The Importance of Erving Goffman to Psychological Anthropology

Author(s): Philip K. Bock

Source: Ethos, Vol. 16, No. 1 (Mar., 1988), pp. 3-20
Published by: Blackwell Publishing on behalf of the American Anthropological Association
Stable URL: http://www.jstor.org/stable/640563
Accessed: 10/11/2009 12:40

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at
http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless
you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you
may use content in the JSTOR archive only for your personal, non-commercial use.

Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at
http://www.jstor.org/action/showPublisher?publisherCode=black.

Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed
page of such transmission.

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact support@jstor.org.

Blackwell Publishing and American Anthropological Association are collaborating with JSTOR to digitize,
preserve and extend access to Ethos.

http://www.jstor.org
 The Importance of Erving
Goffman to Psychological
Anthropology

PHILIP K. BOCK

In the fall of 1963, the American Anthropological Association met

in San Francisco, California.' One of the highlights of that conven-
tion was to be the presentation by Erving Goffman of a paper on
"The Neglected Situation." Many anthropologists were familiar
with his book, ThePresentation of Self in EverydayLife (1959), or with
his early articles, and the small room in the Fairmont Hotel was
crowded as he began his talk. I was standing toward the back of the
hall and I recall being annoyed by the murmurs that threatened to
drown out his voice. The moderator passed him a note and Goffman
pausedjust as the rumor reached me: "President Kennedy has been
shot."
The room emptied quickly. My memories after that are unclear.
I recall walking the San Francisco streets, looking for a newspaper,
still hoping that the rumor would prove false. The convention con-
tinued, but after the death of the president was announced, a feeling
of unreality pervaded the proceedings. Participants spoke of their
"numbness," the feeling of "going through the motions" without
conviction that Goffman analyzed in Encounters(1961b) as aware-
ness of "role distance."

PHILIP K. BOCK is Presidential Professor of Anthropology at the University of New Mex-

ico, Albuquerque, and editor of the Journalof Anthropological
Research.

3
4 ETHOS

Most Americans can remember where they were when Kennedy

was shot, and I had many occasions to think of the events of Novem-
ber 1963. How fitting that the smooth functioning of a focused gath-
ering devoted to a lecture on "The Neglected Situation" should be
disrupted on that Friday and radically "re-framed" by events taking
place thousands of miles away. What better demonstration that our
social action is situated within numerous hierarchical and overlap-
ping frames, any one of which can suddenly become salient, affect-
ing thought, action, emotion, and the interpretation of ongoing
events (see Goffman 1974).
I continued to follow Goffman's work closely. At first I was in-
trigued by his elaboration of the role concept, and I incorporated
some of his ideas about social roles and behavior regions into my
own approach to the formal description of social structure (Bock
1964). His brilliant analysis of "spoiled identity" in Stigma (1963a)
also impressed me, and I tried to teach my ethnology students how
attention to problems of information control and identity manage-
ment could enrich their understanding of social life.
By the early 1960s, other anthropologists were also attempting to
"apply Goffman" to ethnographic problems. The most notable of
these was Gerald Berreman. In his monograph, Behind Many Masks
(1962), he discussed the ways in which his field experience in a Hi-
malayan village was altered when he had to change from a Hindu
to a Moslem interpreter. Berreman expected that the Hindu vil-
lagers would become more reticent in the presence of the Moslem
and he was thus surprised to find many of his informants becoming
much more open and talkative. Reflecting on this experience, he
found Goffman's ideas about "face," "front," and "backstage" use-
ful, and he suggested that ethnographers pay greater attention to the
selvesthat they and their informants try to present to one another.
American psychological anthropology of the early 1960s was
dominated by the cross-cultural correlational method, developed at
Yale by Whiting and Child (1953) and brought by John and Bea-
trice Whiting to the Laboratory of Human Development at Har-
vard. They and their students used an approach that, I have argued,
derives directly from the work of Kardiner and Linton on "basic
personality structure" (Bock 1980b: 150), combined with the corre-
lational methods of G. P. Murdock and applied to data from the
Human Relations Area Files. In this approach, causal hypotheses
linking "antecedent" conditions of childhood experience to "con-
 IMPORTANCEOF ERVINGGOFFMAN 5

