Goffman on Surface and Interior

Goffman on Surface and Interior

Thomas J. Scheff

Studies of human conduct usually focus on either surface or interior, but not both. Behaviorists emphasize surface, psychoanalysts, the interior. Social science concentrates on either one or the other, with attention to surfaces predominating. I show that Goffman touched both bases, an attractive feature of his work. Less obviously, his leaning toward positive science was another. In his attempt to explicitly define the concepts of intersubjectivity and embarrassment, he may have taken a step toward an effective social science, one that that links surface and interior.

Goffman’s Way

If one goes by the indicator of attracting readers, Erving Goffman was easily the most successful social scientist of this or perhaps any century. His first book sold a half million copies, and is still selling. His work has also excited voluminous responses from scholars, many of which are now collected in four volumes (Fine and Smith 2001). Without necessarily urging that his work should be a model for social science, it might be of interest to examine how he dealt with the surface/interior problem.

At first glance, it might seem that he was concerned with surfaces. Much of his work is a description of behavior in social interaction. His most widely read book, Presentation of Self in Everyday Life (1958), begins with a quote:

"…Living things in contact with air must acquire a cuticle, and it is not urged against cuticles that they are not hearts…(Santanyana)

This quotation is a tease, it seems to suggest that the book will deal with the surface of behavior, the human cuticle. Suggestive also is the way that Goffman often attempted to avoid the psychology of individuals, focusing instead on generic social components of behavior. Still more direct are his challenges to the Western idea of the self as an isolated, self-contained unit. Indeed, this challenge is the main thread that connects much of his writing. The alternative that he presents is that the origins of human conduct lie less within the self than in social arrangements, especially patterns of social control, on the surface of everyday life.

But a close reading suggests that Goffman explored hearts and minds as well as hands. In particular, there are two interior components that are either explicit or implied in his descriptions of social interaction: embarrassment or its anticipation, on the one hand, and thoughts about the thoughts of others (intersubjectivity), on the other. These two interior components are closely related: feeling and avoiding embarrassment implies thoughts about others’ thoughts about self.

Goffman’s instinct was to escape entrapment in conventional viewpoints. Becker (1999) has shown how Asylums (1961) invented a new set of concepts that would avoid suppositions about madness hidden in ordinary language. Here I show how he escaped entrapment in surface and interior approaches by combining elements from both.


Unlike most social scientists, Goffman explored emotions as well as thoughts and actions. However, there is a difficulty. An immediate sticking point is that most of Goffman’s treatment of feeling concerns only one emotion, embarrassment. This emotion plays an important part in most of his studies, especially the earlier ones, both explicitly, and in much larger scope, by implication. But why only one emotion? What about other primary emotions, such as love, fear, anger, grief, and so on?

To the average reader, the exclusive focus on embarrassment might seem arbitrary. An exception is the great English comic writer Allan Bennett, who appears to take Goffman’s way with emotions in his stride. He sums it up in this way: "We must love one another or die – of embarrassment (2001, p. 353)." This short sentence packs a lot of information: what Goffman has left out (love) and what he has included (embarrassment). It also wittily alludes to a 60’s song by Crosby, Young, Stills and Nash: "we must love one another or die." But Schudson’s reaction (1984), I think, is more typical. He devoted an entire article to questioning Goffman’s exclusive concern with embarrassment.

Explicitly, Goffman gave only one justification. He argued that embarrassment had universal, pancultural importance in social interaction:

Face-to-face interaction in any culture seems to require just those capacities that flustering seems to destroy. Therefore, events which lead to embarrassment and the methods for avoiding and dispelling it may provide a cross-cultural framework of sociological analysis (1956, 266).

Heath (1988 137) further justifies Goffman’s focus:

Embarrassment lies at the heart of the social organization of day-to-day conduct. It provides a personal constraint on the behavior of the individual in society and a public response to actions and activities considered problematic or untoward. Embarrassment and its potential play an important part in sustaining the individual’s commitment to social organization, values and convention. It permeates everyday life and our dealings with others. It informs ordinary conduct and bounds the individual’s behavior in areas of social life that formal and institutionalized constraints do not reach.

