' 940

I? JUDGMENTS OF THE JUSTICE OF

‘ PUNISHMENT BY ADOLESCENT DELINQUENTS
\
I.

Thesis for the Degree of M. A.

MICHIGAN STATE UNIVERSITY

JEANNETTE MARIE HAVILAND
1968

 

- A ---_~_-f;'. L-W’W~-9J

“ALIAHAI- - ~ -,
Michig;‘..1 State
University "

     

 

 

IIIIIIIIIIIII

m IIIIIIIIIIIIIIIIIIIIIIIIIIIII III III: IIIIIIIIIIIIII ...

3 1293 10646 6349

 

 

 

 

 

 

 

’1: 4—1-1 ‘

‘V'FYTVV‘U‘AT‘I
m-d—c’~A.é-~J-

T“T"‘""'p r1“ CT“ yr'Y‘p T ‘T'T'f‘fi (\‘

L'vyas...&L~J 1‘ .l-Q J MJLJ.L.L‘J._'J p‘
T‘,""‘T’"'T‘ 1' \V'fl ‘_‘17 II HAT r‘v‘n '1" 1 '11- T-74\7*T1' '71:“.
- . . r.
.L ‘uJ.IA.~) .-_~J.I.-'L U‘ ‘XAI'LLJAIJ 4.4-; 4. “4“-o‘4x. H‘I-‘J

1 T'

. 'r .- ‘ l. U, —' .\
oJ uoauLette H. “dVL13hd

.L

«"y o , ‘0 H v r" c .
iois S;Uflj inteati

ates the relevance of Jean Piaget 5

FJ

theory of the (19.9 03:;an of rtlral Juicgrmnt to the probloz: of

1 .0. . ..‘ , -. ° ,3, , A
too dbSLiCd of punlegmunt.

@
C +-
’1
(r;
f f-
to
b'
$11
( I‘
*4 I
O
”‘3
O
F
'3
O
h)
3
I-—’

: ,1.
g v . -

u.11.,“
L 4 .1- . -,-.‘ , ,—

UV 3 to’ts i? T813LJQ to a1013399rt

‘h '.-_ ., .‘. L, - A .'L a”: v a -
QullLCUBHCJ. lug >“LCleuhofl #3 were KMQU\H dealinqu: nLo in a

*‘S
g.)
f)
1..
5
L;
J‘
O
O
PLJ
("f’
O
’)
S
r+'
"S
O
H
u)

‘s were matched Vntz; the :xpori-
mental g; on intellirenoo and géncral background, but wcre not

m . 1 0' ‘ I a“! F! ' . r! "r: . "Ir
kxmo n d91inqu9nts. gals anotnebla N38 taqtco in two “95-3:

" n . A r 1 ‘\ VI\ ‘V v- . ‘ V J 9 - . (-g .-‘ “
toe lITSt quSdlC has tbs intelV I in an LR g_uudooo a thazac-

' .4» . .M ..,. . 1. rt- . _, ... . n”. x, o.
ter in a 590?] .uLo tranbbrosses; tue QCCOLd mcabuxe win L90 4 3

use of punishment and reward in training rats.
The first hypothesis that adolescents who are delinquent

wouli give imr‘ ature moral rospons:s was not significantlw

1

l

suppol‘tcd by the interViow, but wa" S‘gnificantlv supported by
the results of the behavioral method. While training a rat which

0 _ _ ' . .
had L99n pretraire l to make a wron ' regionse, the delinquent

{‘0

83 used sienificancly more punistumenc and si
reward than 01d one control g5.

ns were in one moral udg-

U}

0

zince all the interview quo ti

"I
u

C.)

ment cateeory, the secondb Jpotne sis was that interview response
would not di ffer sigxtifu Gently among tie interview questions.
However, this hypothesis had not previously been tested by
S"stenaticall y alter'ngo nly one variable in each interview
sdfiry. The variable in th's studv was the relationship between
the characters in the interWVi Istory. This Piagetian hypo-
thesis wzs not supported. 319 results su*‘9s ed that moral

judrment of punishment differs witn respect to the re lat 10 rs ip

between people. The §s in the delinquer t and match -ed non-

P‘

icrntly more mature punishment
responses when the interaction was reported to be between adoles-
cent peers than when the interaction was reported to be between
an adolescent and his mother or an adolescent and a teacher

Tnis result doc s not surport Piag et' 3 h5pothecis that moral
judgment of punishment is cognitively 'global', but ratner

would support social learning theorists who maintain tnat moral
judgment is situation dependent.

The correlations between int ervie w responses and punis

(D

ment esponses in th rat training situation also tend to support

the social learniz ' hypotheses. Tor the control is on the adoles-

1C)

cry, punishment re sporses were si

and pOS'tively correlated with interv1ew responses. for the

k»)

’—

JA 2,, L .9. ,._,. ”1,. L __ .4 ”3,11“, .4» ,. A
delimiuens s on L.-.':;’ adolostent pee .3 cry, puiiis_--ne?--s resporsas

were ‘llflCahoIJ and n9 53 tivle correlated with the

Po

nterview
responses.

The results, while not disputin' tLo the oev9lorn91 of

1

. - _ -. ° : .2, .—‘. r. -
moral 3uc5ment 15 related to co5n1tive oz? Jelopment, sagbest that
puni ishmcnt behavior for adolescents 13 Situation oriented,

rathe “than part of a global cognitive code. Beliefs about
”er 1 suit 1 Suit Junislimnnt, and animal punis nucdr are not
9 uri crt, a1

hi5hly related.

.\
p /7 1/
I' - Q I
Approved /)7 Zfi? ‘£/7{:L:>/
/7 I" ,7 r—

Drte g2z¢7/ 4?// ,Kkgijp/

 

'V‘"T‘?l*“0 t“ "ti“ -vurnv*n:1 "V
JIKJWJ'L-Jl-.~L) 'L1 1.1.1.5 (J UL) ' .LKIA‘J \JI

rV-rTf\T_-V?'I'\"IT‘ 1'er n fl'f'm'flhfj‘c,’
PUL..L).L.L'.;.31.¢ 4.1.1. AJCL' b.) Ln

Jeannette Karie Uaviland

A mw"0 :

L..._L"

,qigan State University
in partial fulfillment of the requirowo

vuvnts

for the de~“mc of

wnfimT“ Ch ‘fiMH
«L

l x»; “41» 111.14

Department of Psychology

,“ (“7“: 'f‘\'_'_f'f lrr‘f'?‘ r611!“ 'fif‘t
A'LViLL.‘\-' {LA—iuULIA 0.21 . - s.)

‘l
l

nelps= in tle prd-

If):
()

”fpreoiation is owed to many paople who

V

paration of this thesis. Special thanxs is riven to pg ccwnit-

yr 7r 1‘

1- 5 A. 3 L .: .,. w . ‘ ._, _.',~ ,.-,; . t ,
tee, Dr. Joan P; Lbnlnny, Csaighan, Er. Laarsn Ldlzls, and or.

Dozier Thornton, *nose patience and direction made the prepara-

1

U 0
tion a fCflaTth? trsk.

I would als

the staff and boys of the Loys

O
i. .l
:X“
\ )
r?“
O
(4-
v v‘
i
;.-l
:3“

u.-

an, and of the Korthsrn high

‘ - 1| '
I

r L . ‘ 7‘ ‘. . I —=
905001, LaHSinb, slog;

roit, Kichigan, frr their coopsratitn and assistance.

H.
H-

Introduction.

i?‘3;wlll" (-
.Lk../g) UJOII.

h:,. ”v'

References.

_L..lo

or

P.

P.

f‘. :.

*Jk-Ltl..._-J.I

'rfiI--‘\ . .
n

l .

k.

LIST CI“ TAB L3

The t‘ e of resoonse to uestion one in the
a r
interVicw.

Fcan transformed sc
story on a cale of zero

¢

Percentage of punlsning, rewarding, and
attractin» in rat training procedure
Correlations between interview data and

' 1 -‘ ‘m‘n
punisnmcnt responses

Introduction

Theories of moral and character development are an ancient
tOpic of study, traditionally more of interest to the philoso-
pher than to the psychologist. Only in the last few decades
have psychologists begun empirical studies in this area. In
so doing they have found that 'morality' and 'character' cover
many facets of psychological development, not all of which are
congruent. The topic of this study is the develOpment of
attitudes toward and beliefs about the justness of punishment.

Many contemporary ideas about moral judgment find their
origins in writings ranging from the Nichomachean Ethics to
Augustine, from Dewey to Freud, and, with particular respect
to theories of the development of morality, from Piaget.

This discussion will be limited first to a general review of
the literature on moral judgment and then to the narrower

tOpic of the development of beliefs and practices about punish-
ment.

