basis than that of the shorter person because of the smaller loss

of heat energy from the smaller surface area in proportion to body mass; e.g.,

surface area would be 44% larger for a 73% larger body mass. Empirical data

show that the total body RMR increases at a lower rate than increases in body

mass (RMR « M-”).

There are many other physical characteristics that change disproportionately

with increasing height but a full discussion of these would require a separate

paper. Further information on how the body changes with increasing height or

HOW BODY HEIGHT AND WEIGHT AFFECT PERFORMANCE 135

length can be found in The Truth About Your Height (Samaras, 1994), Astrand &

Rodahl (1986), Haldane (Newman, 1956), and Wentworth (Newman, 1956).

The focus of this report is on measurable quantities related to human size,

performance, health, and survival. However, psychosocial issues need a brief

discussion because of their strong influence on how we perceive larger stature

and body size. Most societies tend to equate taller stature with higher status and

power. Studies reported by Samaras (1994) have shown that taller people receive

more pay for the same work and are favored by employers during hiring and

promotions. John Kenneth Galbraith (2.03 m) observed that height bias was one

of the ‘‘most blatant and forgiven prejudices’’ our society has (Keyes, 1980).

Other researchers have found that most people, especially men, wish they were

taller, and many men feel that they are not as big as they would like to be.

Virtually all of us are imprinted with this bias at an early age, and it makes it

difficult to be receptive to the possibility of smaller human size. Although it took

hundreds of years to convince people that the earth revolved about the sun,

modern communications can accelerate our understanding of new and unusual

ideas, such as described in this paper.

Methods and Materials

Material on how increasing height affects various body characteristics was

obtained from a literature review, using sources involving human physiology,

human engineering, and biomechanics. Other material on mortality was obtained

from sources, such as scientific journals, textbooks, the World Health Organiza-

tion, National Center for Health Statistics, and the California Department of

Health.

Longevity data were obtained from various sources. Since very little formal

research data were available relating body height and longevity, a variety of

sources were used, such as San Diego Veterans Administration Medical Center,

The Baseball Encyclopedia, and biographical data. The Pearson product-moment

correlation coefficient was used to determine the degree of correlation between

height, weight, and longevity.

The basic approach for evaluating the impact of increasing body size on the

quality of life and future human survival was based on configuration management

principles (Samaras, 1988). A key aspect of configuration management involves

assessing the ramifications of changes in the design of a product (human for this

paper) from a multifaceted perspective. Thus, proposed changes are evaluated in

terms of their impact on performance, durability, maintainability, cost, survivabil-

ity, and safety.

136 THOMAS T. SAMARAS

Results

The following material covers key areas related to the average size of human

beings—performance, mortality, longevity, animal research, nutrition, and sur-

vival. Each is critical in determining whether smaller human size would provide

mankind with a better chance of long-term survival on this planet.

Performance

Size does not appear to affect human performance in most areas of human

activities. Both large and small people have demonstrated great abilities in sci-

ence, the arts, technology, and leadership. It is certainly true that people in the

most highly industrialized nations are taller and bigger than most of the people

in less developed countries. Many executives are taller than their subordinates

and tall athletes excel in basketball, football, baseball, and heavyweight boxing.

However, in all these areas, shorter and smaller people have also demonstrated

great capability throughout human history in spite of cultural biases favoring tall

people. Since the achievements of tall people are widely known and accepted,

the following discussion will focus mainly on shorter or smaller people.

History is replete with great civilizations made up of fairly small people. The

ancient Greek males averaged about 1.63 m. Yet, they were great scientists,

athletes, warriors, and artists. The Romans were only a few centimeters taller

than the Greeks and also produced a civilization of impressive proportions. The

outstanding contributions to mankind of the ancient Egyptians, East Indians, and

Chinese were also made primarily by small people. The Maya, Aztecs, and Incas

averaged about 1.6 m and displayed great scientific, artistic, architectural, and

organizational achievements.

More recently, extraordinary Japanese industrial developments were initiated

and implemented by people much shorter than the Japanese youth of today. One

of the leaders of Japanese industrial development was Matsushita, the father of

the electronics industry. He was 1.63 m and 58 kg. Many of the Japanese industrial

leaders were likewise small men by American and Western standards. Highly

successful American or European businessmen or industrialists have included

Ross Perot (1.68 m), Armand Hammer (1.65 m), Aristotle Onassis (1.65 m),

David Murdock (1.63 m), Herbert Haft (1.52 m), and Ronald Perelman (1.73 m).

Ray Kroc and Dale Carnegie were also on the short side.

A study by Walter Bowerman (1947) found that the among 1000 British men

of genius, men = 1.62 m represented over 30% of the population. My studies

have also revealed that famous short or small men represented about 30% of the

famous people listed in Current Biography Yearbooks (1940-1992). A sampling

of famous short people includes Michelangelo (1.57 m), Joan of Arc (1.52 m),

HOW BODY HEIGHT AND WEIGHT AFFECT PERFORMANCE 137

Buckminster Fuller (1.60 m), Martin Luther King (1.70), Picasso (1.63 m), Mother

Teresa (1.52 m), Mahatma Gandhi (1.60 m), Voltaire (1.60 m), Mahler (1.62 m),

Stravinsky (1.54 m), and Kant (1.52 m).

In terms of physical productivity, certain work, ith as sugar cane cutting,

favors taller, bigger workers. Valery Venda (1993) reported that bigger men were

more efficient at shoveling dirt than smaller men. However, Rene Dubos’ study

of Guatamalan peasants found them to be capable of much more rigorous physical

work output than expected for larger North Americans (Dubos, 1980). Balke and

Snow (1965) found that the small (<1.65 m) Tarahumara Indians of Northwestern

Mexico expended over 41900 kj over a 24-hour period during kickball racing

covering over 160 km. The researchers reported that this figure is considered the

upper limit of human voluntary work effort. For small men, the Tarahumara also

exhibited great strength. One young Indian carried a 45 kg burden for 176 km

in 70 hours.

Anthropologists Edmundson and Sukhatme (1990) found the smaller people

of Asia highly productive. They reported that poorer peoples on low food intake

are not physiologically under productive. In fact, most researchers in the devel-

oping world have observed that workers with low-energy intakes are often as

productive or more productive than workers with high intakes.

Some studies have found that shorter children score lower on IQ tests. Recently,

David Sandberg (1995) reported that an on-going study in Britain found that

much of below par performance of short children could be traced to poorer

socioeconomic backgrounds. In addition, anthropologists have found that smaller

brain size is not an indicator of lower intellectual performance. Stephen Jay

Gould (1981) found that brain size has no relation to intelligence. Also women

have smaller brains (as would be expected due to their smaller bodies) but are

not any less intelligent than men. Walford (1980) did not find a decline in mental

acuity among smaller animals subjected to dietary restrictions.

While the most popular sports, such as football and basketball, favor tall men

and women, short people have demonstrated great achievements in other sports.

Sports scientist and physician Robert Cantu (1984) has pointed out that shorter

size is an advantage in gymnastics, long distance running, diving, skiing, figure

skating, and ballet dancing. The reasons for this are the greater aerobic capacity

of lungs with larger surface areas in proportion to smaller body mass, higher

strength-to-weight-ratios, and smaller rotational inertia in comparison to the

body’s ability to produce torque to initiate or counteract rotational movements.

(Rotational inertia is « L° while torque produced by body muscles is « L’.)

The recent winner of the New Orleans gymnastics championships was Domi-

nique Moceanu, a 1.32 m, 32 kg gymnast. Kim Zmeskal, at 1.4 m also demon-

strated exceptional abilities in past gymnastic events. And in spite of basketball’s

138 THOMAS T. SAMARAS

bias, Muggsy Bogues of the Charlotte Hornets, is the shortest man to ever play

professionally. He is 1.6 m and 59 kg. Both he and 1.68 m Spud Webb, who

won the National Basketball Association slap dunk championship, were better

vertical jumpers than teammates who towered over them. In martial arts, Bruce

Lee at 1.68 m was an outstanding champion and many of the martial artists tend

to be small. A review of the heights of professional boxers also revealed that

most were short or small. The November 1995 New York Men’s Marathon was

won by 1.6 m, 50 kg German Silva under the coldest and windiest weather

experienced during the last 40 years. Tegla Loroupe, about 1.52 m tall and almost

36 kg, won the women’s marathon. Both had won the same marathon the previous

year. A few years ago the Olympic weightlifting champion, Suleymanoglu, at

1.52 m and 59 kg was described as the best weightlifter kilogram for kilogram.

Polednak (1979) reported that the U.S.S.R. weightlifters averaged 1.67 m and

77.3 kg while Finnish champion lumberjacks averaged 1.74 m and 72.7 kg.

In warfare, short men have demonstrated high courage and ability. Examples

include the ancient Greek and Roman soldiers. In recent times, the small Gurkhas

of Nepal and WW II Japanese have demonstrated impressive fighting abilities.

Early American Indian warriors, such as the Apache, Comanche, and Navajo,

were short (Samaras, 1994). Military historian Sydney Allinson (1981) described

the physical advantages of short soldiers and reported that the WW I English

Bantam division was cited for valor many times and took the previously impreg-

nable Bourlon Heights from the enemy. These small soldiers (less 1.6 m) also

showed ability in hand-to-hand combat when they engaged the enemy in trenches

and managed to kill 30 enemy soldiers while losing only eight of their own.

During WW II, the most decorated American soldier was Audie Murphy, who

was 1.65 m and about 50 kg.

Mortality and Longevity

This section will review findings on mortality, longevity, and body size. While

the findings may be disturbing, it is important to realize that body size is only

one factor affecting an individual’s longevity. Nutrition, smoking and drinking

practices, heredity, socioeconomic status, medical care, satisfaction with work,

and social support probably account for 80 to 90% of one’s life potential. Addi-

tionally, findings should be evaluated in relation to related studies, potential

confounding factors, and animal studies. For this reason, a number of sources

relating human size, death rates, and longevity are presented for a broader per-

spective.

The following studies refer to body mass index (BMI) as a measure of obesity.

High BMI numbers indicate an overweight condition and low ones a lightweight

condition. BMI’s can be calculated using different formulas. The mortality studies

HOW BODY HEIGHT AND WEIGHT AFFECT PERFORMANCE 139

described in this report used W/H’, where W = weight in kg and H = height in

meters. For example, the typical American male and female have BMI’s slightly

over 26 kg/m’. This is quite high compared to research findings indicating that

one’s BMI should be closer to 20 or 22.

Body weight and mortality. A number of studies have found that the lower

one’s body weight for height, the lower the mortality. For example, the Harvard

Nurses’ Health Study (Manson, 1995) found the thinnest women had the lowest

death rates from heart disease. Over 115000 female registered nurses (30 to 55

years old at the start of the study) were tracked for 16 years. After controlling

for age, smoking, menopausal status, parental history, etc., women with a BMI

< 19 (kg/m*) had the lowest incidence of mortality. Researchers also found that

only a 5 kg rise in weight after the age of 18 increased risk from all causes. The

minimum risk BMI of <19 is a very low body mass, and a 165 cm woman

should weigh only 51.8 kg. This is lighter than most champion marathon runners

(BMI ~20) who are a very thin breed to start with.

Harvard male alumni were evaluated over a 27-year period by Lee, et al.

(1994) for mortality risk with increasing weight. After adjusting for smoking and

other factors, the study of almost 20000 found that men who were 20% below

the U.S. average weight for the same height and age had the lowest mortality.

For example, a 177.8 cm male weighing 82 kg would have to reduce to 65.6 kg

_ to reach ideal weight for height.

Hoffmans’, et al. (1988) study of 78000 Dutch youth tracked these men from

their 18th birthday. They found that men who had a BMI of 19 to 19.99 at the

age of 18 had the lowest mortality over a 32-year period. They also found that

healthy, educated:men who had a BMI of 19 during their youth had the lowest

death rate from all causes.

Larger body size was also implicated in a long-term study being conducted in

the People’s Republic of China by a team of American, British, and Chinese

researchers. They found that larger body size was correlated with higher incidence

of heart disease and cancer. For example, the positive correlation coefficient

between height and cancer was r = .44 (p < .001) and r = .47 (p < .001)

between weight and cancer. For heart disease, they found r = .33 (p < .05) for

height and r = .39 (p < .01) for weight.

Polednak (1979) reported that larger bodied athletes had a slightly higher risk

of dying than smaller athletes. More recently, Baron (Audible, 1994) found in a

study of 6848 National Football League players that the largest players had 6

times the death rate of the smallest players. Baron found a very strong association

between body size, heart disease, and death and recommended that players avoid

bulking up to play football.

Samaras (1994) reported that the California Department of Health found Asians

140 THOMAS T. SAMARAS

and Hispanics had a higher life expectancy than whites. They were also shorter

and lighter than whites except for Hispanic women who were .9 kg heavier than

white women. On a national scale, the National Center for Health Statistics (1995)

reported Asians and Hispanics had considerably lower death rates from cancer

and heart disease compared to bigger whites and blacks.

A review of World Health Organization (WHO, 1992) data indicates that

shorter southern Europeans have significantly lower death rates from cancer and

heart disease after the age of 65—the period when most people die of cancer

and heart disease. Northern countries have 7.2% to 40% excess cancer and heart

disease mortality. Life expectancy for age 65 and over is 1.2 years greater (8.5%

longer) for the southern countries.

Cancer mortality vs height and body weight. A number of major studies

have found a correlation between body height and weight and mortality from

cancer. Giovannucci, et al. (1995) studied 47723 male health professionals be-

tween 40 and 75 years old. They found that the risk for colon cancer increased

with BMI. Independent of obesity, they also found that colon cancer risk was

1.76 times higher for men =1.85 m compared to those =1.73 m. This finding

' was previously reported in other studies and was attributed to the longer colon

and its greater number of stem cells at risk for transformation into cancer cells.

Albanes and Taylor (1990) also found a relationship between adult height and

weight and cancer incidence. They studied the cancer incidence of 24 nations in

Asia, Europe, and North America. All sites cancer incidence was highly correlated

with height among both men (r = .54, p = .001) and women (r = .74, p =

0001). Weight was also related to cancer incidence although higher correlation

coefficients were found for females compared to men. Stronger positive correla-

tions were found for height than for weight.

A large study of 570000 women in Norway found that the tallest women in

all age groups had the highest breast cancer risk for both morbidity (sickness)

and mortality. Overweight was also a risk factor for breast cancer mortality.

However, Tretli (1989) pointed out that while there is a consistent positive relation

between height and breast cancer, it is not a large one. |

The National Cancer Institute found higher cancer risk to be related to taller

stature. Albanes, et al. (1988) studied 12554 men and women 25-74 years old

and found that for most cancer sites in men, and particularly colorectal cancer,

the lowest incidence was among those in the shortest quartile of stature. Taller

women were found to have a higher relative risk for breast (2.1) and colorectum

cancer (1.6). The researchers attributed this higher risk to over nutrition during

early life.

In a study by Swanson, et al. (1989) of 5239 women, an increasing BMI (using

1.5 instead of 2 as a power of height) was associated with increasing risk of

HOW BODY HEIGHT AND WEIGHT AFFECT PERFORMANCE 141

breast cancer. The population was broken up into four BMI groupings. The

relative risk of breast cancer with increasing quartiles of BMI was 1.00, 1.04,

1.40, and 1.29. Independent of weight, height was also found to increase the

relative risk of breast cancer with these values: 1.00, 1.07, 1.15, and 1.27.

Another study relating breast cancer and height was conducted by a Dutch epide-

miologist, de Waard (1975). He found that body mass rather than obesity is an

independent risk factor. The study found that the crude effect of height (including

the increased weight that normally goes with it) accounted for all the risk attributed

to body size. De Waard concluded that Western nutrition alters endocrine and

metabolic processes and predisposes women to increased breast cancer risk.

Other studies have found a correlation between cancer risk and height.

Fraumeni (1967) found that children admitted to the Children’s Hospital Medical

Center in Boston with osteogenic sarcoma (cancer started in bone) were signifi-

cantly taller than a control group of children with non-osseous (nonbone) cancers.

He pointed out that these findings are consistent with the known high risk of

canine bone sarcoma among larger breeds of dogs. Lee and Kolonel (1983) found

taller men had more lung cancer than shorter ones, and Hancock, et al. (1976)

found taller males and females had a higher incidence of Hodgkin’s disease.

Body size and longevity. In 1992, The World Health Organization published

a research paper describing the negative correlation between increasing height

and longevity. The study (Samaras and Storms, 1992) provided longevity data

on several thousand decreased men. A group of 373 decreased veterans revealed

that the shorter half of veterans averaged about 5-year longer lifetimes compared

to the taller half. The correlation coefficient for height vs longevity was r = —.23

(p < .001) and for longevity and weight r = —.20 (p < .001). Evaluation of

over 3200 decreased professional baseball players found a similar pattern. Abso-

lute body weight was also found to be correlated with lower longevity. The

weights studied for baseball players were during their playing years and did not

account for weight changes in later years.

Miller (1990) studied data on 1679 decreased men and women provided by

the Cuyahoga County Coroner’s Office of Cleveland. He found a statistically —

significant inverse relationship between height and longevity. Miller also found

that each additional centimeter in height resulted in a reduction of .47 year in

average age of death. He observed that this reduction was the same as the

difference between men and women based on a 12.7 cm height difference and

7-year longer life expectancy. When he compared men and women of the same

height, men appeared to live as long as women. He noted, however, that his

statistical analysis indicated that height contributed only 10% of the total variation

in longevity. This he found to be expected because there are many other factors

that affect human longevity besides height.

142 THOMAS T. SAMARAS

<160 cm, n=39

2171 cm, n=85

50 =175.cm,n=29

2165 cm, n= 178

2177 cm, n=14

Average Age at Death in Years

155 (61) 160 (63) 165 (65) 170 (67) 175 (69) 180 (71)

Height in Centimeters (inches)

Notes: Each bullet represents the average age for a group of men equal

to or above a certain height or below a certain height; the sample

size (n) is shown to the right of each height.

Error bars = X +/— sx, where X = mean age, sx= sin'/2 ;

Sx = Standard error of the mean, s = standard deviation, and

n = sample size

Fig. 1. Average life span in years vs height in centimeters and inches for 19th Century deceased French

men.

An evaluation of longevity data on over 400 decreased French men and women

(Topinard, 1888) showed that the average age at death declined with increasing

height. Figure 1 depicts the average age at death versus height ranges for men.

(This data was originally collected by Broca, a French surgeon.) Men and women

showed in excess of a 6-year difference between the shorter and taller halves of

men and women. The longevity vs height correlation coefficient for men was r

= —.21 (p < .005) and for women r = —.14 (p < .05).

In 1995, a follow-up study on decreased professional baseball players was

conducted using The Baseball Encyclopedia (Reichler, 1993). The study pre-

viously reported (Samaras and Storms, 1992) covered players who had died up

to 1975. This follow-up study covered 450 players who died from 1976 through

HOW BODY HEIGHT AND WEIGHT AFFECT PERFORMANCE 143

80 Deaths: 1976-1992

Age 70 Deaths: 1910-1975

(Yr) (adapted from previously

published data in the 1992

63 Bulletin of The World Health

Organization)

162 170 178 188 193 cm

64 67 70.5 74 76 inches

Heights of Players

Notes: Error bars = X +/— sx, where x = mean age, sx=s/n12, sx =

standard error of the mean, s = standard deviation, andn = sample size

For players that died from1976 -1992, longevity for 67” players is the average

longevity for players 5’5” - 5’9”; longevity for 70.5” player is average longevity

for players 510” - 5'11”; and longevity for 74” players is the average longevity for

players 6’ - 6’4”. |

The sharp drop for deaths between 1910 and 1975 appears to be an anomaly

based on the death pattern for players who died after 1975.

Fig. 2. Reduction of baseball player average life span with increasing in height.

1992. The results were similar to those of the previous study as shown in Figure

2. Note that the average longevity has risen considerably for all height groups.

However, the shorter players still had significantly greater longevity. For height

vs longevity, the correlation coefficient r = —.27 (p < .005) and for weight r =

—.30 (p < .005).

Animal Research

Walford (1988) reported on numerous studies that have been conducted since

the 1930’s when McCay, Sperling, and Barnes (1943) found that they could

double the maximum lifespan of mice by placing them on an energy restricted

144 THOMAS T. SAMARAS

but well-balanced diet soon after birth. Albanes (1988) also reported that dietary

restriction slows the rate of cell division and this may extend an animal’s lifetime.

These studies have been duplicated with various animals by Masoro (1985) and

others. Another recent study at St. John’s University in New York City found

that by reducing the amino acid methionine in the diet of rats resulted in a size

reduction of 20 to 45% and an increase in lifespan of 45% (Jhanjiani, 1995).

Ross (1976) indicated that body size at maturity was a good predictor of lifespan

independent of dietary variables. He found that larger body size at maturity was

correlated with increasing incidence of degenerative diseases as was found by

Chen, et al. (1990) in the Chinese study previously described.

Gerontologist A. Comfort (1961) also found that within a species, the larger

animals usually had shorter lifetimes. Common examples include horses and

dogs. The larger African elephant also lives about 20 years less than the smaller

Indian elephant. The engineering-biology team of McMahon and Bonner (1983)

also reported that within a species, the very tall and heavy individuals generally

have more foot, leg, and spinal problems and tend to have shorter lives. |

Nutrition, Size, and Health

As was mentioned before, nutrition and growth tend to go together (Tretli,

1989). We have assumed that because our nutrition has sped up the growth

and development of our children, it must be good. However, the World Health

Organization (WHO) found just the opposite to be true. A WHO panel of 11

international nutritional scientists reviewed the literature and 150 background

papers and concluded that radical changes in our food supply have been at the

root of our modern epidemic of chronic diseases (Nutritional Reviews, 1990).

They also observed that in almost every developing country in the world, diet-

related chronic diseases are evolving as the new health problem as people abandon

traditionally healthy diets in favor of affluent foods. This pattern is aggravated

by the public’s perception of what a good diet consists of and the belief that

‘‘good’’ food is the same as ‘‘rich’’ food. The affluent diet of developed countries

was described as one with high consumption of energy-dense foods of animal

origin and foods processed or prepared with added fat, sugar, and salt. The panel

recommended a BMI of 20—22, a value much lower than our present value of

26. The panel concluded its report with the following caveat: °°. . . chronic

diseases are to a large extent manifestations of nutrient excesses and imbalances

in the diet and are thus largely preventable. An epidemic of cancer, heart disease,

and other chronic ills need not be the inevitable price paid for the privilege

of socioeconomic progress.’’ These findings were echoed by Colin Campbell

(Radetsky, 1994) when he stated that we could eliminate up to 90% of our heart

disease and cancer by going on a high quality, low-fat, plant-based diet.

HOW BODY HEIGHT AND WEIGHT AFFECT PERFORMANCE 145

The Tarahumara Indians previously discussed also present an interesting illus-

tration of great physical strength and health with marginal food intake. Health

and fitness physician, Harold Elrick (1978) studied these remarkable people with

a team of other scientists. He found these small people had low-to-normal blood

pressure, low resting pulse rates, optimal cholesterol and triglycerides levels, and

no heart disease or diabetes. Yet, their diet was far below American standards

for protein, kilojoules, vitamins, and other nutritients. Elrick also studied the

inhabitants of a small village in Ecuador. These long lived people were free of

heart disease and consumed only 5028 kj per day with a fat intake of 12 to 19%.

As a result of studying thousands of healthy people throughout the world, Elrick

found that the optimum diet was low in kilojoules, fat, carbohydrates, salt, choles-

terol, and protein.

Survival

Industrial production has increased by 40 times since 1950 and a future popula-

tion of 10 billion people living at current U.S. standards will generate about 400

billion tons of solid waste per year (MacNeill, 1990). At today’s rate, global

resources will become very scarce in future decades. If people continue to increase

in body size, this will compound our waste and resource problems. An overview of

how a population of larger people threatens our survival follows. In all examples, a

population of 1.83 m, 86.4 kg people is compared to a population of 1.52 m, 50

Kg people. It is assumed that buildings, cars, and furniture are scaled to their size

and lifestyles are the same for both populations. Details are given in The Truth

About Your Height (Samaras, 1994).

The consumption of food and water is proportional to body size. Taller people

will consume roughly 73% more food and water. For the present U.S. population

of 265 million, this would require an additional 180 million acres of farmland

for the increased needs of taller people. With about + of the world’s arable land

lost over the last 40 years (ZPG, 1995), the combination of population growth,

body size, and reduced farm land creates a very dangerous scenario for mankind’s

future, which is projected to be 12 billion by 2050.

The annual increase of several key resources, such as aluminum, copper, coal,

and steel, would be about 600 million tons per year for the taller U.S. population.

This estimate would be much larger if all resources were considered. U.S. energy

needs would increase by 40 quadrillion kj per year. Of course, the increase in

resources would be much greater for the world at large, especially in coming

decades when developing countries increase their consumption of resources above

current modest levels.

The additional trash generated by a U.S. population of taller people would be

80 million tons per year. In addition, municipal wastes poured into our coastal

146 THOMAS T. SAMARAS

waters would increase by a several trillion liters per year. Pollution would increase

by 50 to 70%, and the taller U.S. population would pump an additional 3 billion

tons of carbon dioxide into the air each year.

The economic impact of a world of tall people will also be enormous. For the

U.S. alone, about $3 trillion per year would be required to feed, house, entertain,

and transport 265 million people. For example, a typical coast-to-coast airline

flight will cost an additional $33000 (Samaras, 1994). The reason for this added

cost is the lower number of taller passengers that can be carried on an airliner.

Entertainment costs would also go up because fewer seats would be available

for each performance due to larger bodies and longer legs. For example, the

Yankee Stadium was built in the 1920’s and renovated in the 1970’s. The up-

graded stadium contained 9,000 fewer seats because engineers had to account

for an increase of 7.6 cm in the width of the fan’s bottoms.

We now spend about $1 trillion in medical care. This is about 1/7 of our GDP.

How long can a society continue to spend increasing amounts on health care

when huge amounts of money are needed for scientific research, education, new

energy sources, urban renewal, repair of aging city infrastructures, environmental

clean-up, and restoration of forests and degraded farmland. Cleaning up polluted

rivers, lakes, and oceans will also cost huge amounts. Based on the health record

of Asians in California, smaller people raised on healthful diets and lifestyles

could reduce medical costs by many billions per year.

The present picture of unbridled consumption of our resources poses a great

threat to our survival. Over 50% of the original rainforests are now gone. We

are cutting down 40 million acres of rainforest every year. At this rate, the San

Diego Zoological Society predicts that there will be no rainforests in 50 or 60

years. Studies by the Intergovernmental Panel on Climate Change (Monastersky,

1995) has just shifted the threat of climate change from tentative predictions to

certain forecasts. The panel projects a 1°C to 3.5°C increase in the world’s

temperature by 2100 with a rise in the sea level of 15 to 90 cm. As a result,

flooding will displace up to 92 million people each year. Climate change will

also cause significant loss of life through the spread of malaria and other diseases.

Discussion

While not many scientists favor slowing growth, some have begun to see the

problems caused by the diet that is associated with increasing human size. For

example, Walker, Walker, Glatthaar, and Vorster (1994) recently challenged the

belief of many nutritional scientists that maximization of human growth and

stature is a desirable goal. They pointed out that there is evidence which demon-

HOW BODY HEIGHT AND WEIGHT AFFECT PERFORMANCE 147

strates that lower energy intakes with resultant smaller body size may actually

confer a degree of future protection against degenerative diseases, such as athero-

sclerosis and cancer. The WHO panel of 11 international nutritionists reported

similar findings as described in the previous section. Additional support for their

positions comes from our experiences during the Korean and Viet Nam wars.

Autopsies of young American soldiers killed in battle showed that 40% of these

men had significant clogging of their arteries. In contrast, native Korean and

Vietnamese soldiers were found to be free of this problem. American researchers

have found accumulation of fatty deposits in the arteries of children as young as

10 years old.

Confounding Factors and Epidemiological Studies

Many problems exist concerning epidemiological studies as described by Har-

vard professor Charles H. Hennekens. The health of human subjects reflects a

mix of socioeconomic, hereditary, lifestyle, and nutritional factors. In addition,

the mother’s diet and alcohol consumption during pregnancy, congenital heart

problems, rheumatic heart disease, ethnic background, and exposure to toxins

and carcinogens play a role. For example, highly educated people tend to live 4

or 5 years longer than less educated people. Married people in some cases live

10 years longer than those who aren’t—though single people with good social

support networks probably do as well. Government and other researchers have

found that there are more tall people in highly educated classes, and the educated

smoke much less than those with less education. Educated people, such as Harvard

alumni, also tend to have lower BMI’s and better medical care. Also some short

people experience growth problems due to congenital heart problems and other

childhood diseases that curtail their life expectancy without any relation to their

actual height. These factors can sometimes bias the findings of epidemiological

studies in an unexpected way. However, most of these confounding factors would

favor greater longevity for taller people. Therefore, the longevity findings of

Storms, Samaras, and Miller are probably not biased in favor of short people due

to confounding factors.