sequent" beliefs and customs present in a society are tested by ex-

amining a sample of cultures described in the Files.
Mediating the antecedent/consequent relationship was a set of
assumptions about learning processes (Hullian) and personality dy-
namics (Freudian). In one well-known study by Whiting, Kluck-
hohn, and Anthony (1958), exclusive mother-child sleeping ar-
rangements (especially when combined with a long postpartum ta-
boo on resumption of parental intercourse) were shown to be asso-
ciated with painful initiation rites for males at puberty. Mediating
this correlation were the familiar dynamics of the Oedipal situa-
tion-anxiety and hostility, accentuated by the special experiences
of childhood.
Unfortunately, as in most studies using this approach, no data
whatsoever were presented (or even available) concerning individ-
uals or their social interaction. Rather, for a given sample of eth-
nographic cases an association was demonstrated between extended
periods of nursing with infants sharing the mother's bed (often ex-
cluding the father for one to three years) and male puberty rites in-
volving "tests of manliness" and painful genital operations. Even if
we accept this correlation as valid (and there are serious problems
of sample bias; see Bock 1967), the mediating processes are entirely
hypothetical. Some theorists who accept this approach emphasize
the need to "break"a boy's (presumed) excessive identification with
his mother, while others emphasize the functions of such initiation
rites in establishing male solidarity. (Each group can point to eth-
nographic facts and partial correlations supporting its position.)
Although the research I have just described is not a kind of work
to have greatly interested Goffman, there are at least three ways in
which his emphasis on "the neglected situation" has (or should
have) influenced this research tradition. First, the Whitings (espe-
cially in the "Six Cultures" study) and their later collaborators in-
creasingly placed stress on what they called the "learning environ-
ment in which the child grows up" (Whiting and Whiting 1974:4).
Many of their later studies used direct observations of situated in-
dividual behavior and included at least a record of who else was
present (parents, other adults, or children) at the time the obser-
vations were made. This increased sensitivity to situational varia-
bles may not be directly traceable to Goffman's influence, but it rep-
resents a healthy acknowledgment that behavior in public places
has both interactive and reflexive meaning.
6 ETHOS

Second, Goffman has taught us to examine various forms of in-

teraction, especially rituals, from dramaturgical and ethological
viewpoints. We now tend to analyze initiation rites in terms of iden-
tity management, asking about the dramatic means by which a clus-
ter of terrified little boys is transformed into a solidary group of
proud young men. How is the "show" staged? With what kinds of
"teamwork" and "collusion"? And for whose benefit (see Turner
1974:200-201)? We are especially curious about the "dark secrets"
of manhood and their function in producing a new self (see Herdt
1981).
The third influence derives from the phenomenological streak in
Goffman's work, especially in Asylums (1961 a), which has made
many of us curious about the situations in which classifications and
labelings are carried out (see MacCannell 1983). After all, it is acts
of categorization that produce the apparent correlations between
variables such as child-training practices and adult beliefs or cus-
toms. Who decidesthat a given society "has," say, a long postpartum
sex taboo, or a painful initiation rite, or a type of toilet-training
likely to produce a high level of "anal socialization anxiety"? This
question involves something different from the problem of "ethno-
graphic authority" (Clifford 1983). It calls for us to examine the
"backstage" processes of reading behavioral records, accounts, or
interview protocols, and reducing their often ambiguous language
to categories, scales, and computer codes. When we allow ourselves
to remember, we all know that the "dirty work" of social analysis is
frequently done by exploited personnel under various pressures
(and with varying degrees of knowledge of the hypotheses being
tested). It is further complicated when-as in at least some studies
I know of-graduate students' wives are employed (at minimal
wages) to do much of this work. Here is a "neglected situation,"
indeed!

SITUATION AND BEHAVIOR

As is well known, Goffman's work emerged (and then diverged)

from the social anthropology of Lloyd Warner and the occupational
sociology of Everett C. Hughes at the University of Chicago. I have
referred to this approach as "positionalism" (Bock 1980b:189),
viewing it as a refinement of social class analysis that took seriously
Marx's emphasis on the "conditions of work." Chicago students in-
 IMPORTANCEOF ERVINGGOFFMAN 7

vestigated the specific effects of occupational positions on the be-

havior and consciousness of their incumbents. In The Presentationof
Self in EverydayLife (1959), Goffman used many of these empirical
studies (for example, of morticians and musicians, cab drivers and
furniture salesmen) for his own purposes, incorporating them into
his dramaturgical approach.
The importance of this work to psychological anthropology can
best be appreciated when we first recognize two persistent assump-
tions that have limited research and understanding. One of these is
the "uniformity assumption," that iA, the notion that all or most
members of a society, be it as small as Western Samoa or as large as
Great Russia, must share a common, typical, "basic," or "modal"
personality. The persistence of this assumption in the face of all
kinds of counterevidence is truly remarkable. It is easily traced back
to ancient national and ethnic stereotypes that have been translated
into the jargon of personality theory. In the hands of naive anthro-
pologists it also involves a confounding of sociocultural patterns
with psychological ones, and the reification of both (see Bock 1983).
The founders of the Culture and Personality school in American
anthropology-Benedict, Sapir, Mead-were by no means naive
(Bock 1980b:57-82, 1984), but despite themselves, they frequently
fell into the error of neglecting intrasocietal variability and of assum-
ing a high degree of psychological sharing. Cora DuBois's useful
corrective concept of "modal personality" was not sufficiently rad-
ical to break down the uniformity assumption. It required the crit-
ical analysis of Anthony Wallace, who had worked within the modal
personality tradition, to point out the serious fallacy involved in
much of this research (Wallace 1961 :Ch. 4).
Wallace phrased his critique in terms of the empirical "distribu-
tion of psychological characteristics" in society; but Goffman's po-
sitionalism offered a further challenge to the uniformity assumption.
If our behavior and self-concept respond as rapidly and radically to
changes in our position within interaction situations as Goffman in-
dicates (especially in Asylumsand Stigma), it is nonsense to generalize
about the shared personality characteristics of members of any size
social group. This does not make psychological anthropology (or so-
cial psychology) impossible; it simply means that we must always
attend to situational variables. Significant commonalities in the be-
havior of groups of persons, whether they are "nationals," colonists,
peasants, bureaucrats, or mental patients, are more likely to be due
8 ETHOS

