RATIONALE

The

Rorschach, like other scientific techniques, has certain limita- tions

which

need to be recognized before it is used.

For

one thing, the Rorschach does not profess to give reliable information about eventsinthelifeof theindividual

which

havecontributedtotheforma- tion of his personality.

In

this sense, therefore, it does not provide data

which

are entirely comparable to the kind sought in psycho- analysis (although a skilled interpreter,

working

with a rich record, can often arrive at an understanding of

some

symbols

and

traumas significant in a subject'spersonality).

The

Rorschach reveals some- thingof the structure of

an

individual'spersonality; it indicates cer- tain patterns of psychic

economy

_; proportionately

how much

energy

isdevoted torespondingtostimuli

from

the social environment,

how

much

to fantasy, etc. It does not claim orneed to elucidate content primarily. It does not tell the interpreter directly

how

the subject feels about his mother, for instance, orhis father, although a clever reader

may make shrewd

inferences.

Thus

the Rorschach provides data

on

onlyselected areas of the pei*sonality; orrather, it views the personality ina certainway,

and

gives information

which

isrelevant only to that point of view. It gives an

"X-ray"

of the personality ratherthana"complete" anatomicalanalysis.

56 BUREAU

OF

AMERICAN ETHNOLOGY

[Bull. 150 It

may

be felt that in trying to define

modal

personality structure one cannotget

from

the analysis of

Rorschach

materialanything more, or perhaps not as

much,

as can be inferred

from

a

knowledge

of the culture itself.

In

a sense this is true.

A

^standard

^approach

^{to the}

determination of the personalitycharacteristics of groupsis todefine the culture,

and

then deduce,

from

a

knowledge

of

some

system of psychology,

what

sort of personalitiespeople

must have who

behave in the

way

prescribed as

normal by

theculture. But, aside

from

the advantage of quickly getting insights

from

the

Rorschach which

are difficult to secure in

any

other way,

and

the advantages of using a relatively standardized

form

of observation

which makes Rorschach

results directly comparable cross-culturally, one

major

reason for using

Rorschach

(and other projective techniques) is that it avoids the circular

argument

of describing culture first in culturological terms,

and

then redescribingit inpsychological terms.

The

psycho- logical redescription

may

indeed be entirely adequate,

and

^fit in per- fectlywithother sources of information,suchas projectivetests.

But

these culturological analyses are^difficulttouse in

any

butapurelyde- scriptive way.

One

cannot reallytest hypotheses aboutthe relation- ships

and

interactions of personality

and

culture

by

usingthe

same

data forthe description ofbothculture

and

pei-sonality. Ifonetries

it,oneinevitably

comes

outwitha zero statement.

A

^refinedversion of the cultural-deductive

method was

thatadopted

by Cora DuBois

in

The

Peopleof

Alor

(DuBois, 1944).

In

this

book

there appeared,in addition to a careful description of the culture, a psychoanalysis of the cultural

forms and

of certain biographical

ma-

terials,

by Abram

Kardiner; and, as asort of control data, a "blind"

analysis ofa small

sample

ofdrawings, the Porteus

Maze

test,

word

associations,

and

37

Rorschach

records.

The Rorschach

data espe- cially provided a convenient control for the culturological analytic inferences

and

deductions,

and

the substantial agreementofbothsets of conclusions

was

mutually validating. ⁱ

There

^is,however,

an

even betteruse for projective techniques than simply toconfirm opinionsreached

by

other methods.

One

can ask,

and

^{answer, questions}

from

a ^set of projective data,

which

can never be answered

from

a simple generalization about national character, derived

from knowledge

of culture pattern.

Such

questions are:

"What

isstatistically

modal

(most frequent) ^?" "Is thestatistically

modal

personality structure

what we might

predict

from an

analysis ofthe culture?"

"Who

are the psychologically ^typical,

and

psycho- logically deviant

members

of the

community?" "What

are their social roles?"

"What

are the kinds of deviation?"

"What

is the incidence of the

modal and

the other various personality types?"

"How

variable are the personalities of a

community

of people with oneculture?"