In the United States in the early part of this century
interest in children's moral behavior was confined largely to
the practical aspects of training to conform to society's
notion of 'goodness'. Dewey, a member of the pragmatist
tradition which greatly influenced the historical development of
behaviorism, clearly summarizes and heralds the thought in this

area. He explores the belief that moral ideas are learned

Ux

through acting 'morally'. He means by this, acting according
to society's prescription in social siuations, in the family,
at school, and so on. Dewey appeals to situational variables
rather than to general ideas about morali y; he suggests

that social reinforcement and task accomplishment are he tter

J.

reinforcers than punishment, restrictions, future goals, or

n- 10 -~

personal recognition. Dewey also foreshaidws Ilaélg a ivf

1.9

stvmt th: importance of judgment in the learning of moral
behavior.
In this tradition Hartshorne an Pia (1928-19°O) defined
J
character as social habits of ccnformdty, honesty, sclf--con-
trol, and service and tested these habits with te mp ation
D 1‘

situations. Their study found little“ relationship between

day School, Boy Scouts or character -- education classes. In
general hartshorne and Kay conclude that honesty is dependen
upon situatior a1 factors such as the Opportunity to be dis-
honest and remain undetected and the value of the possible
reward for undetected dishonesty.

Several writers (e.g., Kohlberg, 1963) have interpreted
the reported failure of training for honesty to mean that
learning theory concepts were not applicable. However, ano+ rer

interpretation would be that learning ato1t honesty did not

 

generalize to behaving honestly. To test ‘tr e effectiveness

of learning theory with reapect to honesty one should reinforce

0\

subjects for resisting temptation. Then after training to resist
temptation a post--test in other situations would represent a
more adequate test of learning theory rinciples.

Further studies (Grinder, 1962; Sears, Alpert a Rau, 1965)
on the teaching of moral behaviors have shown relatively little
relationship between training and children's behavior. Harts-
horne and Kay found that punishment inhibited or elicited
behavior in a particular situation when the punishment was
immanent and realistic, but little generalization was found to
other situations in which punishment was not immanent. Host
reviewers agree with Hartshorne and hay that these results are
evidence against the existence of a generalized conscience even
though there is no evidence about the effect of positive rein-
forcement on the development of conscience.

Psychologists such as Piaget, 1932, and Kolhberg, 1963,
would interpret results of such studies as Hartshmne and
Kay's differently. They would maintain that general moral
schemas do exist but that moral behavior and judgments are
difficult to change with reinforcement techniques.

Psychoanalytic interpretations of the develOpment of
moral judgment center about the concept of the superego.

Freud argued that the mechanism of identification with care~
takers accounted for the energizing of the superego system.
The young child is completely dependent upon his caretakers.

Because of this, he identifies with them; that is, he matches

‘J

his behavior with the sanctions and prohibitions laid down by
them.

In Freudian theory the superego is usually in direct oppo-
sition to the id, since society's moral code, which the child
adopts, is directed towards inhibiting the id's primitive drives.
According to the Freudian point of view main determinants of
moral behavior are and can only be established in early child-
hood when the important figures in the child's life are all
powerful. Rewards and punishments applied later in life are
not believed to be strong enough to deve10p moral behavior.

Piagetian and psychoanalytic views of moral development
touch on a few issues, but for the larger part are parallel.

For example, both theories agree that the autocratic, all-seeing,
punishing caretaker contributes to the development of certain
immature types of moral behavior. Piaget asserts that most
children pass through a stage in which their cognitive develOp-
ment is dominated by views of parental authority and its essential
validity, even its divinity. He does not hypothesize that this
is the basis of the conscience, bvt rather that this aspect of
moral develOpment changes and matures, just as beliefs about
physical phenomena, such as the movement of clouds, change and
mature. Piaget seemingly ignores consideration of emotional
content in develOpmentally earlier beliefs. Theoretically

this emotional content may not change but may lie dormant, since

it is neither replaced nor punished.

Further differences in Piagetian and psychoanalytic inter-
pretations lie in their assertions about the origins of moral
behavior and judgments. Freud believed that the origins lay in
the resolutions of childhood crises such as toilet tanning, sex
training and identification. Piaget maintains that the dominant
origins are multiple and interacting but include such factors as
parent authoritarianism, cooperation among peers and the changing
patterns in cognitive develOpment.

In summary, psychoanalytic interpretations of moral develop-
ment involved a different aspect of development from that
pursued by Piagetians, the former being more oriented toward the
emotional aspects and the latter entirely cognitive. This is not
to say that a rapprochment in unnecessary or impossihle. In
fact, since many studies of antecedents of moral development have
not upheld Piagetian predictions, we may need to combine the
theories in this area.

Social learning theory has combined to some extenLthe con-
tributions of behavioral and psychoanalytic interpretations. In
the area of moral judgment, however, social learning theory
interpretations have lead to new interpretations and the posing
of new problems.

An extensive study by Sears, Rau, and Alpert (1965) illustrates

both the relation of social learning theory to psychoanalytic

theory and its unique controbutions. Sears et al. attempted

to intercorrelate measures of dependency and identification. The

data revealed a lack of intercorrelation, thus inconsistant with
psychoanalytic predictions. It is therefore clear that the
relationship between identification and dependency must be
reformulated.

Another study (Bandura & Kupers, 1963) shows how social
.learning concepts ofhmitation are related to development of
the ego ideal. The seven-year-old subjects learned to reward
their own behavior through observing a model who rewarded
himself. A model with 'high standards' influenced the child
to have high standards, while a model exhibiting 'low standards'
influenced the child to have low standards. Peers and adults
were equally effective as models.

Further studies by social learning theorists have inves-
tigated the effects of punishment and the timing of punishment
upon resistance to temptation (Parke & waiters, 1967: Aron-
freed, 1961,1963). These will be discussed with other studies
directly related to punishment behavior.

The studies cited above generally tend to uphold the
social learning hypotheses that the Freudian concepts of con-
science, dependency, and identification are very global and
that elements of any of these may be independently acquired
through direct reinforcement,punishment, or imitiation.

Piaget's work in the field of moral judgment is a small
interlocking segment of his life's work in cognitive develOp-

ment. According to Piaget the development of the two moralities --

10

the morality of constraint (the ethics of authority) and the
morality of cooperation (the ethics of mutual respect) --
is congruent with areas of judgment.

Piaget's theoretical position (195M) is called "equili-
bation theory". Equilibfation means that logical cognitive
structure develOps as a function of an internal process
dependent upon activity and experience. The repetition of
acts is not assumed to act through external reinforcemnt
but by a process of mutual influence of the child's activities
upon each other. Contradictions in mental structure bring
about reorganization of concepts and relations (Smedslund, 1961).
Moreover, it is only in the area of moral judgment that
Piaget and his co-workers have specified which of the child's
activities might have dominant influence. In doing this they
obviously leave themselves Open to proof of error, but they also

give other experimenters an opportunity to start the process

U)

of eliminating and building on a strong theoretical basi .
Piaget studies moral judgment through watching and partici-

pating in children's marble games and through interviewing

with stories. In the former case he studies children's notions

of rules and their origins and use; in the latter he notes

children's reactions to stories which involved a moral judgment.

Piaget recognizes the technical drawbacks in his methods; his

results may reflect his own interpretaions and could be influ-

enced by the experimenter's personality and training. Never-

11

theless, Piaget's methods have been fruitful for him and other

C)
(1‘-

In his 1)32 study on lower-c1 ss Swis chwil ren Pia“

5

identified severalc dif mr ent M nds of moral 31d gment ThCS;

included a) immanent justice -- each act has reacting con-

sequences upon the actor, b) moral realism -- an act is judged
according to its external effecm. , c) beliefs about collective
and individual responsibility for punish;cln acts, and e)
beliefs about types and values of punisim ent. It is priz narily
the last are which is of concern in the present study.
Piaget's study of the jufi es of puris h,3nt revealed
striking developmerital Chan es: younger children think that
punishment is just and necessary; the sterner it is, the more
just. Younger children also be ieve that punishment is effec-
tive in the sense that the child who has been dulv chastised
will in t‘n future obey better than one not severely chas~
tized. Older children tend to suggcet punishments that put
things rigat and restore tie status quo; pinishment as such
is regarded as useless, reproach and exT lanation being deems
more profitable than chastisement" (p. 201). However, as is
true of most cognitive structures, immature forms m.y and often
do survive at all ages and may be found among adults as well as
children. Piaget hypothesizes that a change in attitude

toward punishment occurs as the child becomes less egocentric

and recognizes the reciproca effects of causing sufferin

'3'
1
Q.

*4.
{‘0

The child comes to realize that wrong actions are

(7)
C
O
’53
(‘9'-
’1
9.:
<3.
d.
0

rules of cooperation and break the bond of solidarity that
exists within a group, especially a peer group.

Certain social and philosophic questions are raised by
this theoretical conception. Should one accept the results
related to justice of punishnent as having implications for
moral behavior, conscience development, guilt and other non-
behaviorally defined concepts related to good and evil?