to their shared social position than to group personality or national

character.
A similar argument reveals the flaw in the "continuity assump-
tion" (see Lindesmith and Strauss 1950). This is the notion that
adult personality is necessarily determined by early childhood expe-
rience. Although the kinds of longitudinal studies that mightvalidate
hypotheses based on this assumption are sadly lacking (and often
conflicting when they do exist), true believers in such continuity are
still found in abundance. Goffman and other interactionists alert us
to the fact that apparent consistencies in behavior and character are
due to repeated involvements in similar interaction games as much
as to any underlying dispositions established in the early months or
years of life.
Goffman's own skepticism about depth psychology is quite clear;
yet there is a close affinity between his outlook on human action and
that of Sigmund Freud.2 I have gone so far as to write that Goffman
is to everyday interaction as Freud is to the individual psyche. Each
of them took seriously the "ordinary"phenomena within his chosen
domain, interpreting speech, behavior, and fantasy as signs of un-
spoken processes and conflicts. The title of Goffman's first book
seems, to me, a thinly veiled allusion to Freud's ThePsychopathology
of EverydayLife. In the opening pages of Behaviorin Public Places
(1963b), Goffman acknowledged, in a rather left-handed manner,
the mutual relevance of psychiatry, anthropology, and his distinc-
tive sociological perspective. The descriptive work of psychiatrists
who are concerned with behavior that is "inappropriate to the sit-
uation" can be useful, he felt, even though they attribute such de-
viance to intrapsychic factors rather than examining the social
norms that create "deviance" (see Jackson 1978).
Goffman is frequently cited in the psychiatric literature, usually
in relation to his concept of the "total institution" (for example,
Douvan 1974:23), or (with Szasz and Scheff) as a proponent of "la-
beling theory" (for example, Begelman 1971:38;Roman 1971). I am
not able to evaluate his actual influence on psychiatric practice,
though it appears that his work has stimulated some rethinking of
institutional arrangements and staff/patient relationships (for ex-
ample, Rockwell 1971; Stannard 1973).3
The growing interest in the concepts of "self" and "identity" has
also led to increasing use of (or at least references to) Goffman's
early works. For example, in Ethos,the journal of the Society for Psy-
 IMPORTANCEOF ERVINGGOFFMAN 9

chological Anthropology, during the five years from 1980 through

1984 (vols. 8-12), more than 10 percent of the published articles
cited works by Goffman. Some of these citations were only passing
references to a concept or idea, but others showed how deeply his
approach has penetrated the thinking of anthropologists. For ex-
ample, in the special issue devoted to "Self and Emotion" (Levy and
Rosaldo 1983), Edward L. Schieffelin described the public situa-
tions within which the Kaluli of Papua New Guinea express grief,
anger, or shame. Expression of anger and shame rest on shared as-
sumptions about legitimacy and reciprocity. "Thus Kaluli emo-
tions, however privately experienced . . . are socially located and
have a social aim. To this degree they are located not only in the
person, but in the social situation and interaction which, indeed,
they help construct" (1983:190-191). In his analysis of Kaluli situ-
ations of shaming, Schieffelin draws on Goffman's InteractionRitual
(1967). He particularly uses the concepts of "face" and "line," in-
dicating the importance of the "flow of events in the encounter"
(Schieffelin 1983:188) to understanding the social basis of emotion:
Shame is revealed here as a situation as much as a private emotion: both its exis-
tence and meaning are products of interaction, not loss of reputation or a sense of
right and wrong. Like grief, shame involves a sense of vulnerability; but unlike
grief, which has a claim to social support, shame has none. It representsa situation,
or a state, of powerlessness and rejection. The legitimacy of one's basic posture of
assertion or of appeal [that is, the "line" one has adopted] has been removed.
[Schieffelin 1983:189;cf. Nachman 1984]