Wallace]

MODAL PERSONALITY OF TUSCARORA

INDIANS

57 As was

mentioned earlier,

few

anthropologists have published a systematic description

and

analysis of their Rorschach material.

Four major

exceptions to this are Hallowell (1942, 1945 a) _; Oberholzer (1944) ; Billig, Gillin,

and Davidson (1947^8)

^;

and Honigmann

(1949).

Hallowell,Oberholzer,

and Honigmann employ

a substantially sim- ilar approach. This approach consists of assembling

and

scoring the series of protocols,

and

calculating

mean

(average) scores for the several factors (determinants, locations, content categories,

and

ratios).

These mean

scoresarethenassembledtopresent

what might

betermeda

"mean

profile"or a"profileofmeans." This

mean

profile isinterpretedasifitwerethe Rorschachof asingleindividual.

The

psychological structure deduced

by

the interpretation is ascribed to all

"normal" members

of the group.

In

addition to this basic pro- cedure, the authors enter into refinements

and

nuances, particularly Oberholzer,

who

calculates standard deviations,

and

investigates va- rious permutations

and

combinations in his material

by

ad hoc pro- cedures designedto answer very specific problems. Oberholzer, fur- thermore, uses a sample of Swiss

normal

records as a contrasting

group upon whom

toprojectthe Aloreseprofile.

Billig, Gillin,

and

Davidson'smonogi-aphincludes a large quantity of ethnological material; Billigisresponsible for theRorschachinter- pretation. Billigusesa

somewhat

different

manner

ofapproach

from

Hallowell,Oberholzer,

and Honigmann.

Inadditiontogiving

group

averages for

some

factors, he calculates the percentage of cases in the sample

which

fall within certainlimits (in oneto threeattributes simultaneously) with supposed diagnostic significance,

and

presents thedata intabular form.

Although

he

makes no

use ofmeasures of correlation

and

contingency, or ofstatisticalsignificance,heneverthe- less observes frequencies of association

among two

or three factors.

These

diagnostic categories areconceivedas directlyrepresenting per- sonalitytraits,

and

statements are

made

aboutthe relativeincidence of varioustraits:such

and

suchapercentage of thepopulation areintro- versive, such

and

such a percentage are rigid; a certain percentage of the rigid ones are also introversive,

and

^{so on.} In

summary,

he pullstogether thetraitswithhighest incidence

and

^fits

them

together intoa synthetic structuralpicture. This approach,liketheonebefore

it, isnot convincing in itstreatment ofstructure, because thereis

no way

of proving that the traits associated in the

group

picture are really associated in

many

individual records. Billig's

method

does, however, havethegreatadvantageofallowing

him

to discussthe

wide

variability of the population in regard to the various traits.

At

the beginning of the study, the writer expected to handle his data (see table 2)

somewhat

after the

manner

of Hallowell, Ober-

906418—51 5

58 BUREAU OF AMERICAN ETHNOLOGY

[Bull. 150

OOOOOOOOOOOOOOOOOOOOOOOOOO OOOOCOOOO—iOOOOOOOOO

Me«.-io -<>-<«.-(Oil.-Hr-

ON oo^M^

^1-1o?-ic»oc»o o«c<ocie<e<5c<oN-«<e<nc^r-c<cec.-<w ON^^ooc»oc»»Hi-ioooo^r-i.-ioocioo>-<i-iNn^ciicfronncMDC^c^coo^-i^o^

©0-<000000.-iO--(.-iOC»00

00 00 00000

OOOC.-iOt-<0'-i'HO>0>«<OCJ'*00

OOOOOOOOCOOOOt-iOOOOOOOO^^OOC OOOOOOOOOOOOOOOfMOOO

t-iO>-l—iOOOOi-i>-lC<'-<Oi-iM«OCT-iC<lC»5i-<i-it-iM^ »-iOCeC"-<C»P»00^>-l>ONlO'*^T}'»0«0 cot^^c<seo>oe»>oe>5»-<«^'»|^"<»«io>o«0'*>'5t--»"^'*ect>. •^e^tr^-vuix-Vt-i-v^c^o^m'V^a: at~