Beyond these considerations, there is another, broader one that is implied in Goffman’s ideas, particularly the idea of impression management. Most of his work implies that every actor is extraordinarily sensitive to the exact amount of deference being received by others. Even a slight difference between what is expected and what is received, whether the difference be too little or too much, can cause embarrassment and other painful emotions.

In an earlier article (Scheff 2000), I followed Goffman’s lead by proposing that embarrassment and shame are primarily social emotions, because they usually arise from a threat to the bond, no matter how slight. In my view, attunement, the degree of social connectedness, of accurately taking the viewpoint of the other without judging it, is the key component of social bonds. A discrepancy in the amount of deference conveys judgement, and so is experienced as a threat to the bond. Since even a slight discrepancy in deference is sensed, embarrassment or the anticipation of embarrassment would be a virtually continuous presence in interaction.

In most of his writing, Goffman’s Everyperson was constantly aware of her own standing in the eyes of others, implying almost continuous states of self-conscious emotions: embarrassment, shame, humiliation, and in rare instances, pride, or anticipation of these states. Their sensitivity to the eyes of others, and to the judgments of self they imply make Goffman’s actors seem three dimensional, since they embody feeling as well as thought and behavior.

Goffman on Mindreading

The focus on embarrassment as a response to the views that others have of self also implies the second interior component of Goffman’s analysis, intersubjectivity. His work suggests that we spend much of our life living in the minds of others. In this respect, he follows in the footsteps of Cooley (1922), G.H. Mead (1936), and Blumer (1986). What might be called mutual mindreading was central to their perspectives. Mead called it "taking the role of the other."

Mead’s description of taking the role of the other initially gives the impression that he is referring to role behavior. Indeed, he sometimes uses the phrase in that way: in order to coordinate one’s actions with another, say in dancing a tango, one needs to learn not only one’s own role, but also the role enactment of one’s partner.

But in reading further, it becomes clear that Mead is referring not only to behavior, but also, more frequently, the perspective and thoughts of others as well. The concept of "taking the role of the generalized other" clearly means that one takes on the perspective of an imagined person or group of persons, even a fictitious group. Similarly, his definition of a social institution involves each participant knowing not only her own perspective, attitudes, and actions, but also those of the other participants. He gives the example of the institution of private property. To steal a purse effectively, a thief must know the perspectives of the owner, the police, the judge, etc. Mead’s theory of role-taking clearly involves the concept of intersubjectvity, the sharing of subjective states by two or more individuals.

Cooley’s idea of social life was also built around intersubjectivity. But he carried the implications of the idea further than Mead. Cooley argued that intersubjectivity is so much a part of the humanness of human nature that most of us take it completely for granted, to the point of invisibility:

As is the case with other feelings, we do not think much of it [that is, of social self-feeling] so long as it is moderately and regularly gratified. Many people of balanced mind and congenial activity scarcely know that they care what others think of them, and will deny, perhaps with indignation, that such care is an important factor in what they are and do. But this is illusion. If failure or disgrace arrives, if one suddenly finds that the faces of men [sic] show coldness or contempt instead of the kindliness and deference that he is used to, he will perceive from the shock, the fear, the sense of being outcast and helpless, that he was living in the minds of others without knowing it, just as we daily walk the solid ground without thinking how it bears us up (Cooley 1922, 208)

This idea is profoundly significant if we are attempting to understand Goffman’s work. Intersubjectivity is so built into our cultural make-up that it will usually be virtually invisible. It follows from it that we should expect that not only laypersons but even most social scientists will avoid explicit consideration of intersubjectivity.

Although human communication is built upon intersubjective accord, it is learned so early in infancy it goes unmarked in most discourse. Occasionally it will be referred to, but only casually and in passing. For example, one might say to a friend, "We both know that….". The idea occurs more elaborately in the popular song (from the 30’s?) whose lyric was something like:

I know that you know that I know that you know….[that we’re in love?] (I recall this song from my childhood, but so far have been unable to locate it .)

The reference in the song to the cascading levels of reciprocating mindreading is only jokey. But the same idea of knowing that the other knows, etc. was the basis for one of my early articles (Scheff 1967) on a sociological model of consensus. I proposed that the concept of consensus implies these cascades, but they are seldom acknowledged, and have never been explored. Probably by coincidence, Goffman’s Strategic Interaction (1969) took up this very issue. I don’t think I was aware of his forthcoming book at the time I wrote the article.