9 'I

furthermore, what 18 the meaning of tne terms "mature"

a nd
"immature" as they have been applied to moral behavior?
Piaget, drawing principally from the philOSOphies of
Durkheim, Eovet, and J. K. Baldwin, clearly intends to Speak
of the justice of punishment in terms of good and evil and to
imply that the mature judgment is qualitatively better than
the immature judgment. To Piaget expiatory punishment is a
primitive social method of restoring social order by objectively
hurting anything or anyone associated with the disorder or
crime. Theoretically, this r sponse substitutes for the
impossible task of attacking the desorder itself. Punish-
ment is considered a primitive and immature way of responsding
because it lays the burden of restoration of order not on the
.ause of the disorder, but on an individual or group regardless
of its ability to be responsible. This objective cxpiation

would not lead to development of subjective or autonomous

responsibility or authority, but rather to heteronomous respon-

sibility. For example, such an expectiion could lead to the
concept that the evil lies in being 'caught' or being associated
with crime rather than with one's own volition in committing a
criminal act.

Durkheim discusses comparatively the use of expiatory punish-
ment in primitive and more modern democratic societies and
Piaget draws from this material to make analogies to the child's
deveIOpment of concepts concerned with the justice of punishment.
In both the case of primitive societies and in the child's case
Piaget notes the relative importance of omnipotent and omniscient
authorities. For the child these authorities are the parents.
Because of the authoritative relationship between the parent
and the child, Piaget believes that the child adOpts the philoso-
phy of moral realism or non-autonomous authority in his moral
judgment. In one sense, then, the child is simply expressing a
belief in the justice of his own situation, that is in the
relationship between himself and his parents. He expresses the
belief that punishment that emmanates from an external source
is just punishment, and that such punishment prevents further
unwanted (by the authority) behaviors.

The mature alternative to expiatory punishment is censure
or explanation or both. This is a reasonable, though still
idealistic, alternative only in the social situation in which
all the interacting members are able and willing to c00perate.

According to Piaget, COOperation implies the ability to take

another person's point of View. If it is true that each mem-
ber of the society takes autonomous responsibility for his own
behavior, only then do conscience and the ability to censure

or correct the self develop. Although Piaget does no himself
make the point, the implication is that all members of the
society have an implicit contract to cooperate. When the
contract is broken, censure alone restores the bond of coopera-
tion because the individual will correct his behavior when

he is able to observe or is instructed of its consequences on
others and on the contract.

To Piaget, censure and explanation rather than expiation
is 'mature' rather than merely idealistic because he assumes
that the child in transition becomes able to understand and
reason through the concepts of equality and cooperation. This
cognitive restructuring is aided, according to Piaget, by
peer c00peration through which the concepts are presented at
first concretely in the child's experience, although at a
later stage he assumes that the concept would generalize and
be formulated rationally.

A number of studies have tested Piaget's assertions that
immanent justice, collective responsibility, moral realism,
and punishment and other such instances of moral judgment do
exist and that these areas are related as Piaget hypothesized
they would be. Other studies have extended Piaget's findings

to children of different social classes and nationalities.

Well before Piaget, Barnes (1902) presented comparative
data on 8 to 15 year old school children in the United States
and Great Britain. This extensive study is remarkably similar
in method to Piaget's studies and the results may be inter-
preted in exactly the same way. Barnes found that the recom-
mendation of corporal punishment for "a young child who mis-
takenly painted some chairs” decreased from about #0; at age
8 to less than 10% at age 15, while he recommendation of
explanation for the same story increased from about zero at
age 8 to almost 40% by age 15. At all ages girls gave more
mature (explanation) responses than boys, although the signi-
ficance of this difference, of course, was not reported (there
were no tests).

NacRae's (1954) extensive study of antecedents of moral
judgment updates and in part replicates an earlier study by
Lerner (1937). Lerner was primarily interested in replicating
Piaget's results in the United States and in general succeeded.
NacRae tested 2&4 boys ages 9 to 14 by questionaire and inter-
view. He found no significant correlation between maturity of
moral judgment and his measures of extent of parental authority,
interaction with parents, extent of parental discipline and
control, attachment to the more or the less strict parent, or
the extent of internalization of parental prescription. The
boys" perceptions of their own participation and role in peer

groups did not show a significant relation to moral judgment

either. Thus Rachae's findings do not support Piaget's hypo-
theses about possible antecedents to mature and immature moral
judgments.

The factor of peer influence on values and attitudes remains
comparatively unexplred in the psychological literature. The
Sherif, Harvey, White, Hood, an: Sherif (1961) study stands out
as one of the only studies in this field and does support the
hypothesis that peer interaction contributes to the establish-
ment of self-concept, role, and status. A study by Wilson (1959)
lends further indirect support to the Piagetian hypothesis.
Jilson studies high school students" educational, occupational,
and political values. Statistically partialling out such paren-
tal influences as occupation and education made it apparent that
individual values were influenced by balues held by the
majority of the students in the high school. Although these
studies indirectly support Piaget's hypothesis, in contradic-
tion to XacRae's data, which do not support it, the evidence
for either case is meager, and no conclusions should be drawn.

NacRae also correlated responses in several areas in moral
judgment. The results indicated that there are "two distinct
clusters of areas. The first is concerned with intentions and
consequences,...the second is concerned with punishment --
with the expectation of it or the feeling that punishment is
tight" (p.112). This result is not consistent with Piag t's

hypothesis, supported by his studies, that moral judgment is

17

a unified area of cognitive development. XacRae interprets

the two clusters 3 posteriori as indicating the existence of

 

emotional and cognitive levels in moral judgments. 'Within
clusters of questions, though, hacRae found the same deveIOp-
mental trends toward a mature answer as did Piaget and Ler-
ner.

Johnson's (1962) study of children's moral judgments is
also primarily an attempt to measure the consistency of mature
or immature responses to moral judgment questions and is an
investigation of antecedent conditions. In contrast to NacRae's
study, Johnson test both boys and girls across a broader age
range (grades 5,7,9, and 11). The variables in the tests of
antecedent conditions were adult constraint, egocentricity,
age, sex, IQ, and parental occupation, each measured with a
standard written questionaire or test. Of these variables,
chronological age correlated positively and significantly
with all areas of moral judgment. Socioeconomic status and IQ,
themselves positively correlated, were highly correlated with
responses in the areas of moral realism, retribution vs.
restitution, and the efficacy of severe punishment. The eight
significant correlations of parent attitudes out of a possible
b5 occurred most frequently in the areas of immanent justice and
communicable responsibility. Other measures of antecedent
conditions did not show significant non-random correlations

with moral judgment. None of the significant correlations »

18

f

directly contradicts the findings in IacRae's study, since the
measures in the two studies are not the same.

Johnson also included a correlational measure of maturity
of response among questions within an area and among several
areas of judgment. The data seem to support, in contrast to
MacRae's data the hypothesis that a general factor of moral
judgment influences all areas with the possible exception of
communicable responsibility. The correlations are low but in
the expected direction. Since the responses are influenced by
IQ and socioeconomic level, some correlation is expected;
whether it is fully accounted for by these factors is not shown.
The correlations of responses to questions within moral judg-
ment areas show high consistency within each area. An examina-
tion of the stories used in the interview reveals a few logical
inconsistencies in the narrative which if resolved might yield
more impressive correlations.

In a short study on one aspect of moral judgment, that of
immanent justice, Nedinnus (1959) tested the reliability of two
Piagetian story-—questions. He concluded that children are
more likely to give the more mature response if an alternative
rational explanation is provided in the story. Kedinnus suggests
that the results in any particular study of moral judgment will
depend on a) the meaningfulness of the story, b) the concreteness
of the stories in the child's experience, and c) the prescence

or absence of rational explanation in the story. That these

(i

factors make a significant dif1erence is not shown conclusively
11!

use -innus' small san1pla In fact Johnsons' correlational

data would sug est that these factors do not make a significant

3
difference.

Two stu ies of the influence of social class on the deveIOp-
ment of moral judgment maturity are Harrower' s (19°“) and
Boehm and Nass' (1962). Eeo 1m andl sdid not find social class
differences; however, their sample of working Class and middle
class children all came from the same school and presumably the
same neighborhood, so the results of this study are proeabl y
contaminated.

Harrower found large diff mr nces in apps arent development
between upper and lo. Jer clas sEri ish caildren. The lower class
children showed the same developmental rate as the lower class
children in Geneva tested by Pi get and his collaborators.
However, the upper class children did not show a phase of pre-
ferring expiatory pun'snuent over reciprocity or restitution

even at the youngest ages of 5 and 6. Tfii s has some interes tin:r

t.)

but as yet untested implies tie ens for Pia‘e 's theory. It 15

not at all clear what factors mig ht be e erating to produce thi

’1')

schism in social class, and Harrower does not ffer any hypo-
sis. It is unlik (ely that peer interaction prelud es the

r clac

(D
U)

.aturity of moral judgment for upp children. Piaget's
hypothesis that p:ar interaction is a major factor in the
develOpment of mature moral judgment may have to be restated.

