Livelihoods of theory: The case of Goffman’s early theory of the self

Context of emergence

Postwar social science was influenced by the Cold War as its practitioners were summoned and rewarded for research that served the state’s consuming interests. Social science was informed as well by the cultural atmosphere: what has come to be called Cold War research was permeated with the politics of power jockeying, surveillance, brinkmanship, and deterrence; even social conformity was cast as a domestic enemy. Yet the war and its after effects prompted other concerns: the life world of social scientists was not only inhabited by the specter of a global enemy, an apparently formidable one, but also haunted by dilemmas of race, gender, family relations, and democracy. Hovering over these problems, all generating plentiful empirical studies, loomed a momentous matter of concern: what is the nature of human nature? As the social psychologist Gordon Allport recounted the history, the First World War,

followed by the spread of Communism, by the great depression of the 1930s, and by the rise of Hitler, the genocide of the Jews, race riots, the Second World War and the Atomic threat, stimulated all branches of social science. (1954, p. 2)

A superordinate question circled around the limits and potentials of human capacities asking, as Allport did, “how is it possible to preserve the values of freedom and individual rights under mounting social strain and regimentation?” (1954, p. 2). Many intellectuals shared Allport’s concern about human nature in the 1950s, a concern evident in the theories of Hannah Arendt and Eric Fromm, the social criticism of Philip Riesman and William Whyte, and ethological work of Konrad Lorenz. Whether framed as matters of the individual, citizen, or social self, these theories sought to articulate the nature of what increasingly appeared to be a perplexing human nature.

Theories and theorizing about human nature (or the human condition) that emerged during the two decades following the war took no singular or categorical form; there was no simple consensus about the ontology of being, origins, or prospects. Some scholars spoke of a vulnerable human kind: their tonal fear and pessimism figured in the writings of Arendt and Milgram. Some expressed hope and even optimism, promoting humanist ideas of malleable or protean personhood along with nascent cognitive psychologists’ tropes of creative, rational, and flexible beings (Cohen-Cole, 2005; Morawski & St. Martin, 2011). According to one of the actors, scholars were challenged to understand the modern person whose “inner self is no longer fixed and immutable,” and to investigate how “psychic mobility liberated man from his native self” (Lerner, 1959, p. 22). Yet others found the charge of reinventing selves to be inextricably wrapped up with their own identity as scientists and engineers of the social world; they suspected that not only views of human nature but also those of the very scientists of human nature demanded a makeover, thereby complicating the purportedly simple models advanced before the war (Capshew, 1999; Cohen-Cole, 2005; Hunter, 2007).

Recent studies document postwar efforts to remake the self (Engerman, 2010; Heyck & Kaiser, 2010; Isaac, 2007, 2009; Lunbeck, 2000) and trace the ways that social scientific practices, technical tools, and often elite self-conceptions were integrated into and shaped models of human nature (Cohen-Cole, 2005; Gigerenzer, 1996; Lubek & Stam, 1995; Stam, Lubek, & Radtke, 1998; Stam, Radtke, & Lubek, 1995; Stark, 2010). They detail as well how social scientists traded concepts and fact claims across related disciplines (Erickson, 2010; Fontaine, 2010; Vicedo, 2010) and incorporated cultural imaginaries into their models (Bayer, 2008; Lutz, 1997; Martin, 1994; McCarthy, 2005; Nicholson, 2011). These studies register variations in thinking about human nature and revealing in social scientists’ otherwise diverse projects an aim to formulate coherent statements about the human condition. Yet they focus mainly on specialized practices and tools of social science and not on theory or the ways that theory itself comprises a practical, portable tool. Likewise overlooked, as historian Joel Isaac has argued, are the ways that social scientists’ “theoretical practices are entangled in special ways with the world they seek to limn” (2009, p. 416).