An example of scholars taking mindreading in humans for granted occurs in a recent treatise (O’Connell 1998). The author reviews a fairly large body of experiments that show that small children, animals such as primates, and autistic persons are very poor at reading the minds of others. But neither O’Connell, nor any of the studies she reviews acknowledge a clear implication of the findings: children, primates and the autistic are poor at mindreading, but normal human adults are good at it. No studies are reported which test the accuracy of normal adult intersubjectivity. Even studies of mindreading seem to take for granted Cooley’s idea that human adults spend much of their life living in the minds of others.

A flagrant instance involves one of the central doctrines in postmodern theory, Derrida’s proposition that the meaning of all texts is fundamentally undecideable. At the most atomic level, this proposition is true, since all commonly used words, in all languages, have more than one meaning. Multiple meanings lead to inescapable ambiguity in the meaning of individual words, whether spoken or written.

But the leap to the idea of universal undecideability is erroneous. The meaning of individual words is undecideable only if the context is shorn away. Consensual meanings are arrived at by referring to the context in which words occur, both the local context, and the extended context (Scheff 1990). To be sure, interpretation in context is a complex process, fraught with the risk of error. For this reason, there is considerable misunderstanding in communication, even when messages or texts are skillfully constructed.

But, by the same token, there is also considerable consensual understanding about the meaning of messages and texts, even complex ones. Else the social order would immediately collapse. The idea of undecideability seems to be based on a mechanical model of the communication of meaning, as if it were determined by rote responses to individual words. In particular, undecideabilty ignores the possibility that communication involves at its very core the process of taking the role of the other, of understanding the meaning of messages or texts not only from the receiver’s point of view, but also from the sender’s.

Another example is what is called the Problem of Other Minds in the discipline of philosophy. Like much of postmodern theory, it is built entirely upon abstract reasoning rather than on systematic observations. Given this approach, it is not surprising that the contributors to this field have decided that no one can ever really know the mind of another person. This belief reflects the Western insistence on individualism, that each of us is essentially alone. But it seems bizarre in Eastern cultures, with their insistence on the group over the individual. In these settings, each mind is thought to be a fragment of one supermind, the Great Cloud of Unknowing. The concept of intersubjectivity offers a middle ground, in that one can evaluate the accuracy of mindreading, without automatically assuming or rejecting the idea.

A final, and arguable example is the work of Cooley, Mead, and Blumer themselves. Although role-taking is central to their visions of human nature, they deal with the concept only in the abstract. Goffman’s treatment of intersubjectivity, (and its correlate, the pervasiveness of embarrassment or anticipations of embarrassment), begins to fill in the interstellar void between the abstraction and its omnipresent everyday meaning in human conduct.

Goffman’s treatment fleshes out the idea of role-taking itself, and just as importantly, its close kinship with embarrassment, in a way that Mead and Blumer never did. To be sure, Cooley clearly states the kinship to embarrassment/shame in his discussion of the looking glass self, but he doesn’t give the kinds of concrete examples that would allow one to see implications. Goffman uses many, many examples that bring the ideas to life in living color. In this sense, the earlier writers found a vast, unknown continent, the emotional/relational world, but didn’t get off their ships. Goffman not only disembarked but explored parts of the interior.

Even Simmel, who was aware of intersubjectivity and of shame, didn’t link them, and gave few concrete examples. Goffman stands alone. In my opinion, it was he, and not any of his antecedents, who discovered the emotional/relational world. We all swim in this world all day, every day, but Goffman was the first to notice and describe it. For this service to humankind, we should award the supreme medal of honor to his memory.

Goffman and Positive Science

Unlike most analysts of interior life, Goffman was not content to leave his basic concepts undefined. Although he casually uses the term "mystic union" several times to refer to speakers who are talking to each other, he also offers a fairly elaborate and complex definition of "being in a state of talk." Since his definition requires an entire page of text, I will not repeat it all here. Suffice it to know that it contains phrases that imply mutual mindreading: "…An understanding will prevail [among the speakers] as to how long and how frequently each speaker is to hold the floor…" (1967, 35; a similar formulation occurs earlier, on p. 34). The definition comes closest to explicitly describing intersubjective accord in this line:

"…A single focus of thought and attention, and a single flow of talk, tends to be maintained and to be legitimated as officially representative of the encounter (Goffman 1967, 34, emphasis added)."