_ o. .. _ o . :_-- ~ ' .° I r _ _ - o
It 1, ,urgillw shat adults 1nteract1on with and expectations

from children are more imp ortant than Pia'et thou 'Et
"“ r: \/
toc1of1 (lSUO) has correlated ego dew clor‘ezt as mea surcd

by the TAT Wibll socialization as measured by knowledge an:

J.—

use of rules in marble games. Tlie correi atien was Small but

0

%1 nificant. Two of his subject groups were mentally retarded

' - - -a ". ~ ’01 , .v~ . n-'~‘.
ch1ldren. In these broups he MOS r-rved that the same sta cs f

O

L?

developznert in moral judgment occurred at about the 5 me mental

ag e as in he norznal children.

0 , . 1 I .. .-
AU 1 s (1 hi) studr extended to roff s finding 5 on the

CD‘;

ffect of mental recarcation on ochlop- ent of Jerald ud gment in
the area of immanent juscice. All his subjects were girls. Cne
group lived in an ins itution, wt ilet the control group lived

in the community. Cn the te ts of moral jud*mez the control
subjects rave res onses apprOpriate for their mental age, whereas
those livin1 in the institution were more retarded in moral

D

judgment. Abel attributes this difference in the maturi.y of

H
(D

sponses acou t immanent justice to th strict control exerted
ver the girls in the institution. In many respects they are
as controlled as very young children. Thus Abel's study may
indicate the p0531blc influence of authoritative caretakers on

the maturity of moral juu

A Review cf the Literature on Punishment Behaviors 9HQ their Correlates
The problem of punishment has been studied in many different

ways. In his study of moral judgment Piaget included some

questions about children's notions of the justice, severity, and
types of punishment that parents use. The results indicated that
the young child believes punishment to be right and just if an
adult indicates that it is so. The young child also believes
that severe expiatory punishment is more effective. In con-
trast older children take motivation into account and other
mitigating factors, such as the age of the wrongdoer, the
responsibility of the caretaker; furthermore, the older hhild
indicates that a punishment which somehow restores the ela-
tionship, either physical or inte'personal, that existed
before the 'naughtiness' is more effective (Piaget, 1932; Barnes,
1902). Other studies such as those of XacRae (1952), Johnson
(1964), and Harrower (1934) mentioned above have substantiated
Piaget's findings in this area and extended the studies to
cover correlated variables of IQ and social class.

A recent study by Parke and'flalters (1967) investigated
the factors influencing the effectiveness of punishment on
first and second graders in a temptation situation. This
study is related to Piaget's, though it was directed toward
a different question. Piaget asks what does the child believe
is effective; Parks and Walters ask what is effective.
However, the results of their study indicated in part that
deviation in the temptation situation is related to the nature

of the relationship between the agent and recipient of punish-

ment. The subjects in the high nurturanoe group deviated less

often and for a shorter period of time than did subjects in

the low nurturance group. Also high intensity punishment
inhibited deviancy to a greater degree than low intensity
punishment. If behavior in a temptation situation were indi-
cative of moral behavior in general, then the results of this
experiment would have strong implications for the develOpment
of beliefs about punishment. It high intensity punishment from
a nurturant person is effective in inhibiting first and second
graders' deviant behavior, one might predict that children
would come to expect it to be effective.

Other types of information about punishment behaviors are
far less in agreement than those above. For example, results
of studies of correlates to various types of punishment behaviors
reported used by parents have been inconclusive and often con-
flicting.

Aronfreed (1961; 1963) presents sociological and experi-
mental evidence for the hypothesis that disciplinary differ-
ences are related to differences in moral response. He found
that in a story completion task lower class children were more
likely to show "external guilt" such as fear of consequences,
whereas middle class children tended more to use self--evalua-
tion. Aronfreed views this difference as consistant with other
findings (Bronfenbrenner, 1968; Haccoby, Gibbs et al., 195M;
Littman, Koore, & Pierce-Jones, 1957) that lower class parents
are more likely to use direct ag

ression in dixcipline than

0‘
C.)

are middle class parents.

To test eXpe~imentally one aspect of this presumed differ-
ence in parental behavior, in a training procedure Aronfreed
varied the amount of control that the child had in structuring
his own punishment. He hypothesized that externally adminis-
tered punishment such as corporal punishment or scolding does
not lead to self--criticism to the same degree as self-"admin-
istered punishment. The results indicated that the child who
could structure his own punishment was more likely to engage
in self--criticism than the child who was punished by the
experimenter. This result suggests that self-~criticism generalizes
from situation to another, so that if a child receives training
in self--criticism in one situation, he is more likely to
engage in self-—criticism in a different situation than $11

a child who is not trained in self--criticism. 'Whether training

.4.

n self--criticism generalizes to other aspects of moral
behavior and moral judgment cannot be gauged from this study.
For example, one would question whether self~~criticizing chil-
dren would be less deviant on some measure than the child who
is punished by someone else.

Aronfreed interprets his results as contradicting Piaget's
main thesis that moral judgment is a feature of cognitive develop-
ment, since the results indicated that moral responses are a
variable function of the social roles and cultural settings

which the child has experienced. However, equilibrium theory,

while not readily admitting moral judgments as purely situational,
certainly does recognize that cognitive deVelopment is stronly
influenced by social roles and cultural settings. For exmmple,
Piaget notes that eye-hand coordination in the infant is
strongly influenced by cultural and climactic customs of swad-
ling, mittening, providing play things and so on. Cognitive
develOpment is a function of integrated individual capabilities
and environmental Opportunities or restrictions. Since the
young child believes in the stage of egocentrism that adults
are always correct and hence, just, he could reflect parental
Opinions, behaviors, and expectations about the efficacy of
punishment and thus seeningly give a response incongruous with
cognitive level, but congruous with situational variables.

The Stanford studies (Sears, Whiting, Newlis, & Sears, 1953;

ears, 1961; Sears, Rau, & Alpert, 1965) are another well—

U)

known example of an attempt to relate moral responses to early
behaviors and parental disciplinary practices. Although the
1953 study (Sears et al.) found a clear positive relation be-
tween overt aggression in preschool and maternal punitiveness,
testing the sa.e subjects at age twelve showed a negative
correlation between anti-social aggression and severity of
punishment as reported by the mothers when the children were
five. There are several methodological problems which con-
tribute to the likelihood of inconsistent results without

really offering very satisfactory explanations of the discre-

N
Kn

ancy. For example, different measures of aggression were used
in the different studies and the measures of punishment were
data from interviews with mothers only. Furthermore, it is not
clear how or whether behavioral measures of aggression are
related to moral responses.

If delinquency may be used as a behavioral indicator of
immature moral judgment, then studies on the behavioral anto~
cedents of delinquency are relevant to the present discussion.
In particular the effects of parental authority as manifested
in disciplinary practices would be related to Piagetian theory
and to other studies of the efficacy of punishment. The
Gluecks' data (1950) are the most reliable on this; point,
having come from several sources: a psychiatric interview
*‘th the child, interviews with the parents or caretakers, case
reports by social workers, and school records.The results dis-
closed that the

...most marked difference between the disciplinary practices

of the parents of the delinquents and those of the non-

delinquents is found in the considerably greater extent

to which the former resorted to physical punishment and

in the lesser extent to which they reasoned with the boys

about misconduct. (PM 85')

Other revealing points were that disciplinary practices of
parents of delinquent boys were lax or erratic and that fathers

of delinquent boys were particularly likely (over two-thirds) to

F.)
o\

use physical punishment. Although there are many other factors
involved in delinquency, other studies (Bandura & Walters, 1959;

NeCord & KGCord, 1956) have substantiated the Gluecks' results.

Statement of Purpose of the Present Stuiy

The present study was designed to investigate the appli-
cability of the Piagetian theory of the develOpment of moral
judgment to delinquent children and to test behavioral tech-
niques in the area of the justice of punishment.

Many investigators have replicated Piaget's original
study using lower, middle, and upper class children ranging in
age from 6 to 15. All have used some written or oral form of
Piaget's original stories and questions (Harrower, 1934; Lerner,
1937; MacRae, 1954; Johnson, 196M). However, no one has treated
this area of moral develOpment as a cognitive phenomenon using
subjects who one would predict would be advanced or retarded
developmentally on the basis of some social or experimental
treatment. One possible exception is the use of mentally
retarded subjects. Several investigators (Bobroff, 1961; Abel,
1941) have reported mentally retarded children to be retarded
in other developmental areas such as conservation acquisition
(Goodnow & Eethon, 1966) and acquisition of immanent justice.
However, Piagetian theory predicts that other facets of the
personality and environment also have strong effects on develop-

ment of moral judgment. This study is concerned with one of these.