Erving Goffman’s postwar theorizing of social life and the self exemplifies how theories can function as tools, and how theories-as-tools operate in an economy of cross-discipline, cross-project exchanges. As tools, theories can facilitate instrument development, methodological design, and data analysis. Goffman’s theory of the social self and social situation was used in such ways: it aided psychologists in constructing models of the self, handling problems of method, and even critically interrogating experimental method. Three psychologists’ reliance on Goffman’s theory illustrates such practical uses as well as the lively if sometimes precarious life of theory in postwar social science. The three cases illustrate how theories, like facts, travel and serve as polyvalent and mutable resources for social scientists working on distinctly different projects to configure the nature of the human. They reveal how a theory’s textual openness, incompletion, or ambiguity provides opportunities for subsequent theory development. In this regard Goffman’s theorizing afforded at least two dissimilar readings: as description of a self comprised of public and private parts, on the one hand, and a performative self, on the other. These readings, or the theory’s openness suggested by them, map onto different points on the real/artificial and individual/social binaries that chronically challenged postwar social science. The cases of trading considered here illustrate as well the work that theory can do: at some times theory affords the stuff of conceptual foundations while at other times functions as a practical instrument. This latter function indicates how theory can become a technical tool, the obverse of the “tools to theory heuristic” identified by Gerd Gigerenzer (1996). The divergent appropriations and extensions of Goffman mark the ways theory can be circulated, exchanged, and even transvalued, sometimes with the consequence of advancing a kind of human nature distinctly unlike that presumed in the original theory. Owing to the relative abeyance of theories’ functioning in psychology, such exportations and reconfigurations have yet to be systematically examined.

Reconsidering theories in terms of the ways they thrive—their transport, translation, application, extension, and re-reading—is a crucial step toward recovering the presuppositions (and event fantasies) about human nature that social scientists privileged in a given era. Attention to theory’s generativity enables excavation of what tenets about the human were disdained and makes evident aspects of theory that were dispensed, forgotten, or suppressed.

To varying degrees, the psychologist “adopters” of Goffman tried to circumvent notions of the self as performative and social interactions as productive of the self. In other words, they avoided or missed the very aspects of Goffman’s early theorizing that informed performance theory: emphases on the “effect” of action rather than its intention; on the relationship of audience and actor; and on an existential if morally neutral perspective on the actor’s taking responsibility. Yet, these sidelined features are what connect Goffman’s work with later developments in performance studies, and he has been cited as “among the first scholars to make comprehensive studies of performance in everyday life” (Bial, 2004, p. xv; Carlson, 1996; McKenzie, 2001). These vital features of performance proved to be problematic for or at least incompatible with the three psychologists’ ideas about human nature. Instead, they selectively drew upon Goffman’s thinking to solve then-pressing problems in psychological research, and certain implications of this selectivity remain detectable in contemporary methods (techniques for managing or stabilizing the subject’s actions) and theory (positing a distinct cognitive self beneath social interactions). The divergent understandings of Goffman not only demonstrate the vivacity of theory but also suggest a genealogy that tracks where psychology has eschewed what performance theory engages, namely a perspective on the self that emphasizes the social situation, cultural practices, representations, and the effects of action. Such a genealogy likewise might track where performance theory overlooks cognition and interior activities of the self. ¹

Appreciating the practical usages of Goffman’s early sociology of the self first requires familiarity with his theory, particularly those ingredients that distinguish it from role theory with which it has been all too frequently associated. That introduction prepares the way to follow his theory as three psychologists working in the postwar decades engage it as a practical resource: Henry Riecken, Robert Rosenthal, and Edward Jones. Their transplantations and translations indicate that theory exchange requires no distinguishing common language or consensual interpretation. These scientists engaged Goffman’s work to graph three different valences in the nature of human nature matrix, and applied it to better understand particular human actors—the research subject, the researcher, or both. If Goffman’s formative influence on these psychologists has been forgotten or perhaps never appreciated, then it nevertheless persists as vestiges in contemporary psychology’s ontology of the person.

Goffman’s postwar theory: Promiscuous or ambivalent?

Erving Goffman stands as a tall figure in mid-century sociology, one whose work continues to guide inquiries, garner impressive numbers of citations, and retain eminence in contemporary social science curricula. In many of these posthumous appearances, Goffman is acknowledged for his ethnographic demonstrations of role taking, for describing how humans manage the impressions of themselves differently in different contexts. He famously used the analogy of the stage, likening to actors the way all persons present (or have) public (stage) and private (backstage) selves, the latter frequently assumed to be the real self. To this dramaturgical metaphor has been added another: Goffman has been understood as representing humans as confidence men or imposters who, in taking on situationally determined roles, deceive their audience through role performances (Pettit, 2011). If Freud is credited with discovering humans’ self-deceptions, then Goffman, whose first book, The Presentation of Self in Everyday Life (1959; PSEL hereafter), gestures to Freud’s (1901/2003) Psychopathology of Everyday Life, reveals the deceptions permeating social life just as it registers an ambivalence toward psychoanalysis shared by many of Goffman’s contemporaries. When portrayed as such, Goffman’s theoretical endeavors are considered to have importantly advanced role theory that emerged in the 1940s and was refined after the war. Interpretations of Goffman’s theory of self that emphasized the inner, real self, protected or hidden by role enactments, also gained support by a “new sociologically and psychologically inflected discourse of the self” that materialized in postwar America (Lunbeck, 2000, p. 321).