The use of the BMI is also another potential confounder. The BMI which is

computed by dividing body weight in kilograms by height squared in meters has

been found to be relatively independent of height in terms of reflecting obesity

or Overweight as excess fat. (Not all researchers agree that the BMI formula

actually does this.) However, when short and tall people are compared using this

standard, the result is to compare people that are not geometrically similar. For

example, the ideal weight (midpoint) for a male 1.83 m is 73 kg. The weight of

a 1.52 m person, however, is set at 51 kg. If geometrically similar men were

compared, a 1.83 m male should weigh 15.4 kg more or 88.4 kg. Thus, most

148 THOMAS T. SAMARAS

studies compare short mesomorphic-endomorphic people to tall ectomorphs in-

stead of comparing short and tall people of the same geometrical configuration.

Short people should be compared to tall people with BMI’s 2 to 4 points higher

or a different BMI formulation should be used, such as the Samaras Index (Sama-

ras, 1994) or the Ponderal Index, which was developed years ago but has now

been discarded in favor of the BMI formula used in most modern mortality and

morbidity studies.

Conflicting Findings

As most people know, conflicting findings exist for many studies. Earlier

European and American studies found short people had a high coronary heart

disease mortality. These studies; e.g., Barker (1990), found that taller people had

a lower mortality from heart disease and strokes. However, recent studies by

Hebert et al. (1993), Kannam et al. (1994), a Swedish study, and an unpublished

report by the University of Tennessee found essentially no difference in heart

disease mortality for short and tall people. Similarly, not all cancer studies have ~

found positive correlation between height and cancer mortality or morbidity.

There is a plausible explanation for the idea that short people have more

heart attacks. For example, short people in wealthy countries overeat due to the

abundance of food and the fact that portions tend to be designed to meet the

needs of average size people. Thus, small people tend to become overweight.

Also, smaller people tend to have smaller diameter arteries and blood vessels.

These vessels would tend to become blocked more quickly due to high fat diets.

Another factor may be the relatively larger surface area of the shorter person’s

small intestine is likely to absorb more nutrients from food than a tall person.

(As mentioned before, the surface areas, including the intestine, of a 1.83 m

person are only 44% larger compared to a body weight of 73%.) Thus, short

people are at increased risk for heart trouble if they eat excessively and put on

weight compared to when they were 18 years old. But shortness by itself does

not appear to be a problem in view of the lower heart disease mortality for the

Japanese, Chinese, American Asians and Hispanics, Tarahumara Indians, and

women in general.

Possible Causes for Increased Cancer in Taller People

A 1.83 m person has about 100 trillion cells in his or her body. A 1.52 m

person would have only 60 trillion cells. Thus, a tall person has an additional 40

trillion cells where cancer can be initiated. Free radicals caused by metabolic

processes, radiation, toxins, and physical and emotional stress would have more

Opportunity to initiate a cancer event. Perhaps the large increase in skin cancer

is partially related to our larger skin surface which is exposed to UV radiation.

HOW BODY HEIGHT AND WEIGHT AFFECT PERFORMANCE 149

High-energy cosmic ray exposure of an additional 40 trillion cells may also add

to carcinogenic mutations. Albanes (1988) proposed this explanation for increased

cancer in taller people several years ago and theorized that nutritional restriction

early in life permanently reduces organ cellularity. Perhaps this is why cancer

researcher Robert Good (Segerberg, 1974) found Australian aborigines on low

protein diets showed strong resistance to cancer. However, he found that the

protein restriction had to be severe enough to retard growth and maintained

throughout life.

Possible Reasons for greater heart problems

For geometrically similar short and tall people, the heart and blood volumes

are proportional to body weight. However, the taller person’s heart has to do

disproportionately more work than the shorter person’s heart. The reason for this

is that the tall person’s heart must not only pump 73% more blood through the

body, but must pump it 20% farther. Thus, the taller person’s heart does 2.08

times more work but is only 1.73 times larger. Another possibility may be that

the taller person’s relatively smaller lung surface area may not always provide

as much oxygen as needed to the heart with possible long-term damage.

Theory predicts that blood pressure should be the same for geometrically

similar people of different heights. However, some studies have found a slight

increase in systolic blood pressure or in both systolic and diastolic pressure with

height. It is thus possible that a slight life-long increase in blood pressure could

be, harmful to the heart.

On the other side, taller people should have lower heart rates than shorter

people according to Astrand & Rodahl (1986). Lower heart rates have been found

to promote longer life and would appear to be a positive factor. Fitness experts

recommend resting heart rates of less than 60 beats per minute. This advantage

would reduce the work load on a taller person’s heart.

Brain Size and Intelligence

As mentioned before, some scientists have found that brain size and intelligence

are positively correlated and others have challenged this position. If a correlation

is found, it will most probably be related to socioeconomic class and taller stature

as was mentioned before (Sandberg, 1995). Even if a real difference in IQ were

found, it would not be an important factor in survival of humankind. Our problems

stem mainly from lack of wisdom and self-discipline, self-absorption, and igno-

rance. IQ is probably not the root cause of human failure because people with

normal intelligence have a great capacity to function well if they have the proper

motivation, education, and training.

150 THOMAS T. SAMARAS

Health Risks of Dietary Restriction

Traditional beliefs indicate that undernutrition is dangerous to the health of

our children and ourselves. Certainly malnutrition is not good for anyone. But

there’s considerable evidence to indicate that a sparse but well-balanced diet

promotes good health. For example, during WW II Strom and Jensen (1951)

reported that food in Norway became scarce, and the shortage was primarily of

high fat and cholesterol foods, such as meat, milk, butter, cheese, eggs, and sugar.

The average energy intake dropped 18%. At the same time stress increased due

to the German occupation. Yet, there was a sharp fall in mortality for males and

females in all age groups. After the war, mortality rose sharply when traditional

dietary practices were resumed. Food restriction during the war resulted in similar

patterns, including drops in breast cancer, for Denmark, England, Wales, and

Holland.

Another study by Spanish physician Vallejo (Walford, 1988), reduced the

energy intake of 90 patients over 65 years old at a nursing home in Madrid,

Spain. One day he fed the group 9637 kj and the other day 3708 kj. A control

group of 90. patients was fed 9637 kj every day. This 3-year study found that

the lower kilojoule patients had half the mortality of the normally fed group, and

the members of the lightly fed group spent 123 days in the infirmary compared

to 219 day for the group with a higher food intake.

Dubos (1980) participated in a nutrition research in Guatemala. He found the

natives were very sparsely fed compared to our standards and many children and

youth died of infections. However, those who survived to adulthood were capable

of much greater work output compared to larger Europeans and North Americans,

such as *“‘carrying heavy burdens on their backs over long distances up and down

mountains’’. They were short and seemingly frail but they were healthy and lived

to a good old age.

Kagawa (1978) found that Okinawan children ae, 40% fewer kilojoules

than mainland children. Adults consume about 20% fewer kilojoules. Yet, Okina-

wans have about half the mortality of mainland Japanese and more centenarians

than anywhere else in the world. They are also the smallest of the Japanese.

Kagawa concluded: ‘‘. . . westernization of the Japanese diet may be beneficial

provided rice, fish and vegetables are kept as major foods and salt, sugar, choles-

terol, saturated fats and total energy are restricted.’’

During a 12-year study, Murray et al. (1992) found that chronic undernutrition

in adult Masai did not negatively affect their health. The undernourished of group

403 Masai had BMI’s of about 18.5 for men and 16.6 for women. The normally

nourished group of 386 Masai had BMI’s of about 20.4 for men and 18.2 for

women. While none of the Masai (18-35 years of age) died of illness, the

HOW BODY HEIGHT AND WEIGHT AFFECT PERFORMANCE 151

normally nourished had 4 times the malarial attacks and 2 times the infections.

Five normally nourished Masai were struck by malignant tumors whereas none

of the undernourished had them. The normally nourished were also incapacitated

2.4 times more days per year compared to the undernourished.

Environmental Threats

Benjamin H. Alexander (1989) observed that the world’s population is our

number one pollution problem. He also observed that smaller size humans would

help conserve our resources including water, which is so vital to our existence.

He believes that if we do not reduce the size of humans, the most powerful

people in our world will take control and billions of people could be killed

outright to reduce the world population. Billions more, he feels, could be enslaved.

Although this is an unpleasant scenario, the works of many autocrats, such as

Hitler and Stalin, provide ample evidence that it can happen very quickly when

conditions are right.

The former chief scientist of the National Oceanographic and Atmospheric

Association, Sylvia Earle (1995) observed that ‘“‘Depending on the choices we

make, our species may be able to achieve a viable, sustainable future or we may

continue to so alter the nature of our planet that our kind will perish.’ We

are all aware of the need to reduce population growth, pollution, overfishing,

deforestation, and loss of arable land. One way to minimize the damage that

Earle is concerned about is to reduce the size of the average human on earth.

How Should People Respond to These Findings

The value of any person is totally unrelated to his or her size. Therefore, no

one should feel superior or inferior due to their height. Our value to society and

self respect should depend on how we behave towards our fellow humans and

other life forms and what contributions we make to improve our society and

world. Since those of us living today can’t change our heights, our response to

these findings should be to encourage a change in our current attitudes so that

the world will be receptive to future generations of smaller people.

As mentioned, stature is a small part of the total picture affecting an individual’s

longevity. Regardless of one’s height, good health practices are much stronger

factors in determining how long anyone will live. It appears that the recommenda-

tions of epidemiologist Tim Byers (1995) apply to how we should view our

stature on a personal level. Except for large weight gains, he observed that studies

demonstrating the impact of small weight increases on mortality have a relatively

small impact in terms of an individual’s lifespan. Unfortunately the public over-

152 THOMAS T. SAMARAS

reacts to various health and nutrition findings, and most people become frustrated

with their attempts to lose weight because they cannot attain or maintain the very

low body weights recommended. Byers rightly recommends that we should be

focusing On more important aspects of good health, such as regular exercise and

good nutrition. However, as described in this paper, additional emphasis is needed

on low-energy and high fiber diets, avoidance of smoking and excessive drinking,

stress management, and work satisfaction. Leading nutritional scientists, such as

Harvard’s Walter Willett, recommends that we should eat lots of fresh fruits and

vegetables because extensive research shows they protect against cancer and heart

disease. He also recommends eating red meat no more than once a week. These

practices are probably more important to one’s health than focusing solely on

attaining one’s ideal weight. However, lower weight is also important, but it

should be achieved by a permanent change in lifestyle. Therefore, being very

tall should not be viewed with alarm if good health practices are followed.

Samaras’ studies found many tall and large men who lived into their 80’s and

90’s. However, weight control is a very important strategy for tall people.

Impact of Longer Living People on Environment and Economy

Analysis of the impact of longer living short people indicates that the cost

would be almost $400,000 less for a postulated 12-year longer lifespan (Samaras,

1994). Resource consumption would be about 50 to 73% less over a lifetime,

which would more than compensate for the 17% longer lifetime. Therefore, a

population of longer living short people would not neutralize their benefits. In

addition, able bodied retired people could support many worthwhile environmen-

tal and social projects.

Delaying Puberty

According to Foster, a neuroendocrinologist, a high kilojoule diet is propelling

American girls and boys into sexual maturity sooner (Discover, 1995). The aver-

age age for the start of menstruation is now 12.5 years, down from 14 years in

1900. Among hunter-gathers societies menstruation occurs at 16. In our complex

world, early sexual maturation aggravates our social problems because the chil-

dren have not had enough time to gain in wisdom and self-control. As Benjamin

H. Alexander (1989) has said, their lives could be improved by delaying the

onset of hormonal changes that can be distracting from school work and a more

gradual maturation process. This delay would probably also help parents maintain

better control of their children for a longer period than they now have. The

HOW BODY HEIGHT AND WEIGHT AFFECT PERFORMANCE 153

possible benefits would be fewer teenage pregnancies, less anti-social behavior,

and fewer ruined lives.

Conclusion

Virtually all of us have been conditioned to believe that taller and bigger is

better than shorter and smaller. Parents often brag about their children growing

rapidly and being taller than their peers. Yet, we never hear parents proudly

proclaim that their children are growing up short and light. Unfortunately this

deep-seated bias is not based on good science. Smaller people can be just as

physically and mentally fit as taller people. In addition, a population of smaller

people offers one path to reducing the damage we will wreak on the planet in

_ the coming years. As illustrated in the previous sections, smaller people require

much less of almost everything and can help assure a good life for ourselves,

children, and grandchildren. If we plan to populate the planet indefinitely, smaller

people will allow us to achieve this goal.

William Conner (Liebman, 1995), a heart disease expert at the Oregon Health

Sciences Center, reported that pre-affluent Japan had ; the heart disease death

rate that we have now. Today the Japanese are bigger and taller and have closed

the gap so that they have only 5 the death rate of what the U.S. has. Is this a

good trade-off? Wouldn’t our children be better off in a society where everyone

was smaller and healthier? Perhaps the only one who can answer this question

is someone who has suffered from cancer or heart disease, its treatment, and

after-effects.

We have the ability to reduce the size of future generations now. According

to Walford (1986), we could reduce height by up to 20 cm through dietary

restriction. However, psychosocial resistance to evolving smaller size children is

very strong. Achievement of this goal will require wide-scale public education

to achieve an appreciation of the benefits of smaller human size.

The choice to become smaller may not be a future option, and we may simply

become smaller due to the scarcity of rich or energy-dense foods. The American

Association for the Advancement of Science (Krajick, 1995) reports that food

shortages in the next 50 years will become quite severe. Meat, milk, eggs, and

butter will become rare, and we will have to eat more simple foods that come

directly from the earth. In view of this anticipated shortage, food scientists have

an important mission to develop highly palatable substitutes for regular and fast-

food fare which are low in kilojoules and fat but high in vital nutrients and fiber.

They have already achieved some outstanding results in producing good tasting

vegeburgers and by substituting cellulose in baked goods, reducing the energy

content significantly.

154 THOMAS T. SAMARAS

Scientists concerned with our survival say that we can’t continue to expand

our population on an earth of finite size. Actually, the earth is getting smaller.

Certainly its mass is not changing, but the arable land and terrestrial and oceanic

resources are declining as massive human consumption converts more and more

resources into degraded materials that are no longer available for human use.

Economist Georgescu-Roegen (1971) said many years ago that the entropy of

the earth is irreversibly increasing as we convert natural resources into disordered

matter. He also pointed out that ““bigger and better’’ washing machines, automo-

biles, and superjets lead to ““bigger and better’’ pollution. It appears that “‘bigger

and better’’ people also increase the earth’s entropy and reduce the human race’s

chances for long-term survival. Our planet cannot afford the luxury of supporting

12 billion people averaging 77 kg—the average weight of today’s middle-aged

Americans. |

Postnote

The research presented in this paper was initiated as a byproduct of the applica-

tion of the second law of thermodynamics (law of entropy) to human organizations

and their tendency to become disordered with time (Samaras, 1973). This law

was interpreted to mean that as an organization grows in size and energy content

(money), it tends to become increasingly disordered unless energy is expended

on internal operating systems to restore them to proper order. This application

of the entropy law was then applied to human aging (Samaras, 1974). This

theory hypothesized that aging (progressive disorder) is accelerated over time as

a function of the mass and energy consumption of the body. The thermal physics

expression o = log g (N, U) was used as a basis for increasing disorder. The

symbol o refers to entropy and, g, the number of possible states that a system

(the body) can be in. The number of states that a system can be in increases with

the number of body cells (N) and the average daily energy of the system (U).

Thus, to minimize aging, the optimum body configuration would be one in which

the combination of body mass and energy needs would be minimized. From this

beginning, research proceeded from longevity studies to evaluation of body size

on performance, resource, energy needs, and survival (Samaras, 1978).

Acknowledgments

Special thanks are given to Benjamin H. Alexander, Ph. D., Rick Puetter, Ph.

D., Lowell H. Storms, Ph. D., Dennis D. Miller, Ph. D., Harold Elrick, M.D.,

Bernard Rimland, Ph. D., and Claude Hayes, J. D. for their support and sugges-

tions before preparation of this manuscript.

HOW BODY HEIGHT AND WEIGHT AFFECT PERFORMANCE 155

References

Alexander, B. H. (1989). Why is the environmental crisis happening?, Vital Speeches of the Day, LVI, 124-

128.

Alexander, R. M. (1989). Dynamics of dinosaurs and other extinct giants, New York: Columbia University

Press.

Albanes, D. (1987). Total calories, body weight, and tumor incidence in mice. Cancer Research, 47:1987-

1992.

Albanes, D., Jones, D., Schatzkin, A., Micozzi, M., & Taylor, P. (1988). Adult stature and risk of cancer.

Cancer Research, 48:1658—1662.

Albanes, D., & Taylor, P. (1990). International differences in body height and weight and their relationship

to cancer incidence. Nutrition and Cancer, 14:69-77.

Albanes, D., & Winick, M. (1988). Are cell number and cell proliferation risk factors for cancer? Journal of

the National Cancer Institute, 80:772—-775.

Allinson, S. (1981). The Bantams, the untold story of World War I, London: Howard Baker.

Astrand, P., & Rodahl, K. (1986). Textbook of work physiology. New York: McGraw-Hill.

Balke, B., & Snow, C. (1965). Anthropological and physiological observations on Tarahumara endurance

runners. Am. J. Phys. Anthrop., 23:293—302.

Barker, D. J. P., Osmond, C., & Golding, J. (1990). Height and mortality in counties of England and Wales.

Annals of Human Biology, 17:1-6.

Bowerman, W. G. (1947). Studies in genius. New York: Philosophy Library.

Byers, T. (1995). Body weight and mortality. The New England Journal of Medicine, 333:723-724.

Cantu, R. (1984). Clinical sports medicine. Massachusetts: The Collamore Press. p. 12.

‘Chen, J., Campbell, T., Li, J., & Peto, R. (1990). Diet, life-style and mortality in China. New York: Cornell

University Press.

Comfort, A. (1961). Life span of animals, Scientific American, 205:114.

Current Biography Yearbooks (1940-1992), C. Moritz (Ed.), The H. W. Wilson Company.

de Waard, F. (1975). Breast cancer incidence and nutritional status with particular reference to body weight

and height. Cancer Research, 35:3351-—3356.

Discover (1995). Health, sweet 13, Discover, 16:18—19.

Dubos, R. (1980). Nutritional ambiguities. Natural History, 89:14—-21.

Earle, S. (1995). Sea change. New York: G.P. Putnam’s Sons.

Edmundson, W., & Sukhatme, P. (1990). Food and work: poverty and hunger? Economic Development and

Cultural Change, 38, pp. 263-280.

Elrick, H., Crakes, J., Clarke, S. (1978). Living longer and better— guide for optimal health. Mountain View,

CA: World Publications.

Fraumeni, J. 01987). Stature and malignant tumors of bone in childhood and adolescence. Cancer, 20:967-

973.

Georgescu-Roegen, N. (1971). The entropy law and the economic process. Massachusetts: Harvard University

Press.

Giovannucci, E., Ascherio, A., Rimm, E., Colditz, G., Stampfer, M., & Willett, W. (1995). Physical activity,

obesity, and risk for colon cancer and adenoma in men. Annals of Internal Medicine, 122:327-—334.

Gould, S. J. (1981). The mismeasure of man. New York: Norton.

Haldane, J. B. S. (1956). On being the right size. In J. R. Newman (Ed.), World of mathematics, Vol 2, pp.

952-957, New York: Simon & Schuster.

Hancock, B., Mosely, R., & Coup, A. (1976). Height and Hodgkin’s disease. The Lancet, 2:1364.

Hansen, R., & Holley, M. (1967). Is there a better human size? Technology Review, July, 14-16.

Hebert, P. R., Rich-Edwards, J. W., Manson, J. E., Ridker, P. M., Cook, N. O‘Conner, G. T., Buring,

J. E., & Hennekens, C. H. (1993). Height and incidence of cardiovascular disease in male physicians,

Circulation, 88:1437—1443.

Hoffmans, M. D. A. F., Kromhout, D., & De Lezenne Coulander, C. (1988). The impact of body mass

index of 78612 18-year old Dutch men on 32-year mortality from all causes, J. Clin Epidemiol, 41:749-

756.

Jhanjiani, M. (1995). Low protein, long life? Longevity, 7:14.

Kagawa, Y. (1978). Impact of westernization on the nutrition of Japanese: changes in physique, cancer,

longevity and centenarians, Preventive Medicine, 7:205—217.

Kannam, J. P., Levy, D., Larson, M., & Wilson, P. W. F. (1994). Short stature and risk for mortality and

cardiovascular disease events, Circulation, 90:2241—2247.

Keyes, R. (1980). The height of your life. Boston: Little, Brown & Company, p. 25.

Krajick, K. (1995). Will your long life be a good life? Longevity, 7, Sep., 42, 66, 70, 72, 74.

Kroemer, K. H. E., Kroemer, H., & Kroemer, K. E. (1990). Engineering physiology, bases of human factors/

ergonomics, 2nd edition, New York, Van Nostrand Reinhold. p. 17.

156 THOMAS T. SAMARAS

Kunitz, S. (1987). Making a long story short: a note on men’s height and mortality in England from the first

though the nineteenth centuries, Medical History, 31:269—280.

Liebman, B. (1995). Heart disease, Nutrition Action Health Letter, 22:4—7.

Lee, I-M., Manson, J. E., Hennekens, C. H., & Paffenbarger, R. S. (1993). Body weight and mortality-a

27-year follow-up of middle-aged men, JAMA, 270:2823—2828.

Lee, J., & Kolonel, L. (1983). Body height and lung cancer risk (letter to the editor). The Lancet, 1:1662.

Manson, J. E., Willett, W. C., Stampfer, M. J., et al. (1995). Body weight and mortality among women,

The New England Journal of Medicine, 333:677—684.

Masoro, E. (1985). Nutrition and aging—a current assessment. Journal of Nutrition, 115:642—648.

McCay, C., Sperling, G., & Barnes, L. (1943). Growth, ageing, chronic diseases, and life span in rats.

Archives of Biochemistry and Biophysics, 2:469—479.

McMahon, T. A., & Bonner, J. T. (1983). On size and life. New York: Scientific American Books, pp. 53 &

122:

MacNeill, J. (1990). Strategies for sustainable economic development, In Managing planet earth. (p. 109).

New York: Freeman & Co.

Miller, D. (1990). Economies of scale. Challenge, May—June, 58-60.

Monastersky, R. (1995). World climate panel charts path for action. Science News, 148, Nov. 4, 292.

Murray, M. J., Murray, A. B., and Murray, N. J. (1992). Does chronic undernutrition in adult Masai influence

morbidity and survival? Proc. Nutr. Soc., 52:107A.

National Center for Health Statistics. (1995). Health, United States 1994. pp. 109-111. Hyattsville, Maryland:

Public Health Service.

NIOSH study refutes myth of early death (1994). The New Audible, 17:1-—2.

Nutrition Reviews (1991). Diet, nutrition and the prevention of chronic diseases—A report of the WHO study

group on diet, nutrition and prevention of noncommunicable diseases, Nutrition Reviews, 49:291—301.

Pilbeam, D., & Gould, S. (1974). Size and scaling in human evolution. Science, 186:892—901.

Polednak, A. (Ed.) (1979). The longevity of athletes (Ed.), Springfield, Illinois: Charles C. Thomas, p. 47.

Reichler, J. Ed. (1993). Baseball encyclopedia, 19th edition, New York: Macmillan.

Ross, M., & Lustbader, E. (1976). Dietary practices and growth responses as predictors of longevity. Nature,

262:548—553.

Samaras, T. (1973). Measuring organizational entropy. Akron Business and Economic Review, 4:15—21.

Samaras, T. (1974). The law of entropy and the aging process. Human development, 17:314—320.

Samaras, T. (1978). The stature factor—how important is human size in the energy, pollution, and economic

picture? Electric Perspectives, 78/6:8—16.

Samaras, T. (1978). Short is beautiful. The Futurist, X1I:252—255.

Samaras, T. (1988). Configuration Management Deskbook, Annandale, VA: Advanced Application Consul-

tants, Inc.

Samaras, T. T., & Storms, L. H. (1992). Impact of height and weight on life span. Bulletin of the World

Health Organization, 70(2):259—267.

Samaras, T. (1994). The truth about your height—exploring the myths and realities of human size and its

effects on performance, health, pollution, and survival. San Diego: Tecolote Publications.

Sandberg, D., Brook, A., & Campos, S. (1994). Short stature: a psychosocial burden requiring growth

hormone therapy? Pediatrics, 94:832—840.

Segerberg, O. (1974). The immortality factor. New York, E. P. Dutton, p. 210.

Strom, A., & Jensen, R. (1951). Mortality from circulatory diseases in Norway 1940-1945, Lancet, 1:126-

Bagh

Swanson, C., Brinton, L., Taylor, P., Licitra, L., Ziegler, R., & Sh aie C. (1989). Body size and breast

cancer age assessed in women participating in ine breast cancer detection demonstration project. American

Journal of Epidemiology, 130:1133-1141.

Topinard, P. (1888). Le poids de l’encephale d’pres les registres de Paul Broca, Memoires de la societe

d’anthropologie, de Paris, 3, 2nd series, 1—41.

Tretli, S. (1989). Height and weight in relation to breast cancer morbidity and mortality. A prospective study

of 570,000 women in Norway. Int. J. Cancer, 44:23-30.

Venda, V. F. (1993). Work efficiency vs complexity: introduction to ergodynamics. J. Wash. Aca. Sci., 83:9-

Sl iaee

Walford, R. (1986). The 120 year diet. New York: Pocket Books, p. 126.

Walker, A. R. P., Walker, B. F., Glatthaar, I. I., & Vorster, H. H. (1994). Maximal genetic potential for

adult stature: Is this aim desirable? Nutrition Reviews, 52:208—215.

Wentworth, D. (1956). On magnitude. In J. R. Newman (Ed.), World of mathematics (Vol 2, pp. 1001-1046).

New York: Simon & Schuster.

WHO excerpts (1991). Diet, nutrition and the prevention of chronic diseases—a report of the WHO study

group on diet, nutrition and prevention of noncommunicable diseases. Nutrition Reviews, 49:291—301.

WHO 1992 World Health Statistics Annual (1992). Geneve: World Health Organization, pp. D140—D367.

ZPG Reporter (1995). 27, July/August, 12.

Journal of the Washington Academy of Sciences,

Volume 84, Number 3, Pages 157—167, September 1996

Locality, Realism, Lorentz Invariance

and Quantum Mechanics

Joseph Di Rienzi

Mathematics/Physics/Computer Science Department

College of Notre Dame of Maryland

ABSTRACT

Quantum mechanics is a powerful predictor of physical phenomena, but all attempts to

incorporate it with traditional characteristics of scientific theories have proved unsuccessful.

This paper reviews the incompatibility of quantum mechanics with so named local realistic

theories of nature. It goes beyond the Bell statistical inequality to demonstrate that the Hardy

two particle condition produces logic contradictions that prevent realistic theories from being

consistent with quantum mechanics if they are not only non-local, but also non-Lorentz

invariant. The paper questions these contradictions. In particular, it examines two experimental

conditions, state preparation and detector efficiency, to demonstrate that under less than

ideal conditions, the Hardy contradictions reduce to, at best, the less direct Bell statistical

contradictions. The paper concludes with the suggestion that the failure to experimentally

reproduce these logical contradictions may be yet another manifestation of the Uncertainty

Principle protecting nature from humanity’s discovery of its secrets.

Quantum mechanics is a major cornerstone of twentieth century science. By

all standards, it has had astonishing success in predictions of behavior on the

atomic, nuclear, and elementary particle scale. Recent attempts have been made

to extend quantum mechanics to a quantum cosmology to understand the creation

of the Universe itself. Despite its scope and achievements quantum mechanics

remains an unsettling description of nature, placing demands that appear counter

intuitive and producing results that seem baffling to all who try to use it to

comprehend the physical world. This paper will illustrate the contradictory nature

of quantum theory when attempts are made to incorporate it with traditional

attributes of physical theories. The paper will examine experimental limitations

that present obstacles in demonstrating these contradictions and suggest, perhaps,

there is a fundamental limitation in our human ability to observe these bizarre

behaviors.