Goffman has been used in similar ways by cultural anthropolo-

gists in 5 of the 12 articles that make up a recent collection on
"Mind, Self, and Emotion" (Shweder and LeVine 1984). In that
volume, Shweder and Bourne, for instance, comment that it is
"tempting to argue that Western individualism has its origins in the
institution of privacy." However, they continue,
It is sobering to acknowledge that our sense of personal inviolatability [sic] is a
violatable social gift, the product of what othersare willing to respect and protect us
from, the product of the way we are handled and reacted to, the product of the
rights and privileges we are granted by others in numerous "territoriesof the self'
(Goffman 1971). [Shweder and Bourne 1984:194]
There are, of course, different kinds of privacy, and the need of the
self for a backstage area (especially in this electronic age), is a point
that Goffman made in several contexts. My own field experience in
a Mexican pueblo (see Bock 1980a) made me particularly aware of
10 ETHOS

the distinction between visual and auditory privacy. The villagers

displayed, I thought, a high degree of sensitivity to visual intrusions:
high walls face the streets for security, but when a house was occu-
pied, doors and curtains were frequently left open on the assumption
that no proper person would deliberately look in. (Until I learned
to obey this norm, my curious glances were met with shock or sur-
prise.) Auditory privacy was another matter entirely: the sounds of
voices or music and the noise of fireworks might invade one's audi-
tory space at any time, and the villagers appeared to take little no-
tice. (After numerous interruptions of work and sleep, it was with
considerable satisfaction that, at 2 A.M. on my final night in the
pueblo, I set off a barrage of large rockets that I had accumulated
for the occasion.)

DRAMATIC REALIZATION

Privacy is a matter of great concern to the Mehinaku Indians of

central Brazil. They have been studied for nearly 20 years by
Thomas Gregor, a psychological anthropologist trained at Colum-
bia University. In his first book, Mehinaku (1977), Gregor drew on
Goffman's early work in order to describe "The Drama of Daily
Life" in the villages. He showed th6 Indians as self-conscious role
players and he discussed how the layout and architecture of their
villages-the men's house, the residences, paths, and open areas-
contribute to the "theatrical" quality of Mehinaku life.
The Mehinaku villager must... be skilled in performanceand stagecraft. The ba-
sis of this requirement is that Mehinaku social life is highly observable, the gossip
network extremely effective, and privacy very scarce. In such an environment a
man's identity as a good citizen may be jeopardized by a reality that is difficult to
conceal. His laziness, his unwillingness to share food, his failure as a hunter or fish-
erman, his activities as a thief or adulterer, his distrust of his fellows, or even his
sexual inadequacy-all can quickly become public knowledge.... The openness
of the community to the flow of information may thereby endanger vital social re-
lationships. [Gregor 1977:358]
The preoccupation of Mehinaku adults with extramarital sex was
clear. Gregor (1977:93) described the "alligator places" (yaka epuga)
near the villages where a lover lies in wait for his sweetheart, some-
times for hours, until she can slip away from her chores for a quick
tryst. Adultery is considered the spice of life ("pepper" is the Me-
hinaku metaphor); but it may be severely punished, so lovers face
the dramaturgical problem of appearing faithful to their spouses-
 IMPORTANCEOF ERVINGGOFFMAN 11

difficult in this small community, especially if one attempts inter-

course in a hammock a few inches away from one's sleeping hus-
band!
In his recent book, Anxious Pleasures (1985), Gregor writes in great
detail about the sexual lives of the Indians. He has dropped the dra-
maturgical framework of the first book (indeed, there are no refer-
ences whatever to Goffman), substituting a Freudian analysis; but
I shall draw mainly on Anxious Pleasures to suggest the continuing
relevance of Goffman's situational approach to psychological issues.
As Gregor observes,
The public life of Mehinaku society is men's life. Acted on the plaza, the men's
space, it is composed of men's oratory, men's songs, men's rituals, men's wrestling
matches. Women have a culture that parallels that of men, but its public displays
are pale reflections of a masculine image. [1985:110-111]

Mehinaku women endure much obscene teasing. The most im-

portant men's rituals, those involving the sacred flutes and bull-
roarers, function in part to intimidatethe women, who are forbidden
to see the flutes on pain of gang rape. Although this punishment has
not actually been carried out for over 40 years, it is still feared by
the women, and assertive females can be put in their place merely
by rumors that they may have seen the sacred flutes (Gregor
1985:110).
As in many other tribal societies, however, there exists a myth
that, in ancient times, women "were matriarchs, the founders of
what is now the men's house and creators of Mehinaku culture"
(Gregor 1985:112). In a violent revolution, men usurped these in-
stitutions and dominated the women; but other myths, customs, and
above all men's dreams, reveal their persisting anxieties: "fear of
their own sexual impulses and fear of women" (1985:115). Gregor
illustrates the theme of "punishment for sexual wishes" by the myth
of Katsi, a man who, against repeated warnings, had intercourse
with a beautiful, sensual woman who was really a lizard spirit. As a
consequence, his penis grew so long that he had to weave four large
bags in which to carry it about with him; it was given the name "Ka-
pukwa," and took on a life of its own.
At night, when everyone was asleep, Kapukwa woke up. He slithered out of his
sacks and along the floor of the house. He opened the door, snaked across the plaza,
and entered each of the village houses. He slid up the house poles, wound his way
along the hammock cords, and slipped into the vaginas of the sleeping women. But
none of the women knew that Kapukwa was having sex with them.
12 ETHOS