OOOOONOO^OOOOOOOOOO.->000000

ONOOOOOesCO<OOr-lC10^000 OOOOOOOt-lO>-<OOOOOOOOi-iOOOOOOO

000 00

000000«l«Oe>Jt-lO'-i OOOOOOf-fOfHt-lf-fOOOO^^OOT-fOOOOOOOOOOCOOO»-<Oi-h^hO^OOOOOO ,-ioo>-iooo>-ioe<iooooi-io.-<ONOr-ioooi-io ooi-cooccoi-icjestoNifi-iroNMoo

,-lf-4C^iO*-«^CC'*C^^»-t*^'^»-^*-t»^C^OCO»Of-<r-tC4.-l.-c^CO*0<C>0^^'*"f-<C^«^>Ot^C<IOOOI--t^OOO

OOf-iOOOOOOi-x-cOOOOOOOOOOOOOOO0<-<OOOOOOr4MOOi-cr«i-icl>00

00

oooP>oooo-<c<.-io-4>-ioooooNe<soe*ooO'-i®i-i©P»e<iocDeoo3-»»iNotoNO>ooi-ir-i

OOOOOOOOOi-iOOi^OOOOOOOf-HOOOOOOCO*-«OOOOCOO»-<^OCCCO•-4^000

t—

I

rH f-* *-HCi »—IT-l CS rH rjCC CS uiNI—(c^ cc

oooosaocDOaoooit^c&QOO^'^OdoaoiOOOoac^'^c^ooac^co oa)koocDccrHoor^oor«coiOf-H(okcco<oo^

wtf-tOi C*w^i-iWCOi-H-^ -^ CO CO^-Vb- CC t^ »0 CO'•t*to^ w C<»C0C0i-ii-<OCNOONWNi-Ht-tOi-tOW^^

Ot*<0OOt*^t--O*-<C0OOOOC0OcC'C0Q0t»OC0'^Oe^ QO'««<COOO«D"ff"*l^O*Or».»OCO'^»-iOiOCO otDco»oOi-Ht^i-^0'^t^QOOQcoO'^coh^«PQeo'^»ci-^ OM'od«^»coc(OOecQo6i^oo(5oc^ooco C0'«*'^C<iC-^'*'^C0-^C^^'<J<'*'^CQ*O>O(NC^^'^CC'*CS'^ CCCO'*C5C^C<CO-^'*J'SOO*OU5C^CQ'V;5^^^

sssssssssss:sssssf-<t^fefefefep^f^f^i^s;ss(^f^f^^s;ssssfefefef^sss

t^C>f-»<-<C^t*OC^01CSC^000001'<*'COOOi-'COCO

5S

i-H.-iC*C^C4C<COCOeC'<*''*'^*OiO;D«D⁰⁰i-TOCOop opOQOOi00 CC QO CO r* CO C5 CO OiOOep eO 00 Oi Ol 00 05 »c .-<cicO'*»o<Ot*oocbO»-'C^ec'**ncDr*coa)Oi-ieswciNMWw^ciwCO CO cccoececcccccoco-w^'^'"^'^

Wallach]

modal personality OF TUSCARORA

INDIANS

59

OOOOOOOOOi-lOOOr-i-iO

OOOOOOOO-i

i-ioo?<oooo^O'-imO'*(No '-oo'*«-<oooio ooo--iooo>rac^OrtOcit~o^ ciooo-nrH—io-h

«-<<Moooooo'4<ooooooo"OooiMtoi-cmoo

OOOOOOOOOO

OOOOPJO OOOOOOOi-H^

eo^^oo-<«"-io^c«5c»o«i'*>-i r-e«rt«ONOtO'*«

<t~>«<a>t-(«ii>,-HtDr~ooc<ooo>co»'-i'-o^e^ortoousor-

CO-hm»-hoc^ O00>C«»-c ^ l-H -WiM--<-<« ooooooooMOi-iooe^oo oioo-^mocs^o OOOCJOOO.-COOOOOOOO—lO—fCMOOOO