Corresponding Author: Joseph Di Rienzi, College of Notre Dame of Maryland, 4701 North Charles Street,

Baltimore, MD 21210, Voice: 410-532-5319, FAX: 410-532-5793, jdirienz@ndm.edu

157

158 JOSEPH DI RIENZI

I. Local Realism and Quantum Mechanics

At the center of quantum theory is the Heisenberg Uncertainty Principle which

sets restrictions on the ability to precisely measure conjugate variable pairs, such

as position and momentum or spin in two orthogonal directions, simultaneously.

For example, if Ap is the uncertainty in momentum and Ax< is the position’s

uncertainty, the product of their uncertainties is expressed as the following in-

equality:

ApAx = h/2 (1)

where fh = h/2p, h is Planck’s constant (6.63 < 10°** joule-sec).

Kg. (1) illustrates the inverse relation between the two observables’ uncertain-

ties. The smaller the uncertainty in one of the variables, the larger the uncertainty

in the other so that their product satisfies the inequality. Therefore, the Uncertainty

Principle claims making measurements to determine precisely a physical observ-

able always disturbs its corresponding conjugate variable. Due to the intrinsic

vagueness, a probablistic interpretation is used that replaces the classical deter-

ministic view of understanding nature.

In the development of quantum theory, several of its founders were troubled

by these aspects of it, none perhaps more than Albert Einstein. After first trying

to refute quantum mechanics on consistency, Einstein looked to demonstrate

quantum mechanics was not a complete theory of nature. In a paper with col-

leagues Boris Podolsky and Nathan Rosen, Einstein (1935) suggested a thought

(Gedankan) experiment that forced quantum mechanics to violate a seemingly

irrefutable assumption of nature.

The EPR paradox, as it became known, employs a system consisting of two

particles that are strictly correlated (entangled) with each other, such that properties

about the entire state are known (i.e., total momentum). This property of the entire

system enables one to use measurements on one particle in one place, to deduce

the corresponding property of the other without disturbing its conjugate observable.

For example, consider two particles (1 and 2) which are initially together at

rest. If they are forced apart by some reaction, we can still conclude that their total

momentum is zero. If the momentum of particle 1 is measured, by conservation of

momentum, the momentum of particle 2 would be known, without disturbing its

position. Quantum mechanics would not allow this to happen. It would, therefore,

claim measuring particle 2’s momentum disturbed its position, which affected

particle 1’s position. Thus, in order to prevent a violation of the Uncertainty

Principle, there would have to be an instantaneous transfer of information from

one particle to another affecting the corresponding observable.

EPR considers this a violation of locality in that ‘‘no reasonable’’ theory of

nature would allow a measurement of one object at one place to affect the

QUANTUM MECHANICS 159

measurement of another object at a different place. EPR then claims, to obtain

the predictions of quantum mechanics, objects must have pre-existing conditions

or hidden variables that determine the outcomes of measurements. These are

called hidden variables or realistic theories. These hidden variables are consid-

ered elements of reality that exist independent of a physical measurement. Quan-

tum mechanics is thus considered a statistical approximation to much more com-

plete realistic theories that do not have to resort to non-locality.

It was assumed by EPR that these local realistic theories once found would

be consistent with the predictions of quantum mechanics. That assumption was

shown to be untenable by John Stewart Bell (1964). Bell proved mathematically

that no local realistic theory could always give the same predications as quantum

mechanics in a two particle, strictly correlated state. An example of such an

entanglement would be two spin 1/2 particles with total spin 0. Bell’s Theorem

1S an inequality relationship among probabilities of measurement of observables

for this system along different directions, derived under the assumption of local

realism. Quantum mechanics exhibits a statistical disagreement with this inequal-

ity, indicating that local realistic theories would differ with quantum theory a

certain fraction of the measurements. These violations set up a criteria for testing

local realistic theories against quantum theory. In a series of experiments (Freed-

man & Clauser (1972), Clauser (1976), Fry & Thompson (1981), Aspect et al.

(1981), Aspect et al. (1982), Ou & Mandel (1988), Lamehi-Rachti & Mittig

(1976)) quantum theory has consistently been validated.’

A strengthening of the conditions producing a violation was derived by

Greenberger, Horne, Shimony, and Zeilinger, GHSZ (1990). In looking at three

or more particles in strictly correlated states, GHSZ obtained, assuming local

realism in quantum mechanics, statements in direct contradiction without the use

of Bell inequalities. The violation is such that it would occur with each individual

measurement, and not, as in Bell’s proof, only be evident on a statistical basis.

Lucien Hardy (1992, 1992a, 1993) has devised thought experiments on two

particle systems which demonstrate the contradiction of local realistic theories

with quantum theory also without inequalities, but, unlike GHSZ, occurring some

of the time. In order to achieve this violation, Hardy prepares states or arranges

apparatus of experiment such that one term is missing from the complete state

expressed in some basis. He calls these states non-maximally entangled. In

addition, Hardy is able to demonstrate that realistic theories are not Lorentz

invariant.

In order to demonstrate the previous statement, consider these definitions taken

from Hardy (1992):

—Locality: Measurements in one place on a single object can not affect measurements

at another place on another object simultaneously.

160 JOSEPH DI RIENZI

—Realism: If you can predict with certainty (probability 1) the result of a physical

measurement, then there exists an element of reality corresponding to this physical

quantity and having a value equal to the predicted measurement result.

—Lorentz Invariance: The value of an observable is frame independent. The value of

an element of reality corresponding to a Lorentz invariant observable is itself invariant.

Hardy proceeds to demonstrate realistic theories must be non-local if they are

to agree with quantum mechanics for thought experiments involving positron-

electron pairs (1992) and photon pairs generated by degenerated parametric down

conversion (1992a). Clifton and Niemann (1992) generalize this argument to two

entangled spin singlet states.

Hardy (1992) also shows realistic theories of quantum mechanics are not

Lorentz invariant for an interference experiment involving electrons and posi-

trons. In this study I will generalize by using Hardy’s non-maximally entangled

state vector (1993) to demonstrate non-Lorentz invariance for realistic theories

of quantum mechanics. I will then examine some of the implications of trying

to avoid this contradiction.

II. Non-Lorentz Invariance for Hardy State

Consider a particle that has an observable that can take on one of two values,

say t+ Or —.

| + >, represents particle 1 in state with observable value +

| — >, represents particle 2 in state with observable value —

These states are represented in the traditional Dirac notation |>.

Assume in an entangled condition two such particles must have the same value

of this observable, either both + or both —. They can never be in the condition

where one of the particles has this value of this observable + and the other’s

corresponding value is —. Therefore, the states | + >,| — >, and | — >,| + >.

can never occur.

Looking at Hardy’s state for any two particle entangled system with basis

states |+>; (i = 1, 2)

Ui Oe | tee a ee aD (2)

where yw is the complete state of the entangled system and

e+ B=) | (3)

| + > and | — > can be decomposed into basis vectors |u,) and |v;> which

represent intermediate states that can be subject to observation.

QUANTUM MECHANICS 161

| + >; = blu;> + ia*|v,;> (4a)

| — >, = ialu> + b*lv,> (4b)

where |a|? + |b|* = 1

Prepare the state such that the coefficient in front of the |u,;> |u.> term is set

equal to zero.

ab? + Ba’? =0

thus

ala = —b/6

Using Eqs. (4a) and (4b) in Equation (2)

b = —[V(aB)|u,>|vo> + VeB)|vi>|w> + (lae> — |B>)|vi>|v2>]

(5)

Then vectors |u,> and v,;> can be decomposed into basis vectors |c,;> and |d,>

which represent final measurement states.

ws = A*lc> = Bld> (6a)

iv. = B*ic¢> +. Ald> (6b)

where

A = V(aB)N — af) (7a)

B =(la| — |B))NG — af) (7b)

and

JA|? + |B]? =1

Now consider physical observables U; and D, with corresponding operations

U; = |u;> <u; | and D, == | dite <d;|

Since they are projection operators, their respective physical quantities can

_only take on values 0 and 1 corresponding to the eigenvalues of U; and D;. From

the definition of realism, we claim that if

Wu, =u. then, UW = 21

where the symbol U represents with probability 1 that the value of measuring

the corresponding U equals 1.

Now consider a measurement in the laboratory reference system such that

measurements on particles 1 and 2 occur simultaneously. From Equation (5)

162 JOSEPH DI RIENZI

U,U,|y> =

since, as designed, there is no |u,>|u,> term in Equation (5).

Then from the ideas and notation above, we can conclude with probability 1

that

UjUn= 0 (8)

This result implies that U, and U, can not be both 1 simultaneously.

Now, change reference systems into one in which particle 2 splits into |d,>

and |c)> before particle 1 splits into |d;> and |c,>. Using Equations (6a) and

(6b), the state vector Eq. (5) is written as

y = —[B*V(af)|u>|o> + aBAlu>|d> + ((aP)A* + (lal

— |B|)B*)|vi>|ce.>] (@)

There is no |v,>|d,> term, since its coefficient (|a| — |@|)A — \(@B)B = 0

From Equation (9), if we measure D, to be 1 on particle 2, then U, must be

1 on particle 1 since only the |u,>|d,> contains |d,>

In other words if D, = 1 then U, =1 (10)

Similarly, if we choose a reference system such that particle 1 splits before

particle 2 splits, the state of the system can be written as

y= —[((e@p)A* — (Ja| — [B))B*le:>|v2> + V@B)B*|c1>|v2>

+ V(aB)Ald>|uw>) UD

In the case there is no |d,;>|v.> term because coefficient

—By(a8) + (la| — [BIA = 0

By the same reasoning as in Equation (9), we see from Eq. (11)

IfD, =1 then U,=1 (12)

where in Equation (12) the same condition of reality is used.

Quantum mechanics is Lorentz invariant at the level of its statistical predictions

in any frame. Detection (or non-detection) of an observable should be independent

of the reference frame the observable is measured in. Since these observables

exist independent of the measurement system, we can compare them.

To this end, consider an experiment in which D, equals 1 and D, equals 1.

From Equations (5) and (6a-b)

QUANTUM MECHANICS 163

Wb = —[2V(aB)B*A* — (la| — [BBP] |e1>|e.> + W(@B)A*A

— \(aB)B*B + (ja| — |B|)B*A]|c:>|d.> + [-V(@8)BB*

+ V(aB)AA* + (la| — |B|)AB*]|d\>|c2>

+ [-2v(aB)BA + (lal — |B))A7d>|d.> (13)

Since the coefficient in front of |d,>|d,.> is not zero, there is a finite probability

(maximum about 9% (Hardy, 1993) of this state in Equation (13) occurring.

For experiments that satisfy Equation (13), we obtain from Equations (10)

and (12)

MoD — 1. and.) Dodo, then jy U), =. /and:Up,= 1 (14)

therefore

O05. al (LS)

which is in direct contradiction to Equation (8).

Therefore, realistic theories can not be consistent with quantum mechanics

unless they are non-local and non-Lorentz invariant.

II. Examination of Experimental Conditions

To avoid this contradiction, some difficult choices have to be made. One

possibility is to abandon realism and with that the notion of elements of reality

existing independently of measurement. This has been called the Copenhagen

Interpretation. The adoption of this positivist view of nature remains a reluctant

alternative to many.

Another possibility is to assume instantaneous transfer of information between

objects in spacelike intervals. Abner Shimony (1990) argues that quantum me-

chanics does not violate parametric independence which forbids superluminal

transmission of signals but violates outcome independence which allows sublumi-

nal transmissions. On the other hand, Ferrero, Marshall and Santos (1990) claim

that special relativity not only forbids faster than light signaling but also forbids

faster than light actions at a distance.

An alternative is to suggest from the non-Lorentz invariance there exists a

privileged frame of reference in which realistic (hidden variable) theories are

always applied. Thus, hidden variable theories would be in violation with the

tenants of Special Relativity. How this pEngeeet frame would be determined

remains an open question.

Another approach, more pragmatic, is to examine the experiments themselves.

164 JOSEPH DI RIENZI

Can experimental limitations avoid the contradiction? There are two possible

mechanisms for this to occur. |

A. Difficulty in preparing the state properly: This would involve guaranteeing the Hardy

state vector, both in the set up of the apparatus and the preparation of the state vector.

B. Detector inefficiencies: This involves the ability for real laboratory instruments to

detect outcome pairs coincidentally.

Failure of either A or B would not ensure the conditions expressed in Equations

(8), (10), (12) and (14) to be met exactly. It will be shown that these failures

reduce the Hardy contradictions to, at best, Bell statistical inequalities and may

indicate a fundamental limitation in comparing local realistic theories with respect

to quantum mechanics.

To demonstrate the conditions when realistic theories give a contradiction with

quantum mechanics, let \ be the set of hidden variables. Then for

U,()U,(A) = 1 (16a)

U,(A)U,(A) = 0 _ (16a’)

D,(A)U2(A) = 1 (16b)

U,()D2(A) = 1 (16c)

D,(A)D2(d) = 1 (16d)

Let

Suiu2 be the set of all \’s such that Equation (16a) holds.

S’uiu2 be the set of all \’s such that Equation (16a’) holds.

Note, Equation (16a’) allows

U,() = 0 and UA) = 0 = Syy’v’

U,() = 0 and UA) = 1 = Sui'w

WOO! Mand t Cy Oe see

S’uiue = Sui’ue” + Sur’u2 + Suiv2" = Suiv

where Syin> is the complement of Snino (where N; = D, or U; Gi = 1, 2)), the set

of all \’s not belonging to Syinp.

In addition, let

Spiu2 be the set of all \’s such that Equation (16b) holds.

QUANTUM MECHANICS 165

Suip2 be the set of all \’s such that Equation (16c) holds.

Spip2 be the set of all \’s such that Equation (16d) holds.

A. Difficulty in State Preparation: Using Hardy’s analysis (1992) a contradiction would

still be produced if

P(Suiu2/Spip2) + P(Spive/Spip2) + P(Suip2/Spip2) > 2 (17a)

and

P(Spip2) > 0 (17b)

where P(Syin2/Spip2) is the probability of Syin2o given condition Sp;p> is true.

B. Detector Inefficiencies: The second experimental limitation deals with detector effi-

ciency. This loop-hole, as it is referred, can sometimes be ignored by claiming fair

sampling. By fair sampling it is assumed that all detectors are identical and operating

at the same efficiency. Therefore, the number of detections at one detector is propor-

tional to the number that would be measured if the detector was ideal with the same

constant of proportionality for all measuring devices. Under this condition probability

relations among separate pairs of measurements remain the same as under ideal

conditions.

Without fair sampling, the contradictions can still be analyzed using the following

method of analysis by GHSZ (1990).

For no contradiction, relationships in Equations (16a-d) can be expressed as

Suiu2 = (Spiv2 U Suip2 U Spip2) (18)

Using probabilities this becomes

P(Suiu2) S P(Spiu2 U Suip2 U Spine) (19)

Using the condition that (Spip2, Suip2, Spip2) may not be mutually exclusive

P(Suiu2) = P(Spiu2) + P(Suip2) + P(Spin2) (20)

Since

P(Suiu2) = P(Sui'u2") + P(Sui'u2) + P(Suiv2’)

therefore,

P(Suiu2) > P(Sui'v2’)

Then

P(Su1'u2’) = P(Spiv2) + P(Suiu2) + P(Spipe) (21)

To account for detector inefficiencies, assume for simplicity the fraction detected by

all detectors for all outcome pairs is f. For example, assume the rate of particles 1

and 2 are known and for D, and U, there are a fraction of these that are detected

with D, = 1 and U, = 1. Call that fraction f

f = PSpiv2) = 1,

therefore,

P(Spiu2) = f (22a)

166 JOSEPH DI RIENZI

P(Spiv2) < 1 — f (22b)

Use similar bounds for Sy,;p> and Sy;'y2’

P(Suip2) = f (23a)

and

Rui) ve (23b)

P(Sui'u2') > f (24)

For Spip2 we have that the maximum probability is approximately 9% (Hardy, 1993),

P(Spip2) > .O9f (25a)

and

P(Spip2) < 1 — .09f (25b)

Using Equation (21)

SLES sip) Sl tse ei se (26)

fo Oil

To violate this inequality f = .971. Assuming each detector has efficiency €, then

for two measurements (ex. D, and U,), f = &?

Therefore,

€ = .985

IV. Conclusion

Thus, to demonstrate that realistic theories violate locality and Lorentz Invari-

ance, the detectors must have efficiency > 98.5%. Notice that in both conditions

of experimental limitations, Eqs. (17) and (20), Hardy’s direct contradiction is

replaced by a Bell-type inequality. Furthermore, if detector efficiency is not very

high (< 98.5%), no violation of realistic theories with quantum mechanics is

evident.

From this analysis it becomes evident that this contradiction, although stronger

in principle than Bell’s inequality, is very difficult to realize in practice. Further-

more, Emilio Santos (1992) states there is a fundamental incompatibility in exper-

imental limitations A and B. Requirements of high detector efficiency and ideal

state preparation are not simultaneously realizable. Santos argues that in most of

these tests of Bell-type contradictions with quantum mechanics the state prepara-

tion involves the analyzing ability of devices such as polarizing filters for photons

and the efficiency of detectors such as photocells. The polarizers rely on the

wave properties of the detected quantum states, and the photocells rely on the

particle properties. Santos tries to demonstrate that the correlation of the analyzers

QUANTUM MECHANICS 167

decreases with detector efficiency, and the greater the analyzing ability the more

difficult it is for detector efficiency to show a Hardy type contradiction.

Can it be that nature “‘protects’’ hidden variables in a manner similar to

the way the Heisenberg Uncertainty Principle protects quantum mechanics from

attempts to show it is contradictory? If so, then is this just another example of the

profound mystery surrounding any attempt to understand the nature of quantum

mechanics.

Notes

1. In all these experiments except Lamehi-Rachti & Mittig (1976), the system does not consist of two spin

1/2 particles, but instead two photons of correlated polarizations emitted in opposite directions. Polarization

states in photons are the electromagnetic analogues to particle spin states.

References

Aspect, A., Grangier, P. and Roger, G. (1981) Experimental realization of Einstein-Podolsky-Rosen-Bohm

gedanken experiment: A new violation of Bell’s inequalities. Phys. Rev. Lett., 47, 460-463.

Aspect, A., Dailbard, J. and Roger, G. (1982). Experimental test of local hidden-variable theory. Phys. Rev.

Lett., 49, 1804-1807.

Bell, J. S. (1964). On the Einstein-Podolsky-Rosen paradox. Physics, 1, 195-200.

Clauser, J. F. (1976). Experimental investigations of a polarization correlation anomaly. Phys. Rev. Lett., 36,

1223-1236.

Clifton, R. and Niemann, P. (1992). Locality, Lorentz invariance and linear algebra: Hardy’s theorem for the

entangled spin-s particles. Phys. Lett. A., 166, 177-184.

Einstein, A., Podolsky, B. and Rosen, N. (1935). Can quantum mechanical descriptions of physical reality

be considered complete? Phys. Rev., 47, 777-780.

Ferrero, M. T., Marshall, W. and Santos, E. (1990). Bell’s theorem: local realism versus quantum mechanics.

Am. Jl. Phys., 58, 683-688.

Freedman, S. J. and Clauser, J.F. (1972). Experimental test of local hidden variable theories. Phys. Rev.

Lett., 28, 938-941.

Fry, E. S. and Thompson, R. C. (1981). Experimental test of local hidden variable theories. Phys. Rev. Lett.,

47, 465-468.

Greenberger, D. M., Horne, M. A., Shimony, A. and Zeilinger, A. (1990). Bell’s theorem without inequali-

ties. Am. JI. Phys., 58, 1131-1143.

Hardy, L. (1992). Quantum mechanics, local realistic theories and Lorentz invariant realistic theories. Phys.

Rev. Lett., 68, 2981-2984.

Hardy, L. (1992a). A quantum mechanical experiment to test local realism. Phys. Rev. Lett. A, 16, 17—23.

Hardy, L. (1993). Nonlocality for two particles without inequalities for almost all entangled states. Phys. Rev.

Lett., 71, 1665-1668.

Lamehi-Rachti, M. and Mittig, W. (1976). Quantum mechanics and hidden variables: a test of Bell’s inequality

by the measurement of the spin-correlation in low energy proton-proton scattering. Phys. Rev. D, 14, 2543-

2555.

Ou, Z. Y. and Mandel, L. (1988). Violations of Bell’s inequality and classical probability in a two-photon

correlation experiment. Phys. Rev. Lett., 61, 50-53.

Santos, E. (1992). Critical analysis of the empirical tests of hidden variable theories. Phys. Rev. A, 46, 3646—

3656.

Shimony, A. (1990). Exposition of Bell’s Theorem. In A.I. Miller (ed.) Sixty two years of uncertainty (pp.

33—43). New York: Plenum Press.

Journal of the Washington Academy of Sciences,

Volume 84, Number 3, Pages 168-178, September 1996

A Translation of a Zosimos’ Text in an

Arabic Alchemy Book

H. S. El Khadem

The American University, Department of Chemistry, Washington D.C. 20016

Received February 13, 1996

ABSTRACT

In a recent paper (E] Khadem 1995), it was reported that an Arabic translation of a Greek

text by Zosimos was found in a copy of a book entitled ‘““Keys of Mercy and Secrets of

Wisdom,’’ written by the twelveth century alchemist Al-Tughra’i. Reported here is a descrip-

tion of this rare book, which has recently been added to the Library of Congress’ Near East

Section collection.

Tughra’i, Author and Translator

The copy of “‘Keys of Mercy and Secrets of Wisdom’’ under consideration

was written in two parts designated, **Part One, Introduction’ by Al-Tughra’i’’

and Part Two, *“‘From Keys of Wisdom by Zosimos’’ translated to Arabic by

Al-Tughra’i. The author and translator’s full name is Mu’ayed-ul-Din Abu Ismail

Ibn Al-Hassan Ibn Ali Al-Tughra’i. He was born in 1062 A.D. in the city of

Asbahan in Persia and was later appointed ‘‘Katib’’ (secretary) in the court of

the Seljug Sultan Malik-Shah and that of his successor, Sultan Muhammad.

Because of his skills in calligraphy, he was assigned the duty of affixing the

royal signature *““Tughra’’ to the sultan’s writs (hence his name, which means

the writer of Tughras). After several years, Tughra’i moved to Mosul in Iraq

where he was appointed Vizir to Emir Ghiyat-ul-Din Mas’ud. When the Emir

died, uncertainty regarding his successor led to a palace revolt. Tughra’i sided

with the oldest son, Mas’ud, who subsequently lost the power struggle to a

younger brother, Mahmud. The latter, angered by his support of his brother,

Dr. H.S. El Khadem College of Arts and Sciences Department of Chemistry The American University 4400

Massachusetts Avenue N.W. Washington, D.C. 20016-8014

168

ZOSIMOS’ TEXT 169

arranged to have him accused of heresy and then had him beheaded in the year

1121 A.D. Tughra’i’s execution caused dismay among the learned community

in the region and prompted many publishers to delete all what they considered

controversial from his books.

Tughra’i was a statesman, an alchemist and a poet, considered by many as

one of the key literary figures of his time (see Nicholson 1941). The present text,

which according to an annotation on its title page, was also known as: “‘Key of

Mercy and Lantern of Wisdom’’ and “*Key of the Treasures and Lantern of the

Symbol’’, has been cited by authors, such as Ullmann (1972), who lists it as

““Keys of Mercy and Lanterns of Wisdom’’, and Sezgin (1971) and Kraus (1943)

who list it as “‘Keys of Mercy’’. None of these authors, however, mentions that

this, or any other book by Tughra’1, contains a translation of a text by Zosimos.

A possible explanation of this absence is that the translation of Zosimos’ text

was deemed sufficiently controversial to delete it from many copies of ‘‘Keys

of Mercy.’’

Zosimos:

Zosimos, the author of Part Two of the present text, was the most famous

alchemist of his time. He was a gnostic philosopher, born in the city of Panopolis

(present day Akhmim) in Southern Egypt around the year 300 A.D. He lived in

Alexandria, and traveled to many parts of the Hellenic world (see Read 1937,

and Hopkins 1967). Although Zosimos was a prolific writer, all his books have

been lost and what remains of them today are mere passages and quotes written

in the original Greek language, or translated to Syriac or Arabic. The Greek and

Syriac texts have been translated to French by Berthelot (1885, 1888, 1893) and

discussed in detail by Halleux (1979) and Mertens (1990).

Arabic translations of Zosimos’ work are listed by Sezgin (1971, p. 73) and

by Ullmann (1972, p. 160). They are also listed in the Arabic encyclopedia,

‘*Kitab al-Fihrist’’, published in Baghdad in 987 A.D., by Ibn Al-Nadim (1872).

In Section Ten of this book, Nadim gives the titles of four books authored by

Zosimos (see Flugel 1872); they are: ‘‘Keys of the Craft,’’ by ‘‘Rimos,”’’ (its

title was translated by Berthelot 1888, p. 28) as ‘‘Keys of the Work’’); ““Keys

of Magic,’’ by ‘‘Thosimos’’; ‘‘The Book of Elements,’’ by ‘‘Dosimos’’ and

‘Book to All the Wise of the Craft’’ also by Dosimos. The inconsistency in

spelling Zosimos’ name can be traced to two reasons: (a) Arabic vowels may be

deleted, altered, or transposed, according to certain rules, to render foreign names

easier to pronounce; (b) the pairs of Arabic letters “‘Ra’’ and ‘‘Za’’ that produce

the sounds ‘‘R’’ and ‘‘Z,’’ and ‘“‘Dal’’ and ‘‘Thal,’’ that produce the sounds

170 KHADEM

‘*—D,’’ and ‘“Th’’ (as in “‘the’’) are identical in shape, except for a dot on top of

the second letter of each pair. A dot on the letter “‘Ra’’ changes it to ‘‘Za’’ and

a dot on ‘‘Dal,’’ forms the letter ‘“Thal,’’ In the writings of Geber, Avicenna

and Tughra’i, Zosimos is referred to as Rismos or Zismos, depending on whether

the copier of the manuscript remembered to put the dot. For example on p. 2 of

the Introduction of the present text, Zosimos’ name is spelled with a dot, whereas

on the title page of Part Two it is spelled without the dot. Nadim, probably did

not realize that the authors he lists as Rimos, Thosimos and Dosimos were one

and the same person. Furthermore, because ““The Craft,’’ *“The Work,’’ and

‘‘Magic,’’ were synonyms used to describe ‘‘Alchemy,”’ it is quite possible that

Zosimos’ books listed as, ““Keys of the Craft,’’ and ‘‘Keys of Magic,’’ were one

and the same book which Tughra’i later referred to as “‘Keys of Wisdom,”’

because he did not wish to use of the word “‘craft’’ or “‘magic’’ lest he be

accused of heresy. Nadim describes ‘‘Keys of the Craft’’ as a collection of letters,

numbered one through seventy, and states that the book was also called the

‘*Seventy Letters.’’ Another book having the word “*Keys,’’ in its title namely,

‘‘The Book of Keys,’’ also known as ‘“‘The Little Key of Zosimos,’’ is more

difficult to relate to the present text, because it was not listed by Nadim; it was

cited instead by the Byzantine monk, Michael Psellus (see Berthelot 1885).

The Text

The present book contains extremely valuable historical information about the

chemical knowledge available in Tughra’i’s time. Unfortunately, Part Two is not

a verbatim translation of Zosimos’ book ‘‘Keys of the Craft,’’ since it offers

comments without specifying whether they belong to Tughra’i, or to Zosimos.

It does however, give a detailed summary of Zosimos’s text, and contains innu-

merable direct quotes of Zosimos and many philosophers of antiquity.

The Preface of ‘““Keys of Mercy and Secrets of Wisdom’” lists the chapters of

both parts of the book. Part One or “‘Introduction’’ is divided into five chapters:

I. The science, and its Materials; II. Mixing and its Ways; III. Fire and its Nature;

IV. Balances (of properties); V. Metals and Plants and how to Recognize them.

Part Two, entitled ‘‘From Keys of Wisdom,’’ is divided into seven sections: I.

Definitions and Symbols; II. Promotion, and what can be Promoted; III. Distilla-

tion, what can and cannot be Distilled; IV. Conversions and Synthesis; V. Degra-

dation and Decomposition; VI. How Chemists Deduced these Facts; VII. The

stages of the Work. Even though the chapters of the two parts of the book have

different titles, they are similar in content and present the subjects in roughly the

same order. Two sections of Part Two, namely Section Five and Section Six

ZOSIMOS’ TEXT 171

were missing from the copy studied. However, because their subject matter had

been previously discussed in Chapters Four and five of the Introduction, it was

possible to comprehend the text without much difficulty.

