196 The Lucifer Effect
SUMMING UP BEFORE DIGGING INTO THE DATA
As you have seen, our psychologically compelling prison environment elicited in-
tense, realistic, and often pathological reactions from many of the participants.
We were surprised both by the intensity of the guards' domination and the speed
with which it appeared in the wake of the prisoner rebellion. As in the case of
Doug-8612, we were surprised that situational pressures could overcome most of
these normal, healthy young men so quickly and so extremely.
Experiencing a loss of personal identity and subjected to arbitrary continual
control of their behavior, as well as being deprived of privacy and sleep, generated
in them a syndrome of passivity, dependency, and depression that resembled what
has been termed "learned helplessness."
1
(Learned helplessness is the experience
of passive resignation and depression following recurring failure or punishment,
especially when it seems arbitrary and not contingent upon one's actions.)
Half of our student prisoners had to be released early because of severe emo-
tional and cognitive disorders, transient but intense at the time. Most of those
who remained for the duration generally became mindlessly obedient to the
guards' demands and seemed "zombie-like" in their listless movements while
yielding to the whims of the ever-escalating guard power.
As with the rare "good guards," so too, a few prisoners were able to stand up
to the guards' domination. As we have seen, Clay-416, who should have been
supported for his heroic passive resistance, instead was harassed by his fellow
prisoners for being a "troublemaker." They adopted the narrow dispositional per-
spective provided by the guards rather than generate their own metaperspective
on Clay's hunger strike as emblematic of a path for their communal resistance
against blind obedience to authority.
Sarge also behaved heroically at times by refusing to curse or verbally abuse
a fellow prisoner when ordered to do so, but at all other times he was the model
obedient prisoner. Jerry-486 emerged as our most evenly balanced prisoner; how-
ever, as he indicates in his personal reflections, he survived only by turning in-
ward and not doing as much as he might to help the other prisoners, who could
have benefited from his support.
When we began our experiment, we had a sample of individuals who did not
deviate from the normal range of the general educated population on any of the
dimensions we had premeasured. Those randomly assigned to the role of "pris-
oner" were interchangeable with those in the "guard" role. Neither group had
any history of crime, emotional or physical disability, or even intellectual or social
disadvantage that might typically differentiate prisoners from guards and prison-
ers from the rest of society.
It is by virtue of this random assignment and the comparative premeasures
that I am able to assert that these young men did not import into our jail any of
The SPE's Meaning and Messages 197
the pathology that subsequently emerged among them as they played either pris-
oners or guards. At the start of this experiment, there were no differences be-
tween the two groups; less than a week later, there were no similarities between
them. It is reasonable, therefore, to conclude that the pathologies were elicited by
the set of situational forces constantly impinging upon them in this prisonlike set-
ting. Further, this Situation was sanctioned and maintained by a background
System that I helped to create. I did so first when I gave the new guards their psy-
chological orientation and then with the development of various policies and pro-
cedures that I and my staff helped to put into operation.
Neither the guards nor the prisoners could be considered "bad apples" prior
to the time when they were so powerfully impacted by being embedded in a "bad
barrel." The complex of features within that barrel constitute the situational
forces in operation in this behavioral context—the roles, rules, norms, anonymity
of person and place, dehumanizing processes, conformity pressures, group iden-
tity, and more.
WHAT DID WE LEARN FROM OUR DATA?
The around-the-clock direct observations that we made of behavioral interac-
tions between prisoners and guards, and of special events, were supplemented by
videotaped recordings (about twelve hours), concealed audiotape recordings
(about thirty hours), questionnaires, self-reported individual difference person-
ality measures, and various interviews. Some of these measures were coded for
quantitative analyses, and some were correlated with outcome measures.
The data analyses present a number of problems in their interpretation: the
sample size was relatively small; the recordings were selective and not compre-
hensive because of our limited budget and staff, and because of the strategic deci-
sion to focus on daily events of high interest (such as counts, meals, visitors, and
parole hearings). In addition, the causal directions are uncertain because of the
dynamic interplay among guards and prisoners within and across guard shifts.
The quantitative data analysis of individual behavior is confounded by the obvi-
ous fact of the complex interactions of persons, groups, and time-based effects.
In addition, unlike traditional experiments, we did not have a control group
of comparable volunteers who did not undergo the experimental treatment of
being a mock prisoner or mock guard but were given various pre-post assess-
ments. We did not do so because we thought about our design as more a demon-
stration of a phenomenon, like Milgram's original obedience study, than as an
experiment to establish causal associations. We imagined doing such control-
versus-experimental group comparisons in future research if we obtained any
interesting findings from this first exploratory investigation. Thus, our simple
independent variable was only the main effect of the treatment of guard-versus-
prisoner status.
198 The Lucifer Effect
Nevertheless, some clear patterns emerged that amplify the qualitative nar-
rative I have presented thus far. These findings offer some interesting insights into
the nature of this psychologically compelling environment and of the young men
who were tested by its demands. Full details of the operational scoring of these
measures and their statistical significance is available in the scientific article pub-
lished in the International Journal of Criminology and Penology
2
and on the website
www.prisonexp.org
.
Personality Measures
Three kinds of measures of individual differences among the participants were
administered when they came in for their pre-experiment assessment a few days
prior to the start of the study. These measures were the F-Scale of authoritarian-
ism, the Machiavellian Scale of interpersonal manipulation strategies, and the
Comrey Personality Scales.
Do'stlaringiz bilan baham: |