This common reading of Goffman’s work is (and perhaps was thus when PSEL was published) not the only one. Then and now, Goffman’s theorizing has been alternatively understood not as describing role taking that masks an inner self but, rather, as actually reneging “on the image of the hidden manipulator” (Manning, 1989, p. 343). By this account, Goffman conceptualized not an independent self behind the curtain but rather self as “social process”: the self is an accomplishment of social interactions. Philip Manning has traced the ways in which this “Second Goffman voice intrudes in a quietly disruptive fashion” into the “two selves thesis” (1989, p. 342). When Goffman’s stage metaphor is not taken literally, the private and public can be understood to represent “different kinds of stage,” neither one more real than the other. Defying any categorical distinction between the real and apparent, Goffman strived to show how humans “make their performances convincingly real” whether those performances be trustworthiness or betrayal (Manning, 2000, p. 288). This conception of the social self connected backward to George H. Mead’s interactionist self, attending to the effect of performances and constraints on action. It connected, too, with existentialism’s emphasis on persons’ responsibility in interactions, a responsibility of communication. Thus understood, his theory nurtured development of performance theory and performance studies (Bial, 2004; Carlson, 1996).

These two versions of Goffman’s theory of the social self—the role taking, inner and outer self and the performing self—are rarely presented together. Social theorists who do discuss both interpretations typically attribute the seeming discrepancy between them to Goffman’s own modifications, his evasive if masterful writing, or his amoral stance (Carlson, 1996; Manning, 1989, 2000; Tseelon, 1992). For instance, it has been claimed Goffman was so committed to “roles” persons undertake that he failed to appreciate “that the ascription of interiority is itself a publically regulated and sanctioned form of essence fabrication” (Butler, 1988, p. 528). Whatever the reason, the theory’s interpretive plasticity was to be consequential and these consequences, both immediate and far-reaching, warrant attention.

Those who adopt the first, most common version of Goffman’s social self typically associate him amongst a cohort of sociologically minded critics arriving in the 1950s who labored to expose the reified omnipotence of a bureaucratic culture, the artifices of conformity, and the mechanical determinism of oversized organizations. As historian Jackson Lears argued, these social critics were so fixed on a reified idea of culture that they “dismissed the human agents of power altogether” (Lears, 1989, p. 42). According to this view, Goffman’s central contribution is an ill-fated analysis of how institutional structures cause deterioration of autonomous selfhood. Lears ascertained, “The only solution Goffman could imagine was an ironically detached acceptance of the fragmented role-playing self. … The self-conscious awareness that one was in fact engaging in a theatrical social performance constituted at best a Pyrrhic victory” (p. 45). Goffman and his fellow social scientists, as this account goes, were merely warning consumers of the seemingly inescapable perils of conformity. More critical social analysts have found Goffman’s theory of the social self not as describing but essentially serving late capitalism. Such grave assessment takes the self described by Goffman “as pure commodity, utterly devoid of any use-value: it is the sociology of soul-selling” (Gouldner, 1970, p. 383). Focusing on only this sense of Goffman’s social self, these interpreters allege his deep reliance on two common Cold War tropes: ubiquitous deception and perilous conformity.