The subjects were adolescent dolinouent boys, most of them
IN:gro. They were chosen on the hypo thesi is that factors which
contribute d to their delinquenc:y might also be factors wE .ich
contribute to developmental retardation in moral judgment. If
this notion is correct one would pre edit t that loser class,
Negro, delinquent adolescents would have im1nature conceptso

the justice of punishment. There are three as uuptions behind

t to assume t-1at the parents of

H .
U)

01

this hypothesis. The fir

’Jo
U}

these boys probably used severe expiator* pu1i5nments in d
cipl ining their children (Glneck & Glueck, 1950). The second
is to assume that mid Ml class soc 1e ty does not accept the

Ne gro and in particular the lomvvr clas stegro on equal terms;
it expects the lower class Negro to be submissive, respectful,

and unable t s on an equal emotional and

O
'4
"3
{0
t)
ho.-
f)
O
Q;
(3
(1
"U

ntel cctlal level. In short it clearly views the lower class

SJ.

Negro as 'child-like' in the: e respects. A third assumvtion

is that because the delinquent sample is being actively punished
hy society, they will resnond as children do to e:(pia
punishment, even though a part of the punislz en is ostensibly

rehabilitation.

1ne first prediction is that
1. Drlinllent aioles 11e ts rill give1 immature rjspons sa) to

the intervie w q1.1estio11.1aire and b) on a behavioral meaSI..1'e in a

.41..

9

\_I

133”?

a
\

v

If”;

‘0

'1ll

V

Ir

J.

response

e

‘
I

t}

The ma

minor p

\
III-‘3

story--question prooed
C

the

.1
.1

rt to el

S 51138

(SCd to
the

‘
1

C5.

5 1
stuiy ar.

Le

V"

J.

\
4

etc
nc

."J
‘ J

Fer
re

.L A

111..

4.
L;
the

(1:(1‘
///“).
that

1
a differen

V
J

L

ld predict

1
x

1

1 stor

A 1,
\TC

\.

“J7

-

n ea
5, adults, arr

A.
’70

—‘

A

J.
V

J-

an t}

n 309?
t1

.4

t r-
ut +113.

J.
»l

3
Cl
was
3“:

.18..

’.J
I.

\J

o

n '.'p

L.

w
I

1
ct

f3,
»/,1.V
V

v

64

ral

i0

’
r'. H V"
1.11:; V

nistra-

'13

1 b
3»-"
A};

1
.A C
ad

l

.4
.1

\l

-n
O

.
'V'If’.‘I
“b ‘

tori

("0

o

VflV‘ ~57 1‘. 14 'r“.

0111:,311.) 1115 p1 cc
‘J ‘—

r.

i

113

3.

l'

f

C C
9

l
\

ion
e a
q ,1. '
L(~L—LV

+L
bh‘

11d the pr
val

all

1.1
on

as a funct

cter
MAJ 1.

up.

tions
1;ar1
1
.J.
v.’

1.
’CCJU

.4
SU

(.

dolcsccnt c
‘v’ 1‘: 7b"?

story--que

a

.‘ t . .,.,‘2 .- 1 n a. A _‘
*tJ-sikiulfieht, OI‘ I'QL-Itif'u, {JaCC 0.. Wii Cf] TUSUOHL‘B

('3
'3.
{L
t4
0..
(
(D
0
ET"
(.1.
O
C:
1

’fl‘.-

3- I v“ V" J ,.—\ ' r‘r‘ ‘ -‘= .u '41” "- "O "n ~ 3 '2 .'- 'afiJ' 1
b-VU1J SCOTBm. inc use oi tnis measure prac itsllj Climluabgq

(4‘

‘ I
cc; 3 response was dc513nci to

same tins the measure is contaminated by the possibilit" tnat

‘ 1‘ vr - r it“ :A .-" \ I ~ IN" ".'~ -. u . r‘ . r H
Lnere mad to a 1ohl iifcloncc ostwscn tAu subduct s idcds

about the justice of pinishmcnt, snticcuont, or rcwsrd, and his

\ O

own behavior in a punisnin? situation. However, the third pre-

d s that

iJ-

ction

‘J.

‘ O

'ne maturity of the responses given in the intervicw

*3

q
)0

will correlate significartly with the resnonses in the rat-

training.

If these predictions are confi“med, then further study and

experimentation can proceed with these mothods.

Method
£329.12

Experimental 35. From the population on work programs
at Boys Training School (BTS) in Lansing, Michigan, a school
for delinquents, 22 male adolescents were randomly selected.
The mean age of the §s was 16.2 (standard deviation = 0.38).
Most of the §s had attended high school, all had attended
eighth grade, but school attendance was erratic and grade
level was not always available in institution records. The
range in IQ was within normal limits 80 to 105, (i = 92).
The IQ scores were obtained from various standardized indi-
vidual tests administered to all boys at the school.

The S5 were at BTS for a variety of reasons, all for
incorrigibility, with other misdemeanors and crimes ranging
from truancy and car theft to rape and armed robbery.

Sixteen of the §s were from the Negro slum area of
Detroit, three were from poor sections of Jackson, Michigan,
and two were from smaller southern Michigan towns. Sixteen
of the Ss were Negro, five Caucasian and one Mexican.

Contrqlzgs. Twenty-two male adolescents were randomly
selected from study halls in a downtown Detroit high school.
The school pOpulation is almost entirely Negro and is located
in a Negro slum area. No attempt was made to select control
gs without police records, the only restrictions on selection

were that §s had low-normal IQ's and were 16 years of age.

:1

The E was not permitted access to school records, and indi-
vidual IQ tests could not be administered. Instead, the study
hall supervisors checked scores on standard group IQ tests
and listed the boys who fell into the low-normal range (80
to 105). The mean age of the control Ss was 16.2, standard
deviation 0.10.
Procedure

Pretest and familiarization. The E visited BTS twice a
week for a month. During each visit she talked informally
with the boys individually and in groups, soliciting their
opinions about the interview and animal training procedure
and assuring them of her goodwill and ability to keep con-
fident any of their responses. The boys remained defensive
throughout most of the study, but were not hostile after the
first few visits. They became more Open and friendly towards
the end of the study when word spread that E had been bitten
by one of her own rats during a trial run at the school.

Eight volunteers took part in a pre-test of the interview.
The results of the pre-test were used to reformulate the
story-questions and to gain fijfi rapport. The pre-test dis-
closed the importance of emphasizing the fictional aspect of
the stories and the importance of the fact that the rats
were laboratory bred, relatively tame animals, not pets of
the E on the one hand, nor dangerous and wild on the other.

Animal training. Twelve laboratory bred white rats

were trained to run a T-maze prior to the study. Using

shaping and training procedure with partial reinforcement

the rats were trained to make a left turn in the maze to
obtain delayed reinforcement. The criterion was seven correct
trials in succession. The Ss were not told that the rats

had been previously trained to make this consistently wrong
response. After each trial run with an SJ‘E retrained the
rat's left turn response.

Interview. Each §_was interviewed individually. When
he entered the testing room either at the high school or at
BTS, he was requested to pick an identifying number rather
than to introduce himself. The E explained that §_would
remain anonymous. Then.§ spoke the following instructions
from memory:

I have asked you to take part in a study about learning.

we are interested in finding out how you think peOple

learn not to do things. I am going to read you some
made-up stories about boys your age and then ask your
opinion about certain parts of the stories. There are
no right or wrong answers, I just want to know what

you think about them. Afterwards I would like you

to help train a rat (points) but I‘ll explain that

when we are finished with the stories.

The stories were presented in a random order and E
recorded §_'s answer without comment. Occasionally an _S_
would ask to have the story repeated or a question rephrased,
in which case this was done.

The three stories each concern theft. Each story
involves a male adolescent thief and one other person:

a teacher, mother or male peer. The stories are similar

in style to those used by Piaget but are written for adolescents

and are relatively comparable in content except for the

relationship between the characters.

Story I

Jim was alone in study hall and decided to look in
the teacher's desk. There was some money in the tOp
drawer and he took it. The teacher found out and
punished him.

1. How do you think the teacher would punish him?

Another time Alan did the same thing. He took some
money from a teacher's desk. But the teacher only
talked to him and explained that he was taking some-
thing that someone else owned. Later Jim and.Alan
found a teacher's wallet in the school parking lot.
One of them wanted to take the wallet to the teacher.

2. Who was it?

3 o Mlyr ?
Stogy II

Tom knew that his Mother kept money in the bedroom.
One day he took some, but his Mother discovered it
and punished him.

1. How do you think she would punish him?

Sam stole some money from his Mother, too, but she

did not punish him. She talked to him about how wrong
it is to steal. A few days later Tom and Sam found
somethings that their Mothers had left on the front
step. One of the boys wanted to sell what they had
found and keep the money. The other boy wanted to
give them back.