‘“Keys of Mercy and Secrets of Wisdom’’ is written in the format of lectures.

The narrator in both parts of the book seems to be Tughra’i since he refers to

Zosimos in the third person. Furthermore, Part Two often contains references to

things that had not occurred, or did not exist in Zosimos’ time. Example of these

are statements like: ‘‘the Moslem philosophers said . . .’’ (Islam came three

centuries after Zosimos’ death), and “‘gun powder’’ (a product that was not

known in Zosimos time). Similarly, in a dream depicting *“Cinnabar,’’ as a giant

sitting on a throne reached by nine steps (the number of steps needed to prepare

the elixir), the person relating the dream praises the prophet Mohammed, and

invokes the archangel ‘‘Israfil’’ (the angel who blows the horn on judgment day,

according to certain Islamic writings). In both parts of the text, the narrator ends

each paragraph with the typically Islamic cliche: ‘‘God is more knowledgeable,’’

which Zosimos, a Christian, would not normally say. However, it is also quite

possible that these pious words were intentionally added by Tughra’i to abate

criticism by the religious leaders of his time.

Although some might suspected that Zosimos’ name was added to the book

in order to enhance its value, this possibility is remote for two reasons: (a)

Tughra’i was a successful author and an influential statesman, who did not need

such a practice to promote one of his book, and (b) Zosimos’ name is not

displayed prominently, but seems instead to be intentionally hidden; it does not

appear on the book’s title page, but is relegated to the title page of Part Two,

which comes after hundreds of pages belonging to Part One.

The Quotes

Among the many quotes attributed to Zosimos, some are in the form of letters

addressed to women. One is addressed to a certain Maria (probably Mary the

Copt), and stresses the importance of rigorously following procedures in any

chemical work. Zosimos says: *“You may think, Maria, that all the balances and

the ten laws that pertain to the Substance (the elixir) need not be rigorously

followed, and that some may be altered a little, while others may be totally

ignored. It is not so; never disobey any of the rules, otherwise you will not

succeed in your preparation and all your efforts will be wasted.’’ In turn Mary

asks: “‘Can you produce gold but from gold, or can you form a metal from a

non-metal? Can you produce a man save from a man; a plant except from a plant

and an animal but from its own kind?’’

172 KHADEM

In another section of the book Zosimos is quoted as saying: ‘“Knowledge is

treated with great honor, because only a philosopher, who has acquired Wisdom,

scientifically and practically, is able to use it. An experimentalist may obey his

master when he tells him: Take this and do such and such a thing, evaporate it,

dissolve it, distill it, and so on till the end of the work. That aide does not

understand anything beyond how to do things; whereas the person who compre-

hends the science and the practice, knows how and why something is evaporated,

i.e. the purpose of the evaporation. This is why, to become a philosopher, one

must know the aim of Wisdom in each step of the work.’’

The book also contains several quotes made by famous Greek philosophers,

such as Aristotle, whose discussions with Plato are reproduced in some detail,

and Democritus, who is quoted as saying: ‘“The stone is not formed until it has

gathered all the colors that exist in the universe, and until it has been colored

with all the simple and complex colors.’’

Galen is quoted as saying: ‘“To determine the amount of a drug to be used as

an antidote, select three doses; one in great excess, let it be forty eight (48) units

of weigh; drams, iotas, or any other unit; the second in the middle, which is

twenty four (24), and the second (third), which is the least, six (6) parts. To

determine which of these to use, consider all the variables, the powers, and the

reasons, as well as the benefits gained by increasing the dose to forty eight units

or decreasing it to six. If one condition requires an increase and one condition a

decrease, then you use twenty four.”’ |

The book also contains quotes from lesser known personalities such as Androma-

chus (a contemporary of Galen), Heracles, Tamagus and “‘Balinas’’ (Apollonios

of Tyana). Also quoted, are mythical characters, such as Hermes Trismegistus, his

son Tata, and his daughter Queen Cleopatra. For example, Hermes is quoted as

saying: “‘A body will not accept a soul that is not its own, and a soul will not

reside in a body that is not its own. Thus a human body will not accept the soul

of a bird, and the soul of a bird will not reside in a human body.’’ Unfortunately,

some Greek philosophers quoted could not be identified because of lack of records,

or because transliteration had altered their names beyond recognition.

Among the Greek literature cited in Part Two of the text are: ‘“The Book of

Revelation’’ (Istigla’) by Aristotle; ‘“The Basics’’ by Apollonios of Tyana (Bali-

nas), from which a passage is quoted, describing how to dye elixirs with yellow

colors extracted from a sun flower like plant called in Greek *““Lumenia’’, and

‘‘Letters from Ostanes to Cleopatra’’, which are discussed in some detail.

Content of the Book

Both parts of the book start with a discussion of the ‘‘four elements’’ (fire,

air, water, and earth) and the ‘‘four natures’’ (hot, cold, moist, and dry), and

ZOSIMOS’” TEXT 173

continue with their quantitative estimation. This is followed by a determination

of their ratios and how to amend these to form the elixirs. The book ends with

the use of the white elixir in the transmutation of copper to silver and the red

elixir in the conversion of silver to gold.

Both parts of the book contain detailed accounts of dreams that reveal the

secrets of Alchemy and long sections dealing with astrology and the role of the

seven ‘‘planets’’ (the sun, the moon, Mercury, Venus, Mars, Jupiter, and Saturn)

in each stage of the work. They abound with diagrams depicting benefic and

malefic configurations of ‘‘planets’’ and their effect on the work.

Nomenclature is quite confusing because the chemical names bear no relation

to composition. For example, lead sulfide is referred to as the *“Tree that Grows

in the Black Soil of India’’ because of its color and the heat used in its preparation

- from sulfur.

Most of the conclusions reached by the authors of the text are today invalid

because of two major flaw in reasoning: The first is the belief that there are only

four elements, and the second, that metals are not elements, but compounds. In

spite of these shortcomings, Alchemists have succeeded in producing yellow

colored alloys made of silver and gold, and white ones made of copper and silver.

Avicenna (Ibn Sina) correctly warned his contemporaries that it was not possible

to produce real gold chemically, saying: ‘“Only imitations of gold can be formed,

because the essential nature of a pure metal can never be altered’’ (see Holmyard

1928). His ideas were unfortunately disputed by Tughra’i in the book ‘‘Facts

about Martyrdom.’’ Zosimos believed in transmutation, but he correctly states

that to prepare gold (alloys) out of silver one must start with gold, and to make

silver out of copper one needs silver. He is quoted as saying: ‘‘He who sows

silver reaps silver, and he who sows gold reaps gold.’’ His mistake was to

think that the amount of silver or gold added increases like that of yeast during

fermentation.

Some of the important contributions of alchemists, discussed in the present

text are the distinction between distillation, and pyrolysis (which they called

smoking). Their success in making stills and constant temperature reactors, can

be seen in the illustrations depicted: Thus Fig. 1 shows a sublimation apparatus;

Fig. 2 the precursor of the modern Kugelrohr; Fig. 3. shows a reactor warmed

with what is described as ““moist heat’’, and Fig. 4. an incubator warmed by

fermenting garbage and burning coal. In a remarkable statement, Zosimos ex-

plains why vapor rises against gravity during distillation; he says: ‘“Motion is

due to heat for without heat there would be no motion’’.

Most alchemists rejected the idea of ‘‘spontaneous creation’’. Thus Tughra’i

says: ‘“Try as they may the wise were never able to form something from some-

thing other than what it is normally generated from; humans from human semen;

174 KHADEM

LIYSU!

Bcyeat

wey

AGED

WVUDOL

YEOMON

REPS

Soap ae 7

a tt

baler

Fig. 1. Sublimation apparatus.

wheat from its grains, etc. They tried to produce snakes and asps by fermenting

human hair; bees and wasps from putrefied horse meat; humans from human

flesh as well as from innumerable other things, but they all failed.”’

Format of the Book

The manuscript ‘“Keys of Mercy and Secrets of Wisdom’’ is written in black

and red inks; black ink for text, red ink for punctuation and both inks for art

ZOSIMOS’ TEXT 175

Fig. 2. Upper figure: A fractional distillation apparatus, made up of three glass flasks connected by metal

and sealed with clay. Lower figure: a modern Kugelrohr.

Fig. 3. Coal heated water bath to produce “‘moist heat’’ for a reactor.

176 KHADEM

Fig. 4. Upper Figure: An incubation bath; Lower Figure: The bath placed in a pit warmed from two sides

with fermenting garbage and from the other sides with coal.

work. The text was originally made up of 157 folios (314 sheets or 628 pages);

of these 24 folios (48 sheets or 96 pages) of Part Two are missing. The Introduc-

tion, or Part One, is in 60 folios (120 sheets or 240 pages) grouped in 13 signatures

(booklets), that contain 28 Tables and 23 Figures. Part Two is presently composed

of 73 folios (146 sheets or 292 pages), arranged in 15 signatures, that contain

10 Tables and 42 Figures. Although the pages of the book are not numbered, it

was possible to determine the existence of a gap, because the first word in each

verso is entered at the bottom of the preceding recto. This made it possible to

ascertain that there are missing pages between Sections Four and Seven. The

size of the gap was determined from two annotations found at the end of each

Part. These give the number of sheets that were present in the Part in question.

ZOSIMOS’ TEXT 177

On the last page of Part Two, a recent owner of the book wrote: ‘‘Owned by

legal purchase by so and so, son of so and so, in the holy month of Zul-Que’da

1148 H. (1735 A.D.); 194 sheets in 19 booklets.’’ A count of the actual number

of sheets present in Part Two, revealed that 48 sheets (24 folios) in 4 booklets

were missing. A similar annotation found at the end of Part One, revealed that

this part contained 120 sheets in 13 booklets, which is the actual number of

sheets presently found.

The pages of the book (204 x 147 mm) contain about twenty five lines of

text (less if a page contains an art work). Tables and Figures are often surrounded

by frames made of double red lines. The margins are generous; top margins are

24 mm, and bottom ones, slightly larger (26 mm). Side margins vary in width;

the right margins in rectos and the left margins in versos are wider (60 mm) than

the margins opposing them (18 mm). It seems that the two parts of the books

were not kept separately because the last signature of Part One and the first of

Part Two have identical worm holes and water stain marks.

Many of the side margins of the book bear annotations, written in ink by

successive owners of the manuscript. An annotation on the title page states that

the present text was also known by the two other names mentioned earlier, and

lists the title of three other books by Tughra’i. They are: (a) ‘‘Collection of

Secrets and Compositions of Lights’’; (b) ‘Introduction to the Book of Healing,”’

and (c) ‘“‘Facts about Martyrdom.’’ There are also two biographical notes; one

located on the title page, dealing with the biography of the author, and the other,

on the last page of the book, which promotes the work of the fifteenth century

alchemist, Al-Jildaqi, and names six of his books. In addition, there are several

short notes in the margins, dealing with errors in the text and their corrections,

and longer ones dealing with explanations and interpretations.

The present text must have been copied after Tughra’i’s death, since his name

is followed by the words ‘‘May God have mercy on him.”’’ It is in a reasonably

good condition, thanks to some restorative treatment administered in France after

world war II. At that time the copy was dated, circa fifteenth century, which is

consistent with the fact that the book was first published in the twelveth century,

and has since then been periodically recopied. The restoration was made at the

request of Dr. Puy-Haubert the war time Director of the French Hospital in

Alexandria (Egypt). It involved cleaning the pages and spraying them with insecti-

cides to kill the worms that had damaged the margins; then spraying the damaged

pages with a varnish.

References

Berthelot, M. (1885). Les Origines de |’Alchimie, Steinheil, Paris, Reprinted by Librairie des Sciences et des

Arts, Paris, 1938, pp. 177-187.

178 KHADEM

Berthelot, M. (1888). Collection des Alchimistes Grecs, Steinheil, Paris, Vol. 1, pp. 119, 127-174, 209, 250;

Vol. 2 p. 28, 117-120; Vol. 3, pp. 117-242.

Berthelot, M. (1893). La Chimie au Moyen Age, Steinheil, Paris,; Reprinted by Philo Press, Amsterdam, 1967,

Vol. 2, pp. 203—266; Vol. 3, pp. 28, 30, 41.

El Khadem, H. S. (1995 Sept.). ‘‘A lost text By Zosimos Reproduced in an old Alchemy Book,’’ J. Chem.

Education, 72, No. 9, p. 774.

Flugel, G. Index to Kitab Al-Fihrist, Vogel Leipzig, 1872, p. 353.

Halleux, R. (1979). Les Textes Alchimiques, Typologie des Sources du Moyen age Occidental, Turnhout, fasc

32 pp. 61; see also Compte Rendu du 104e Congress National des Societes Savante, Bordeaux (1979), fasc

4 pp 169-180.

Holmyard, J. The Great Chemists, E. Methuen, London. 1928, p. 24.

Hopkins, A. J. (1967). Alchemy Child of Greek Philosophy, AMS Press, New York pp. 8, 49, 69-77, 117,

124, 182.

Ibn Al-Nadim, M. (1872). Kitab Al-Fthrist, Edited by G. Flugel, Vogel Leipzig, pp. 419, 420 (Although more

recent editions are available, the one cited has an excellent index in German). For a good text in English,

see B. Dodge, The Fihrist of al-Nadim, New York, 1970.

Kraus, P. (1942). Contribution a l’histoire des idees scientifique dans |’Islam. 1. Le Corpus des ecrits Jaberiens,

Cairo 1943; II. Jabir et la science Greque, Cairo.

Mertens, M. (1990). Alchemy Revisited, Edited by Z. von Martels, Leiden.

Nicholson, R. A. (1941). Literary History of the Arabs, Cambridge University Press, Cambridge., p. 326.

Read, J. (1937). Prelude to Chemistry, MacMillan, New York, pp. 9, 14, 33, 40, 41, 129, 154.

Sezgin, F. (1971). Geschichte des arabischen Scrifttums, Leiden, Vol. IV, pp. 46, 69, 107, 159, 231, 236, 256;

266.

Sezgin, F. (1971). Loc. cit. pp. 73-77.

Ullmann, M. (1972). Die Natur- und Geheimwissenschaften im Islam, (Handbuch der Oreintalistik. Erste

Abteilung: Der nahe und mittlere Osten. Erganzungsband VI. Zweiter Abschnitt), Leiden, pp. 83, 227, 229,

230 29252595

Ulimann, M. loc. cit. pp. 160-3.

DELEGATES TO THE WASHINGTON ACADEMY OF SCIENCES,

REPRESENTING THE LOCAL AFFILIATED SOCIETIES

Bimoscphical Society sor Washington): 2.2. .2 65. 6. ce cee bee wees Thomas R. Lettieri

Puubropolosical Society Of Washington... 22... ke ee ee ele eee wees Jean K. Boek

EvaineiGal SOCICLY Of, WASHINGTON: 28). 8 oe le ee eee ee lnc eee odes Kristian Fauchald

Mitemicalsocietyol Washinpton .. de ie esc ee ee Dee ee ede tale ees Elise A. B. Brown

EPaomolorical Society of Washington ..............-...422822-0)8: F. Christian Thompson

Ree CrCOOTAPINC SOCIELY 9) 5.5... eel dnc ed fee kde ech gue ea eles Stanley G. Leftwich

Sema Medi society Ol WaShINGtON: 3.2.6.6 6. eee ee ee ee een dh eee ees VACANT

meee society of the District of Columbia: .....2. 2.5.5.5. adie eee eee es John P. Utz

Peeaaned society of Washineton, DC, 2.0.0.2 02 eee ee ee lee eee woes VACANT

Pmecae Society Of Washington 22.22.26 ieee Gee eck cee k ee cae e etn es Muriel Poston

Society of American Foresters, Washington Section .....................4-- Eldon W. Ross

Peminmerom Society Of ENMSINEEIS iets bn Send eels edien bed ede we eee es Alvin Reiner

Institute of Electrical and Electronics Engineers, Washington Section ........ George Abraham

American Society of Mechanical Engineers, Washington Section .......... Daniel J. Vavrick

Helmuimthological Society of Washington .............0. 0.0 e cece cece eens VACANT

American Society for Microbiology, Washington Branch ...................2.0204- Ben Tall

Society of American Military Engineers, Washington Post ............... William A. Stanley

American Society of Civil Engineers, National Capital Section ................... VACANT

‘Society for Experimental Biology and Medicine, DC Section ............. Cyrus R. Creveling

Per rumenmanonal, Washington Chapter .:.......... 6.650500 c eee ee eee Richard Ricker

American Association of Dental Research, Washington Section ............ J. Terrell Hoffeld

~ American Institute of Aeronautics and Astronautics, National Capital

Sie in MISES Fee Ee Nae cll as eeu hy GOL eed a ab bl a leva: wees. br ssslaei aie Reginald C. Smith

American Meteorological Society, DC Chapter .......................... A. James Wagner

Peer setence society Of Washington ............2-. 2-26 et ee To be determined

Acoustical Society of America, Washington Chapter ...................... Richard K. Cook

panctican Nuclear Society, Washington Section ....... 2.0.2.2... eee eee Kamal Araj

Institute of Food Technologists, Washington Section ..................2--4-- Roy E. Martin

American Ceramic Society, Baltimore-Washington Section ................. Curtis A. Martin

SSS RE BUEN CEI SCC 1S1R yaaa eRe es es Ge er ee Regis Conrad

Py asiimngion History of Science Club >.........22.25.52.0 2 5c e eee eee ee Albert G. Gluckman

American Association of Physics Teachers, Chesapeake Section ............ Robert A. Morse

Optical Society of America, National Capital Section .................... William R. Graver

American Society of Plant Physiologists, Washington Area Section ........... Steven J. Britz

Washington Operations Research/Management Science Council ............... John G. Honig

Instrument Society of America, Washington Section .................00 00sec eeee VACANT

American Institute of Mining, Metallurgical and Petroleum Engineers,

Cees TOMMSCCUIOM 5,5 2 fos) hehe ees ss MON ok Nees a alewials Anthany Commarota Jr.

National Capital Astromomers .. 2.6... cc ce ee ee enna eee Robert H. McCracken

Mathematics Association of America, MD-DC-VA Section ................ Sharon K. Hauge

District of Columbia Institute of Chemists ....................2.02205- William E. Hanford

District of Columbia Psychological Association ...................2.-5. Marilyn Sue Bogner

Pyasmnecvon Paint Vechnical Group... 2... . 6.6.26. cc es sce ee ee ke bee dees Lloyd M. Smith

American Phytopathological Society, Potomac Division ................... Kenneth L. Deahl

International Society for the System Science, Metropolitan Washington

co ITE USTEE ih RR EE are A Re Sf oe David B. Keever

Human Factors Society, Potomac Chapter ..................-.00 eee ee eee Thomas B. Malone

mucican Fisheries Society, Potomac Chapter ................02 06050006 Dennis R. Lassuy

Association for Science, Technology and Innovation ................... Clifford E. Lanham

Easton SOciolopical: Society 2. 2 ).2 522 sn e holes fsa beseech ca eis Ronald W. Manderscheid

Institute of Electrical and Electronics Engineers, Northern Virginia

DEEMONM Ge NA LM RU NMEE inn r Lr h mn teen aU au G Blanchard D. Smith

Association for Computing Machinery, Washington Chapter ............. Charles E. Youman

Pasiine owe Statistical SOCISIY(). i he Galera cs she oe ibis Sie 6 ke dieae 3g de eee David Crosby

Society of Manufacturing Engineers, Washington, DC Chapter .............. James E. Spates

Institute of Industrial Engineers, National Capital Chapter ................ Neal F. Schmeidler

Delegates continue to represent their societies until new appointments are made.

Washington Academy of Sciences Periodicals Postage Paid

Room 811 at Washington, DC

1200 New York Ave. N.W. and additional mailing offices.

Washington, DC 20005 ;

Return Postage Guaranteed

w3l7

i VOLUME 84

Number 4

Jour nal of the December, 1996

WASHINGTON

ACADEMY .. SCIENCES

ISSN 0043-0439

Issued Quarterly

at Washington, D.C.

SMITH SO ny >

/ Wy,

CONTENTS

Obituary:

FATE RCC WATE ite tive slet tess n i tice Lats NAY lute wig auaketalar ty eta aia nus Y Mies

Annual Report:

“‘Annual Presidential Report, Washington Academy of Sciences’’ ...........

Articles:

ELLIS L. YOCHELSON, ‘‘The Washington Academy of Sciences:

Backerouna,Origm, and Early Years’? 0) .b. os sccdeeseek ances wade

Washington Academy of Sciences: Bylaws ..............0. cece eee ences

Announcement:

‘‘Special Offer on Biography of Charles Doolittle Walcott, President of the

Washington Academy of Sciences, 1899-1910 ............ 0... ccc eee eee

Washington Academy of Sciences

Founded in 1898

EXECUTIVE COMMITTEE

President

Rita R. Colwell

President-Elect

Cyrus R. Creveling

Secretary

Michael P. Cohen

Treasurer

John G. Honig

Past President

John Toll

Vice President, Membership Affairs

Clifford Lanham

Vice President, Administrative Affairs

Phil Ogilvie

Vice President, Junior Academy Affairs

W. Allen Barwick

Vice President, Affiliate Affairs

Peg Kay

Board of Managers

Elise A. B. Brown

Jerry Chandler

Rex Klopfenstein

John H. Proctor, Chair, Centennial Com-

mittee

Eric Rickard

Grover Sherlin

James Spates, Chair, Joint Board

REPRESENTATIVES FROM

AFFILIATED SOCIETIES

Delegates are listed on inside rear cover

of each Journal.

ACADEMY OFFICE

Washington Academy of Sciences

Room 811

1200 New York Ave. N.W.

Washington, DC 20005

Phone (202) 326-8975

FAX (202) 289-4950

Email was @aaas.org

Web www.inform.umd.edu/WAS/

EDITORIAL BOARD

Editor:

Thomas Bottegal, Arthur D. Little, Inc.

Associate Editors:

Milton P. Eisner, Mount Vernon Col-

lege

Albert G. Gluckman, University of

Maryland

Marc Rothenberg, Smithsonian Institu-

tion

Marc M. Sebrechts, Catholic University

of America

Edward J. Wegman, George Mason

University

The Journal

This journal, the official organ of the Washing-

ton Academy of Sciences, publishes original

scientific research, critical reviews, historical

articles, proceedings of scholarly meetings of

its affiliated societies, reports of the Academy,

and other items of interest to Academy mem-

bers. The Journal appears four times a year

(March, June, September, and December). The

December issue contains a directory of the cur-

rent membership of the Academy.

Subscription Rates

Members, fellows, and life members in good

standing receive the Journal without charge.

Subscriptions are available on a calendar year

basis, payable in advance. Payment must be

made in U.S. currency at the following rates:

U.S. and Canada «2 ......\... 5. eee $25.00

Other counthies \.. 2.0.0.5 see 30.00

Single copies, when available ......

Claims for Missing Issues

Claims will not be allowed if received more

than 60 days after the day of mailing plus time —

normally required for postal delivery and

claim. No claims will be allowed because of

failure to notify the Academy of a change of

address.

Notification of Change of Address

Address changes should be sent promptly to

the Academy Office. Such notification should

show both old and new addresses and zip

codes.

POSTMASTER: Send address changes to

Washington Academy of Sciences, Room 811,

1200 New York Ave. N.W., Washington, DC

20005.

Journal of the Washington Academy of Sciences (ISSN 0043-0439)

Published quarterly in March, June, September, and December of each year by the Washington

Academy of Sciences, (202) 326-8975. Periodicals postage paid at Washington, DC and

additional mailing offices.

Journal of the Washington Academy of Sciences,

Volume 84, Number 4, Page 179, December 1996

Obituary

T DALE STEWART, 96, physical anthropologist, died October 27 in Bethesda, MD. Born June 10, 1901

in Delta, PA, Stewart began a career association with the Smithsonian Institution in Washington, DC, in 1924

as an aide to physical anthropologist Ales’ Hrdli¢ka. He took premedicine courses at George Washington U

(AB, 1927). After studying medicine at the Johns Hopkins U (MD, 1931), he returned to the Smithsonian,

succeeding Hrdlicka in 1942 as curator of the Division of Physical Anthropology. He served as head curator

in 1961 and museum director (1962-66). Stewart formally retired from the Smithsonian in 1971 but continued

professsional activity for another two decades.

Stewart taught at Washington U Medical School (1943), in Mexico City (1945), and at the George Washington

U School of Medicine (1958-67). In 1954 and 1955, he assisted the US Army Quartermaster Corps in Japan

in the identification of human remains recovered from the Korean conflict. This work led to the classic 1957

monograph (with Thomas McKern) Skeletal Age Changes in Young American Males. Additional research-

related travel included work in East Africa, China, Guatemala, Iraq and Peru.

Stewart published over 200 articles and books, focusing mostly on interpretation of human skeletal remains

but including topics in archaeology, forensic anthropology, history of anthropology, paleoanthropology (espe-

cially issues related to Shanidar Neanderthals), human and primate comparative anatomy, paleopathology,

dental anthropology, general human skeletal biology and physical anthropology. His major works include The

People of America (1970), Essentials of Forensic Anthropology (1979), the 1952 edition of Hrdli¢ka’s Practical

Anthropometry and Personal Identification in Mass Disasters (1970). In 1992, at the age of 91 years, Stewart

published a Smithsonian monograph on his 1938—40 archaeological work at the Virginia site of Patawomeke.

Stewart is well known for his many problem-oriented articles, attention to detail, clarity of expression, meticu-

lous scholarship and professional dedication. He was also an accomplished portrait artist who played the piano,

loved to entertain and recount his world travels and always found time for students and friends.

For two decades beginning in 1942, Stewart served as the regular consultant in forensic anthropology for

the FBI. In this capacity he reported on many cases and regularly testified in murder trials as an expert witness.

Stewart was awarded an honorary doctorate degree from the U of Cuzco (1949). He received the Wenner-

Gren Foundation’s Viking Medal in Physical Anthropology (1953) and the Smithsonian’s Joseph Henry Medal

(1976). In 1962, he was elected to the National Academy of Sciences. He became an honorary member of the

American Orthopaedic Association (1963) and the American Academy of Forensic Sciences (1974). Stewart

served as president of the Anthropological Society of Washington, vice president of the Washington Academy

of Sciences, president of the American Institute of Human Paleontology, member of the Committee on Research

and Exploration of the National Geographic Society, editor (1952—48) of the American Journal of Physical

Anthropology and president (1950-52) and treasurer-secretary (1960-64) of the American Association of

Physical Anthropologists. In 1993, he received the Charles R Darwin Lifetime Achievement Award from the

AAPA.

Stewart was preceded in death by first wife Julia Cable Wright Stewart (1951) and second wife Rita Frame

Dewey Stewart (1996). He is survived by his daughter from his first marriage, Cornelia Stewart Gill, three

grandchildren and 7 great-grandchildren. (Anthropology Newsletter, Jan 1998)

179

Journal of the Washington Academy of Sciences,

Volume 84, Number 4, Pages 180—183, December 1996

Annual Presidential Report

Washington Academy of Sciences

May 16, 1995

Naval Medical Center, Bethesda, Maryland

The most famous performer in the history of American musical theater was

Al Jolson. One of Jolson’s great attributes was his supreme chutzpah. The classic

example came at a war-bonds benefit at the Metropolitan Opera House in New

York City in 1917. Performing on the bill just before Jolson was Enrico Caruso,

the greatest tenor in the history of opera at home in his own house. Caruso chose

one of the supreme numbers for a tenor: Vesti la giubba from the opera I Pagli-

acci. The applause was thunderous. As it finally died down, Jolson ran on stage

and sang out “‘You ain’t heard nothing yet.’’ From Caruso’s own stage. No

wonder the audiences loved him. (THE ECONOMIST, April 22, 1995, p. 88.)

Such is the state of American science at this moment as we look to the future.

‘*You ain’t heard nothing yet.’’

What do I mean? Let me give you an example. ?

A few years ago I was asked to deliver the centennial address at a little shrine

in central New York State. Actually, it is the Shrine of the North American

Martyrs in Auriesville, New York. The date was 1985. As I thought over the

talk I did some research on how times had changed since the Shrine was originally

dedicated in 1885. I found that the members of the original audience had arrived

by one of three means of conveyance: river barge, horse and buggy, and passenger

trains. As I looked out over the crowd in 1985 I was amazed to realize that not

one of the guests in 1985 arrived by a means of conveyance that even existed

in 1885. There are no more passenger barges on the Hudson for routine travel.