But intellectual life in the immediate postwar years brought more than Cold War thinking; the period also resonated with uncertainty about the human condition. Historian Carl Schorske recounted the intellectual chaos he entered as a young academic after the war when certainties were in crisis, when “after Nietzsche, whirl was king, and I felt rudderless” (1989, p. 99). Doubts about intellectual foundations (epistemic, methodological, theoretical, and practical) circulated through writings and conversations. Accompanying these dilemmas of doubt were imaginings of a more positive vision or, in Schorske’s words, attending doubts were “new horizons that opened with them” (p. 99). Even the “beat generation” were at once disillusioned by the war and bomb yet at the same time yearned to understand how “life should be lived” (John Clellon Holmes, as cited in Menand, 2007). For young intellectuals, including the beat generation, estrangement from what was once taken as normal opened way for imaginative aesthetics of resistance and change. Alongside the rhetoric of Cold War defense and hyper-conformity, then, some social scientists reassessed the nature of humans. They found in human nature both paradox and irony, glimpsing if not embracing inventive ideas about subjectivity that were taking form in the arts and humanities. Some came to comprehend human nature as contingent, multifaceted, performative, protean: it is not “natural” as conventionally understood. These alternative conceptions of personhood often involved a paradox, inheriting from existentialism, as Ian Hacking has described,

a vision—of the whole of human nature, while denying, in a sense, that human beings have a nature at all. It is in the nature of a human being to have no intrinsic nature, but to live one’s life constantly choosing who one is. (Hacking, 2004, p. 281)

They were ironic as well, intimating, to paraphrase Kurt Vonnegut Jr., that one is what one pretends to be, not more or less than that pretend performance.

Erving Goffman, whose academic career began in the early 1950s, can be understood as advancing one such alternative version of personhood. Paradox and irony are not incidental in his early writings but, rather, are central to an inventive re-conceptualizing of the binaries of self and society, authenticity and artifice, real and the apparent. In his 1955 essay “On Face-Work” Goffman declared, “Universal human nature is not a very human thing.” The paragraph with which this statement opens ultimately concludes, “the human nature of a particular set of persons may be specially designed for the special kind of undertakings in which they participate, but still each of these persons must have within him something of the balance of characteristics required of a usable participant in any ritually organized system of social activity” (1955, p. 231).

These passages have been used to support the interpretation of Goffman’s theory that foregrounds a within-ness of persons that is distinguishable from outer presentations that may or may not accord with that inner self. But not all readers have shared this reading; some, notably Karl Scheibe (1995), have appreciated the complexities of Goffman’s conception of the self (see also Walsh-Bowers, 2006). Others situated his intellectual ambitions in a post-Enlightenment project to engage (not remove) the paradox of “natural” man and the “social self.” In his comprehensive history of the modern idea of authenticity, Marshall Berman (1970) preferred this latter interpretation. Howard Brick similarly found in Goffman’s project “demands for authenticity and the inevitability of artifice were combined in a theory of social interaction that sharply criticized social institutions but also recognized them as the basis of personal integrity” (1998, p. 67). These allied interpretations counter the idea that Goffman’s interests lay in studying “appearances”—that he was merely describing the masks of our self-presentations and not reality. Instead, they appreciated Goffman’s proposal that the self is an effect of performances and that the person usually has choices in those performances. They notice as well the flexibility and protean agency afforded in “personal integrity.” When Goffman observed that a person “cathects his face,” he was contending that “the person’s face is something that is not lodged in or on his body, but rather something that is diffusely located in the flow of events in the encounter and becomes manifest only when these events are read and interpreted for the appraisals expressed in them” (Goffman, 1955, p. 214). Such readings consider how Goffman dismissed binary notions of real and appearances and, instead, posited a notion of the self as an effect of social interaction. As referred to above, Goffman wrote in “On Face-Work,” “Universal human nature is not a very human thing,” adding that “the person becomes a kind of construct, build up not from inner psychic propensities but from moral rules that are impressed on him from without” (1955, p. 231; see also Burns, 1992, p. 109). Here Goffman reflexively admitted the “double mandate” he imposed on the self—as a player of ritual games and as an image effect of the flow of events (1955, p. 225).

Shortly after publishing that essay, Goffman proposed (in PSEL) that the self as character is not, as it is usually assumed, “something housed within the body of its possessor … this self does not derive from its possessor but from the whole scene of his action” (1959, pp. 252–253). The self is a “product” or “effect” and not the “cause” of a scene. Although eschewing the idea of self as simply an organic entity, he conceded that the attributes of performances are “psychobiological in nature, and yet they seem to arise out of intimate interaction with the contingencies of staging performances” (p. 254). The dichotomous being so often theorized and assumed in everyday life (the real and the feigned, the honest and the contrived) is simply “the ideology of honest performer, providing strength to the show they put on, but a poor analysis of it” (p. 70). The amalgamation of what is taken to be real and staged is implied in the now famous line that follows, “All the world is not, of course, a stage, but the crucial ways in which it isn’t are not easy to specify” (p. 72). When understood in this way, Goffman’s notion of “realness” or “authenticity” clearly departs from the ontology presumed in positivist epistemology and adopted by many of his contemporaries.