The significance of the phrase "a single focus of thought and attention" becomes more apparent if it is compared to a similar phrase, "joint attention" used by the psychologist Bruner (1983), when he is explaining how an infant learns to become attuned with its caretaker. The mother, he says, is only trying to teach a new word. She places an object (such as a doll) in her own and the baby’s line of gaze, shakes it to make sure of the baby’s attention, saying "See the pretty DOLLY." In this situation, the baby is likely to learn not only the meaning of a word, but also, since both parties are looking at the same object, how to have, jointly with the mother "a single focus of thought and attention", to use Goffman’s phrase. A conceptual definition of intersubjectivity is as far as Goffman goes in attempting to explicate this idea.

But with the other interior strand of Goffman’s work, embarrassment, he was not content to give only a conceptual definition, but also offered elements of an operational definition:

An individual may recognize extreme embarrassment in others and even in himself by the objective signs of emotional disturbance: blushing, fumbling, stuttering, an unusually low- or high-pitched voice, quavering speech or breaking of the voice, sweating, blanching, blinking, tremor of the hand, hesitating or vacillating movement, absentmindedness, and malapropisms. As Mark Baldwin remarked about shyness, there may be "a lowering of the eyes, bowing of the head, putting of hands behind the back, nervous fingering of the clothing or twisting of the fingers together, and stammering, with some incoherence of idea as expressed in speech." There are also symptoms of a subjective kind: constriction of the diaphragm, a feeling of wobbliness, consciousness of strained and unnatural gestures, a dazed sensation, dryness of the mouth, and tenseness of the muscles. In cases of mild discomfiture, these visible and invisible flusterings occur but in less perceptible form (Goffman 1967, emphasis added).

This definition links an interior emotion with surface observables. With his usual uncanny instinct, in the last sentence he even seems to hint at the need for further elaboration of the operational definition: "these visible and invisible flusterings [that accompany embarrassment], but in less perceptible form." This clause seems to point toward the development of more elaborate coding systems for the verbal and gestural indicators of shame and embarrassment, such as the one by Retzinger (1991; 1995).

Goffman’s attempt at defining embarrassment is even more extraordinary in the context of contemporary social science. The few social science theorists who emphasize emotions seldom define them, even conceptually. An example would be Elias’s masterwork, The Civilizing Process (1939). His proposition that the threshold for shame is advanced in the civilizing process is the central thread of the entire work. In a later work of Elias’s, The Germans (1996), shame is again frequently evoked, though not explicitly as in the earlier study.

Yet Elias offered no definition of shame in either book, seeming to assume that the reader would understand the concept of shame in the same way that he did. The absence of any definition of shame and a systematic way of identifying it is particularly glaring in The Civilizing Process. This study entails an extensive analysis of shame in many excerpts from advice and etiquette manuals in five languages over six centuries. The analysis of the excerpts is completely intuitive, and in most cases, highly inferential. That is the word shame is sometimes used in the excerpts that he selected, but much more frequently it is not.

Elias relied on intuitive and unexplicated interpretations of what Retzinger would call cue words, in context. Even if his interpretations were fairly accurate, which they might be, he still gave little direction to future research on the subject. Unlike Elias and most other analysts of emotion, Goffman took at least the initial step toward overcoming this problem. By explicitly defining his concepts, he attempted to link interior variables with observable indicators. Perhaps the secret for success in social science is not only to study both surface and interior, but to also provide links that connect them.

The Effect on Goffman’s Approach on Readers

Many authors have noted the intense effects of Goffman’s writings on his readers. Reading Goffman, as Lemert (1997), put it,

…made something happen… a shudder of recognition…The experience Goffman effects is that of colonizing a new social place into which the reader enters from which to exit never quite the same. To have once, even if only once, seen the social world from within such a place is never after to see it otherwise, ever after to read the world anew. In thus seeing differently, we are other than we were…(p. xiii-viiii)

This is a strong claim: our vision of the world, and even of ourselves, is transformed by reading.

In a perceptive review of Goffman’s work, Bennett (1994) made a similar, if less expansive, comment: "…no other writer in this field [sociology] so regularly startles one into self-recognition."