Finally, one man saw what Kapukwa was doing. He told his friends, and on the
following night, they were ready with clubs. When Kapukwa emerged from his
house, they attacked, beating him on the head with heavy sticks. With each blow
he shrank back, until Katsi was left with a tiny, tiny penis. [1985:133]

The final sentence of the myth is accompanied by a gesture: "the

narrator holds up a thumb and forefinger to frame a space of about
one quarter of an inch-the size of Katsi's shriveled and mutilated
penis" (Gregor 1985:135). Gregor believes that "the story of Katsi
and Lizard Woman suggests that a man's sexual impulses may lead
to harm to his penis and even to castration" (1985:135), and he pre-
sents two other myths on the same theme.
It is not my intention to deny the castration fears that are ex-
pressed by these myths; however, I believe that situational analysis
complementsa psychodynamic approach. In particular, Goffman's
concept of "gender displays" (1979:1-2) can help us to understand
what is at stake in the ritual events that characterize Mehinaku so-
ciety. To be sure, the men's anxieties and their antagonism to
women, including their belief in the dangerous qualities of men-
strual blood, have an unconscious basis (Bettelheim 1954; also,
Bock 1967). But as Goffman pointed out,
The expression of subordination and dominance through . . . situational means is
more than a mere tracing or symbol or ritualistic affirmationof the social hierarchy.
These expressions considerably constitute the hierarchy, they are the shadow and
the substance. [ 1979:6]

The selected photos and drawings in Gregor's Anxious Pleasuresare

not really suitable for interpretation in the ways that Goffman used
candid and posed pictures in GenderAdvertisements(1979); however,
if we examine the six pictures in which Mehinaku adults of both
sexes are shown together, it appears that some of Goffman's gener-
alizations may apply cross-culturally-at least through the eyes and
camera of an American ethnographer.
In Figure 21 of Anxious Pleasures, a large male with extended arm
is shown giving instructions to a group of women dancers in the
background (Gregor 1985:111; cf. Goffman 1979:54). The native
drawings reproduced by Gregor on pp. 122-127 show various ritu-
als of the pequi harvest season, during which men and women take
turns attacking one another, both symbolically and physically. For
example, in the Kiriri ritual, women pour foul liquids on the men
who are imitating crickets, and the drawing shows one attacking
woman as much larger than any of the men (p. 123), whereas in the
 IMPORTANCE OF ERVING GOFFMAN 13

drawing of Hopa (the scorpion ritual), the attacking men (who squirt
water from their mouths) are twice as large as the fleeing women (p.
126; cf. Goffman 1979:28).4
A more general point here is that, among the Mehinaku (as in the
United States), male dominance is expressed in the idiom of adult/
child interaction: "ritually speaking, females are equivalent to sub-
ordinate males and both are equivalent to children" (Goffman
1979:5). In the Amazon village, men show their superiority by ex-
cluding women from the clubhouse, by confining them to the resi-
dences when the sacred flutes are paraded, and by repeated teasing,
insulting, and ordering about. These ritual degradations (to which
children are subjected by all adults) are backed by the threat of
physical force in the form of wife beating and gang rape. The women
are permitted periodic retaliation, but though they are "a force to
be reckoned with," it is clear to all that, "in the battle of the sexes,
women are the losers" (Gregor 1985:110, 130).

STRUCTURAL RESTATEMENT

In my own theoretical work, I have consistently used the recurrent

situation as the fundamental unit in a formal description of social
structure (see Bock 1986). Where others have taken the role, the re-
lationship, or the "actone" (Harris 1964) as basic units, I have fol-
lowed Goffman in selecting the more complex situation, which I de-
fine as the expectedoccurrenceof a class of roles within a setting. Settings
are formed by the intersection of a dimension of social space with a
dimension of social time. Contrastively defined roles, areas of space,
and periods of time are described in linear formulas and in two-di-
mensional displays, yielding a highly economical account of social
structure that removes a great deal of ambiguity from the formal
descriptions. (There is, of course, a cost for this gain in precision.)
I will not here attempt to present my "descriptive model" in any
detail, much less to defend it. However, I would like to offer a brief
demonstration of its use in "structural restatement" (Bock 1986:45-
49), so the reader can judge whether it is worth the effort to master
the notation and follow the arguments in my monograph.
I shall use a brief anecdote told me by a graduate student several
years ago as data. This student was an expert skier and supported
himself by giving instruction at a local ski resort. When "on the
slopes," he was a figure of unquestioned authority and high prestige;
14 ETHOS

however, to make extra money, he also worked in the evenings as a

busboy in the dining room of the lodge. He reported (truthfully, I
believe) that when he attempted to serve diners who had been his
students earlier that day, they became extremely flustered: on first
recognizing him, many of them would actually rise from their seats
and try to help him pour water or clear the table!
Goffman would have loved this story. He might have used it to
illustrate a breakdown in role segregation with a consequent loss of
"idealization" (1959:34) due to the divergent values assigned to my
student's two performances. In my formal description of this situa-
between persons in the two
tion I would first focus on the transactions
relationships, instructor/student and busboy/diner. Using the set of
"fundamental role attributes" that I have proposed for the purpose
(Bock 1986:37), I would define these relationships contrastively as
follows:
R1, [gives orders; receives money]/R2;
R3, [provides care; shows deference]/R4.
My formal model requires that these relationships be situatedin
their respective interaction settings (see Table 1): "on the slopes"
during the daytime contrasts with "in the lodge" at night. The two-
dimensional display shows these settings as part of a larger situa-
tion-matrix (M), and indicates a few of the other role, space, and
time units that would be used in a fuller description of the ski resort.
At least some of these units would be relevant to any instance of face-
to-face interaction for, as I have argued elsewhere (Bock 1966), hu-
man behavior takes place in actual situations but also with reference