Criticizing the bald constructionist perspectives of role and labeling theorists, Goffman insisted that he employed the term “dramaturgy” as a metaphor. He defined the self by way of the “social act,” thereby locating the constitution, constraints, and enablements of self intersubjectively, in the dynamic mutuality of interactions. As biographer Tom Burns has suggested, Goffman made the social act his “monad” (1992, p. 109). Yet an interactionist definition of self, a self that is located in social acts, did not foreclose on the project of describing that self; nor did his tacit yet sustained critique of individualism. Rarely acknowledged is the meticulous attention given to emotions, and to the choices the individual has in presenting and internalizing his self. Goffman’s comprehension of choice seizes existentialism’s paradoxical denial of human nature and insists that humans have only choices. Self-presentations, which unfold through the social acts enabling or constraining them, require choices, which, in turn, generate new choices or new ways to choose. Readers of “On Face-Work” are repeatedly reminded of these small choices: they are instructed that a person “will find a small choice of lines will be open to him and a small choice of faces will be waiting for him.” That person is left “free to take a high line that the future will discredit, or free to suffer humiliations that would make future dealings with them an embarrassing thing to have to face” (1955, p. 214). Having “choice” in face-saving practices, an offender can refuse “to heed the warning” of interactants: he “decides at each moment, consciously or unconsciously, how to behave” (p. 227). Authenticity lies, if anywhere, in the dynamic loopings of myriad minute choices and the exigencies of self-presentation unfolding through social interactions. The limits of choice ensue from the fact that a person’s self is formed through reflexive social interactions and, therefore, is bounded by these interactions. ²

Goffman’s avowed commitment to transcending the dualisms of real and apparent, and authenticity and artifice, however, posed a potential contradiction, one that likely made way for his work to be transported and taken up in dramatically different ways. How can one’s action be determined by the interaction demands and also entail free choice? How can one maintain a true self and also succeed as a role player? One means of better understanding these apparent inconsistencies was found in a rubric of personhood that circulated through American culture of the 1950s. Cold War thinking involved what Catherine Lutz (1997) has termed an “epistemology of the bunker” where the pitting of “us” and “them” evolved into psychic strategies of vigilance and deception. Many social science researchers posed deception as a central characteristic of the self: notions of deception traversed the discourses in international relations, the military, the sciences and social sciences, literature, and popular media (see Martin, 1994; Noble, 1991; Robin, 2003). And deception emerged as an accepted aspect of social science research methods (Korn, 1997; Pettit, 2011). The pervasive thinking about and through deception engendered, in turn, suspiciousness that the enemy was everywhere, including in one’s imagination. Cold War thinking, in Frederick Dolan’s words, thus became a “looking glass war” in which the spectral and speculative seemed to merge: real and apparent, and the self as well as other, were subject to interrogation (1994, p. 67).

Yet Goffman’s account of deception differs from the prevailing social scientific notions of conniving and false pretense. He looked beyond conceiving deception as merely guise and disguise to explore how deception opens interactive possibilities. Deception complicates social life, and yet is productive. It impels change and flexibility: deception, potentials for deception, and even self-deception actually make intersubjective actions dynamic, unclosed as it were, and dim if not erase the lines purportedly separating real and apparent. The surface and invisible are dynamically connected such that “Whenever worlds are laid on, underlives develop” (Goffman, 1961, p. 305). Underlives affirm the partiality of the visible. Interaction is constitutive of and constituted through these underlives; therefore, underlives are necessary for social exchange (Goffman, 1959, pp. 143–144).

Whether due to textual ambiguities or to a normative pressure that resisted his radical re-conceptions of self, deception, and conformity, Goffman’s early theorizing made way for discrepant interpretations of his social self, interpretations that ranged from role theory to performance theory, and from emphases on the social situation to mining a hidden self. In their projects to scientifically understand the self, three young psychologists gained crucial insights from Goffman, selectively using his theory to remedy a perplexing problem of the self. Their transportations of theory solved particular research problems, yet also carried vestiges of Goffman’s more radical views even as they presumed the conventional binaries of real and artifice, individual and social, that Goffman sought to transcend in his early writings.