Lofland’s (1980, 47) remarks hint at reasons for this effect:

I suspect I am not alone in knowing people who have been deeply moved upon reading Stigma (1963) and other of his works. These people recognized themselves and others and saw that Goffman was articulating some of the most fundamental and painful of human social experiences. He showed them suddenly that they were not alone, that someone else understood what they know and felt. He knew and expressed it beautifully, producing in them joy over pain understood and appreciated, an inextricable mixture of happiness and sadness, expressed in tears.

Although Lofland doesn’t name specific emotions, his reference to "the most fundamental and painful of human social experiences" might apply to, among other things, embarrassment, shame, and humiliation. Goffman’s focus on embarrassment could be a central cause of the empathic identification described by Lofland, particularly since most social science writing do not involve any emotions. Lofland’s reference of "the most fundamental…human experiences" might also refer, by the by, to Goffman’s frequent evocation of intersubjective understanding.

Implications for Social Science

This essay has proposed that Goffman considered both surface and interior in his analysis of social life. With two critical concepts, he also attempted to provide adequate definitions, so that anyone who wished to evaluate or elaborate on his findings would not have to rely completely on intuition.

Even if it is conceded that Goffman’s approach works for readers, it still might not work for social science. In this section I argue that it might, comparing some approaches that emphasize only the interior or the surface. It seems that results obtained through intuitive analysis of interiors are believed by virtually no one. On the other hand, results that depend largely on the analysis of surfaces have the opposite problem. That is, they turn out to be obvious, only "commonsense."

The plight of orthodox psychoanalytic theory, representing an extreme of dependence on interiors, has already been mentioned. Since there is a wide consensus on this negative evaluation, I will not argue it further. But the response to Elias’s work on the advancing shame threshold deserves further comment.

The Civilizing Process (1939) has been widely hailed as a masterpiece of historical sociology, especially in Europe. But his analysis of the advancing shame threshold, the theoretical and empirical core of the book, has been largely ignored. Surprisingly, none of the many reviews have even mentioned it. None of his followers and admirers have taken up advances in the shame threshold as a central feature of historical change.

The reasons that led readers to ignore the central thread of the book are probably many and complex. But it seems reasonable that one of the reasons is the casual way in which Elias identified shame in the excerpts that form his main body of data. He offered no theory of shame, no method for identifying it in texts, or even a conceptual definition. To be sure, there was virtually no literature on any of these issues available to him at the time he wrote (1939) so he did well to deal with shame at all. Apparently his analysis of shame was, and perhaps still is, well in advance of the understanding of his readers. If this is the case, then Goffman’s attempt to theorize and define two interior concepts is all the more laudable.

If we turn out attention to the whole of social science, the overall picture suggests that most approaches emphasize surfaces. Quantitative social science uses surface variables almost exclusively. At first glance, we might think that the use of psychological scales involves interior more than surface elements. But even a brief review of the methods used in constructing scales suggests the opposite. Since virtually all of the thousands of existing scales are constructed on the basis of paper and pencil tests or verbal responses in interviews, it is reasonable to conclude that scales involve only aspects of experience that are available in consciousness. The outer behavior and aspects of consciousness that are studied quantitatively deal almost entirely with the surfaces of human conduct.

The central problem with research results based on surfaces is not that they won’t be believed, but that they are not news, they are obvious. Surface variables are known to anyone who wants to consider the problem, not only to social scientists. In one of the most widely respected of social science studies, Blau and Duncan (1967) found that in the U.S, for white males, class of origin is related to income as an adult. However, this relationship does not fully explain income; it accounts for only a small amount of the variance. This finding was subsequently replicated several times by other social scientists, one of the relatively few studies that have led to a series of replications.

But knowing that income is due, in part, to class of origin, is not news. It is not even news that class of origin is probably only one of many determinants of income. Not even a fanatical Marxist, if there are any left, would be surprised. Yes, of course. Coming from an upper class helps establish one, and coming from a lower class holds one back. Then there are luck, pluck, and perhaps the degree of preoccupation with money, wealth, or status. But pluck and obsession are not surface variables; far as I know, they have never been studied by those who research income.