TABLE 1
ROLERELATIONSHIPS
SITUATED

Daily Cycle >

M: The Ski Resort T1, Daytime T2, Evening
Ski Area > SI, The Slopes R1/R2
S2, Lift Area
S3, Snack Bar, and so on
The Lodge > S4, Dining Room R3/R4
S5, Kitchen
S6, Lobby
S7.n, Guest Rooms (1-n)
 IMPORTANCEOF ERVINGGOFFMAN 15

to past or future situations, real or imagined (cf. Goffman 1967:98;

Brittan 1973:90).
In the structural restatement of our anecdote, a feature of my "ex-
tended model" (1986:31-40) can be used to display the role conflict
that led to the diners' discomfort. "P-notation" allows us to specify
the actual set of persons, p(Rm), or the particular individuals,
pi(Rn) who perform a given role. The notation may also be extended
to units of space and time when it is desired to specify particular
places, p(Si), or periods, p(Tj), rather than categories. In this case,
the formal statement would be something like:

Sl/Tl:pi (Rl, Ski Instructor)/p(R2, Students);

S4/T2:pi (R3, Busboy)/p(R4, Dinner Guests); but,
pi(Rl) = pi(R3), and prestige of Rl > R2 = R4 > R3!
The peculiarity of the interaction in the second situation (S4/T2)
is understandable only with reference to the structure of the rela-
tionships in the previous situation (S 1/T 1); both of these are encom-
passed within the total situation-matrix (M). Any person perform-
ing the role of busboy, p(R3), takes on a mildly stigmatized social
identity and may be treated as a nonperson; but if his personal iden-
tity becomes salient (as in the anecdote), his recognition as a per-
former of a role with contrasting attributes (high prestige) produces
cognitive dissonance, uncertainty, and a type of situational emotion
("shame"?) similar to that described for the Kaluli by Schieffelin
(1983:189; see above, p. 7).
In the fully elaborated version of my model (Bock 1986:65-73),
entire classes of situations are described as surface structures gen-
erated from underlying deep structures via transformation and in-
sertion rules. Particular interactions can then be shown as realiza-
tions of a limited number of unconscious structures that impose con-
straints on what can be done, just as unconscious grammatical
forms constrain our speech, making possible communication among
persons who share the same "code." The formalization of these con-
straints on interaction permits an economical and rather elegant
statement of the structure of situations and institutions. If I am cor-
rect, learning a social structure
involves the acquisition of a few "deep structures" and the rules for generating a
wide range of more specific, recurrent situations. Human beings are "designed" to
learn such structures, and they interpret their experience, for the most part, by
referring it to shared expectations which mav be economically represented by ma-
16 ETHOS

trices. It is at this level that, I believe, the analogy between language structure and
social structure can be most productively pursued. [Bock 1986:73]

LECTURE AND TEXTURE

In closing, I want to return to the anecdote with which I began

this paper. Goffman's lecture on "The Neglected Situation" was
published the year following the San Francisco meetings in a special
issue of AmericanAnthropologist(Gumperz and Hymes 1964). The
printed version contains no reference to the Kennedy assassination
or to the interruption of its author's presentation. This is not un-
usual, for as Goffman pointed out in his introduction to "The Lec-
ture" ( 1981:161), situational features are omitted "in almost all con-
versions from talk to print." Nevertheless, it is somewhat surprising
that a person as situation-conscious as Erving Goffman would com-
pletely neglect the opportunity to examine issues that he would later
analyze in terms of "framing" (1974) and "footing" (1981:124-
159). Can we recover some of these features from the printed text?
Let us begin with Goffman's definition:
A lecture is an institutionalized extended holding of the floor in which one speaker
imparts his views on a subject, these thoughts comprising what can be called his
"text." The style is typically serious and slightly impersonal, the controlling intent
being to generate calmly considered understanding, not mere entertainment, emo-
tional impact, or immediate action. Constituent statements presumably take their
warrant from their role in attesting to the truth, truth appearing as something to
be cultivated and developed from a distance, coolly, as an end in itself. [1981:165]