The railroad station at the foot of the hill has been closed for years. Horse and

buggies are rare and not one was there that day. No, everyone came in 1985 by

car or bus, internal combustion driven gasoline fired vehicles. They did not even

exist in 1885 except in Henry Ford’s imagination.

If we now try to look forward to the bicentennial of that same shrine in 2085

the one point history tells us to hold for certain is that the audience will arrive

by some means of conveyance that does not even exist at this time. We can guess

180

ANNUAL PRESIDENTIAL REPORT 181

that such a prediction must be true. It is difficult to know with assurance when

the supply of gasoline will finally run low on Earth but no optimist expects it to

last till 2085. There will have to be a change. It is not clear what it will be.

You see, what is worrisome about the status of American science at this

moment is something that the ancients already knew. Words can have opposite

meanings. If I were to say, as Jolson did, that you ain’t heard nothing yet from

American science, I could mean that the future is rosy and resplendent, that the

achievements of the past are only a hint of the wonders to be revealed in the

coming century. The words could also mean, however, that the future would be

bleaker than ever experienced before.

In ancient times this ambiguity of the future was classically expressed to the

poor clients who came to the Delphic oracle for predictions of the future. Croesus,

king of the Lydians, asked if he should attack his enemies. The oracle replied

that he need have no fear, that, if he attacked, he would destroy a great empire.

The point left vague by the oracle was whether it would be the enemies’ empire

or his own. Unfortunately for Croesus, it turned out to be his own.

We know how great our past has been. Just this year your Washington Academy

of Sciences sponsored together with the Smithsonian Institution a lecture series

featuring eight Nobel prize winners, all American. I am sure that our incoming

President, Dr. John Toll, will mount an equally impressive program in the coming

year that will bring us into contact with other scientists of equal stature. | am

sure that you have all heard the wonderful news about Dr. Toll. He has been

named President of Washington College in Chestertown, Maryland. That is won-

derful news for Washington College to have so experienced an administrator,

wonderful news for independent colleges in Maryland, and wonderful news for

American higher education that needs strong leadership in our time of change.

But there are worrisome signs for our future. Consider the sad tragedy in

Oklahoma City. What was the first reply? Of course, it was that one of the

attorneys for O. J. Simpson, Johnnie L. Cochran, Jr., is now initiating a law suit

against the makers of the fertilizers used in the bomb. The first reply was not to

propose a law that the fertilizer, ammonium nitrate, be mixed with some harmless

compounds of potassium and phosphorus that make the fertilizer unusable for

explosives. That step has already been taken in Britain and Germany, as Russell

Seitz of Harvard’s Olin Institute remarked in an OP-ED piece in today’s New

York Times. No, we Americans reply by lodging a lawsuit.

The number of students majoring in the sciences is not encouraging. We can

understand that fact. The future is obscure and it is difficult to know where the

jobs will be. Industries are changing. Those related to armaments, for instance,

are retrenching. Not fast enough, no doubt. The Bosnian factions seem to want

to buy more armaments to an unlimited extent. They apparently experience a

182 HAIG

monstrous perverse thrill in shooting women and children and none of their

leaders has the courage of a Nelson Mendela to seek peace. But, still, jobs for .

scientists are problematic.

It is not easy to know the solution to that problem. One direction of wisdom

seems to be to broaden the training scientists receive so that they have more

flexibility in career choices. There are physicists who have studied no chemistry.

There are biologists who have studied no mathematics. They say that there are

even engineers who have studied no English. We do not want our students to

close the doors to alternative options too soon.

Your Academy of Sciences has always been at work on this problem. One

most evident sign is the magnificent achievement of Marylin Krupsaw with the

Junior Academy of Sciences. Some of those youngsters may even succeed in

solving that transportation problem for us.

As we look to the future with its uncertainties, we do always allow ourselves

to be inspired and even in some sense directed by the past. Our own Academy

is fast approaching its centennial in 1998. It is only right, therefore, that Dr. John

Proctor, one of our past presidents, is chairing a major effort to celebrate our

birthday. You will be hearing more about that program in the future.

One of the most troubling aspects to our future is a concern our friends in the

Russian Academy of Sciences have explored with us already in our famous

Georgetown University television program. But let me put it in a different context.

Last year I studied at the Max Planck Institute for Nuclear Physics in Heidel-

berg, Germany. In preparation for my remarks this evening I e-mailed a question

to some of my friends there. ‘‘What would you say to American scientists about

science in the next century?’’ One of the answers I got was startling.

Dr. Sandy Klevansky is a professor at Heidelberg University with which the

Institute has close connections. She is from South Africa. She is just back from

a return trip there. Much is going well in South Africa because of the excellent

leadership Mandela and de Klerk have provided. If only we could clone those

two. At the same time, it is a period of change. There is, for instance, a strong

movement to overhaul the teaching of science so as to give equal time to “‘tradi-

tional’’ systems. That is, they want the teaching of Zulu science and witch doctor

medicine in the curriculum of the universities.

The movement we call fundamentalism is world wide. Partially, it is healthy

and necessary. We need a sense of values or we will all float rudderless in

dangerous seas. But false and antiquated science, superstition and repression, we

do not need. It is, however, there knocking at the door. It challenges us all

constantly. I do think that American television has gone to absurd lengths, particu-

larly in its daytime programming that is broadcast in a time slot so that young

students coming back from junior and senior high school can see it. But is the

ANNUAL PRESIDENTIAL REPORT 183

Iranian solution the way to go? Teheran has now outlawed the use of satellite

disks because they receive American television and America, after all, is the

great Satan. The desire to retain the traditional art and culture and identifications

of ethnic groups is laudable but it can be murderous. Our Russian friends spoke

of the problem in the abstract with us at Georgetown and then met it in the

flesh in Chechnya. The desire to protect young people from the meretricious

sensationalism of the worst elements of the entertainment industry is praiseworthy

but takes a great deal of thought.

The moment at which we stand is a crossroads for science. We desperately

need research and progress. The strange growls of the wild beasts off to either

side of the road along which we are walking are thoroughly audible. We need

only think of Aids and now the Ebola virus. Such monsters are moving stealthily

through the grass around us. Only intelligence and wisdom are there to protect

us. But with them we can make Al Jolson’s cry be one of joy. As far as American

science is concerned, you ain’t heard nothing yet.

Rev. Frank R. Haig, S. J.

President

Washington Academy of Sciences

Professor of Physics

Loyola College

Baltimore, Maryland

Journal of the Washington Academy of Sciences,

Volume 84, Number 4, Pages 184—220, December 1996

The Washington Academy of Sciences:

Background, Origin, and Early Years

Ellis L. Yochelson

Research Associate, Smithsonian Institution, Washington DC 20560

Introduction

As time proceeds, new organizations appear and old organizations disappear.

Those groups which persist either fill a continuing need or adapt through time

to meet new challenges. The Washington Academy of Sciences (WAS) of today

is not the same as the organization of a century ago, yet its basic structure as

a federation of Washington area scientific societies has remained substantially

unchanged. A centennial is a particularly appropriate time to reflect on history,

for considering where one has come from is one method of guiding movement

into the future.

The background of the WAS stretches far back to the days of the Civil War

in Washington and is linked to the growth and needs of the local scientific

community for avenues of communication. During the last decade of the 19th

Century, this growth reached a ‘‘critical mass’’ and a new form of organization

appeared and evolved. One of the driving forces in this new structure was Charles

Doolittle Walcott (Yochelson, 1967), president of the WAS for more than one-

tenth of its existence and the one who led it into the 20th Century.

The National Academy of Sciences

Although a detailed study of the scientific societies in Washington has been

published (Flack, 1975), a brief review with different emphasis may be appro-

priate. To begin, the National Academy of Sciences (NAS) was founded in 1863;

it is emphatically not a local society, but plays a roll in this account. Although

there were many underlying reasons for its formation, a pragmatic one was that

184

WAS: BACKGROUND, ORIGIN, AND EARLY YEARS 185

Charles Doolittle Walcott, second President of WAS. Photo Smithsonian Institution Archives, Record Unit 95.

Negative No. 82-3143. Before the turn of century, Walcott wore a beard, as seen in his picture in the 25th

anniversary history of the Cosmos Club.

186 YOCHELSON

the NAS did, in some small measure, enlist science in the cause of the Union.

The most important founding member residing in Washington was Joseph Henry

(1797-1878), the first Secretary of the Smithsonian Institution who became the

second President of the National Academy from 1868 until his death.

The Academy met twice a year, with an annual meeting held each spring in

Washington, and a fall meeting at whatever location would host the group. When

Henry became the President of the NAS, he provided headquarters space for that

organization in the Smithsonian Castle. What little day-to-day business there was

came from that address until the NAS building was dedicated in 1924.

In an early effort to vitalize the NAS, Joseph Henry suggested that those

members who resided in Washington meet once a month. Several NAS members

from outside Washington, and especially Louis Agassiz, viewed this suggestion

as a plot to place the NAS under the thumb of the Federal government; Henry

tactfully withdrew his suggestion. Although the early records are murky, despite

Agassiz’ fuss, it is likely that this perceived need for intellectual stimulation and

cross-fertilization led to the founding of the Philosophical Society of Washington

(PSW) in 1871; the PSW included mostly non-members of the NAS.

During Henry’s time, and for a few years thereafter, the PSW met at Ford’s

Theater, then the site of the Army Medical Museum. Science was serious business

and Joseph Henry permitted no nonsense at the meetings. Joseph Henry also saw

science as a broad-gauge activity. During the era that Henry was the principal

scientist in Washington, no one would have dared to suggest that the approach

of a scientific society should not be all-encompassing.

1879-1889

In the years following Henry’s death in 1878, the number of scientific societies

in Washington grew dramatically. Two prime factors in this growth were, first,

the increasing specialization of various scientific disciplines and, second, the

enlargement of the Federal government staff, especially in scientific activities

associated with the Department of Interior, the Department of Agriculture, the

United States National Museum, and the Fish Commission. A third key factor

was the presence of the Cosmos Club in Washington, founded in 1878. Although

this club was not by any means a formal scientific organization, it was an institu-

tion at which scientists rubbed elbows socially; when after a few years the Club

finally settled on Lafayette Square, the assembly hall provided a centrally located

meeting ground for the new scientific societies.

The first new scientific society to appear on the Washington scene, February

17, 1879, was the Anthropological Society of Washington. Its founding during

WAS: BACKGROUND, ORIGIN, AND EARLY YEARS 187

the winter of 1879 may have been pure coincidence, or it may have been a subtle

form of lobbying for the Bureau of Ethnology. The bill which established the

U.S. Geological Survey (USGS) on March 3, 1879, also established that Bureau.

John Wesley Powell, one-armed explorer of the Colorado River, and the prime

mover in founding the Cosmos Club, became head of the new Bureau and within

two years was also director of the USGS.

The next scientific organization to be formed, on December 3, 1880, was the

Biological Society of Washington. This was an organization near and dear to the

heart of Spencer F. Baird (1823-1887), Second Secretary of the Smithsonian.

Just as Henry had assisted the PSW in its early days and helped it start a bulletin,

so Baird assisted the biologists.

The third scientific society to emerge was the Chemical Society, which began

January 31, 1884. This group is considered the model and the ultimate starting

point for the present-day American Chemical Society. One of the key persons in

its formation was Frank Wigglesworth Clarke of the USGS, noted for his keen

and biting wit. According to legend, one spring an out-of-town member of the

NAS attending the annual meeting met him on the street and asked ‘‘How’s the

Comical Society?’’ Clarke reply was ‘‘Fine. How’s the Notional Academy?’’

Apparently this was so upsetting to the academician that for some years thereafter

Clarke was not elected a member of the NAS.

Four years after the start of the BSW, the National Geographic Society (NGS)

appeared January 27, 1888. Despite its grand name, for the first few years of its

existence the NGS was essentially a local body, though one designed to appeal

more to the interested layman than to the professional scientist. It is interest-

ing that this society was founded at just about the same time the USGS was en-

deavoring to increase its annual appropriation so as to produce more topo-

graphic maps.

The Joint Commission

Public lectures sponsored by the Smithsonian Institution which were an early

innovation of Joseph Henry, actually started before construction began on the

‘‘Castle.’’ The great fire of 1865 destroyed the lecture hall in that building, but

completion of the United States National Museum building in 1881 (now the red

brick Arts & Industries Building to the east of the ‘‘Castle’’) infused new life

into that public program. During the spring of 1882, the Anthropological and

Biological societies cooperated in organizing a lecture series on Saturdays. Al-

though the PSW stood aloof for a few years, it became involved as another

sponsor during the 1885 and 1886 lecture seasons.

188 YOCHELSON

Frank Wigglesworth Clarke, third President of the WAS. Smithsonian Institution Archives, Record Unit 7099.

Negative No. 97-1692.

According to Gilbert (1899:2—4), the first secretary of the WAS, the three

societies had a joint committee of conference in 1882, which recommended a

federation of the societies into a Washington Academy of Sciences. This plan

“WAS: BACKGROUND, ORIGIN, AND EARLY YEARS 189

Grove Karl Gilbert, first Secretary of WAS, Photo, Smithsonian Institution Archives, Record Unit 95. Negative

No. 78-15934.

failed because the membership of the PSW voted against the proposal. As noted,

inside of a few years that society did assist with the lecture series. There are no

records, but it is likely that these arrangements for lectures laid the foundation

for the Joint Commission (JC).

All Gilbert (1899:4—5) wrote of the formation of the JC was that ‘‘Early in

the year 1888 the desire for federation which had inspired the attempt to organize

an Academy, led to the movement to secure a permanent committee to deal with

the questions of common interest, and this movement was successful.’’ Inasmuch

as John Wesley Powell was one of the founders of the National Geographic

Society and the JC suddenly appeared a month after NGS was founded, Powell

may have been the proponent of this commission; he was a member of all five

of the cooperating societies which originally formed the JC.

190 YOCHELSON

Many members of the Washington scientific community had eclectic interests

and belonged to more than one of these new societies. The founding of The

National Geographic Society could well have been the final straw in emphasizing

the need for some sort of more general cooperation and regulation, even if Powell

himself was not the driving force. Thus, February 25, 1888, the JC came into

being to represent the five local societies. ““The Commission consists of three

delegates from each of the component societies and its functions are advisory,

except that it may execute instructions on general subjects and in special cases

from two or more of the societies participating’’.'

As an example of multiple memberships, in the first edition of the JC directory

of the local societies Walcott is listed as a member of the Anthropological,

Biological, National Geographic, and Philosophical Societies. The second year

he had dropped out of the Anthropological Society and in the 1894 edition,

naturally enough, he had added the new Geological Society of Washington to

his membership. 7

Garrick Mallory, one of the founders of the Anthropological Society was the

first president of the JC and Marcus Baker was the first secretary. Within a year,

the JC arranged a calendar so that the several societies did not directly conflict

on meeting dates. This calendar was published in the annual directory of mem-

bers of the various societies. The introductory part of the directory gives a

brief account of the JC and each of the five societies, along with the officers of

each group. |

The directory itself is a useful source of local addresses, work affiliation, names

in full, and memberships; the directory also listed the non-resident members of

these various groups. Its increase in size over the years documents the growth

of the local scientific community and of the several societies. The 1889 edition

listed 579 members, the 1890 edition, 662, and 1891, 851. In the fourth edition

of 1892, separate counts were made for members in Washington and vicinity

(684) and non-resident members (278) of the local societies.

In addition to the directory, one of the first actions of the JC was to invite

the American Association for the Advancement of Science (AAAS) to meet in

Washington in 1891. The AAAS had only met once previously in Washington

and that was in 1854. Hosting this meeting was a formidable undertaking, but

‘the JC was able to bring it off successfully. The newly formed Geological Society

of America met in conjunction with the AAAS and immediately following it,

the Fifth International Congress of Geologists convened, so the arrangements

also had -an international flavor.

The JC was a remarkably stable body during its early years with Mallory and

Baker continuing in office for years. Delegates from the societies changed, but

mainly it was by a member shifting from representing one society to representing

WAS: BACKGROUND, ORIGIN, AND EARLY YEARS 191

Marcus Baker, second editor of the Proceedings. USGS photographic library, Denver, Colorado. Negative No.

Za. Portraits.”

another. On December 20, 1892, the Entomological Society was admitted to the

Commission with one delegate, and on December 9, 1893, the Geological Society

of Washington (GSW) was admitted with two delegates. President of the GSW

Charles Doolittle Walcott appointed himself and J. S. Diller, another USGS em-

ployee, as the representatives to the JC. The founding of the GSW so late relative

to the other disciplinary societies has been discussed (Yochelson in Robertson,

1993). There is no obvious reason for that delay, but the society itself came into

192 YOCHELSON

being within a week, during a respite from Congressional inquiry of the USGS

which had begun in 1892.

The milestone of breaking 1000 memberships was reached in 1894 when 1072

members were listed in the Washington area; non-resident members of societies

dropped to 140. More significantly, the JC underwent a major change. On January

25, 1895, the JC adopted a constitution written by J. S. Diller.* The officers and

boards of the several societies formed the JC and it was given the power to

organize joint meetings, arrange public lectures, and distribute notices of meeting,

along with the publication of the annual directory. Walcott was one of seven

members of the Executive Committee, one from each of the adherent societies,

plus the two officers. Gardiner Greene Hubbard, president of the National Geo-

graphic Society, was elected president of the more formalized JC for several

years and led it, or misled it, through the steps of this reorganization and the

early phases of formation of the WAS.

It is not obvious as to what caused this change to a more formal structure.

One possible factor may have been the action of the Women’s Anthropological

Society. The group applied for membership in the JC. The male worthies deter-

mined that the JC had no power to admit them and that only by an affirmative

vote of all the affiliated societies could another group be included within the JC

structure. As might be expected, the membership application failed, but it could

have left in its wake a notion that having a set of rules to refer to might be

appropriate.

The following year, the short constitution was reprinted in the directory along

with bylaws and a list of 82 officers of societies who now became part of the

JC, besides the 11 persons who were officers or on the executive committee.’

This made for what was potentially a most unwieldy group.

The Geike Affair

To backtrack a little, Professor George Huntington Williams the founder of

the Department of Geology at The Johns Hopkins University in Baltimore, died

unexpectedly in 1894. He was a rising star in the study of metamorphic rocks

and his death at 38 from typhoid fever was a tragic loss. Mrs. Williams instituted

a lectureship to commemorate his memory and the first lecturer chosen for the

honor was Sir Archibald Geike, director-general of the Geological Survey of

Great Britain and Ireland and an expert on ancient volcanoes (Pettijohn, 1988:21).

At some time during the early part of 1896, Gardiner Greene Hubbard, then

president of the JC wrote to Geike, inviting him to address the JC at Washington;

Hubbard knew that this distinguished foreigner would be close by and this was

WAS: BACKGROUND, ORIGIN, AND EARLY YEARS 193

too good an opportunity to miss. Geike’s reply, late in June was positive.

“*It will give me pleasure to accept your invitation to address the Joint Commis-

Stemag oO."

In May, 1896, before any reply had been received from Sir Archibald, Hubbard

informed the GSW by letter that he had invited that worthy to Washington the

following year, after Geike had completed his duties in Baltimore. Hubbard also

asked the GSW to second his invitation.” However, GWS president S. F. Emmons

then wrote inviting Geike to lecture before that society. When that letter was

sent is not clear, but what was apparently another letter was sent to him on

December 1, 1896.

On January 8, 1897, Geike finally responded to Emmons: ‘‘Your letter on

behalf of the Geological Society of Washington gave me great pleasure. I have

delayed replying in the hope of being able to say something definite as to dates.

But this I find impossible at present. . . . I shall of course be very pleased to

meet the members of the Society over which you so worthily preside. As for an

address my hands are so full of work that I dare not promise any forward prepared

discourse.°

According to a Washington myth, the JC invited Geike, but upon hearing of

the invitation from the GSW, the JC withdrew their invitation. The geologists

considered this a slight against a distinguished visitor and were incensed. As

with so many myths, the real story is not quite that simple.

As early as December, 1896, the Chemical Society had informed the GSW of

its concerns over the expenditures and functions of the JC. In mid-January, 1897,

after Geike’s somewhat vague acceptance of their invitation was presented to the

GSW council and a committee on arrangements for his lecture appointed, the

feisty council voted that a committee be appointed to investigate the history of

the JC; Whitman Cross and G. K. Gilbert, both USGS, were appointed. Several

times the committee reported no progress in gathering the necessary information,

which certainly poured a little gasoline on the fire.

On March 18, 1897, Sir Archibald, who was becoming increasing cramped

for time, wrote Hubbard: “‘I find that I can be in Washington from the 2nd to

the 7th May and should be pleased if any date within these limits would suit

the convenience of the Members [presumably of the Joint Commission]. The

Geological Society of Washington has also asked me to address its members. I

presume that they are also members of the Joint Commission and in that case it

might perhaps suffice to have a single lecture’’.’

Geike further asked Hubbard to discuss matters with the president of the GSW

and the same day sent another letter to Emmons, whom he thought was still

president. ‘‘I don’t know about a formal discourse that would be possible for

me, my time for preparation being limited. But I would gladly address the Society

194 YOCHELSON

on some subject on common interest to us all. Mr. Hubbard asked me to address

the Joint Commission and I am writing to him this much [?] for the subject. I

hope it can be arranged that one lecture will suffice and perhaps you would

kindly see him on this point’’.® |

A third letter on apparently that same day went to Hague, who as new president

of GSW had re-extended the invitation. Part of Geike’s reply was: ‘‘You suggest

the repetition of one of the Johns Hopkins lectures. I am quite willing to make

that arrangement if you think it best or this address may be merely informal and

be left to be decided on when I get to Baltimore’’.”

These letters, innocuous in themselves— after all Sir Archibald had his hands

full in Baltimore and Washington was simply an extra chore—added gunpowder

to the explosive mixture of concern about activities of the JC. On April 1, 1897,

possibly an appropriate date for this concatenation, the council of GSW voted

almost unanimously that their society should sponsor the Geike lecture, but it

was also agreed that they should consult with the JC.

At the JC Executive Committee meeting of April 6, 1897, President Hubbard

noted that indeed he had corresponded with Sir Archibald, as had later the GSW.

Near the end of the correspondence, Sir Archibald had indicated that he preferred

to give a single lecture in Washington. At this point, the JC insisted that since

Hubbard had written first, the JC was the appropriate sponsor of the lecture.’°

This opinion of the JC was technically correct, but it poured burning naptha on

already troubled waters.

At the April 1, special meeting, the GSW council had appointed Walcott and

its current president Arnold Hague to take up the matter with the JC. In two

weeks, they reported that all had been smoothed over. Just incidentally, because

of the press of other duties, Walcott had resigned from the GSW council in

January and had no part, officially, in the specific and more general concerns

raised by the GSW as to JC activities and the invitation to Sir Archibald.

Geike spoke on the evening of May 5, 1897, under the auspices of the GSW.

The lecture went well, but it did not heal the strained relations between the JC

and the GSW. The Treasurer had no objections to the bill of $7.00 for the lantern

slides used, though earlier he had objected to what he considered excessive

charges by the JC and insisted on itemized bills.

It took almost three months for the two-man investigating committee to gather

the information on the JC which the GSW council had requested. When they

were ready to report, the report was delayed until after the Geike lecture. A

special council meeting was called on May 26, 1897, to discuss the JC. The

council voted to send the report and a circular letter to all the other societies

affiliated with the JC. The sentiment was that the GSW “‘council expresses it

disapproval of the present organization of the Joint Commission as being neither

WAS: BACKGROUND, ORIGIN, AND EARLY YEARS 195

well adapted for performing the business of the societies nor representing them

in scientific matters’’.'' One concluded that the GSW was not happy.

The First Step

Whether the trigger was Geike’s lecture, or that was simply the last part of a

deeper concern, on September 15, 1897, the circular was distributed by the GSW;

it suggested ‘‘that the question of a joint organization should be considered by

the scientific societies with a view to improving the means for furthering their

common interests.’’ The letter proposed that each society appoint ‘‘a committee

of conference to meet similar committees from other societies for the consider-

ation of the general subject.’’ These quotations are from the report of geologist

G. K. Gilbert (1899:7), first secretary of the WAS and one-half of the GSW

investigating committee; more details of this circular can be found in the GSW

minutes.

Gilbert gave a brief summary of the history of the JC and in it noted two

‘political’ actions undertaken in 1896-1897 by the reorganized JC. One was

to endorse the request of the Secretary of Agriculture to Congress for a position

of Director-in-chief of Scientific Bureaus and Investigations, and the second was

to oppose a bill restricting vivisection in the District of Columbia. ‘‘While the

motives . . . were shared by nearly all members of the Commission, there was

serious doubt as to the propriety of permitting a Commission organized primarily

for business purposes to attempt to represent. . .”’ This is a correct view of the

sequence of events and was a good official reason to replace the JC structure;

Gilbert was too much of a gentleman to mention the Geike problem, nor to note

how unwieldy the JC had become.

Organizing for WAS

In the fall of 1897, as requested by the GSW, the seven local societies each

appointed three people to the conference committee. The circular of the GSW,

the conference committee members, and their final action were noted in Science

(Anonymous, 1898). After several quick meetings, the group passed six resolu-

tions, the fifth being the prime point. That one suggested changing the name of

the Joint Commission to the Washington Academy of Sciences, that the WAS

assume independent scientific functions, and that it have the power to add mem-

bers. The resolutions were then sent to the JC and moved on to the affiliated

societies for consideration. In short order, the Biological, Entomological, National

Geographic, Geological and Philosophical societies accepted all six resolutions;

196 YOCHELSON

the Anthropological Society accepted the key resolution, and the Chemical Soci-

ety had not met to consider the matter.

In the midst of assembling WAS, Gardiner Greene Hubbard died on December

11, 1897 at age 75. In addition to other activities, Mr. Hubbard was a Regent of

the Smithsonian Institution and, though he was professionally a lawyer, he was

a respected senior member of the inner workings of the city’s scientific commu-

nity; no one held him personally responsible for the confusion about the sponsor-

ship of the Geike lecture.

By January 11, 1898, the Joint Commission began serious discussion on what

had been started. Major Powell wanted a selected body which was financially

stable. *‘Mr. Gilbert spoke of the functions of the Academy expressing the belief

that (1) it should not duplicate the work of existing societies but should seek to

do what they did not do (2) that it should provide for courses of public lectures

and (3) that provision should be made for patrons’’."”

A committee of eight was appointed by the Joint Committee to draft a constitu-

tion, and they immediately added seven members including Walcott. Interestingly

enough, he did not attend the first four meetings of the constitution committee.

He had other business to occupy himself, with a new session of Congress starting,

but he must have been kept apprised of the discussions.

Enter Walcott

This heading is an oxymoron, for though Walcott was not officially involved

until January 19, 1898, as one of the vice-presidents of the JC, he for years had

had his finger on the pulse of Washington science. In addition to being director

of the Geological Survey, late in January of 1897 he took on the additional task

of Acting Assistant Secretary of the Smithsonian Institution in charge of the

United States National Museum, a position he held for eighteen months. Hardly

had he agreed to that, when suddenly from late February onward he was heavily

involved in the setting up of the National Forest Reserves and when the smoke

cleared in the spring of 1897, the USGS was studying trees, as well as rocks and

water. This was also the year he was vice-president of the Cosmos Club; the

following year he was president.

It was on that January night in 1898 that the JC authorized the Committee on

Constitution to incorporate WAS under the laws of the District of Columbia. The

constitution was distributed to the affiliated societies. On the evening of January

25, 1898, the JC had two main pieces of business. The Medical Society of the

District of Columbia had asked to be admitted to the JC; it was agreed that while

this was an appropriate society to include within the fold of the JC, no action

WAS: BACKGROUND, ORIGIN, AND EARLY YEARS 197

should be taken until the WAS was formed. Then came the more difficult issue

of who should be in the WAS. Several plans had been circulated, other ideas

were proposed by the affiliated societies and still others came from those present.

““Mr. Walcott then offered a substitute for Dr. Merriam’s motion which after

discussion was adopted in the following form:

Resolved That a nucleus of 75 members of the Washington Academy of Science

be formed by election in the following manner:

Each member of the Joint Commission may prepare a ballot containing not

more than 100 names of persons now members of one or more of the scientific

societies of Washington, the ballot to be canvassed by the Executive Committee

of the Commission not later than Monday Jan. 31, and the 75 persons having

the largest number of votes to constitute the original members of the Academy’’.'°

The Executive Committee was given the power to decide on a member in the

event of a tie vote. Other suggested changes in the Constitution were referred to

the fifth and final meeting of the Committee on Constitution, six days later.