In his provocative essays on the ills of sociology, Cole (2001) notes several well-known studies whose results seem to flirt with the obvious. He calls attention to "Skocpol’s (1979) theory that we have a social revolution when the state breaks down. The latter is measured by the occurrence of revolutions." He also criticizes the elaborate theory of organizations developed by Hannan and Carroll (1992) and others:

…even assuming that the theory is true and that the authors have dealt successfully with the large number of …problems… which might be influencing the result, the theory leads to obvious conclusions. To say that there is a greater chance of establishing a type of organization that has legitimacy than one that does not, and that organizations are less likely to survive when there is heavy competition, is not saying anything which intelligent non-sociologists would not know. …not only the main points of the theory… but most of the details. For example, the authors point out that unions which begin as secessions from other unions have a much higher chance of disbanding than unions which begin as a result of a merger with another union (Hannan and Freeman, 1989: 256). I am sure that this finding would surprise no union organizer or official. (Cole 2001, 47, emphasis added).


Cole (2001) uses these and other examples to argue that general theory in social science is impossible, since it succeeds only in stating the obvious. But I reason that the problem is that obvious results are a product of using only obvious, surface variables.

This problem occurs not only with quantitative studies, but also with qualitative ones. Conversation Analysis (CA), which prides itself with dealing only with surfaces, seems to be an example. Sacks, et al, (1974) quote Goffman with regard to a system of turn-taking in conversation that he had sketched earlier (1964):

Talk is socially organized…as a little system of mutually ratified…face to face action…Once a state of talk has been ratified, cues must be available for requesting the floor and giving it up… to ensure that one turn at talking neither overlaps the previous one to much, [etc] (Goffman 1964, emphasis added).

Sacks and his colleagues took Goffman up on the last part of his definition, the observables on the surface of talk, rather than the front part, with its emphasis on ratification, which hints at intersubjective accord. Nor did they avail themselves of his definition of "being in a state of talk" in the 1967 chapter on face-work, discussed above, which refers still more directly to intersubjective accord ("a single focus of thought and attention.") Finally, the Sacks et al study also ignored the way that Goffman’s 1967 chapter on embarrassment drew attention to the part played by emotions in the organization of social encounters.

This study, and others by Sacks, established a pattern that CA researchers have faithfully followed. A large number of empirical studies of conversation have been used to establish a formal system of classification. For example, a question followed by an answer is one of a larger class of sequences that are called, in CA lingo, "adjacency pairs." The use of these and other phrases are universally used within the CA group, but it is not clear that they contribute to our understanding of human conduct.

Although their studies have provided some interesting details about the language used in social transactions, their main results are at least partially obvious. It is no secret that speakers take turns, and that interrupting during another’s turn, or remaining silent during your own, will usually be seen as rude or impolite. In this respect, the findings of the CA studies of turn-taking do not seem to show much advance over Goffman’s early sketch.

As in the disciplines of psychology (experiments and scales) and economics (theory and math. models), the main interest in CA seems to be in formalizing a method, however obvious the results. Modern psychology began when several experiments produced interesting results. Similarly, modern economics was founded on math. models because some of the earliest uses showed potential. CA was founded on the methods used by Harvey Sacks, who appears to have been a creative genius.

But the subsequent researchers in these three fields have simply repeated the methods that seemed successful initially. Over a hundred years ago, Boltzmann (1899), the physicist who created the brillinat theory of heat transport in solids, had a similar complaint. He noted that when a new method yields interesting results, many become wedded to it, they come "to believe that the development of science to the end of time would consist in the systematic and unremitting application of it."

A final example concerns the development of modern sociology, founded on a highly innovative study of suicide by Durkheim (1897). Like Sacks, Durkheim was a creative genius. His study demonstrated statistically that year after year, the suicide rate in Protestant regions, and in traditional societies, was higher than the rate in Catholic regions. He linked this finding to a general theory of social integration: in communities, social bonds that were neither too loose (isolation) nor too tight (engulfment) protected its members from suicide. Durkheim’s theory of social integration (solidarity/alienation) is central to modern sociology, and his statistical method forms the basis for quantitative studies. The theory of social integration implied by Durkheim’s study is not obvious, at least in Western societies. It goes deeply against the grain of the idea of the self-contained individual, an idea that lies near the core of Western culture.