Before he is through, Goffman has made us aware of the various

issues of framing, noise, accessibility, and so on, that are usually out
of consciousness. He also points out an important presupposition
that "supports the notion of intellectual authority in general: that
through the statements of a lecturer we can be informed about the
world" (1981:195). At the same time, his delicate irony undercuts
this assumption, deconstructing his own text.
The printed version of "The Neglected Situation" is brief-fewer
than four full pages. Its style is informal though not quite conver-
sational: "I'm sure these two currents of analysis-the correlational
and the indicative-could churn on forever (and probably will), a
case of scholarly coexistence" (1964:133). I find it impossible to re-
construct with any confidence what kinds of changes have been
made from the typescript used for the original talk, though the single
 IMPORTANCEOF ERVINGGOFFMAN 17

footnote referring the reader to the just-published Behavior in Public

Places (1963b) was almost surely a later addition.
Goffman seems to be addressing an audience that was just becom-
ing aware of the "ethnography of communication" as a possible field
of studies. He cautions them that "once you cross the bridge [from
speaking to social conduct], you become too busy to turn back"
(1964:134). The truth of this observation became increasingly ap-
parent with the establishment in the next few years ofsociolinguistics-
a new discipline that responded to Goffman's rhetorical question,
"Where but in social situations does speaking go on?" (1964:134).
Goffman's own speaking about the "neglected situation" first
"went on" in a situation that was disrupted by national events; the
printed version, however, is framed by a situation-matrix that has
its nucleus in Berkeley, not in Dallas. Of the 12 contributors to The
Ethnographyof Speaking, 7 indicated University of California-Berkeley
as their institutional affiliation, and earlier meetings under the aus-
pices of the Kroeber Anthropological Society are alluded to in the
preface to that volume (Gumperz and Hymes 1964:v). The Berkeley
connection is also betrayed by a peculiar clause in Goffman's arti-
cle-one that beautifully illustrates the interpenetration of situa-
tion, content, and medium. The full sentence reads: "It can be ar-
gued that social situations, at least in our society, constitute a reality
suigeneris as He used to say, and therefore need and warrant analysis
in their own right, much like that accorded other basic forms of so-
cial organization" (1964:134).
I cannot recall whether "as He used to say" was in the original
spoken version or whether Goffman substituted a proper name for
the indexical pronoun. In either case, this passage shows that the
speaker was "alive" to his audience for, in 1963, meeting in the
shadow of the Bay Bridge, any group of anthropologists would
surely have known that the "He" who spoke of culture as "a reality
sui generis" had to be the late Alfred L. Kroeber (d. 1960). But they
could not have known, any more than you could, that this third-
person pronoun was capitalized!

LEARNING FROM GOFFMAN

To conclude: The legacy of Erving Goffman is extremely impor-

tant to psychological anthropology as it tries to escape from invalid
assumptions about the uniformity and continuity of human behav-
18 ETHOS

ior, as it attempts to understand the public character of emotion and

affect, and as it searches for more effective ways to relate the abstrac-
tions of social structure to real persons acting out their relationships
in concrete situations. Whether we work in modern urban society or
in the villages of India, New Guinea, Mexico, or the Amazon, we
must never neglect the situation or its larger institutional frame. To
paraphrase Terry Eagleton (who was writing of Shakespeare), "It
is not a matter of learning from [Goffman] in any simple way: our
formulation of the problem is not his, and our experience of it is
therefore different. But to understand his attempts to grapple with
the difficulties is inevitably to deepen our own understanding, and
to recognize new ways forward" (Eagleton 1967:11).

NOTES

'An earlier version of this paper was presented at the Memorial Conference on the work of
Erving Goffman, University of York, England, July 1986. James Sebring offered useful com-
ments on the article.
2Randall Collins (1980:184) appears to agree.
3According to the Social Science Citation Index, during the 1970s, Goffman's work was
cited about equally in psychiatric, sociological, and anthropologicaljournals, with increasing
referencesin linguistic and communication journals later in the decade.
4These size differences may, however, be attempts to render perspective, for in some other
drawings, on pp. 125 and 127, distance is indicated by interposed forms. The only photo of a
father and daughter together is on p. 175.

REFERENCES

BEGELMAN, D. A. 1971. Misnaming, Metaphors, the Medical Model, and

Some Muddles. Psychiatry34:38-58.
BERREMAN, G. D. 1962. Behind Many Masks. Ithaca, NY: Society for Applied
Anthropology Monograph No. 4.
BETTELHEIM, B. 1954. Symbolic Wounds.Glencoe, IL: The Free Press.
BOCK, P. K. 1964. Social Structure and Language Structure. Southwestern
Jour-
20:393-403.
nal ofAnthropology
1966. Social Time and Institutional Conflict. HumanOrganization
25:96-102.
1967. Love Magic, Menstrual Taboos, and the Facts of Geog-
raphy. AmericanAnthropologist69:213-217.'
. 1980a. Tepoztlain Reconsidered. Journalof Latin AmericanLore
6:129-150.
1980b. Continuitiesin PsychologicalAnthropology.San Francisco:
Freeman.
Re-
1983. The Samoan Puberty Blues. Journalof Anthropological
search39:336-340.
 IMPORTANCE OF ERVING GOFFMAN 19