General Sternberg, Surgeon-General of the Army and the new president of the

JC, had a previous engagement and Walcott was asked to chair this late afternoon

meeting; he had helped write the constitution for the Geological Society of

Washington and earlier had helped to revise the constitution of the Biological

Society. That same night the Executive Committee of the JC met from 8:00 Pm

until 2:00 Am to determine the 75 members. The individuals selected received

from 18 to 32 votes each and, as Walcott had anticipated, there was a tie for the

75th place.

February 2, the Executive Committee of the JC met again. This time the main

agenda items were fine tuning of the constitution and a reading of the proposed

articles of incorporation. Then the document was turned over to the several

societies for their consideration. The established PSW was as reserved as the

new GSW was feisty. Fortunately Walcott was a vice-president of the PSW and

at a meeting two days later, he ‘‘Voted to support Washington Academy of

Sciences’’,'* which was hardly a surprise. In its vote the PSW reserved the right

to consider the final product, a formality which soothed the concerns of a few

members of that august group. In founding the WAS, the aligning of the PSW

was by all odds the most formidable task of salesmanship.

February 18, 1898, the members of the Constitution Committee met to incorpo-

rate the WAS, not forever but for a term of nine hundred ninety-nine years. They

met again on March 5, to give legal status to the elected officers of the new

organization. March 22, the JC held its last minute and adjourned sine die. In

accordance with Murphy’s Law, the early part of the meeting was marked by an

argument that a quorum was not present; fortunately, several members appeared

late and completed the transition from Joint Commission to Washington Academy

198 YOCHELSON

of Science. The functions of the new WAS and its officers were again an item

for “‘Science’’ magazine, (Anonymous, 1898a).

The archives of the JC went to the WAS and their furniture was donated to

the Cosmos Club. In a bizarre sort of justice, the JC archives do not contain the

final resolution dissolving itself, but a copy is pasted in the minutes book of

the GSW.

1898

Unfortunately, the early minutes of the WAS are lost. Gilbert (1899:xii) records

only five items for the first year. The first two are the incorporation and organiza-

tion meetings mentioned above. A second and third meeting for organization

were held on March 17, and March 29, respectively. Finally, a business meeting

was held for election of members on May 27.

However, more was happening: ““The Board of Managers has held numerous

meeting for the transaction of business.’’ The Medical Society was admitted; its

president gave his address under the auspices of the WAS, as had the president

of the Anthropological Society earlier in the year. It may be noted that J. R.

Eastman, of the U.S. Naval Observatory, was the first President of WAS, G. K.

Gilbert, USGS, was the first Secretary, and Bernard R. Green, Superintendent of

Construction, Library of Congress, was Treasurer. These officers, and nine vice-

presidents representing each of the affiliated societies, supplemented by the

elected Board of Managers, were to manage the WAS.

‘*The principal work of the Academy in 1898 was organization. . . . A number

of business meetings were held by the Academy and the Board of Managers in

the spring, and the Board of Managers held another series of meetings in the

autumn and winter. . . the Board has developed the machinery for the publication

of proceedings, the conduct of a lecture course, and the holding of occasional

meetings for the reception of new scientific material. The function which has

received the most attention is publication. A plan for the selection and printing

of papers has been carefully matured and accepted papers will soon go to press.

A joint directory for 1899 is now in preparation under the editorship of Mr.

Marcus Baker. [Issued February 24, 1899]’’ (Gilbert, 1899:13-—14).

The constitution as first written made provisions for three classes of member-

ship, patrons, honorary members, and regular members. In addition to the 75

members first elected, another group was added in mid-March, and later a number

more, mainly as a result of the Medical Society affiliating with the WAS. A few

persons declined membership, a few others did not pay dues, and when all was

settled, the WAS had 147 ‘“‘original’’ members. These paid dues of $5.00 and

were to receive the Proceedings.

WAS: BACKGROUND, ORIGIN, AND EARLY YEARS 199

James R. Eastman, first President of WAS. Photo, Smithsonian Institution Archives, Record Unit 7099, box

32. Negative No. 97-1690.

That spring, in an amendment to the constitution, non-resident members were

added as a category. Dr. Florence Bascom, Professor of Geology at Bryn Mawr

was elected a non-resident member of WAS in 1905. In 1906, the category of

life member was added to the constitution, the requirement being a payment of

$100. A 1907 letter addressed to ‘‘Miss Florence Bascom’’ stated: ‘‘Dear Ma-

dame, I have the honor to inform you that at its meeting of January 8, 1907, the

Board of Managers elected you a life member of the Washington Academy of

Sciences’’.’° For years, thereafter she was indicated on the membership lists as

a life member; for at least the first few decades of the WAS she was the only

life member and may have been unique in that category.

YOCHELSON

Bernard R. Green, first Treasurer of WAS. Photo Library of Congress.

The same year that the first life member was elected, the first honorary member

was also elected. The published minutes do not give the names of either. Who

was the first honorary member is a minor mystery for someone else to pursue.

WAS: BACKGROUND, ORIGIN, AND EARLY YEARS 201

In the separate membership lists of the WAS, produced in later years, honorary

members were named, but there are no dates of election nor any clue in the WAS

minutes as to the process.

Regardless of what plans the WAS might have had, money was vital to finance

the directory and, more importantly, to the start of the Proceedings. The JC noted

a donation from Mrs. Gardiner G. Hubbard in memory of her husband, and in

1899 she made another donation to the WAS. The list of resident members

published in the first volume of the Proceedings lists her as a patron, the first

one. Patrons contributed $1000 or the equivalent in property to the WAS.

The WAS Directories

The housekeeping chores, if you will, of the Joint Commission were carried

on by the Washington Academy of Sciences. The directory with the calendar of

meeting was an important document. ‘‘This Directory, like that for 1899, 1900,

and 1901 (none was published in 1902) has been prepared and published under

the direction and at the cost of the Washington Academy of Sciences. It [the

Directory] may be regarded as the successor to the Joint Directory of the Scientific

Societies of Washington, which first appeared in 1889, and annually thereafter

until 1898’’.'°

Late in 1901, the WAS issued a form to all members asking for details which

might be included in the next issue of the directory.'’ Most members complied

with the request.

In 1903 and thereafter for the next two decades, the directory was issued

biannually; the directory continued into the 1940s. Those issues produced during

Walcott’s tenure as president are listed in the notes.’*

Proceedings, Patrons, Members

Volume 1 of the Proceedings of the Washington Academy of Sciences bears

a title page date of 1899; (the volume was not completed until 1890). Volumes

2-4 were yearly, but 5—9, each crossed into a succeeding year. Volumes 10-—

13 were each within the year 1908-1911, respectively. Officers of WAS are

listed and in all volumes save the last, the president is Charles D. Walcott.

The Proceedings actually say little about the activities and meetings of the

WAS, but serve as a vehicle for publication of papers. The manuscripts accepted

by the publication committee could come either from the WAS or be forwarded

by any of the affiliated societies; the societies were expected to contribute half

the cost of publishing manuscripts originating within their society. Publication

202 YOCHELSON

W ASHINGTON ACADEMY OF SCIENCES.

NOVEMBER 5, IQOI.

DEAR SIR:

The Publication Committee of the Academy has been

directed to prepare a new list of members for insertion in the

current volume of ProcEEpINGs. ‘To this end, members are

requested to fill out the following blank and return promptly to

MARCUS BAKER,

1905 16TH ST., WASHINGTON, D. C.

[Name. WNawe ts preferred in full.]

{Official position or title if any ]

wobec se ceee ces sc oes sees teases cee ere rare Cem ee es eseseneeseasconseacscescece ei eee eee ere rere rere rrr rrr ry ty

[Official or business address. Give also any special address.]

Information form of Charles Doolittle Walcott for WAS Directory

WAS: BACKGROUND, ORIGIN, AND EARLY YEARS 203

was quarterly in brochures which contained one or more papers; each separate

paper within a brochure also appeared as a reprint and was available for sale.

Membership of the Publication Committee and the rules concerning publication

are given in the Proceedings."

Walcott (1900) availed himself of this series twice. The first article is based

on field work conducted in eastern Canada during 1899. He did not return to

Washington until mid-September and was unable to start on the manuscript until

early December, yet it appeared February 14, 1900; as shown below, his speed

in writing matched that in setting up organizations. This paper completed the

volume I of the Proceedings. His article is profusely illustrated and in some

respects resembles articles of that day in the *‘National Geographic Magazine;”’

it cost $.50. Some years later (Walcott, 1905), he published a much shorter article,

which sold for $.10. The proceedings of the commemorative meeting to John

Wesley Powell at which Walcott presided cost $0.75.

One outgoing letter in 1901 indicates that non-members of the WAS could

purchase an unbound volume for $5.00.” A copy bound in red cost another 50

cents for non-members; members received bound copies. The same letter suggests

that an exchange program with other societies was in operation. What happened

to any publications received in exchange is another minor unsolved mystery.

Critical to any publishing enterprise is money, for if one does not pay the

printer, there is no printing; as generous as Mrs. Hubbard had been as the first

patron in 1899, more support was needed. On January 12, 1901, the Council of

the WAS unanimously voted two more persons to patron status, James W. Pinchot

of New York, and his son Gifford Pinchot, by then nominally an employee of

the Department of Agriculture*'; each donated $1000 to the WAS. There are no

hints in Walcott’s diary as to how this came about, but if Gifford Pinchot had

one stalwart supporter in Washington it was Walcott. In 1896 Walcott provided

office space for the NAS Forestry Commission and later managed to obtain the

essentially honorary position for Pinchot as a forester in Agriculture. He arranged

for Pinchot to speak to the NAS and even tried to have him elected to membership

in the Academy. None of this information proves Walcott asked for the financial

support, but it is otherwise inexplicable why the elder Pinchot would support a

local Washington society.

The younger Pinchot, never one to make life easier for anyone, complained

that with Patron status he was no longer eligible to vote and the rules committee

struggled to resolve this problem. Eventually he was permitted to exercise his

franchise without payment of annual dues. On the annual summary of total

membership, there is a category of ‘‘counted twice’’ containing the figure 1; that

is Mr. Pinchot, member and patron.

The timing of notifying these two patrons is a bit strange in that the title page

204 YOCHELSON

of volume 2 of the Proceedings is dated March-December, 1900; likely, the

prefatory material was printed and mailed early in 1901. Preceding the list of

members are the names of five patrons (Anonymous, 1900: xii).

The remaining two patrons listed in that first volume were also elected in 1900,

though no letters give the date of elections or any details. One of the patrons is

Thomas Walch. He began by grubstaking miners in the west and ended up owning

a gold mine in Ouray, Colorado. After achieving monied status, he moved to

Washington; his mansion on Massachusetts Avenue is now the Indonesian

embassy. Walcott could have talked both mining and Washington real estate

with him.

Walch is followed on the list of patrons by Mrs. Henry L. Higginson. Mr.

Higginson was a friend of Henry Adams and in 1902 was a member of the Board

of Trustees of the Carnegie Institution of Washington: he died almost as soon as

the organization was founded. Adams was a friend of Walcott. Bizarre though

it may have been for the senior Pinchot in New York to support WAS, it is even

stranger that this lady from Boston should be involved, yet it is the sort of

arrangement that Walcott would have engineered.

In May, 1901, E. H. Harriman, the railroad magnate in New York became a

patron. In 1899, he financed an elaborate expedition to Alaska and before it

departed, Walcott and C. Hart Merriam of the Biological Survey went to visit

him. In later years, Walcott had frequent contact with Harriman; Walcott held a

patent on a railroad tie spike and promoted it for years. A few years later, a

number of shorter papers resulting from the Harriman Alaskan Expedition were

published in the Proceedings. .

On Friday, December 13, 1901, Walcott ‘“‘Called on Mr. Cleveland H. Per-

kins & talked to him of Washn. Acad. Sci. He agreed to send $1000 to aid in

publication’’.** Walcott must have been a silver-tongued orator, for he did not

know the man; he later wrote the surnames of Mr. Cleveland in his diary and

got them backwards. | 7

At some time during that same year Mrs. Phoebe A. Hearst, of San Francisco,

California, was added to the list of patrons. She was the wife of Senator George

Hearst and a philanthropist in Washington. Thus, the fourth volume of the Pro-

ceedings lists eight patrons. Apart from Mr. Perkins, there is no proof that Walcott

obtained their support, but he is a likely source for all these patrons. However

their support may have been gained, the WAS was now financially stable.

As noted, by the end of its first year, the WAS had 144 members. By mid-

January of 1900, this increased to 159. At a comparable date in 1901, resident

members decreased slightly to 156, but 115 non-resident members are reported.

By mid-January, non-residents increased to 152 (Baker, 1902). Early in 1905,

non-resident members were 167, four more than the resident members. Resident

WAS: BACKGROUND, ORIGIN, AND EARLY YEARS 205

C. Hart Merriam, first editor of the Proceedings. Smithsonian Institution Archives, Record Unit 95. Negative

No. 23966.

members were elected by vote of the WAS membership at the annual meeting;

non-residents were elected by vote of the Board of Managers. As might have

been expected, some non-residents later moved to Washington and created further

problems for the rules committee to sort out.

Grand Plans

After the annual meeting, on January 18, 1899, Walcott recorded: **At survey

office during the day. Meeting of Washn. Acad. Sci. 8—10 Pm Elected president.

Presidents reception 10°—11 pm.’’** At the same meeting Frank Baker, director

of the National Zoological Park, was elected Secretary and he served for years

in this thankless task. Bernard R. Green continued on as Treasurer until 1909,

when Arthur L. Day, of the Geophysical Laboratory, at the Carnegie Institution

of Washington, succeeded him. The voting for Walcott became an annual affair

206 YOCHELSON

Frank Baker, second Secretary of WAS. Photo, Georgetown University, Washington, DC.

which was noted as “‘elected’’ or “‘reelected’’ in his diary; the event became so

sop routine that three times he neglected to write it down.

Several events occurred that year as recorded in the printed minutes. As regards

the first, two special meetings were held. At the second of these, on March 15,

1899, Walcott ‘‘Spoke on the United States National Museum before Washington

Academy of Sciences 8“—9 pm. Reception or social meeting followed’’.” He

had completed his stint as Acting Assistant Secretary of the Smithsonian Institu-

tion, but had begun the process of obtaining a new building from Congress; this

was part of his successful campaign as the presence of the domed Natural History

Building testifies.

Second to be recorded were seven annual addresses of presidents of affiliated

societies that were given under the auspices of the WAS. Third, seven public

lectures were also held under their auspices. Fourth, two receptions were held,

one for the annual meeting of the NAS in the spring, and one in December

when the Geological Society of America returned to Washington. Having an

WAS: BACKGROUND, ORIGIN, AND

Arthur L. Day, second Treasurer of the WAS. From Geophysical Laboratory, Carnegie Institution of Washing-

ton.

208 YOCHELSON

organization prepared to host such social events as a matter of course was surely

an excellent, if unwritten, reason for forming the WAS.

Fifth, three business meetings were held. The first two established the category

of non-resident member, and the third elected more resident members. Members

of the WAS had the tangible benefit of receiving the Proceedings. Far more

important, but again unwritten, as members they received the intangible benefit of

prestige. University presidents, some of whom were administrators not scientists,

joined the non-resident category. Those scientists who were able investigators

but were not in the NAS joined. Various categories of science were set up and

colleagues in Washington were encouraged to submit the names of those who

might be elected as non-resident members. Through this process 208 persons

were elected, though not all were seduced into paying dues. In subsequent years,

recruitment of non-resident members continued, but not at such a pace.

Walcott was not a stranger to the NAS, and had spoken before the group three

times prior to his election in 1896. In 1898, he added to his manifold duties that

of NAS Treasurer, and served in the office until 1902. He knew that the NAS,

while not ineffectual, was seldom consulted by the Congress. For the first fifty

years of its existence, historians remark on only two noteworthy actions by the

NAS —the reports leading to the legislation establishing the USGS in 1879, and

the report leading to the setting aside of the Forest Reserves in 1897.

Walcott wanted a much stronger voice for science in Washington, better to

influence Congress. He was building a power base for science, though it was not

without its toll in the time required. In 1899, the Board of Managers held twenty-

five meetings, the same number the next year, and eighteen during 1900. His

efforts did pay dividends.

One of Walcott’s immediate aims was to further higher education. Some gov-

ernment facilities had been opened to graduate students by Congress in 1892,

but much more could be done in this regard. Walcott had the WAS pass resolu-

tions and he **Talked to a lot of men at A. Graham Bell’s residence on movement

to utilize scientific bureaus in Washn. to aid in Education’’.» By an act of March

3, 1901, ‘‘facilities for study and research in the Government departments, the

Library of Congress, . . . shall be afforded to scientific investigators and to duly

qualified individual students and graduates. . . .’ (Walcott, 1901:1004). The

WAS and particularly Walcott were the driving forces in having this act pass. A

provision for ‘‘similar institutions hereafter established. . . .’’ is the legal basis

for students at Los Alamos, Livermore, and a host of other federal facilities. If

the WAS had accomplished nothing else during its century of existence, the

legacy of this legislation more than justified the formation of the organization.

In 1901, the WAS invited the American Society of Naturalists and their affili-

ated societies to meet in Washington and extended a similar invitation to the

WAS: BACKGROUND, ORIGIN, AND EARLY YEARS 209

AAAS. The meeting in December, 1902, was as much of a success as had been

the 1891 gathering which the Joint Commission had sponsored.

Early in 1901, secretary Frank Baker wrote to a disgruntled medical doctor

whose complaint was that the meetings of the WAS conflicted with those of the

~ Medical Society. After refuting the charge, he went on to explain:

‘‘Perhaps you may not fully realize that the object of the Academy is not

primarily to bring out /ectures or to give entertainments. It is for the general

advancement of science at the national capital, the lectures and entertainments

being merely incidental. We have at present 115 non-resident members who

cannot, of course, attend lectures and entertainments at all.

‘‘During the past year the Academy has progress greatly in several way. It

has secured sufficient money to purchase a site for a building intended to accom-

modate all the affiliated societies, it has published a volume of Proceedings

comprising 710 pages which will compare favorably with the publications of any

similar society in the world’’.*®° The accomplishments of the WAS were golden.

The Attempt for a Building

The lackluster interest of Congress in both the AAAS and the NAS graphically

demonstrated that without a headquarters, an organization lacked a Washington

presence and could accomplish little to influence legislation. Walcott spent part

of 1899 trying to find space for the NAS in the new Library of Congress building,

and elsewhere within the government structure, but without success. It was now

time for the WAS to act and by 1900, the same year that Walcott was President

of the Geological Society of America, he had the wheels in motion; ‘‘Meeting

on Committee on building Washn. Acad. Sci.’’*’

It was about the time that Walcott was prowling Washington to look for a NAS

headquarters that he joined forces with the newly founded George Washington

Memorial Association (GWMA). This was originally exclusively a national ladies

organization whose aims were to honor George Washington’s desire for a national

university and to increase patriotic sentiment. A series of state chapters were set

up to raise funds for this proposed university.

Walcott played both sides of the net in steering toward his particular goal. The

GWMA emended its charter to his specifications, and within a relatively short

time, each organization had appointed a committee to meet jointly. Walcott was

‘‘At Geol. Survey for most of the day except for meeting of Committee Washn.

Acad. Sci. & G.W.M.A. at Mrs. George Hearst’s’’.*® Recall that patron Phoebe

Hearst and Mrs. George Hearst are the same person. Two days later another

meeting involved representatives of the American Association of Agricultural

Colleges, and another joint meeting was held two weeks later.

210 YOCHELSON

Walcott continued to move both groups toward his objective of a headquarters

building. By letter dated May 11, 1901, Walcott appointed himself chairman of

a five person committee authorized: ‘‘Ist to enter into such agreements with

the George Washington Memorial Association for cooperation in the erection,

maintenance, and conduct of a memorial building as may be judge necessary;

2nd to co-operate with the Executive Committee of the George Washington

Memorial Association in selecting and designating incorporators’’.”

It had taken time to get all forces aligned, but it was characteristic of Walcott

that once the items were in place he moved swiftly. Within a week ““The incorpo-

rators of the W. [ashington] M. [emorial] I. [ntitute] met in my office 10 am.’’*°

Despite inertia and disparate objects of two such dissimilar groups, by May 20

when the WMI incorporation papers were filed, they had been welded into an

administrative structure which could undertake a major building project (Walcott,

1901: 1008—1010); naturally, the WMI president was Walcott.

When things began to move, Walcott kept them moving. The National Council

of Education appointed a “‘distinguished committee of fifteen’’ to consider the

need for a national university. They prepared a long report with included ‘‘Re-

solved, That we approve the plan for a non-Govermental Institution, known

as the Washington Memorial Institution, to be established and maintained at

Washington, D.C. for the purposes of promoting the study of science and liberal

arts at the National Capitol and of exercising systematic oversight of . . . students

in the Governmental laboratories and collections,. . .’’>!

The overall scheme was that the GWMA would raise money for a building

and the WAS would manage it. The hope was that the National Educational

Association and the Association of Agricultural Colleges and Experiment Stations

would also cooperate in the venture (Baker, 1902:15). As indicated in the unpub-

lished source noted above, by 1902, the presidents of these groups were on the

advisory board of the WMI, along with cabinet members and other distinguished

officials in Washington. Meanwhile, the WAS had acquired a building lot, though

it may not have been in an ideal location within the city.

The founding of the WMI and its aims and aspirations received three pages

in Science (Anonymous, 1901), plus a letter to the Editor (McGee, 1901). To

keep all parties happy, in December Walcott attended the ‘‘Annual meeting of

G.W.M.A. Talked to the ladies 11°2’’.*”

As a result of the founding of WAS, some members of the NAS became

concerned as to its motivations and sphere of influence; this upstart might do

more than provide a reception at the annual meeting and it could undercut the

NAS. The scientific writings and speeches of Walcott are clear. To sooth the

NAS members he prepared a three-page form letter which is a classic in the field

of obfuscation.

WAS: BACKGROUND, ORIGIN, AND EARLY YEARS 211

“‘The Washington Academy may consider such questions affecting the rela-

tions of science to the government as are not taken up by the National Academy,

and which affect the entire scientific body in its relations to the Government. It

may also consider the interrelationship of the various scientific organizations that

unite to form the scientific body, and give such assistance that may be in its

power to the correlation of scientific work and the advancement of scientific

interests in the country. It was reasons such as these that led me to favor the

development of the Washington Academy of Sciences... . . I have given a great

deal of time and attention to this organization the last year with the belief that

it was well spent. As the same time I should take pleasure in aiding in any way

in my power in the advancement of the interests of the National Academy’’.*°

That did an excellent job of muddying the waters, but the issue continued to

simmer. Walcott wrote John Shaw Billings whom he had succeeded as Treasurer

of NAS: “*The question of the relations of the National Academy and the Wash-

ington Academy can be talked over at the time of the meeting next month. I do

not, however, see any reason for giving the subject serious consideration from

the point of view that the Washington Academy is contemplating any injury to

the National Academy. I think it would be best not to agitate the subject, as the

impression might be given that the National Academy was endeavoring to retard

and injure the Washington Academy’’.™* It is difficult to distinguish between the

views of NAS Treasurer Walcott and WAS President Walcott.

The WAS was not about to supplant the NAS, but by pointing it in the

direction of a headquarters building, Walcott had shocked the members of the

NAS into movement of their own to that end. The hopes for a building led to

more intertwining of the organizations. For example, a donation to the NAS of

$300 was acknowledged by Treasurer Walcott as ‘‘subscribed towards a build-

ing fund for a building of the Washington Academy of Sciences, National

Academy of Sciences, etc.’’.*°

In 1900, Alexander Agassiz had intimated that he would provide $100,000

toward a NAS building. The following year NAS Treasurer Walcott wrote Agassiz

to clarify matters between the NAS and WAS. After discussing Agassiz’ pre-

sumed donation, he indicated: ‘‘My view is that the Washington Academy and

the scientific societies here should raise at least $100,000 and that the building

should cost no less than $200,000 exclusive of the ground’’.*°

These are the dollars of nearly a century ago and a multiplication of 15 or 20

is needed to bring them to 1998 sums; no one ever accused Walcott of thinking

small. If he had any hobby, apart from fishing, it was real estate and he was at

the time involved with his third apartment building in Washington.

Another letter to Agassiz adds more details to the machinations. *‘During the

212 YOCHELSON

past winter the Washington Academy of Sciences made an agreement with the

George Washington Memorial Association after the latter had changed its charter

eliminating ‘‘national university’’ to the effect that the Memorial Association

was to provide the means for the erection of a building which would be the

headquarters for the scientific organizations of Washington, the National Acad-

emy of Sciences and the proposed organization for post-graduate work and re-

search in cooperation with the Government Departments.

‘*Mr. Alexander Graham Bell and myself are members of the Advisory Com-

mittee, and Mrs. Walcott is First Vice-President of the G.W.M.A. so we are in

full touch with all that transpires within the Association’’.*’ Walcott went on to

add that the GWMA had $17,000 in cash and pledges for far more.

In trying to obtain support and funds for a headquarters building to give the

scientific and educational organizations a physical place in Washington and a

higher profile, Walcott had the highest of motives. It was more realistic than

trying to found a national university. Nevertheless, put in bald terms, he had

suborned the original purpose of the GWMA from campaigning for such a univer-

sity to erecting a building mainly for science. To be fair, Walcott never abandoned

the GWMA and as late as 1922 was chairman of their finance committee.

He continued his efforts toward greater support for education, but the 1901

law was the high water mark of that activity. To add one more tangle to the

skein, as a result of the WMI Walcott became a Trustee of Columbian University.

What was originally Columbia College metamorphosed into George Washington

University, hardly a national establishment, but at least a token acknowledgement

in the nation’s capital of the original purpose of the GWMA.

Grandiose Schemes

Early in January, 1901, Secretary Baker wrote a non-resident member trying

to convince him to remain with WAS. ‘‘What we wish to do is build up an

association of scientific men which will be in America what the Royal Society

is to Great Britain, and thus to dignify and consolidate all branches of American

science. We are therefore very careful whom we invite to membership, as you

will see by scanning the list which I inclose [sic] herewith’’.*®

These were bold sentiments, indeed. Four months later, he wrote another letter

to W J (‘no stop’’) McGee as chairman of a five man committee. ‘“The president

of the Academy has appointed the following committee to consider: a. Whether

the Washington Academy of Sciences is now a national organization. b. If not,

should it become such. c. What steps are necessary to accomplish that end’’.”

Bolder sentiments.

WAS: BACKGROUND, ORIGIN, AND EARLY YEARS 213

The term grandiose used in the title of this section may be inappropriate, but

grand schemes succeed, whereas grandiose ones do not; the merit of the proposal

itself may ultimately not be germane to success or failure. What is evident that

from about 1903 onward for the next five or six years nothing much new devel-

oped within the WAS.

Early in 1902, the WAS held a meeting at the Cosmos Club to discuss a

proposed building; they had a lot and the nucleus of a building fund, but that

was all the WAS had.

Another potential initiative began in February, 1903, when Walcott appointed

himself chairman of a committee of four ‘‘On the relations of the Academy to

other organizations’’.*° Presumably closer association with educational groups

was intended, but there is no record to support this assumption.

Matters began to slow down in a number of respects. For example, from 1904

onward the Board of Managers met eight or nine times a year; they had gathered

three times as often when the WAS was growing vigorously. The WAS continued

to host a social event for the NAS, and in 1907 entertained visitors coming

south from the International Zoological Congress in Boston, but the concept

of supporting national meetings in Washington receded. In 1906, the Board

recommended that a natural scientist be appointed to the District of Columbia

school board, an important enough action, but a far cry from national concern

about education.

The WAS never became a national organization, nor even one of national

prominence. What with resident and non-resident members being elected and

resigning, and changes of address, there was a considerable load for the Secretary,

but it was paper activity, not growth. A committee recommended that any member

of a local society be automatically eligible to join the WAS, but nothing came

of that suggestion. It may have garnered a few more supporters, but it would

have cheapened the prestige of being elected to membership in the WAS.

April 24, 1909, Mrs. S. W. Dimock, long-time president of the George Wash-

ington Memorial Association who had succeeded Mrs. Hearst, addressed the

WAS on the aims of the GWMA and the need for a building. It was metaphorically

a cry in the wilderness. The association eventually got to a cornerstone-laying

ceremony, but the building never began, for even with the merits of the cause,

funds could not be raised.

The Proceedings became a major concern as their contents had become increas-

ingly specialized. Add to that from Volume VIII onward, members— whether

interested in the contents or not—received only about half the number of pages

they had been sent in earlier years. Walcott left office just after volume XII was

completed.