Although modern sociologists copy Durkheim’s method in Suicide, they ignore what to me is the essential aspect of this study: he linked a surface variable, suicide rates, to an interior and surface variable, degree of social integration, which is part of a general theory. By and large, facts only become interesting when they are used to test an explicitly stated theory. Interest in facts always arises out of a theory, but it is usually unstated, or even outside of consciousness. Unless a theory is explicit, it cannot be discarded if it is wrong, or improved if it is partially right.

Most studies in current social science, whether quantitative or qualitative, are not undertaken in the service of an explicit abstract theory. Even in studies oriented toward testing a hypothesis, the hypothesis is virtually always a low-level generalization, at best. Modern sociology imitates Durkheim’s statistical method, but ignores the part played by theory in his study.

It is also important to note that although Durkheim’s findings have been replicated, the correlation between degree of social integration, however it is indexed, and suicide rate is always quite small, accounting for less than 10% of the variance. The tiny size of the correlation suggests that many, many other factors are involved. In other words, the theory of social integration does not explain the major causes of suicide.

There was a legitimate reason for Durkheim’s approach: he was establishing sociology as an approach to the study of human conduct separate from other disciplines, especially separate from psychology. But now that sociology, psychology, and economics have all been established as separate disciplines, it seems senseless to try to explain human conduct exclusively within a single discipline.

The Blau and Duncan study (1967) of adult incomes, and the replications of it, are cases in point. These studies copy exactly the methodology of Durkheim’s study of suicide, attempting to explain complex human conduct by using only sociological explanatory variables. Most sociologists believe that they should use only sociological variables. They are particularly opposed to employing psychological variables, which might be a first step toward the interior, because of an erroneous reading of Durkheim’s advice in this matter.

But the upshot of copying what sociologists take to be Durkheim’s method is that their studies, like Durkheim’s, account for only a tiny proportion of the variance. The Blau and Duncan study (1967) may be the high water mark in this regard, since class of origin accounts for 16% of the variance, about twice as high as in the original Durkheim study, and in its modern replications. Still their findings do not account for the major forces at work in shaping income.

The way in which the social sciences emulate the methods of their founders recalls an episode that occurred during my days as a physicist. Our lab was built around a cyclotron, at the time the biggest in the world. When a group of Japanese scientists decided to build their own, they copied the cyclotron mechanically, including the errors. One error, a hole in the magnet, was very expensive to correct for us, and it would be for them. Like this group of scientists, the social sciences copy the errors of their founders, as well as their advances.

The French philosopher Emile Chartier observed that "Nothing is more dangerous than an idea when it’s the only one you have" (circa 1910). He may not have had the disciplines and subdisciplines in mind, but his observation fits exactly. Specializing in a single aspect of knowledge is usually a good idea at the time it is undertaken, but ultimately comes also to be an impediment to advance in knowledge, when the original good idea is used as a shield against other good ideas.

It would seem at this point that the only way to generate results that would explain the major causes of a social phenomena would be to utilize variables from all the relevant disciplines, both interior and surface. If economic, psychological, and social variables each contributed some 12% of the variance, and the interaction between the three 20%, a significant step would be taken toward understanding the major causal forces. Perhaps it is time to begin to build interdisciplinary teams that could carry out such a program.

I have been unable to find a study involving three disciplines in this way. But a recent series of studies of health levels of welfare recipients in Sweden by an interdisciplinary team combines economic and interior psychological variables, an unusual combination in itself (Starrin, et al 1997, 2000, and 2001. There have also been other publications in Swedish). Their basic finding, that shame and financial stress together predict poor health far better than either variable alone, has been replicated in several different populations. Perhaps studies that combine a surface variable (financial stress) with an interior one (shame) can be expanded to include yet other causes in a way that would lead to a viable social science.


Becker, Howard. 1999. The Politics of Presentation: Goffman and Total Institutions. Conference presentation at U. of Grenoble. (http://www.soc.ucsb.edu/faculty/hbecker/goffman.html)

Bennett, Alan. 1994. Cold Sweat. In Fine and Smith, Erving Goffman. (2001).

Blau, Peter, and O. D. Duncan. 1967. The American Occupational Structure. New York: Wiley.