2. Which boy wanted to give the things back to his

Mother?
30 Why?
Stogy III

Peter was over at his friend's place one afternoon
waiting for him to come back. He found some money
in the room and took it. His friend found out and
really gave it to him.

1. What do you think his friend did?

Another time John was at a friend's place and he took
some money from his friend. His friend found out and
explained to John what he was going to do with the
money and what he thought about stealing. Peter and
John found something their friend had left at school
a few days later. One of the boys wanted to take it
back to the friend.

2. Who wanted to take it back?

3. Why?

After the stories were presented and the interview
completed, E continued:

Another part of the study on learning is animal train-
ing. ‘We want to know how fast you can train this rat
to run from here to there and back again (demonstrate).
If the rat goes in these red ends (point) he has made

a mistake. would you like to pick a rat to train?
(While S makes his selection) These are laboratory
rats from Michigan State University. They are born
there and raised by experimenters to be used in studies
like this. (The rat is placed in the maze). You may
use two things to train your rat. Use either this

food that you can drop in -- the rat hasn't eaten since
yesterday -- or this poker. The poker has a rubber

end on it which pulls the rat's hair a little and
bothers it. While you train the rat I have some
figuring to do so I'll sit over here and work and watch
the time. Are you ready? Go.

The E sat behind Ss where she could watch their move-
ments. The E recorded pokes, food drops, and other be-
haviors. In this position §_was out of §fs range of vision
and was relatively unobtrusive. .After five minutes E said
"Time's up, how many times did it run?" and then thanked
§ and walked back with him to his supervisor or study hall.

Data

 

In the interview E recorded st responses verbatim.

0F
.x/

These were later analysed as 'mature' or 'immature' with
respect to Piagetian theory.

In the rat training phrase E also recorded the number
of times § poked the rat, dropped food to it, and exhibited
other behaviors such as encouraging talk, attracting be-

haviors, swearing, etc.

T ‘+ a" 'V‘YV
~L‘z-I- ‘L '-

o
“ |-\ v v
.L 1 y‘-

‘1'" ' ~..~ — -. - < 1 ‘: . ' n.
1he interview responses we-e LlaSSLfleu as mature or

'..,.,3- 0 a
immature as f011wa:

4

J

a
no

r.’

1. How do you thir
Immature 7:

would punish hgm

{I}

FJ. i. .-

m
l—‘O (‘
{D
((3) r+ b—J }—J

c:

punishment
restriction
o authority
oal response
relations .ip
.nation
2. flhich boy would not steal th-e ?
Immature -— the puni 'shed boy
Mature -- the boy given an explanation
J0 may?

$11 *dd '0’
L‘)
[‘11

Vjflé

(D *d

{J 0) F“ CL}
0
H30

Yature --

H “S
H O
’U m (2'

’73
F);

I)
:1)

 

Immature -- revenge for the punished boy to steal a gain
punishment is effective
Nature -- explanation is effective

The distribution of responses to question one is contained in

Table 1. Since many categories had few or no entrim , it is not

possible to conduct an analys1s of variance. However, certain
relations may be noted. The delinquentg "ro oup ref3rs stealing

to a iigher aut} ori ty onl./ one--third as often as t1e control
group. leis may be due partly to the effect of a new prinici-
pal in the Detroit hi 1gh school from which the control group was
selected. The charismatic principal influenced his students and
the students consulted him over many kinds of problems.

1"

seco ndly t‘e uClllau nt group suggests phys

HI

cal punishment
almost twi e as often as the control group in the child-~parent

situation. The third possiele difference in response is to the

adolescent--peer story in which more delinquent tnan control

’3'?

J I

Table l. The type of response to question one in the interview.
The number of §s giving each response to each story is

given in the blocks.

 

 

 

 

Stories

Control Experimental
Response 1 2 _3 l 2 3
physical punishment 2 13 5 - 11 9
physical restriction 15 - 16 21 - 10
refer to authority 5 2 - 1 1 -
{£51,} 15111;; I ' ' '6' " ' I " " ' ' '9" ' 1' ' ' "
reciprocity - - - - - l

explanation - l l - l 2

 

suggested loss of relationship as an adequate response to stealing

by a peer. In cneral the results as expressed in Table 1 show
the relative moral immaturity of response as defined by Piage

in both the experimental and control groups, since more than

half of the responses are in the categories of physical restric-

'3

tion and punishment. .his is particularly well exemplified

the percentage of responses recommendins corporal punish-

. o

wh e n
ment in a parent--child situation in this study is compared
with data obtained by Barnes (1902).

The data for each story were transformed to give one score
as a composite of the scores of the three questions in each
story (Table 2).

The transformed scores were used in an analysis of variance
of the interview data. The prediction that the maturity of
response given to each story would be the same was not SUpported.
The responses to each story were significantly different (i=5.09,
df:2, p‘.02). The mean response scores for each story are given
'n Table 2. This implies that the maturity of the moral judg-
ment is determined to some extent by the relationship between
the sory characters. however, the control and experimental
groups were not significantly di"ferent (9-.u9). The interaction

).

between type of question and group was not significant (5:.7

\x)

Cne anecdote deserves mention here: three of the delin-
quent boys, after making the mature response that the boy who

received an explanation was less likely to steal again than the

'30

I
'

Table 2. Mean transformed scores for responses to each story on

a scale of zero to four.

 

 

 

Stories
Group 1} 2 «3
Experimental 1.41 2.27 1.09
Control 1.36 1.86 1.18

 

(3

boy punbhed, then commented that "this does not work for me or
"t‘at wor} s for some Izids, any} es". There is apparently some
discrepancJ betwefan what tV y thovt 5ht would be effective when
applied to the nselves and what would be effective for others.

hao the "others" might be is not clear. It is possiole that the
delinquents have received instruction about handling offenses
made against them at 3T3 and that they believe that this is the
'right' middle class answer. It is probble that other members of
beta samples s, but probably predominantly in the delinquent greup,
have similar unexpressed thou
to the questions. One must note, however, that there is no

corxelation between such responses on the interview and the

actual punishment belxavier of these thre ‘Es.

r U
4' “weir
V -J- 4....

r13

J

The rat training data should have beenc ichotemous with only
two possible responses -- p0};ing (punishment) and feeding (re-
ward). However, the boys devised a tnirdr Mp ns 0. The sub-
ject would tap the poker en the ca*e orIub a food pellet acres
the wire or put his face close to the caged T--maze to talk to

rat. All these behaviors were labeled "attracting" responses

(+-
0-3“
('0

9.)
:5
:1.

are combined with the feeding responses where indicated.
The percentages of punishment, reward, and attracting
behaviors are presented in Table 3. The number of punishment
behaviors relative to the number of rewarding behaviors gives

thee dis tribution a stron-

1?
(3

negative skew. In this gross sens

\D

the behavioral and interview data are similar with respect to

Lil

Table 3. Percentage of punishing, rewarding and attracting in

rat trainingiprocedure.

 

 

 

Punishment Reward Attracting
Experimental 92% 3% 5%
Control 86% 5% 9%
Level of significant .01 .01 .12 n.5-

difference

V‘—

3‘.
|

'immature' and 'mature' responses.

Since the distribution of the scores was no

C t

skewed and the variance was large, the non-parametric lann--

Whitney test of identical population distributions was used
rather than a Student p—test. All claculations are corrected

for ties in rankdsand the results are shown in Table 7. The
.1

1

ooLesis that the pepuiation distributions are tne same for

1

'51;

r:’

the delinquent and control groups is rejected at the .01 level
of s= ficance for punishment behaviors and for rewarding

behaviors. The distribution of attracting resoonses are not
Q l

'3

(0
p.

gnificantly di-ferent. Therefore, the data on rat--training

’-

indicated that the delinquent and control groups were not the
same in use of these procedures in contrast to the results
obtained in the analysis of the interview data.

Relation between Interview and Objective Data

Tne correlations between mature and immature responses on
the interview and rat-~training task are presented in Table U.
The data used for the correlations are dichotomous. 1he res-
ponses to the second question of the interview ~- "Who would
return..., the punished boy or the bev given an explanation?" --
is correlated with the raw score on the food drops. Any b
who gave two or more food rewards is coded 'mature', an‘g
giving zero or one food rewards is coded 'immature'. This

coding is somewhat arbitrary and awkward, but raw scores or

Jy’)

Table 4. Correlations between interview dataiand punishment responses.

 

 

 

Story
Groups .1 2 3
Experimental .31 -041,“ 002
Control '118 .20 .37*

 

*n<.05

’DI‘C

(D

ntage of rewards are very skewed and have very large vari-

*6

Using this method of coding data, the correlations are not

large, nor are all in the predicted direction. The delinquent

sample's response to the adolescert--pecr story shows a signi-

(

ficant n ga'ive correlation with their punishment behavior. The

centre sample 3 res oases to the adole

cent--parent story is

U]

positively correlated with their punishment behavior.