In 1905, there was discussion of a weekly newsletter, but it never matured, as

214 YOCHELSON

there simply was not that much local society news. However, these discussions

may have laid the seeds for a monthly Journal, and that new series began in

1911. The Board of Managers agreed that the Proceedings would be phased out

with only two issues that final year.

The Board was surprised when a third brochure came out in August and several

of the Managers were quite upset that there was no announcement that the series

was to end. The WAS treasury was no longer fat, and money was grudging

appropriated to pay the bills for this printing. The reaction to this third brochure

was mild compared to that when another issue appeared in November; at least

that one finally announced the end of the Proceedings. There simply was no

money not already committed to the Journal and the Board was asked to make

individual contributions to bail out the editor. It was not a happy time for the

Treasurer, the Managers, or the former editor who was not listed as a member

of the new editorial board for the Journal.

Discussion

Considering what had been accomplished in slightly more than a decade, the

early history of the WAS certainly cannot be deemed a failure. However, the

change between the first few years of Walcott’s tenure and his later years as

president of WAS deserve comment. In a nutshell, Walcott was foundering from

an overload of administrative duties. Unfortunately by reelecting him year after

year the membership of the WAS refused to recognize that fact.

In a reprise of his 1897 activities, in December of 1901, Walcott met Andrew

Carnegie. The Carnegie Institution of Washington (CIW) was teetering on the

edge of never being born, but Walcott rescued it, incorporated it in January of

the following year and became its secretary for three years (Yochelson, 1994);

the secretaryship turned out to be a far more complex job than anyone, least of

all Walcott, had anticipated. Of course, he was still the director of the ever-

growing USGS. Almost immediately after incorporating the CIW, the reclamation

of western lands came to the fore. At the insistence of President Roosevelt, the

1902 Reclamation Service became part of the USGS because Roosevelt wanted

it run right.

Although there are parallels, these duties presented a more complex and de-

manding role than that Walcott faced five years earlier, when he was running

the National Museum and had the Forest Reserves piled on the USGS. To add

to that, in 1903, President Roosevelt had him chair, not one, but two committees.

He was always a prolific scientist, but in 1903, only one two-page paper and

remarks at the Powell Memorial meeting (Walcott, 1903) are on his bibliography.

In 1904, only the annual report of the USGS is recorded.

WAS: BACKGROUND, ORIGIN, AND EARLY YEARS 215

Walcott was swamped, yet he did manage to keep the WAS going, even if it was

essentially marching in place. He simply could not keep the Carnegie Institution of

Washington moving forward and simultaneously pursue the interests of the WMI

and the GWMA. Many onlookers could not figure out what the relationship of

the CIW was to the other organizations. Funding which might have come to the

WAS or WMI dried up in the face of the massive Carnegie endowment and

uncertainty as to what it might support.

To partially balance the scales, in May 1901, Walcott had convinced Daniel

Coit Gilman, retiring president of the Johns Hopkins University to serve as

chairman of the WMI. That fall Walcott wrote to Rockefeller and to Carnegie

asking for money to support the WMI. Unknown to Walcott, Gilman was advising

Andrew Carnegie and it was he who brought the two men together.

In the long run, keeping the CIW on track did much more for science than a

headquarters building in Washington would have accomplished; it would be many

years before “‘public interest’’ lobbying became significant. Although the WAS

played an indirect role through the founding of the WMI and the resulting contact

with Andrew Carnegie, the organization can take pride in at least tangentially

helping to found the CIW.

With all his load, still before leaving the USGS in 1907 and starting another

major chapter in his research, Walcott did lay the foundation for the Bureau of

Mines. What the WAS might have accomplished had president Walcott been less

burdened is another piece in the great history game of ‘‘what if.’’

Of course, one can turn *‘what if’’ on its head. After shedding his responsibly

as president of the WAS, Walcott again became active. In 1911, he helped found

the Research Corporation with F. C. Cottrell (Cameron, 1952) and in 1916 the

National Research Council with George Ellery Hale. He founded the National

Advisory for Aeronautics in 1915, and he was President of the National Academy

of Sciences from 1916 until 1922. All this was done while he became the second-

most effective Secretary in the history of the Smithsonian Institution, and still

carried on a major field research program in western Canada. Who can say what

this paragon might have accomplished had he not been saddled with the WAS

and had turned the helm over to others in 1903. However one wants to play the

‘‘what if?’ game, Walcott continued as a member of the WAS until his death on

February 9, 1927.

Conclusion

Thursday, January 19th 1911, the 13th annual meeting of the WAS was held

in the Cosmos Club; at this point 13 societies were affiliated. Walcott jotted in

216 YOCHELSON

Barton W. Evermann, third editor of the Proceedings. From Division of Herpetology, California Academy of

Sciences.

his diary: ‘‘Meeting of the Washington Academy 4* p.m. This closes my eleventh

(11th) year as President of the Academy’’.*! Frank Wigglesworth Clarke, Chief

Chemist of the USGS, not Walcott, was elected president.

‘‘Especially felicitous remarks were made by Mr. Walcott, the retiring presi-

dent, and Prof. Clarke, the newly elected president. On motion of Dr. Kober, the

Academy extended to Mr. Walcott a hearty vote of thanks for his long and

successful services as president of the Academy. On motion of Mr. Walcott, a

WAS: BACKGROUND, ORIGIN, AND EARLY YEARS 217

similar vote of thanks was extended to all the officers who had contributed to

the successful work of the Academy. At 6:20 the Academy adjourned’’.””

It was the end of an era in several respects. Thereafter, presidents of WAS

changed each year. This also marked the stop of the Proceedings and the start

of the Journal of the Washington Academy of Sciences. Increasing administrative

efforts at the Smithsonian and a strenuous research program with months in the

field each year, limited Walcott’s participation in the affiliated societies. Ac-

cording to the WAS directory, in 1911, he still retained membership in the

Biological, Geological, National Geographic, and American Foresters societies.

From 1903 onward, Mrs. Walcott was listed as a member of the Archaeological

Society; she was killed in a train accident in 1911. In 1913, Walcott was listed

only as in the Archaeological and Geological societies; presumably he carried

on her membership.

By 1919, the directory no longer listed affiliations to other local societies. In

a sense this was an indication of the increasing specialization of science and the

increasingly harried pace of life, both consequences of that great watershed in

Washington, the first World War.

After Walcott’s death, at the 288th meeting of the Board of Managers, February

23, 1927, a committee of three was appointed to draft an appropriate resolution.

C. G. Abbot, Acting Secretary of the Smithsonian Institution, L. O. Howard, the

leading entomologist in the Department of Agriculture, and F. C. Cottrell, chemist

and long-time friend, wrote: “‘the Washington Academy of Sciences hereby

records its profound sense of loss occasioned by the death of its member and

former president, Charles Doolittle Walcott, and its sympathy for his family’’.*”

Acknowledgements

Without the Centennial celebration of the Washington Academy of Sciences,

there would be no reason for this paper. Dr. John Proctor kindly arranged for

me to participate in this event and made pertinent comments on several drafts.

Miss Diana M. Hawkes, Haselmere Educational Museum, Haselmere, England,

took time from her busy schedule to search for Archibald Geike letters and then

to transcribe five of them for my edification. The Smithsonian Institution Archives

continues its tradition as a fine place to conduct research; among other help he

provided, Mr. James Steed arranged for nearly half of the illustrations used herein.

Mr. J. McGregor of the U.S. Geological Survey Photographic Library, Denver,

was his usual helpful self in securing another photograph. Dr. A. Leviton, Califor-

nia Academy of Sciences, allowed copying of a picture of B. W. Evermann; Dr.

H. S. Yoder, Jr., of the Carnegie Institution of Washington, provided the photo-

218 YOCHELSON

graph of A. L. Day. The archives of Georgetown University seemingly are the

only repository in Washington of a picture of Frank Baker, and efforts to copy

it are greatly appreciated.

Notes

In a strict sense, the directories of the Joint Commission and the Washington Academy of Sciences are

publications, not unpublished sources, but they are ephemeral. The archives of the WAS are in the Smithsonian

Institution Archives under Record Unit 7099. Box 43 contains all ten of the Joint Commission directories, and

perhaps the most complete set of the WAS directories extant. Unless otherwise indicated all notes are to

unpublished documents in RU 7099.

'The ‘‘Directory of the Scientific Societies of Washington comprising the Anthropological, Biological,

Chemical, National Geographic and Philosophical Societies’’ is anonymous for purposes of cataloging, although

the secretary of the Joint Commission is included as one of the officers listed. One copy of the third directory

is inscribed ‘“With the compliments of James C. Pilling.’’ Pilling was the Chief Clerk of the U.S. Geological

Survey and also the Bureau of Ethnology and was—literally—Powell’s right hand man, and is a likely

candidate for the person who actually did the work of compiling lists and preparing copy for the printer.

Marcus Baker, the first secretary of the Joint Commission, was also an employee of the USGS.

The first directory was issued in February, 1889 (39 pages) and annually thereafter: February, 1890, (43 p.);

March 18, 1891 (57 p.); March 10, 1892 (58 p.) [This edition carries Entomological in the title; the Entomologi-

cal Society of Washington was founded February 28, 1884.]; February 10, 1893, 71 p. [This edition adds

Geological to the title; the Geological Society of Washington was organized February 25, 1893]; February 10,

1894, 67 p. [Membership continued to grow and the lesser number of pages reflects a smaller type face in the

directory]. February 25, 1895, 44 p. [Information on place of employment was dropped and a smaller type

size used]; February 21, 1896, 53 p.; February 15, 1897, 51 p.; and February 17, 1898. This was the 10th and

final year of publication by the Joint Commission and was a supplement to the 1897 directory of 23 pages. It

consisted of 19 pages of additions and corrections of addresses, and one page of those who apparently no

longer had membership in any of the societies. Record Unit 7009, Box 1. [The cooperative nature of this

directory is indicated in the minutes of the Geological Society of Washington. The first year the GSW was

listed it was accessed $20.70 for 8% of the cost; in 1897 the GSW was accessed $60.75 for 8.33% of the

costs and the council demanded an itemized statement].

* «Directory. . .’’ 1895, p. 9. Box 1.

3. «Directory. . .”’ 1896, p. 10—13. Box 1.

*Geike to Hubbard, 28 June, 1896. Geike letter book, December 1895—October, 1896, pp. 319-320.

Haselmere Educational Museum, Haselmere, England.

> The events are discussed in the ‘‘Geological Society of Washington minutes of the meetings of the council

1893—1900’’ catalogued as G (226) q W 24.98, in the rare book room of the USGS library, Reston, Virginia.

The first step of the drama is on p. 99.

° Geike to Emmons, 8 January, 1897. Geike letter book, October 1896—June, 1897, pp. 188—189. Haselmere

Educational Museum, Haselmere, England.

’ Geike to Hubbard, 18 March, 1897, pp. 375-376.

* Geike to Emmons, 18 March, 1897, pp. 377-378.

° Geike to Hague, 18 March, 1897, pp. 402—403.

'° Minutes of the Proceedings of the Joint Commission of scientific societies of Washington from 1897 to

—. April 6, 1897, p. 15-16. Box 1.

'' Geological Society . . . , May 26, 1897, p. 126.

'? Minutes of the Proceedings. . . , January 11, 1898, p. 21. Box 1.

'S Minutes of the Proceedings. . . , January 25, 1898, p. 41. Box 1.

'* Walcott’s daily diaries are in chronological order within RU 7004. This quote is from February 4, 1898,

Box 13.

'° Outgoing correspondence of the Secretary, January 10, 1907, p. 161. Box 8.

was the brilliant red paper covers.

'? The forms which were returned by members are in alphabetical order in Box 43, along with the directories

of the JC and WAS.

'® «Directory of the Washington Academy of Sciences and Affiliated Societies comprising the Anthropo-

logical, Biological, Chemical, Entomological, Geographic, Geological Historical, Medical, Philosophical’’

{Historical (The Columbia Historical Society, founded May 3, 1894, met at the Shoreham Hotel], and

Medical [The Medical Society of the District of Columbia was first chartered February 16, 1819, and

WAS: BACKGROUND, ORIGIN, AND EARLY YEARS 219

revived July 7, 1838; it met weekly] were added to the original title}, February 24, 1899, 67 p.; February

19, 1900, 63 p.; Thirteenth year of publication, March, 1901, 67 p.; Fourteenth year of publication, March

28, 1903, 68 p. {Archeological [The Archaeological Institute of America was founded in Boston in 1879;

the Washington Society or Chapter was founded April 7, 1902, and met irregularly] and Botanical [formed

November 28, 1901 out of fusion of two informal groups were added to the title. [For the Archeological

and National Geographic only local members were listed; the total of memberships for the year was 2,158.

There were still a large number of multiple memberships; Walcott is listed in the Anthropological, Biological,

Geographic, Geologic, Philos[phical. Mrs. Walcott is listed as a member of the Archeological]; Fifteenth

year of publication, February 28, 1905, 72 p. {Foresters [The Society of American Foresters was organized

November 30, 1900, and originally met at the home of Gifford Pinchot] [Archeological changed to Archaeo-

logical] was added to title}. [This is the first edition that did not include a summary of the society member-

ships]. Sixteenth year of publication, February 28, 1907, 87 p. {Engineers [The Washington Society of

Engineers was organized November 23, 1905, and met at National Geographic Society Hubbard Hall] was

added to title; Seventeenth year of Publication, March 11, 1909, 67 p. [In the 1909 edition Walcott was

listed as also being in the Society of American Foresters].

In addition to the above directories, Box 43 contains biannual directories for 1911—1927 [1921 is missing],

1930, 1932, 1935, 1937, 1939, and 1941, all as part of a continuously numbered series. The 1941 edition is

printed from a typescript and is the first not to include a calendar of meetings. A few directories were published

in the post-World War II years, but mainly the WAS published only a list of its members.

19 Some mechanical details of the Proceedings are summarized below. For volumes I and II, C. Hart Merriam,

Biological Survey, is listed as Editor; IIJ—V, Marcus Baker, then Assistant Secretary, Carnegie Institution of

Washington; VI- XIII, Barton W. Evermann, Bureau of Fisheries.

I. 1899 (April, 1899—February, 1900). Pages xii + 347 numbered pages, including 6 p. index. Eight articles,

26 pls., 11 figs. [Includes as one article “‘First Annual Report of Secretary and of Treasurer,’’ 1—14]. [Rules

for publication are given on xi-x1i, and irregularly repeated in other volumes].

II. 1900 (March—December). xiv + 694 p., including 17 p. index. 36 articles, 37 pls., 43 figs.

IJ. 1901 (March—December). ix + 612 p., including 6 p. index. 22 articles, plus one brief obituary, 65 pls.,

66 figs.

* TV. 1902 (March—October). vi + 573, including 12 p. index. 13 articles, 35, pls. 22 figs. [Includes as one

article ‘‘Organization and Membership of the Washington Academy of Sciences, 1902’’ which contains second—

fourth annual reports of Secretary and Treasurer].

V. 1903. (January, 1903—February, 1904). vii + 429 p., including 6 p. index. Five articles, plus notices of

deceased members, 22 pls.

VI. 1904. (October, 1904—February, 1905). viii + 481 p., including 9 p. index. Five articles, 34 pls. 81 figs.

[Includes as one article “‘Organization and Membership of Washington Academy of Sciences, 1904,’’ which

contains fifth-seventh annual reports, 439-472].

VII. 1905 (June, 1905—March, 1906). xii + 402 p., including 5 p. index. 12 articles, 14 pls., 7 figs. [Includes

eighth annual reports, ix—x1i].

VIII. 1906 (May, 1906—March, 1907). xiii + 491 p., including 4 p. index. 12 articles, 27 pls. [Includes

ninth annual reports, ix—xii].

IX. 1907 (July, 1907—February, 1908). vi + 558 p., including 9 p. index. Nine articles, 15 pls., [Includes

as one article ‘‘Organization and Membership of Washington Academy of Sciences, 1906, which includes

tenth annual reports, p. 523-—547.].

X. 1908 (July-December). x + 248 p., including 3 p. index. Three articles, plus short obituaries, p. 189-—

243, 2 pls. 3 figs.

XI. 1909. (March—December). ix + 299 p., including 4 p. index. Six articles, no illustrations.

XII. 1910. (January—December). xi + 330 p., including 2 p. index. Three articles, 1 pl., 3 figs.

XI. 1911. (February—November). vi + 101 p., including 3 p. index. Four articles, plus price list of 13 p.,

5 pls.

As indicated in the price list, volumes I, X, XI, XII (and presumably XIII) cost $3.00 paper bound and

$4.00 cloth bound; II-IX were priced at $4.50 and $5.00 respectively. One handwritten slip in Box 11, folder

2, summarizes the printing costs of most volumes: I - $1,162.19; II - $2,351.99; III - $1,655.72; IV - $1,707.82;

V - $1,729.53; VI - $1,430.46; VII - $1,333.01; VIII - $1,000.70. The original press run was 1,000 copies; in

volumes IV and VI it is given as 1,200. Some of the variation in price may be related to the number of

illustrations. The stock of Proceedings was stored in Evermann’s basement. There is an inference that many

copies were on hand; where they eventually went is another minor mystery.

° Outgoing correspondence of Secretary, January 4, 1901, p. 35. Box 8.

7! Outgoing . . . , January 12, 1901, p. 45, 46. Box 8.

*? Walcott diary . . . , December 13, 1901. RU 7004, Box 14.

3 Walcott diary . . . , January 18, 1899, RU 7004, Box 13.

4 Walcott diary . . . , March 14, 1899. RU 7004, Box 13.

*° Walcott diary . . . , January 17, 1901, RU 7004, Box 14.

220 YOCHELSON

*° Outgoing . . . , January 12, 1901, Box 8.

*7 Walcott diary. . . , May 18, 1900. RU 7004, Box 13.

8 Walcott diary. . . , March 11, 1901. RU 7004, Box 14.

? Outgoing . . . , May 11, 1901, Box 8.

°°? Walcott diary. . . , May 16, 1901, RU 7004, Box 14.

3! The resolution is quoted from pages 8—9 of an anonymous pamphlet ‘‘History of the organization of the

George Washington Memorial Association and the Washington Memorial Institution’’; a date of 1902 is written

on it. This document occurs in the GWMA file of the Walcott papers, RU 7004, Box 43, Folder 14. Presumably

this pamphlet was prepared for purposes of fund raising.

°? Walcott diary . . . , December 14, 1901. RU 7004, Box 14.

*° Three-page form letter reproduced by mimeograph or similar method. It is unknown how many were

distributed and to whom, but is has been found in the files of several NAS members; it may also have been

sent to prospective members. A copy is pasted in a WAS scrapbook, Box 30.

** Letterbook of Treasurer, C. D. Walcott and S. F. Emmons, to John Shaw Billings, March 22, 1901. National

Academy of Sciences Archives.

°° Letterbook . . . , to W. Sturgis Bigelow, April 22, 1901, p. 129.

*7Letterbook . . . , to Alexander Agassiz, April 1, 1901, p. 124-125.

*8 Outgoing . . . , January 21, 1901, p. 41. Box 8.

*° Outgoing . . . , May 11, 1901, Box 8.

*! Walcott diary. . . , January 19, 1911, RU 7004, Box 15.

*? Minutes of the Recording Secretary, January 19, 1911. Box 1.

*° Minutes of the Board of Managers, January 31, 1910—December 19, 1929. Box 2.

References

Anonymous. (1898). The Washington Academy of Sciences. Science, new series, 7:42.

Anonymous. (1898a). The Washington Academy of Sciences. Science, new series, 7:595.

Anonymous. (1900). [Title page, contents, etc]. Proceedings of the Washington Academy of Sciences, 2:i—

KK,

Anonymous. (1901). Washington Memorial Institute for Post-Graduate Study and Research in Washington.

Science, new series, 13:922—924.

Baker, F. (1902). Second annual report of the Secretary, 1900; third. . . , 1901; fourth. . . , 1902. Proceed-

ings of the Washington Academy of Sciences, 4:11—15.

Cameron, F. (1952). Cottrell: Samaritan of Science. Country Life Press, Garden City, New York. [Reprinted

1993, Research Corporation, Tucson, Arizona].

Flack, J. K. (1975). Desideratum in Washington: the intellectual community in the capital city 1870-1900.

Schenkman Publishing Company, Cambridge, Massachusetts, 192. p.

Gilbert, G. K. (1899). First Annual Report of Secretary. Proceedings of the Washington Academy of Sciences,

1:1-14.

McGee, W J (1901). The Washington Memorial Institute. Science, new series, 14:111.

Pettijohn, F. J. (1988). A century of geology 1885-1985 at The Johns Hopkins University. Gateway Press,

Inc, Baltimore, Maryland, 316 p.

Robertson, E. C. (Ed.). (1993). Centennial history of the Geological Society of Washington 1893-1993.

Published by the Society, Washington, DC.

Walcott, C. D. (1900). Lower Cambrian Terrane in the Atlantic Province. Proceedings of the Washington

Academy of Sciences, 1:301—339, pls. 22—26, figs. 9-11.

Walcott, C. D. (1901). Relations of the national government to higher education and research. Science, new

series, 13:1001-1015.

Walcott, C. D. (1903). John Wesley Powell. Jn Proceedings of a meeting commemorative of his distinguished

services. Proceedings of the Washington Academy of Sciences, 5:99—100.

Walcott, C. D. (1905). The Cambrian fauna of India. Proceedings of the Washington Academy of Sciences,

7:251-—256.

Yochelson, E. L. (1967). Charles Doolittle Walcott: March 31, 1850—February 9, 1927. National Academy of

Sciences, Biographical Memoirs, 39:471—540.

Yochelson, E. L. (1994). Andrew Carnegie and Charles Doolittle Walcott: the origin and early years of the

Carnegie Institution of Washington. In G. Good (Ed). The Earth, the Heavens, and the Carnegie Institution

of Washington. American Geophysical Union, History of Geophysics, 5:1—19.

Yochelson, E. L. (1998). Charles Doolittle Walcott: Collector of fossils, paleontologist, federal geologist and

administrator, being an account of a busy man from 1850 to 1907. Kent State University Press, Kent Ohio.

Journal of the Washington Academy of Sciences,

Volume 84, Number 4, Pages 221-233, December 1996

WASHINGTON ACADEMY

OF SCIENCES

Bylaws

(Approved by the Board of Managers

March 12, 1997 and the Membership

of the Academy by vote in 1997)

ARTICLE 1. OBJECTIVES

Section I. The objectives of the Washington Academy of Sciences (hereinafter

called the Academy) shall be: (a) to stimulate interest in the sciences, both pure

and applied; and (b) to promote their advancement and the development of

their philosophical aspects by the Academy membership and through cooperative

action by the Affiliated Societies.

Section 2. These objectives may be attained by, but are not limited to: (a) publish-

ing a periodical, occasional scientific monographs and such other books or pam-

phlets as may be deemed desirable; (b) conducting public lectures of broad scope

and interest in the fields of science; (c) sponsoring a Washington Junior Academy

of Sciences (WJAS); (d) promoting science education and a professional interest

in science among people of high school and college age; (e) accepting or making

grants of funds to aid special research projects; (f) convening symposia, both

formal and informal, on any aspects of science; (g) calling scientific conferences;

(h) organizing or assisting in scientific assemblies or bodies; (1) cooperating with

other academies and scientific organizations; (j) awarding prizes and citations for

special merit in science; (k) maintaining an office and staff to aid in carrying out

the objectives of the Academy.

Section 3. (Moved from previous Article XIII). The Washington Academy of

Sciences is organized exclusively for charitable, educational and scientific pur-

poses, including, for such purposes, the making of distributions to organizations

221

222 WASHINGTON ACADEMY OF SCIENCES

that qualify as exempt organizations under Section 501(c)(3) of the U.S. Internal

Revenue Code (or the corresponding provision of any future United States Internal

Revenue Law).

ARTICLE I. MEMBERSHIP

_ Section I. The Academy shall be comprised of individuals, Affiliated Societies

and Sustaining Associates. Throughout this document when reference is made to

individuals, the use of ‘‘he’’ shall be interpreted as “‘he or she.”’

Section 2. Members shall be individuals who are interested in and will support

the objectives of the Academy and who are otherwise acceptable to at least two-

thirds of the Committee on Membership. A letter or application form requesting

membership and signed by the applicant may suffice for action by Committee;

approval by the Committee constitutes election to membership.

Section 3. Fellows shall be individuals who by reason of original research or

other outstanding service to the sciences, mathematics, or engineering are deemed

worthy of the honor of election to academy fellowship.

Section 3(a). Nominations of Fellows shall be presented to the Committee on

Membership on a form approved by the Committee. The form shall be signed

by the sponsor (a Fellow who has knowledge of the nominee’s field), and shall

be endorsed by at least one other Fellow. An explanatory letter from the sponsor

and supporting material shall accompany the completed nomination form.

Section 3(b). Election to fellowship shall be by vote of the Board of Managers

upon recommendation of the Committee on Membership. Final action on nomina-

tions shall be deferred at least one week after presentation to the Board of

Managers, where two-thirds of the vote cast shall be necessary to elect.

Section 3(c). Each individual (not already a Fellow) who has been chosen to be

the recipient of an Academy Award for Scientific Achievement shall be consid-

ered nominated for immediate election of fellowship and will not be required to

pay annual dues for the current year.

Section 3(d). Any fellow of the Academy may recommend in writing that an

individual of national eminence be considered for immediate election to fellow-

ship by the Board of Managers, without the necessity of compliance with the

procedures of Sections 3(a) and 3(b) of this Article.

Section 4. Patrons. Members or Fellows who have given to the Academy not

ACADEMY BYLAWS 223

less than one thousand dollars, or its equivalent in property or tangible assets,

shall be eligible for election by the Board of Managers as Patrons of the Academy

(for life).

Section 5. Life Members or Life Fellows shall be those individuals who have

made a single payment in accordance with Article II. Section 11(a) in lieu of

annual dues.

Section 6. Members or Fellows in good standing who are retired and are no

longer engaged in regular gainful employment may be placed in emeritus status.

Individuals in emeritus status shall be designated an Emeritus Member or Emeri-

tus Fellow as appropriate. Upon request to the Vice President for Membership

Affairs for transfer to this status, they shall be relieved of further payment of

dues, beginning with the following January first; shall retain rights to hold office

and attend meetings, shall receive notices of meetings without charge; and at

their request, shall be entitled to receive the Academy periodical at cost.

Section 7. Members or Fellows living more than 50 miles from the White House,

Washington, DC shall be classed as Nonresident Members or Nonresident

Fellows.

Section 8. An election to any dues-paying class of membership shall be void if

the candidate does not within three months thereafter pay his dues or satisfactorily

explain his failure to do so.

Section 9. Former Members or Fellows who resigned in good standing may be

reinstated upon application to the Vice President for Membership Affairs for

approval by the Board of Managers. No reconsideration of the applicant’s qualifi-

cations need be made by the Membership Committee in these cases.

Section 10. Dues. Annual dues for each membership class shall be fixed by the

Board of Managers. No dues shall be paid by Emeritus Members, Emeritus

Fellow, Life Members, Life Fellows, Patrons, or Affiliated Societies.

Section 10(a). Members and Fellows in good standing may be relieved of further

payment of dues by making a single payment that has a value equal to ten years

of dues current at the time of payment. (see Article II, Section 5) Such persons

are to be identified as Life Members for Life Fellows as appropriate. Income

from this source shall be identified as the Life Membership Endowment Fund

(LMEP). All decisions regarding investments in the LMEF will be made by a

two-thirds vote of the Board of Managers, after Board Members have received

advanced notice of such action.

224 WASHINGTON ACADEMY OF SCIENCES

Section 10(b). Individuals whose dues are in arrears for one year (counting from

the “‘dues payable date’’ on the latest dues payment bill) shall neither be entitled

to receive Academy publications nor vote in Academy elections.

Section 10(c). Individuals whose dues are in arrears for twenty-four (24) months

(counting from the ‘‘dues payable date’’ on the latest dues payment bill) shall

be dropped from the rolls of the Academy, upon notice to the Board of Managers,

unless otherwise directed. Those who have been dropped from membership for

nonpayment of dues may be reinstated upon approval of the Board of Managers

and upon payment of back dues for two years together with dues for the year of

reinstatement.

Section II. Affiliated Societies. Bona fide scientific societies may apply for affil-

iation with the Academy by furnishing to the Secretary of the Academy an outline

of their objectives, organizational structure and requirements for membership in

their society. The Secretary will transmit the application to the Policy and Plan-

ning Committee for review and recommendation for action by the Board of

Managers.

Section I1(a). Each Affiliated Society shall select one of its members, who is

also a member or fellow of the Academy, to serve as its delegate to the Board

of Managers. The delegate shall serve until replaced by his society.