The experiment as a social situation

Social psychology benefited materially from postwar support of the social sciences and heeded the promotion of a natural science model of research. With heightened emphasis on natural science methods, experiments had become the sub-discipline’s primary and most revered method, and social psychologists became so engaged with experimentation that by the late 1960s they were making significant contributions to methodology. Toward this methodological mission, social psychologist Henry Riecken (1962) proposed, as his essay’s title pronounced, “A Program for Research on Experiments in Social Psychology.” The paper was presented at a 1958 behavioral science conference on decisions, values, and groups sponsored by the Air Force Office of Scientific Research and later published in a 1962 book featuring the conference papers. The young Riecken arrived at the event with noteworthy credentials for discussing methodology: he had worked alongside Leon Festinger in an innovative field-experimental infiltration of an apocalyptic religious group, a project culminating in the much-respected book, When Prophecy Fails (Festinger, Riecken, & Schachter, 1956).

Speaking at the conference, Riecken did not address field experiments or “external validity” (the accurate representations of social psychological phenomena that field experiments aimed to achieve). Instead, he attended to a known difficulty in laboratory experiments, namely the problem of “unintended variance.” The paper was exceptional in rejecting the conventional view that unintended variance was simply technical error that could (and should) be reduced if not eliminated through adjustments of laboratory technique. In lieu of its conventional definition, Riecken recast this supposedly technical problem in terms of Goffman’s theory of the self in social interactions and, in so doing, challenged the reigning epistemology and presumed ontology of psychology experiments. Experiments should be comprehended as social situations in which participants, both subjects and experimenters, are like everyday actors who aim to perform successfully. This framework, he claimed, derived from Goffman’s insistence (and here Riecken quoted from an unpublished version of PSEL) that we consider the ways a person “guides and controls the impressions they form of him, and the kinds of things he may and may not do while sustaining his performance before them” (Riecken, 1962, p. 27). Adopting Goffman’s performance framework, Riecken extracted the specific features that make up the social situation of the experiment and carefully enumerated “the kinds of things” the subject and experimenter can do while performing the experiment. One distinctive feature of the experiment is that it is “specifically vague,” both in being posed as an invitation to participate where the participants’ rewards are unspecified and in terms of what actually is to transpire in that situation. The social situation of the experiment is also opaque regarding the relations between its participants because the experimenter is often experimenter and teacher, and each of these status positions involves different relationships with the participant. Both, however, deem the experimenter a “powerful figure.” The experimenter “has two kinds of power: as a professor, lie [sic] is a member of the superordinate group that has the power of effective evaluation of students; as a psychologist he has the power of insight into the subject” (1962, pp. 29–30). He has a simpler relational existence in the situation than does the participant: “he usually wants to use the subject as an instance of behavior, use him just one time, and then forget everything about him except the data he has produced” (pp. 29–30). By contrast, the participant, subordinated to the experimenter’s power, has quite different apprehensions of the situation. He aspires to evade the indignity of being seen as just another datum, but also to make impressions in what he views as an encounter that might have consequences for his future. Further, as Riecken reported, the participant’s task is challenging because the situation entails a “one-sided distribution of information,” and thus compels “a set of inferential and interpretive activities on the part of the subject in an effort to penetrate the experimenter’s inscrutability” (p. 31).

These features make the experiment a setting in which what actually transpires in the laboratory is much more (and more interesting) than what is stipulated by the experimental design and hypothesis. Beyond the hypothesized laboratory activities, the experimenter and participant each form an “early definition” and proceed to negotiate the situation. Given the unequal distribution of power and information, the negotiation is one-sided: the experimenter’s inflexibility, mandated by his assumption of standardized experimental procedures, is not shared by the participant who, by contrast, is challenged by his own multiple aims. The aims guiding the participant’s interactions include perceived and actual rewards (pay, course credit, self insight), discovery of the true nature of the experiment, and positive self-presentation. The participant is guided by all these objectives in addition to the experimental task or tasks set before him.