The interview data do not support the prediction that
delinquent adolescents would make significantly more morally
immature responses than a control group from the awe seeio-
economic background. However, both Barnes (1902) and Piage
(1932) reported that for s; 11 to 12 years old, about 80; of
the responses were recommendations of reciprocity or explana-
tion rather than expiatory punishment. 1J9 is in this study
gave fewer than 20 3 reciproci+ y or e Aplanation responses. Toese
results seem to indicate tr at the constra nts on lower class
male adolescents inhibit tr ne co5nit ive restructuring of moral

judgment as defined by Piaget.

m

ot1 the de lir ”an nt and non-delinquent es responded sig-
nificantly more maturely in peer situations than in adolescent-
adult situations. This result contradicts the second predic-
tion and com nls social--learning hypo 1;: sesabout the situation-
centeredness of morality(e.g. Hartshorne & Nay, 1928-1330).

One might even hypothesize discrimination learning because of

the constraints imposed on the adolescent-adult situations.

Tiii sdoes not imply that the 85 are unaole to generalize from
one situation to anether,but that to do so is not adaptable.
Other research does not seem to support the nypothesis taat
adolescents of any social class may have one set of moral values

or peer situations and another for adolescent- adult situations

\ v o / ‘- o ‘ a o
(Peck a hav13hurst 19o0 . Peck and davinnurst d1d not directlv
c.) 3 L) V a

ask specific questions about punishment beliefs and practices,
however. Cohen (1055) has hypothesized that such a schism occurs
among adolescents who belong to a delinquent gang, but notes

that the values expressed by the delinquents did not seem to

form a code which had any validity in itself, but only for what
it could obtain for the delinquents. The comments of the delin-
quent rs in the present study tend to support Cohen's report.

A common response to the second story was "Well, he just wouldn't
be ny'friend no more. You can't be friends with someone what's

going to steal from you". Kany s impl'citly or explicitly

|Cn

stated that it is just to steal from a peer for 'revenge'.
Although the interview method did not discriminate between
the delinquents and non-delinquents, the objective method indi-
cated that the delinquent g5 used punishment significantly more
and rewarded significantly less than the control Es. This
finding supports the first prediction of the study. The inter-
views contain more uncontrollable elements than did the objec-
tive measure of punishment behavior and presemably were con-
taminated by them. Cne such demcnt already discussed is that
the delinquent fie in particular may have responded with a non-
expiatory explanation while privately disavowing its efficacy.
This was mentioned with respect to a few comments about the
ineffectiveness of censure and explanation for themselves in
contrast with its theoretical effectiveness for some unspecified

others.

'1

A: other element of possible contamina ion is the non-ver-

balne es

(1)
O
H)
('7'
O
J
U}
y
:3
CT
0
( 1"
3*
I

broups, but particularly in the dolin-
quent group. Rest is Clearlv had problems ereressihg themselves

and may have simply left out parts of their C; planatior ns. A

L

I-

large )art of the workine vocabulary of boys at E S is comoosed

*‘5
in
U)
(D
U)
’1')
r].
L
O
"S
C
S
p
r)
O
’1’)
*6
DJ
L.
H
(D
(-1"
L
O
*‘5
:3
O
(1'
x:
C.
e”)
"I
n),-
O
0‘
:1
4
(‘3

mildle class women. fhe E 3 presence may t ve con tributel to

+1» o ' 1'1: . + r .. +‘~ ~ 5‘

one is inaoiiitr to express LJCMSL1V3S clearly.
mu «'1 L'.., .1- +1 '4 ..--.. . ‘ .-L .1: ° ,.
in: correlations 1593;10an we lllsfil‘vlevf and the lat t1 lfllllb

data no not support the second ar nd third predictions, but again

1

~ 1"" 4‘.‘~,.~': YR . -. ~. ~A1 r. .- . . ~.J“ 1"- e
1e napoun sis tAat banishment behavior is blouaoiOI

i
r

confirm t

H)

oriented, rather than aris ing rem a global code. The signifi-

enac negative correlation of the secozd story an” the pVDirh-
mat responses for the cvoeriwrntal san ple and the significant
positive corr. 1” -ion botwoe‘. the third story and the punish-
ment responses for the control sample both lead to the same con-

clu51on: rat training bch.aviors are more similar to beliefs about

Ffllt-Chlld -unishment that to beliefs about peer pinishmcnt.

.C’

This finding seems to be related to the absolute authority rela~
tion in each of these two situations in conparison with the peer

o _ o L _: “‘1'". 3'. ‘, -.‘ -_-. A ~. --n"~ . 1
Situation. However, colieis about PO? puni hKSUtv pale“. ”bled!

ild punisnmont, and animal tra:ining are not nigniy

1'... . ..»— . 17-1-
“eh-Ever, 0113 IUQJ 1Cd':c'\d

'Q

L

.1
7 '1 -..
.A.»

CI'

ir
go

4'-
1
to-
r‘ted
o

,1
.' ‘
O
V
I)
A

H I .‘\
| 9—-
V .

Fla;

J.

51
.L
La

(‘53-.

k, a -I-

o N

f)

A-

f‘ .

' Cfp‘flCt-T
gay-warn
—’.l_'.

L:

VLOAI

O-‘I itlv‘“

.0 \I

l

V
nque x

.C‘" 3"
.A.)

,-

Q

,~

,‘1

v

o
l

)"

\~L~¢.
itu

r34
.1

u.) ' "-
J.
lfimflt
0
1| ‘3
—k‘\ *.A—

.r‘h
IJ

Y-_,

15
10

V 1"»- .
1L“;
f r:
f -. r“
. 1-1.01
b
is t?
I".
1
A
1""

1

C'
s.)
C

"I
O
<.

.

~A-h

L

or

110
:.on C’
{1“ _ U.

o
$.L L)

r

‘

L,
)n unknwwl

L
f.
“C‘T-t"

I"
i.
bL.‘.‘J

AtJ
'1

~4-

:1-\.‘
gal-LL
_La
0 ,4—
18C?
.11
1
T.
.5
UPC
7‘ .Lll a I10!

I

U

-T‘
’N.

~"

I

Q
ll)‘-

4 ... :1
”L,

l
.t

“f“ C‘Y‘fi

.Lx)
a
J.
or"

C

V

1.0
H
\J.

1313

1:
1T1
L
.V '
LL:
~ 'f‘\
y 1
'L

.{-',

no
a
-'U
h‘r
._;
N;
A.
‘.

-0

A
r“ .
\nLA

.-J
DiscAfiV‘AI"

‘,-,-~ A

“la
afoun:
»J
L,
5.4

a Asa»

V
o .3 l
fhhlfigm,
+T- ’-
L;
."‘.‘-A
.1 L
L
11;,
“:1"

'1

norm1

9
\ ‘11.};31,
£3
T.
.L-A
L‘
'l .. \J.L
q‘ ‘r‘
ouluzi
t" .
I‘ -lk‘

)‘
n
I 1““

..

U

am

“4-; '4

13 a I"

Y
.2110 .‘I
19'1“
.- 4‘
4P h
l/
.l.
y ‘
v ‘~.*; ‘
335.11,

'5‘

O

.
f“
1 l

"I;
r.

ldr‘p
+1-
h4~.
+1,
VVQ
(“’3‘
U-’

en UV:

non-.
'on to

..
0
VJ.

'3”.
.n.

rather
correla

.l.l

a

 

' T
)A ‘ .‘fr
L‘AL—L— I“-..

in L’

‘7‘
f-‘
-1.

3:?1‘3

l

ins yr

.1

Ev L

trai

,
v.1

.

i lcal to

\

,'. N D ' w
”lb COM;

‘
' .

t

A

U

'-
J.

.J.—a. vd

.olos
atil
so
0-

h

c

-. .J. .2 (L
‘s_&

3! }~ ‘I'!

J3 -

I‘C 577-011

"‘ r-
r
-1. ‘4.

1

r
lirvn,

cl

sou

I
- _o
..L

-6. ‘-
O.
J.
V:
r111
C".

7-71,»:

vl.
I“
gt.

Y';".

I

‘5
.,v‘-
-02.-.
A-Jv

01

.K

’5."L
J.

_‘ l
U

I".

no; 0

J
So
..r\
l...
‘
."n
.L~

1.1..

I
L
.J n

0

cc?

L‘

3.

$-
L -. v

'1

Yva
U

1
' I "I
q.
,.
Ci. L‘-.$
ro

'J.

1 t
I’

"\

tudj would ra

n.

n
C.
\J

v

. , 1

12K) L-

.‘A t "

try-.~: 91.3.0
Q‘J'fi

.41
U

I-

07""
Vufi "

J
c-
U
.3. ’ P P
14111; .L

U ..
.11";

a
I

p
.L

01;“
..
1
.‘-‘,11
‘g'.’