Section 11(b). Each Affiliated Society shall cooperate with the Academy in

sponsoring joint meetings of general scientific interest.

Section 12. Sustaining Associates. Any association, corporation, firm, institution

or subdivision thereof, which has an interest in promoting the objectives of the

Academy may be invited by the President of the Academy, with the approval of

the Board of Managers, to become a Sustaining Associate for the purpose of

supporting the Academy and its programs. The names of the Sustaining Associates

shall be listed annually in the Journal of the Washington Academy of Sciences.

Section 12(a). Each Sustaining Associate shall designate a person to serve as

liaison to the Washington Academy of Sciences. This individual will receive the

Journal of the Washington Academy of Sciences and mailings regarding upcom-

ing technical meetings. The position shall be non-voting unless the liaison is

concurrently an individual Member or Fellow of the Academy.

ARTICLE Il. ELECTED OFFICERS AND BOARD MEMBERS

Section 1. Officers of the Washington Academy of Sciences shall be President,

President-Elect, Vice President for Administrative Affairs, Vice President for

ACADEMY BYLAWS 225

Membership Affairs, Vice President for Affiliate Affairs, Vice President for

WJAS Affairs, Secretary, and Treasurer. All shall be chosen from resident fellows

of the Academy.

Section 2. The President shall appoint all committees and such nonelective offi-

cers as needed unless otherwise directed by the Board of Managers or provided

in the bylaws. He (or his substitute; the President-Elect, the Vice President for

Administrative Affairs, the Vice President for Membership Affairs, the Vice

President for Affiliate Affairs, the Vice President for WJAS Affairs, the Secretary,

or the Treasurer, in that order) shall preside at all meetings of the Academy, the

Board of Managers and the Executive Committee.

Section 3. The President-Elect shall succeed to the office of President following

one term as President-Elect. The President-Elect shall serve as Chair of the

Program Planning Committee to arrange speakers and meeting places for the

following year (the year in which the President-Elect succeeds to President).

Section 4. The Vice President for Membership Affairs shall have general respon-

sibilities for committees related to membership; the Membership Committee, the

Membership Promotion Committee, and the Committee on Awards for Scientific

Achievement.

Section 5. The Vice President for Administrative Affairs shall have general re-

sponsibility for operation of the Business Office of the Academy and the Journal

of the Washington Academy of Sciences. He shall oversee the activities of the

Editorial Advisory Committee, the Home Page Committee, and the Office Man-

ager. In the absence of the Archivist the Vice President for Administrative Affairs

shall assure that the historical records of the Academy are maintained permanently

at a secure and accessible location.

Section 6. The Vice President for Affiliate Affairs shall serve as Chair of the

Affiliates Society Representatives. He shall maintain a current list of Affiliated

Societies, their presidents and representatives to WAS. He shall maintain liaison

with the delegates of the Affiliates and keep them informed of WAS meetings

and events.

Section 7. The Vice President for WJAS Affairs shall have general responsibility

for the committees relating to organizing and maintaining the Junior Academy

(WJAS). He shall have responsibility for the Committee for Encouragement of

Science Talent and the Committee on Grants-In-Aid for Scientific Research. He

shall interface with the Joint Board on Science and Engineering Education.

Section 8. The Secretary shall act as secretary to the Board of Managers and to

226 WASHINGTON ACADEMY OF SCIENCES

the Academy as a whole. He shall record and distribute the minutes of the

meetings of the Board of Managers and such other meetings as the Board of

Managers may direct. He shall conduct all correspondence relating thereto except

as otherwise provided and shall be the custodian of the Corporate Seal of the

Academy. He shall be responsible for keeping the working records of the Acad-

emy current.

Section 9. The Treasurer, in cooperation with the Vice Presidents for the func-

tional areas described in Sections 4, 5, 6, and 7, above, shall be responsible for

preparing the Budget of the Academy and submitting it to the Board of Managers

for approval. The Treasurer shall also be responsible for distributing to the Board

of Managers in a timely manner records of funds received and expended. The

Treasurer shall be responsible for maintaining records of funds deposited in banks

or other savings instruments. The Treasurer and/or other designated persons shall

sign checks for disbursements of funds as directed by the Board of Managers.

The Treasurer shall prepare annual reports as required by the Internal Revenue

Service, the U.S. Postal Service, etc. He also shall deposit and disburse funds of

the Washington Junior Academy of Sciences.

Section 10. The President and the Treasurer, as directed by the Board of Manag-

ers, Shall jointly sign securities belonging to the Academy and endorse financial

and legal papers necessary for the uses of the Academy, except those relating to

current expenditures authorized by the Board of Managers. In case of disability

or absence of the President or Treasurer, the Board of Managers may designate

the President-Elect or another elected officer as Acting President and/or another

elected officer of the Academy as Acting Treasurer, who shall perform the duties

of these offices during such disability or absence.

Section 11. When for approved Academy obligations, circumstances necessitate

payment by persons other than the Academy officers who sign checks, reimburse-

ment to such persons shall be made only when appropriate documentation is

submitted to the Treasurer of the Academy.

Section 12. Two Members or Fellows of the Academy shall be elected each year

to serve a three-year term as Members of the Board of Managers. To initiate

staggered terms or to fill vacancies, additional Members of the Board of Managers

may be selected in the annual election.

Section 13. The newly elected officers and Members of the Board of Managers

shall take office at the close of the annual meeting, when the President-Elect of

the previous year becomes President. |

ACADEMY BYLAWS 227

ARTICLE IV. APPOINTED OFFICERS

Section I. An Office Manager may be appointed by the Board of Managers. The

Office Manager shall be responsible for the routine business operation of the

Academy. The Board of Managers shall determine the type of business activity

(volunteer workers or contract workers) and the amount of funds to be allocated

to the business office.

Section 2. An Editor for the Journal of the Washington Academy of Sciences

shall be appointed by the Board of Managers. He shall receive advice from the

Editorial Advisory Committee. The Editor shall be responsible to the Vice Presi-

dent for Administrative Affairs for administrative policy and related activities.

Section 3. An Archivist may be appointed by the President. If appointed he shall

maintain the permanent records of the Academy, including important records

which are no longer in current use by the Secretary, Treasurer or other officer,

and such other documents and material as the Board of Managers may direct.

The Archivist shall assure that historical records are maintained permanently at

a secure and accessible location.

ARTICLE V. BOARD OF MANAGERS

Section I. The activities of the Academy shall be guided by the Board of Manag-

ers, consisting of the President, President-Elect, immediate Past President, four

Vice Presidents, Secretary, Treasurer, six elected members of the Board of Man-

agers, and one delegate nominated by each of the Affiliated Societies. The Editor

of the Journal of the Washington Academy of Sciences and the Office Manager

shall be members ex officio of the Board of Managers.

Section 2. The Board of Managers shall set the dues for individual members

and the minimum contribution for Life Members, Life Fellows, Patrons and

Sustaining Associates. For prolonged, diligent and well-documented service in

the administrative work of the Academy the Board of Managers may recognize

such service of a Member or Fellow by citation including dues paid for life.

Section 3. The Board of Managers shall transact all business of the Academy

not otherwise provided for in these Bylaws. A quorum of the Board shall be one

fourth of the membership of the Board of Managers. To be eligible to vote, the

officer or member of the Board of Managers must be in good standing, casting

one vote only regardless of the number of offices or Affiliated Societies that he

may represent.

228 WASHINGTON ACADEMY OF SCIENCES

Section 4. The Board of Managers may provide for such standing and special

committees as it deems necessary.

Section 5. The Board of Managers shall have power to fill all vacancies in its

elected membership until the next general election. This does not apply to the

offices of the President and Treasurer or to delegates of Affiliated Societies.

ARTICLE VI. COMMITTEES

Section I. An Executive Committee shall have cognizance of Academy finances

by reviewing the Treasurer’s monthly reports of budgeted expenses and antici-

pated income, and by reviewing the status of several internal accounts; the Life

Membership Endowment Fund, the I.R.S. Form 990 accounts, the U.S. Postal

Accounts, the WJAS Account, etc.

Section 2. The Executive Committee shall meet at the call of the President. It

shall conduct all day-to-day business not requiring Board approval, and it shall

prepare issues and positions in advance of Board of Managers meetings.

Section 3. The Executive Committee shall consist of the incumbent elected

officers of the Board of Managers plus two non-elected members appointed by

the President.

Section 4. Committees under the cognizance of the President are the Executive

Committee, the Nominating Committee, the Policy and Planning Committee,

and the Audit Committee.

Section 5. Committees under the cognizance of the President-Elect are the Pro-

eram Planning Committee.

Section 6. Committees under the cognizance of the Vice President for Administrative

Affairs are the Editorial Advisory Committee, and the Home Page Committee.

Section 7. Committees under the cognizance of the Vice President for WJAS

Affairs are the Committee on the Encouragement of Science Talent, Commit-

tee on Grants-in-Aid for Scientific Research.

Section 8. Committees under the cognizance of the Vice President for Member-

ship Affairs are the Membership Committee, the Membership Promotion

Committee, the Committee on Awards for Scientific Achievement.

Section 9. The President shall appoint from the Academy membership such ad

hoc committees as are authorized by the Board of Managers and such special

ACADEMY BYLAWS 229

committees are necessary to carry out its functions. Committee appointments

shall be staggered as to term whenever it is determined by the Board of Managers

to be in the interest of continuity of committee operations.

Section 10. The President, with the approval of the Board of Managers, shall

appoint a Nominating Committee of six fellows of the Academy, (see Article

VI. Section 4) at least one of whom shall be a Past-President of the Academy,

and at least three of whom shall have served as representatives of Affiliated

Societies for at least one year. The Nominating Committee shall be appointed

no later than the November meeting of the Board of Managers (or Novem-

ler, 1 5:):

Section 11. The President shall appoint a Committee of Tellers, of three Mem-

bers or Fellows no later than the December meeting of the Board of Managers

(or December 15).

Section 12. The Nominating Committee shall prepare a slate listing one or

more persons for each of the offices of President-Elect, the four Vice Presidents,

Secretary, Treasurer, and four or more persons for the two Members of the Board

of Managers whose terms expire after three years at at least two persons for each

vacant unexpired term of such position (see Article III, Section 11). The slate

shall be presented for approval at the meeting in December. Not later than

December 15, the Vice President for Administrative Affairs shall forward to each

Academy member and fellow an announcement of the election, the Committee’s

nominations for the offices to be filled, including biographies of nominees, and

a list of incumbents. Additional candidates for such offices may be proposed by

any Member or Fellow in good standing by letter received by the Vice President

for Administrative Affairs not later than January 3. The letter shall include the

concurrence of each nominee and the names of at least 15 members or fellows

making the proposal. Upon verification by the nominating committee the names

shall be entered on the ballot. The Vice President of Administrative Affairs shall

remind Members and Fellows of the foregoing option with the distribution of

the preliminary slate.

Section 13. Not later than February 15, the Vice President for Administrative

Affairs shall prepare and mail ballots to members and fellows. Independent nomi-

nations shall be included on the ballot, and the names of the nominees shall be

arranged in alphabetical order. When more than two candidates are nominated for

the same office, the voting shall be by preferential ballot in a manner prescribed

by the Board of Managers. The ballot shall contain a notice to the effect that

votes not received by the Vice President for Administrative Affairs before the

230 WASHINGTON ACADEMY OF SCIENCES

first Thursday of March, and votes of individuals whose dues are in arrears for

one year or more, will not be counted. The Committee of Tellers shall count the

votes and report the results at the April Meeting of the Board of Managers.

Section 14. The President shall, in advance of the Annual Meeting, appoint an

Auditing Committee consisting of three persons, none of whom are a current

officer, to audit the accounts of the Academy.

ARTICLE VII. MEETINGS OF THE ACADEMY

Section 1, The annual meeting of the Academy shall be held in the Washington,

D.C. area each year in May. It shall be held on the third Thursday of the month

unless otherwise directed by the Board of Managers. At this meeting, the reports

of the President the Secretary, the Treasurer, and the Audit Committee shall be

presented.

Section 2. Regular meetings of the Board of Managers shall be set preferably

for a fixed place, hour, day of week, and sequence of months excepting July and

August. Other meetings may be held at such time and place as the Board of

Managers may determine.

Section 3. Special Meetings of the Board of Managers shall be held as called

by the President, or in his absence by the President-Elect, or within ten days

after a written request by six members of the Board of Managers has been

received by the Secretary, he is required to convene a special meeting of the

Board of Managers to address the specific issues for which the meeting was

requested.

Section 4. The rules contained in ‘‘Robert’s Rules of Order Revised’’ shall

govern the Academy in all cases to which they are applicable, and in which they

are not inconsistent with these Bylaws or special rules of order of the Academy.

ARTICLE VII. REMOVAL FROM OFFICE

Section I. Members of the Board of Managers and the Executive Committee

shall assure that all business of the academy is conducted in the highest spirit of

ethics and integrity.

Section 2. If any member of the Board of Managers or the Executive Committee

is found by a vote of two-thirds of the Board of Managers to have violated the

ACADEMY BYLAWS 231

spirit of ethics and integrity or the conflict of interest requirements, he or she

shall be removed from office.

Section 3. The position vacated by such removal shall be filled temporarily by

appointment by the Board of Managers until the next general scheduled election

or regulation appointment to the affected position.

ARTICLE IX. COOPERATION

Section I. The term ‘‘Affiliated Societies’’ shall be held to cover the current

Affiliated Societies and such others as may hereafter apply for affiliation, are

recommended by the Policy Planning Committee, approved by the Board of

_ Managers and elected by two-thirds of the members of the Academy voting, the

vote being taken by correspondence. A society may be released from affiliation

on approval by the Board of Managers, upon a written request to the President

or the Vice President for Affiliate Affairs.

Section 2. The Academy may assist the Affiliated Societies in any matter of

common interest, as in joint meetings, or in the publication of a joint directory;

provided it shall not have power to incur for or in the name of one or more of

these societies any expense or liability not previously authorized by said society

and societies, nor shall it, without action of the Board of Managers, be responsible

for any expenses incurred by one or more of the Affiliated Societies.

Section 3. No Affiliated Society shall be committed by the Academy to any

action in conflict with the charter, constitution, or bylaws, of said society, or its

parent society.

Section 4. The Academy may establish and assist a Washington Junior Academy

of Sciences for the encouragement of interest in science among students in the

Washington area of pre-college and college age.

ARTICLE X. AWARDS AND GRANTS-IN-AID

Section I. The Academy may award medals and prizes or otherwise express its

recognition and commendation of scientific work of high merit and distinction

in the Washington area. Such recognition shall be given only on approval by the

Board of Managers of a recommendation by the Committee on Awards for

Scientific Achievement.

Section 2. The Academy may receive or make grants to aid scientific research

in the Washington area. Grants shall be received or made only on approval by

232 WASHINGTON ACADEMY OF SCIENCES

the Board of Managers of a recommendation by the Committee on Grants-in-

Aid for Scientific Research.

ARTICLE XI. AMENDMENTS

Section I. Amendments to these bylaws shall be proposed by the Board of

Managers and submitted to the members of the Academy in the form of a mail

ballot accompanied by a statement of the reason for the proposed amendment.

A two-thirds majority of those members voting is required for adoption. At least

two weeks shall be allowed for the ballots to be returned to the Secretary.

Section 2. Any Affiliated Society for any group of ten or more members may

propose to the Board of Managers an amendment to the Bylaws in writing. The

action of the Board of Managers in accepting or rejecting this proposal to amend

the bylaws shall be by a vote on roll call, and the complete roll call shall be

entered in the minutes of the meeting.

ARTICLE XII. DISTRIBUTION OF FUNDS ON DISSOLUTION

In the event of a liquidation, dissolution, or termination of the Washington Acad-

emy of Sciences (whether voluntary, involuntary, or by operation of law), the

total assets of the Washington Academy of Sciences shall be distributed by the

Board of Managers, provided that none of the property or assets of the Washington

Academy of Sciences shall be made available in any way to any individual,

corporation or other organization, except to one or more corporations, or. other

organization which qualify as exempt from federal income tax under Section

501(c)(3) of the Internal Revenue Code of 1954, as may be from time to time

amended.

ARTICLE XII. CONTROL OF FUNDS, ACTIVITIES

No part of the net earnings of the Washington Academy of Sciences shall inure

to the benefit of, or be distributable to its members, trustees, officers, or other

private persons, except that the Washington Academy of Sciences shall be author-

ized and empowered to pay reasonable compensation for services rendered, and

to make payments and distributions in furtherance of the purposes set forth in

Article XII hereof. No substantial part of the activities of the Washington Acad-

emy of Sciences shall involve the carrying on of propaganda, or otherwise at-

ACADEMY BYLAWS 233

tempting to influence legislation. The Washington Academy of Sciences shall

not participate in, or intervene in (including the publishing or distribution of

statements) any political campaign on behalf of any candidate for public office.

Notwithstanding any other provision of these Articles, the Washington Academy

of Sciences shall not carry on any other activities not permitted to be carried on

(a) an association excempt from Federal income tax under section 501(c)(3) of

the Internal Revenue code of 1954 (or the corresponding provision of any future

United States Internal Revenue Law) or (b) by an association, contributions to

which are deductible under Section 170(c)(2) of the U.S. Internal Revenue Code

of 1954 (or the corresponding provision of any future United States Internal

Revenue Law).

Journal of the Washington Academy of Sciences,

Volume 84, Number 4, Page 234, December 1996

Special offer on Biography of

Charles Doolittle Walcott,

President of the Washington Academy

of Sciences, 1899-1910

A biography of Charles Doolittle Walcott, Paleontologist, by Ellis L. Yochelson,

will be published in June 1998 by the Kent State University Press. Charles

Doolittle Walcott (1850-1927) is one of the most important, but little known

figures in American geology. This in-depth biography documents his career and

life from birth to retirement from the U.S. Geological Survey in 1907, when he

became Secretary of the Smithsonian Institution. Throughout much of his adult

life he was a federal scientist, yet his efforts as a consummate administrator of

government scientists and engineers were even more significant.

Regularly priced at $49.00. Special Prepublication offer: $40.00, shipping and

handling inclusive.

Send check or money order in the amount of $40.00 (Ohio residents must add

6.25% sales tax) to: The Kent State University Press, P.O. Box 5190, Kent, OH

44242-0001. Include book title and address to which it should be sent. For credit

card orders, call 330-672-7913. Offer expires 6-30-98.

234

DELEGATES TO THE WASHINGTON ACADEMY OF SCIENCES,

REPRESENTING THE LOCAL AFFILIATED SOCIETIES

EepBUNIS HIER SEHCICEN ION FNIMERICA) occu) Selon k Fic bedi wie cldle dee Slee sesanw saa tees Tim Margulies

Peewean Associaion Of Physics Teachers... 2... 1... i. ce ec cee eee ee eee Frank R. Haig

ES BE, QUSISETS UG S30) STLEt) ep NE 8 ar e Laurie George

Ree REE AM ISHERICS SOGICLY 55445 <iq Fare snes oie Men ed ie gens ba es eemmee sles Ramona Schreiber

American Institute of Aeronautics and Astronautics ..................... Reginald C. Smith

American Institute of Mining, Metallurgical, and Petroleum Engineers ....... Michael Greeley

a tee TCLE OTC CICAL SOCICUY | )./2.-.5 y./205 2 ee hv icatie Sly wie oe esd bees ee wee A. James Wagner

ee mE PCR ECTS OC ICEY 0G 2s pees ope alae Od 6 boa a ieee Sere es bee a eee es kee ae Paul E. Thiess

pee Ee nytopatholosical Society .. 2.22... ie. ce ee nce eee ness Kenneth L. Deahl

Pemeamesocicry OF Microbiology 2.5... 2... ce he sve a ee eae en be wees awes VACANT

eemmeomctety Of Civil PNOMEEIsS .... 2.2.2.2. 2. ea elke ee ee kee sees John N. Hummel

Eememean saciety Of Mechanical Engineers ..................02 02 c eu eee Daniel J. Vavrick

Peaaemesecicny OF Plant Psychologists... 2.2... 0... 60.2 ee eda eee eee eees VACANT

me PMcrseal Society Of Washington .......:.........06..02s cece ecees Marilyn London

ERE LTRS C02 RM ia Ca ee Toni Marechaux

Pemetmnniat Computine Machinery .2.. ..... 2020)... 5. ec ee de cee cesses Lee Ohringer

Association for Science, Technology, and Innovation .....................00000: Isaac Welt

PeeeeeeEeIcty Cl WaSHINSION .. 2 <6. 8 ee eke we heed cence eens nees Janet W. Reid

eS EIE TE OE WASPHIMOTON 200 22 oe swig Sse eas Haid we we seb a cee be one Ray Peterson

ame mcicry Of Washincton |... 2. 28 se ne ee eee ces Elise Ann B. Brown

Beenie arumnia institute OF Chemists: 1... ee ee ee cee cee VACANT

eee Columbia Psychology Association ...............5.00s0sce cece es David Williams

PEL RCSONICE: SUEVEYS PEAMN 6 = 602500. 2 2). cee le bas So oe les ee es VACANT

Pa peOPiCal SOCICLY .... 26. 2. cee ee eee eee eee leew eee Ronald W. Mandersheid

Renn ARIES MIELE SN eo ee eee tcc 2 ihe Seas kia iein’ Sain cid dae sda eile e bac VACANT

Pmommerical society of Washington ............2.......2..06.05. F. Christian Thompson

Dee S@GIcly OF WASHINGION . 2 652. isn sin ee ee ee tee ea ca neees Bob Schneider

ffeemanenoeical Society Of WaShingion ......... 2... 2. ese e cee cece wees VACANT

ome mesecicty Of Washington, DC... 2... 2... ee cc ce eee ee Phillip Ogilivie

Bema Factors and Frogonomics Society ..................... 0.5 0n eee Thomas B. Malone

femme of electrical and Electronics Engineers ......................-- Rex C. Klopfenstein

MEME TPCCRNOIOPISES 26 oe kc ck a ene ede dae eect cawlensie sce Isabel Walls

MME IOMMPHOMSIFIAl PMOIMCEDS < . oo... oc 68 oe co ee ee ee beeen Neal Schmeidler

_ 02 PLEEEL) Dd BVS F000 Ya rg les (oe: OU ee ee eee John I. Peterson

International/American Association of Dental Research ................... J. Terrell Hoffeld

eae asieal Association Of America’ ..... . 2.6... eee ee eee Sharon K. Hauge

Paemeal society of the District of Columbia ......... 2.202.620 2. ee ck eects Duane Taylor

meee AAU ANSITONOMEIS: cs 6 iis oe doin sce hn Seb bcd eee se elec e ec ee ae Harold Williams

MMR PGE UPEANNICSSOCIELY! 6 2.5025 4c) ies is LAG sisi ole Bb sinning Belda ae Hoan Sie eee VACANT

SE Seeeeee SUCICLY Ol VV ASMIISIOI) 25 5 oc ig eos cick sein b26 ds eis nies wie ee #4) ela apdeerelens Si VACANT

EMME AL SOCICLY OF WaASHMISION 325 6 54's2c. 2 acc lee cles sleeves seen bale es oe James Goff

epee RMR GCHET Al SYSECIMS IRCSCAFGH) 5 225 6010) a sla ta livid Sid cain alae tie eee diene ke ss VACANT

ne EeeN Ie SNICHE ANCE ORESIERS | 02 en Pu sl wk Ole Dawe sek pas 2 Belgie Michelle Harvey

Society of American Military Engineers ........... Set id NOt oe OR EARS geo VACANT

Society of Experimental Biology and Medicine (SEBM) .................... C. R. Creveling

eee Gr MantiiaChirine ENGINEENS: .. cic. la. Sctec wee soe ceed dete cle cee eweens VACANT

ie eremetienyy MrATISICh: SOCIELY. Goi. 2 5 6 Se isivicia ae soos semien de nldine ale sae Sees Clifford Lanham

Pasminewoa tiStory OF science Club) ¢......6... 6.5. cc a seae sadness eee ee Albert G. Gluckman

Washington Operations Research/Management Science Council ............... John G. Honig

StmMeton Edit Mechaical GIOUD \..)2).00 5-256 sd dea ek nae ceca we wed Robert Kogler

1 SUTEE, ORS UTES Tc) 0" oT 2) | 0 ae Michael P. Cohen

Delegates continue to represent their societies until new appointments are made.

Washington Academy of Sciences 2nd Class Postage Paid

Room 811 at Washington, DC

1200 New York Ave. N.W. and additional mailing offices.

Washington, DC 20005

Return Postage Guaranteed

a»

" +

3 228

N INSTITUT

STENT IANA Ey AEN

jt ye

cunt ener

at sees n aes

2 Bend A

Dn aha

yaNte sys

ae TS,

peate

A ens R eyelet at kage gg ago aye

sete

SEU Nate Xk en eaiteate tegen en eng

Far oN t

PANGS SRE BRE AE SAT Uo ie Oho A

LEN ai eet cage eran : ae

wins me aber Ratna ee a

LVR aie ke fe are

wet

Senn pnreett

er areas

Seren tote net

Meme ed eager tgieauiy®

UNAM egeaNem ede

one

ApAgtin eras

ae Huse

pope aaa gaat’ 3 So

FORE Ri omens yng Cher

rere ny

cEN Ay ty tnd

SA alta bp

wee

Joye

Ye TUE ae TH

Saran

aitin ruses

3 5 ekeK, oh

* abt Pn Oe an aha eae at ieee

rms mae ee EN Les oth ant HOTERAS and

Tenn te

’ AUR Ane san Ee BENS PACAg cava ger dnstee

Howraytatie gee eg nig

her!

OPN We

OM BOUES SH CTs TeEL AS

WEEN Oo AP IER TA A AEE

Eee MA yeh it SEER Sk OAT WE hey,

AS Ae ten rar ad

Ah Rite Bis, PCN oe Set hn Ret

PATS Md

eM POL D

eT a!

Serer ert ery

PR op Fe ihe

PERS AY tery

LINES ETFs

Spt

PD PA es

PAO ERE NE pha te Soe A

COU Kye ay 0A

BAe bel he Ua a ont Peale yey Ce ete e

THR Sh wR ON EE BE RE Be nat : . Pe aoe coitens

: Sri BREM ee eyes, my ne tf

WATE ok Me con ME ILLS ke

eden ede oe Wa er fea

ADE aR tasiolgns

ONENESS ©

se grassed

Se G Me

Viera,

SOT Ora ae pa

SA HRA Ta 8)

Wey i

ADE RUM Re

NOD Ory

oar ea oes

Peer yon

Pa ee?

EDS

Se ae

VON aes eee

DAMP Reyer A

Wee feta ; 4 x CPOE PT SOA UR sep

Yup ase

PRP oe

POU pie Ry aesa gee Ue GILG

mene pS

see

eM Aipe ea MEP Ay Ay ake

eh en A

Apter see pais 4

WS eat)

evi oly ke og

enh MeN ee, Mate

Mr ye ke at

BRAN EN af ey AOINAN-8

eee ere meee th Th

NP tee (Wena ty Gage a

Fre nig

We ewes Zim

ASN Oe

WERE STNG LAO ATE NEES lee tr

ete Be NN

syee

Wha ptee sy

west

WP eas tet

PUT yPuk yas tal ae

er We ete er ag

Pyeny ery set

ete

Wesiny hayes?

Men Rare SAMUS WESLEY ye mga ee nt te SATA Sh

erie eee SEK SAS ae WAU Sh ye

ate FAlen S Ber ria | SERLALPaME DS:

- AW RENAL ONT ALE BOSSA UT RIEU TASES S Fit Wear ce he or vs

Sree Bites teen gee Es PEER ara se

sheamnnalbnes fala geen weeeteacien SATA AS oA SE matty’

LOSS CAMO D ATR eg rf gene Wt Leal eee ta gn

HUA UN na ENN PNT a Bie ae ge IY EEN NIC Tye getinuee t= toa ke py AAS Le wrt

fea Mg eras PP PROT ORC ULE UMN Gar &

Uy orey

CaaS

ahs ene

Naan yey

DEAT ERG ne ty

ONS EMRE ATE

TRAE

miei pas

: SAE SOMERS UPSET WD

. My VP ae i L

ih DWI ps Lay

: My AERO US UAE

CEN 4

VPA ty

Lisa een

eMpoe tne hye

Fee gt ™

cody TL ONAN RUN

SKIS aA E CLEA CS URIS wg iis

8 ot ing aus MNT AL eR Poo o

A lee Se

wes Aa ly

ROA fev pM

tate VLE Ly

Thy

ane era

’

NOE EAD atta

On a

«i