Scrutinizing the experiment as a social situation, Riecken described laboratory selves as negotiated and performed, and he uncovered the ways in which performing those selves entails choices as well as constraints. At the outset of the paper he announced that his analysis implied neither an “interest in the correspondence between ‘appearance’ and ‘reality’; nor a necessary implication that people in ordinary social interaction are deliberately and self-consciously ‘playing a part’” (1962, p. 27). By deferring any epistemic claim about the “real” or assumptions about persons’ possible manipulations, Riecken intimated agreement with Goffman’s belief that the self is constituted and comprehensible only through social interactions. But this stance is seemingly belied in the paper’s conclusion, where he recommends an empirical program that would more thoroughly investigate participants’ understandings of experiments. This proposal for systematic study of how participants respond to various experimental conditions implies a discrete, stable participant who can be objectively observed and analyzed. Yet again, even this recommendation for systematic inquiry into a calculable participant is punctuated with doubt, for Riecken next wondered whether studies that varied experimenters’ characteristics and performances could adequately identify and control the myriad subtle cues perceived by participants, and strongly suspected that experimenters’ “anxiety” and their confederates’ “tensions” and “guilt” might prove difficult to eliminate. The chapter concludes, then, not with confidence but rather with an air of resignation if not confusion and a stark admission that methodological refinements might merely shift “the burden of communication” to another dimension of the experiment, thus creating an unending regress of negotiations between experimenter and participant. Just as Riecken began by bracketing the question of the relation of “appearance” and “reality,” so his final summation abstains from fully believing the realism of experiments.

Whether Riecken was reflecting an ambiguity he might have discerned in Goffman’s writings or his own ambivalence cannot be ascertained as he did not, with minor exception, continue his inquiry into the social situation of experiments. Despite intimating a social self whose ontology exceeds what can be observed within the parameters of experiments, he returned to an ideal of a consistent, autonomous self. Riecken’s chapter ultimately was read as distinctly celebrating this ideal: for decades it was cited as a warning call for devising technical controls that would reduce or eliminate experimental “artifacts” as most unintended variance in experiments came to be called.

From the unconscious to (managed) performance

Just two years prior to Riecken’s conference presentation, another psychologist reported serious methodological problems in experimentation. Working on his dissertation in clinical psychology at The University of California at Los Angeles, Robert Rosenthal (1956) initially encountered not conceptual or epistemic problems but empirical ones. Unlike Riecken, Rosenthal initially sought explanatory clarification of these problems in psychoanalysis, the very theory that his experiment was testing. His dissertation study of projection includes a didactic review of the research on that defense mechanism, a phenomenon he described as so elusive that a clinician who interprets a patient’s fantasies could himself be projecting. Aiming to avoid “knowledge by revelation, by edict or by authority,” as well as “epistemological meanderings,” he undertook a controlled experiment to observe the “process of projection by inducing it” (p. 10). Disguising the experiment’s true purpose, Rosenthal then again deceived the “subjects” (“Ss”) into thinking they performed either successfully or unsuccessfully on an intelligence test. Both before and after the test the participants were instructed to rate portraits in terms of the pictured person’s success or failure. Rosenthal hypothesized that participants who believe themselves to be failures as measured by the rigged intelligence scale would project their negative self-feeling onto the individuals represented in the portraits. The hypothesis was not confirmed. Even with several different statistical analyses, the measures failed to reach statistical significance, save one: the pre-test assessment of the portraits by participants assigned to the success test group differed significantly from those of participants assigned to the unsuccessful group. These data indicated an experimental effect had occurred before there was an experimental intervention. Given that both experimental groups had received identical pre-test instructions, Rosenthal deduced that the only explanation for this pre-test effect was unconscious processes—processes not merely in the participants but also, and more importantly, in the experimenter (“E”). The pre-test outcome was just what the experimenter would have wanted in order to obtain his predicted experimental effect. Rosenthal concluded that this outcome was due to “unconscious experimenter bias” that influenced the participants. “There are subtle, important processes occurring within the experimenter which bias the outcome of his research” he cautioned, adding “Thus it is felt in this research that the experimenter’s hopes as to the outcome led his treating the experimenter groups differently in subtle ways even while reading them identical instructions” (p. 64).