Blumer, Herbert. 1986. Symbolic Interactionism. Berkeley: University of California Press.

Boltzman, Ludwig. 1899. The Recent Development of Method in Theoretical Physics. Monist 11: 229-30.

Cole, Steven (Editor). 2001. What’s Wrong with Sociology. New Brunswick, NJ: Transaction.

Cooley, Charles H. 1922. Human Nature and the Social Order. New York: Scribner’s

Durkheim, Emile. 1897. Suicide. Glencoe, Ill.: Free Press (1951.

Elias, Norbert. 1939. The Civilizing Process. Oxford: Blackwell.

____________ 1996. The Germans: power struggles and the development of habitus in the nineteenth and twentieth centuries. New York: Columbia University Press.

Fine, Gary, and Smith, G.W. 2001. Erving Goffman. Newbury Park: Sage.

Goffman, Erving. 1959. The Presentation of Self in Everyday Life. Garden City, NY: Doubleday Anchor.

_____________1961. Asylums. New York: Anchor.

_____________1964. The Neglected Situation. The American Anthropologist. 66 (6, part 2), 133-136).

____________ 1967. Interaction Ritual. New York: Anchor. .

_____________ 1969. Strategic Interaction. Philadelphia: U. of Pennsylvania Press.

Hannan, Michael T. and Glenn R. Carroll. 1992. Dynamics of Organizational Populations: Density, Legitimation, and Competition. New York: Oxford University Press.

Hannan, Michael T. and John Freeman. 1989. Organizational Ecology. Cambridge, MA: Harvard University Press.

Heath, Christian. 1988. Embarrassment and Interactional Organization. In Paul Drew and Anthony Wooton (Editors), Erving Goffman: Exploring the Interaction Order. Cambridge: Polity.

Laing, R. D, H. Phillipson and A. R. Lee. 1966. Interpersonal perception: a theory and a method of research, London, Tavistock.

Lemert, Charles. 1997. Goffman. In Charles Lemert and Ann Branaman (Editors), The Goffman Reader. Oxford: Blackwell (pp. Ix-xlviii).

Lynd, Helen. 1968. On Shame and the Search for Identity. New York: Harcourt.

Lofland, John. 1980. In Ditton, Jason. (Editor) 1980. The View from Goffman. New York: St. Martin’s Press.

Mead, George H. 1936. Mind, Self, and Society. Chicago: University of Chicago Press.

O'Connell, Sanjida. 1998. Mindreading: an investigation into how we learn to love and lie. New York: Doubleday.

Retzinger, Suzanne. 1991. Violent Emotions. Newbury Park: Sage.

______________ 1995. Identifying Shame and Anger in Discourse. American Behavioral Scientist 38: 104-113.

Sacks, Harvey, E. Schegloff, and G. Jefferson. 1974. A Simplist Systamatics for the Organization of Turn-taking for Conversation. Language 50: 696-735.

Scheff, T. J. 1967. Toward a Sociological Model of Consensus. American Sociological Review. 32: 32-46.

___________1990. Microsociology: Discourse, Emotion, and Social Structure. Chicago: U. of Chicago Press.

________2000. Shame and the Social Bond: A Sociological Theory. Sociological Theory. 18: 84-99.

Schudson, Michael. 1984. Embarrassment and Erving Goffman’s Idea of Human Nature. Theory and Society 13: 5, 633-648.

Skocpol, Theda. 1979. States and Social Revolutions. New York: Cambridge University Press.

Starrin, B, Rantakeisu, U, Forsberg, E & Kalander-Blomqvist, M, Understanding the Health Consequences of Unemployment - The Finance/Shame model. In Kieselbach, T (Ed.) Youth Unemployment and Health - A Comparison of Six European Countries. Leske + Budrich. Opla-den, 2000.

Starrin, B, Rantakeisu, U & Hagquist, C, In the wake of the recession - economic hardship, shame and social erosion. Scandinavian Journal of Work and Environment, 1997, 23, 47-54.

Starrin, B, Jönsson, L R & Rantakeisu, U, Unemployment and sense of coherence. International Journal of Social Welfare, 2001:10, 107-116.

Stern, Daniel. 1977. The First Relationship. Cambridge: Harvard University Press.

Surfaces Oct 8 , 2001 6, 703 words