1984. Edward Sapir as Psychological Anthropologist. Paper pre-

sented at the Annual Meeting of the American Anthropological Association,
Washington, D.C.
1986. TheFormalContentof Ethnography. Dallas, TX: International
Museum of Cultures Publication 20.
BRITTAN, A. 1973. Meaningsand Situations.London: Routledge and Kegan
Paul.
CLIFFORD, J. 1983. On Ethnographic Authority. Representations 2:118-144.
COLLINS, R. 1980. Erving Goffman and the Development of Modern Social
Theory. TheViewfromGoffman(J. Ditton, ed.), pp. 170-209. London: Macmillan.
DOUVAN, E. 1974. Commitment and Social Contract in Adolescence. Psy-
chiatry37:22-36.
EAGLETON, T. 1974. Shakespeare andSociety.New York: Schocken Books.
GOFFMAN, E. 1959. ThePresentation of Self in EverydayLife. Garden City, NY:
Anchor Books.
1961a. Asylums.Garden City, NY: Anchor Books.
1961b. Encounters. Indianapolis, IN: Bobbs-Merrill.
1963a. Stigma.Englewood Cliffs, NJ: Prentice-Hall.
1963b. Behaviorin PublicPlaces.Glencoe, IL: The Free Press.
1964. The Neglected Situation. TheEthnography of Communication
(J. Gumperz and D. Hymes, eds.), pp. 133-136. Special Publication of American
Anthropologist66 (6), Part 2.
Ritual.Garden City, NY: Anchor Books.
1967. Interaction
1971. Relationsin Public.New York: Basic Books.
1974. FrameAnalysis.New York: Harper and Row.
1979. GenderAdvertisements. New York: Harper and Row.
1981. The Lecture. Formsof Talk,pp. 160-196. Philadelphia: Uni-
versity of Pennsylvania Press.
GREGOR, T. 1977. Mehinaku:TheDramaof Daily Life in a BrazilianIndianVil-
lage.Chicago: University of Chicago Press.
1985. AnxiousPleasures:TheSexualLivesof anAmazonianPeople.Chi-
cago: University of Chicago Press.
GUMPERZ, J., and D. HYMES, eds. 1964. TheEthnography of Communication.
Special Publication of AmericanAnthropologist 66(6), Part 2.
HARRIS, M. 1964. TheNatureof CulturalThings.New York: Random House.
HERDT, G. 1981. Guardiansof the Flutes: Idioms of Masculinity.New York:
McGraw-Hill.
JACKSON, B. 1978. Deviance as Success. The ReversibleWorld(B. Babcock,
ed.), pp. 258-275. Ithaca, NY: Cornell University Press.
LEVY, R., and M. ROSALDO, eds. 1983. Self and Emotion. Ethos(special is-
sue) 11(3).
LINDESMITH, A. R., and A. L. STRAUSS. 1950. A Critique of Culture-Per-
sonality Writings. AmericanSociologicalReview15:587-600.
MACCANNELL, D. 1983. Erving Goffman (1922-1982). Semiotica45:1-33.
NACHMAN, S. R. 1984. Shame and Moral Aggression on a Melanesian Atoll.
Journalof PsychoanalyticAnthropology7:335-365.
ROCKWELL, D. 1971. Some Observations on "Living In." Psychiatry34:214-
223.
ROMAN, P. 1971. Labeling Theory and Community Psychiatry. Psychiatry
34:378-390.
SCHIEFFELIN, E. L. 1983. Anger and Shame in the Tropical Forest: On Af-
fects as a Cultural System in Papua New Guinea. Ethos 11:181-191.
20 ETHOS

SHWEDER, R., and E. J. BOURNE. 1984. Does the Concept of the Person
Vary Cross-culturally? CultureTheory:Essays on Mind, Self and Emotion (R.
Schweder and R. LeVine, eds.), pp. 158-199. Cambridge: Cambridge University
Press.
SHWEDER, R., and R. LEVINE, eds. 1984. CultureTheory:EssaysonMind,Self
andEmotion.Cambridge: Cambridge University Press.
STANNARD, D. L. 1973. Ideological Conflict on a Psychiatric Ward. Psychia-
try36:143-156.
TURNER, V. 1974. Drama,Fields,andMetaphors:SymbolicActionin HumanSoci-
ety.Ithaca, NY: Cornell University Press.
WALLACE, A. F. C. 1961. CultureandPersonality.New York: Random House.
WHITING, B. B., andJ. W. M. WHITING. 1974. Childrenof Six Cultures.Cam-
bridge, MA: Harvard University Press.
WHITING,J. W. M., and I. CHILD. 1953. ChildTrainingandPersonality.New
Haven, CT: Yale University Press.
WHITING, J. W. M., R. KLUCKHOHN, and A. ANTHONY. 1958. The
Function of Male Initiation Rites at Puberty. Readingsin SocialPsychology(3rd
edition) (E. Maccoby, T. Newcomb, and E. Hartley, eds.), pp. 359-370. New
York: Holt.