OOOOOOOOOOOOOt^

—O000-<-H000-^

OOOOOOO—ii-iONOOCiOO >00-HOC<50rHO>0

^o^{*H1-^}o^{c*i *-<}o Mwt^o^«-i»o »o ecCD1-1loMo^{c^}r>.ko CO

i-H .-H -*^T-^

*-iO O^t-<O O^»-(CO CO»-tCO^O0»»-^C*C^ lO »0 t^OCO CO »o ^^

OOOi-iOO.-lt~OOOOOM^O

OOOOONO—

lO Mf-lOOOO'^N'-lCOOCOOCOCOO 00 •H CO 0=«Tf Tfr~-

M

o o oCOo «^1-1*^«<1-co o o^•>»•eoo^ooo^coi-<o ^coi-i-h

^ ^1-1 eoi-i

CO -^ c^ 00 r^ CO r^ CO^{•CO r^ CO}-"^^t^1-tCD *o^CqCOl-lCOi-< i-H coci t^ cs c^ *h oi e^ -^i-H-ICONrHNrH to "5 -^ CO^»C -^*-<O^"5enO^COO00 t^»o o> CO ^h t- t* »CC*<co i-iOONoooc^c^i-<i-icoi-iojco"\ cjoot^.—(-J^-Ao^- 1^o o^{1^}oo"^ CO CO i^MMooi«?i^^ lo COt"o mCO<y1^

-^O^t*cooOS COOOt^COOCOQOeO0«CO CO^1-11^ CO CO CO^ ^

cioco-<t*OQdeococ4i-ioc^c4c*icoi-H cococood-^*—hccJooq t>-COCOo⁰⁰Kr* CO CO1-tCO f*MCO CO »o coi*'co-^^»o eo 25 f5 lO CO CO COO^{CD1-1t^ CO 00'«(;}MCD CSOW5 ^H CO 1-tO^{CO CO}—^t* *0 CD CO COO^{CD1-1t^ CO 00'«f'}

M

»-(i-( ^^H C^1-1t^ Tf^ N-< -H CO -^i-<(>»^CM CO t^CO -^

1-1 CO 5§3

§SS§SSS^P=*f^f^^pHfepMfeg;gix,fep^fefe^CH CO00 CD 0> -^ -^Ot^ Ol r^ C^ OS 0>O^{f-5t^00"tf}CD <N -^ "2 »0^U3 COTj<»0 CO 00 00 t^i-it-tCS CCCCC«3'f3t—^" 00•'^CD C^^""^ ^»0^'^U3 CO.^ CO«-*C^ C^W-^ CO U3 CDb* 00 a»o«-ics

^-^^^lO ic »o

ss

»0 CO b- GO 0>O^{»-< WCr?}-^ "^ COt-»00 OSO

IC iO lO lO U) CO CO CD CO to CO CO CO CO CO toS:

as

.ag

60 BUREAU

OF

AMERICAN ETHNOLOGY

[BULIy.150 holzer,

and Honigmann,

calculating a profile of

means and

in- terpreting

from

that the

modal

personality structure. Fortunately, however, his dissertation adviser, Dr. A. I.Hallowell, suggested that he apply for advice about the statistical problems involved. Dr.

Malcolm

G.Preston, of the

Department

ofPsychology,University of Pennsylvania,

was

asked

what

he thought of the proposed procedure for defining the

"group

personality." Dr. Preston

was most

discour- aging, atfirst.

He

considered that calculating

mean

scores,

and

then interpreting the

mean

profile as ^if it characterized everyone in the group,

was

psychologically meaningless. ^{It also}

committed

the fal- lacy of operationally

assuming

that scores

have

standard

meanings

regardless ofcontext.

He

pointedout furthermore, thatsincethefre-

quency

distributionsof various scores in

any

one^category

were

almost

all heavily skewed, the

mean was

a very poor

measure

of central tendency. (See fig. 3, p. 60, for the frequency distribution of

M.)

About

the

same

time. Dr. ^Julius Wishner, also at

Penn

in the psy-

18- 16-

%

^14-

^

lO

I

^-

^-

O

^I

23 4S67Q9tOllt2