.01

t

V

‘-
.

O
o punLS'

A

u

do.
n0”

"’.C‘.‘.’lOI‘S O

‘
.' X

.L) 1‘

I

'. i
T i.

I

ull

+1
.
d-L'—'-L
l.

\I
"f\
fly

I!

L:

1

.Tv
Jo D
“H“ on PL
E n 0
1w . o
3 Ju VJ
3 Q n U .. o
3 O .J .l
. .1. :1. F
_ 3 LL 1 U
a a 1 o
c f1 r mi nm
m. an“ mnr mu .1
.L 1 a .1
C .0 C C
$L ~.1Ih I‘M LU 1!.
to C ,1. 4v
3 . :1
VJ n“ P; t o n
.TV may 0 .1
.J. 1.1
T S O S
O O .1 {U W?
Va. ..J ru 31,“ ,U
o. a 1...~. .‘nw 11* V“
wn. u an 1 ”3..
L .. .1 k r
f S O l C
C ML... 1 01.
I? V v J “9 I.
S L ‘3 o .1
“I. ..... .11 j

1
U
._
.3
I.‘
u
t‘

1h“. .Hu - n...
W. w. 1b S O
xv. p 5.4 o L ..

M .o r S .r.
9 1 ,J B l

i i. u 31* .l w 0 Ha
.5 lb P v. m, 0 CJ

4» a U as

11 a. W .a s r
U 0 Q

1 an \r. O a
a 3 S .. 1.

1; “1 Lb 1.... mi.

Au 1 O n a,
no r A...

:1 1.. r ml.” ”ad 0
3 3 9 L. 1,1 1

11 ms x” S u a
n... L v A _ .1. .C C
.9. C n c O
n O u m ”a .,l_
. +v h p S «L

.1 a

I: 11 LI“ 8 1n] u

we no is to $1 Lo

P. 3 G C .l

O J to C S
CU “UL AL L

. + “(I O f .b
"V O L 3 .1

the1

nurke

‘J

[‘8

-
VAL

V

‘ to

"i
-1.

e .

J‘.‘ 1‘
L
3. L.

-1 1

l

n
1

A
k.

of
r:

v
.LVJ‘
.L

U

o

‘

ur'

4.
L.

w

:5

1T;

7a
the

,

n‘.

QLLLFJII'
.:

(3 C‘

.1, .Lg.)

ol

A

..
. A

~11;

JL

T,+-
actoian

‘I'C‘

Pi

I;

.19

abiil

C‘
U

" -
w’.-.)
-.L\z

L

it'd to

1n
4.5.

re

it

is

volvu

O
-. v.
J_.'i

‘ .‘I
1"“: )14 1 xi

V‘-

U’JJ

‘

t

C
.J

1 o
7‘ “. F
ULl—hsJ

1.

p
l.

01"; O

for

ions

‘1

J

-J-
“Upa'

n1

a‘~4~c
aJ-5b‘

nt x

v
1.1”11"’y."7.
‘Jk —RQ;‘_«

. ‘

a1

1"

ture mo

'7"

n
L..'J.

‘4'!

b—:‘ l‘

1.

‘ .
n
L

w

,a‘

-‘

L
L’s

”-
.b

10

s
\L

Pa

4

.1"; .1- 31‘?

OH

tur

a,
UL»

(n X

T
r

O

on.

'n

0631

‘I

d

woul

fi+

(g‘I.u

(“Y-'1

‘4
du~ LO-“

I
-~

moral

O
5"055 in

\
.

'Ear

C

References

 

Abel, T. M. Moral judgments among subnormals. Journal of Ab-
normal and Social Psychology, 19u1,13§, 378-392.

Aronfreed, J. The nature, variety and social patterning of
moral response to transgression. Journal of Abnormal and
Social Psxchology, 1961, 6 , 223-241.

Aronfreed, J. The effect of experimental socialization para-
digms upon two moral responses to transgression. Journal of
Abnormal and Social Psychology, 1963, 66, #37-448.

Baldwin, J. M. Social and ethical interpretations in mental
development. New York: Macmillan, 1906.

Bandura, A., & Kupers, Carol J.. The transmission of patterns
of self-reinforcement through modeling. Journal of Abnormal

and Social Psychology, 1964, 6 , 1-19.

 

Bandura, A., & Walters, R. H. Adolescent aggression. New York:
Ronald Press, 1960.

Barnes, E. Growth of social judgment. Studies in Education,
1902, g, 203-217.

Bobroff, A. The stages of maturation in socialized thinking
and in the ego development of two groups of children. ,thlg
Development, 1960,13l, 321-338.

Boehm, H., & Nass, M. L. Social class differences in conscience
develOpment. Child Develonment, 1962, 33, 565-575.

Bronfenbrenner, U. Socialization and social class through time

and space. In E. E. Maccoby, T. M. Newcomb, and E. L. Hartley

(Eds.), Readings in social_psyphology. New York: Holt, Rine-
hart, & Winston, 1958. Pp. 400-425.

Cohen, A. K. Delinquent boys. Glencoe, Ill.: Free Press, 1955.

Dewey, J. Moral principles in education. New York: Philoso-
phical Library, 1959.

Festinger, L. A theory of cognitive dissonance. Evanston, 111.:
Row, Peterson, 1957.

Glueck, S., & Glueck, E. Unravelling juvenile delinquency. New
York: Commonwealth Fund, 1950.

Goodnow, J. J. & Bethon, G. Piaget's task: The effects of school-
ing and intelligence. Child Development, 1966,‘jz, 573-582.

Grinder, R. Parental child rearing practices, conscience, and
resistance to temptation of sixth grade children. Child Devel-
gggggt, 1962, 33, 802-820.

Hall, C. S. A primer of Freudian psychology. Cleveland: ‘World
Publishing Co., 1954.

Harrower, M. R. Social status and the moral develOpment of the
child. British Journal of Educational Psychology, 193%, 3,
75-95.

Hartshorne, H., & May, M. A. Studies in the nature of character.
Vol. I. Studies in deceit. Vol. II. Studies in self-control.
New York: Macmillan, 1928-1930.

Johnson, R. C. A study of children's moral judgments. .thld
Development, 1962,‘3}, 327-354.

Kohlberg, L. Moral develOpment and identification. In H. Steven-

son (Ed.), Child psychology, 62nd Yearbook of the National Society

 

 

[”3
.2,

for Studies in Education. Chicago: University of Chicago
Press, 1963.

Lerner, E. Constraint areas and the moral judgment of the child.
Menasha, Wisc.: George Banta Publishing Co., 1937.

Littman, R. A., Moore, R. A., & Pierce-Jones, J. Social class
differences in child rearing: a third community for comparison
with Chicago and Newton, Massachusetts. American Sociology
Lem 1957. 2. 694-704.

Maccoby, E., & Gibbs, P. K. Iethods of child-rearing in two
social classes. In W. E. Martin.& C. B. Stendler (Eds.),
Readings in child development. New York: Harcourt Brace,
1954. Pp- 380-396.

MacRae, R. A test of Piaget's theories of moral development.
Doctoral dissertation, Harvard University, 1952.

McCord,'W., & McCord, J. Psychonathy and delinquency. New
York: Grune & Stratton, 1956.

Medinnas, G. R. Immanent justice in children: a review of the
literature and additional data. Journal of Genetic Psychology,
1959. 24. 253-262.

Parke, R., & Walters, R. Some factors influencing the efficacy
of punishment training for inducing response inhibition. MQQQ:
graphs of the Society for Research in Child Development, 1967,
3g, No. 109.

Peck, R. F., & Havighurst, R. J. The psychology of character

develOpment. New York: Wiley, 1960.

Piaget, J. The moral judgment of the child. New York: The
Free Press, 1965 (originally 1932).

Piaget, J. The origins of intelligence in children. New York:
Basic Books, 1954.

Sears, R. R. Relation of early socialization experiences to
aggression in middle childhood. Journal of Abnormal and Social
Psychology, 1961, 63, #66-492.

Sears, R., Rau, L., & Alpert, R. Identific§tionggpd child rearing.
Stanford, Ca1if.: Stanford University Press, 1965.

Sears, R. R., Whiting, J. W. M., Nowlis, V., & Sears, P. 3. Some
child-rearing antecedents of aggression and dependency in young
children. Genetic Psychologicgl Monographs, 1953, Q2, 135-234.

Sherif, M., Harvey, 0. J., White, B. J., Hood, W. R., & Sherif,
C. W. Intergroup conflict and 000pergtion: the robbers c532
experiment. Norman: University of Oklahoma Press, 1961.

Smedslund, J. The acquisition of conservation I. Introduction.
Scandgnavian Journgl of Psychology, 1961, g, 11-20.

Wilson, A. B. Residential segregation of social classes and
aspirations of high school boys. American Sociology Review,

1959. as. 836-845.

 

"11.111111111111111I