Experimenter biases, Rosenthal (1956) conjectured, could occur everywhere in the scientific process—from choosing a theory and experimental design all the way to interpreting data. However, the bias operating “during the performance of the experiment itself” (p. 67) was categorically different: experimenter bias was not public (it was not accessible to public scrutiny) and it even might be inaccessible to experimenters themselves who do not apprehend their own psychological involvements in experiments. Upon acknowledging the seriousness of this form of experimental bias, Rosenthal proposed several methodological solutions. Among these was the recommendation that experimenters submit themselves to experimental measures by participating in empirical studies that would expose them to the disguises, duplicity, and surveillance in experimental “as if” worlds. As he admitted, this solution dramatically altered the experimental situation by reversing the experimental candidates such that “the Es used would really in a way be Ss and it would be necessary for them to remain in ignorance of the real nature of the research” (p. 71). Rosenthal ultimately reassured his readers and himself “That the infallibility of the experiment has been questioned need not be disconcerting even to a science which is often self-conscious in regard to its scientific status” (p. 71). He concluded his dissertation by urging greater “ontological control” of the experimenter.

Rosenthal’s fascination with the methodological problem arising in his dissertation was to fuel his entire research career. Leaving aside the study of projection, although not immediately discarding psychoanalytic ideas about the unconscious, he proceeded to design and execute experiments exploring the effects of the experimenter on participants. Within a decade Rosenthal was recognized as one of psychology’s arch methodologists, and his work continues to be cited routinely in experimental reports and textbooks. His 1966 Experimenter Effects in Behavioral Research and 1975 The Volunteer Subject, the latter co-authored with Ralph Rosnow, became go-to guides for researchers who sought to better control the experimenter and participant respectively. Rosenthal’s life-long dedication to matters of research methods has significantly informed experimental practices, including now-routine techniques for “blinding” the experimenter from knowledge about the experiment. His research also begat new avenues of research, notably the renowned studies of how a teacher’s expectations about students influence their later academic performance (Jacobson & Rosenthal, 1968).

These esteemed contributions to experimental design owed not to the jettisoning of psychoanalytic notions of conscious and unconscious motives but to utilizing a version of Goffman’s theory of the self in social situations, and this definitive shift in theorizing ultimately yielded a model of experimenter–participant interactions. It was early in a stellar, career-making research program that Rosenthal changed his theoretical orientation. Several early publications related to the dissertation research engaged psychoanalytic notions of the unconscious, namely the experimenter’s unawareness of his own mental processes and how they shaped consequential behaviors. But the unconscious soon disappeared from his reports altogether, and was replaced by accounts of social interactions and negotiations. When Rosenthal first employed this new theory stance, he credited not Goffman, but Riecken’s 1958 paper ³ (Riecken, 1962). Drawing upon Riecken’s perspective (on Goffman), Rosenthal re-conceptualized his objects of analysis: the experimenter and participant were no longer described as possessing internal, unconscious properties. Albeit without considering the matters of power, anxiety, and guilt that Riecken had addressed, Rosenthal adopted Riecken’s Goffmanesque analysis of the social situation as a guide to empirically identifying the non-hypothesized, unacknowledged communications between experimenter and participant that occur in experiments and potentially affect experimental outcomes. By 1963, just 5 years post-dissertation, Rosenthal cited Goffman directly, substituting the language of invisible and unconscious forces with those of “the E–S dyad as signal exchange system” and the “business of impression management” (1963, p. 280). He sought to isolate the variables that would predict why “In the normal course of behavioral research, different experimenters (Es) often obtain different data from comparable groups of subjects” (p. 280). Persisting to interrogate the very problem found in his dissertation, but without the aid of psychoanalysis, Rosenthal commenced an extensive research program that would simplify Goffman’s theorizing. Impressions came to be all that mattered, and any “underlife” or traces of unconscious, anxious, or hidden selves disappeared, leaving only surface performance of behaviors.

Techniques for stabilizing impressions, namely by instituting better controls of the social situation of the laboratory, would eliminate or reduce the bias and unintended communication that produced them. A specific, narrow usage of Goffman’s social self was essential to this transformation in Rosenthal’s theorizing: by eschewing anything but appearance (behavior) he derived a conception of the subject, or subject’s self. Impression management was made equivalent to experimental technique. In turn, the social self that was observed under such management was taken to be adequate material for psychological inquiry. If the participant’s experimental behavior is sufficiently managed, and the experimenter’s sufficiently controlled, then experiments would yield valid knowledge about the participant. Carefully designed studies would test this assumption by varying conditions of the laboratory (the social situation, to use Goffman’s term).

Rosenthal’s adoption of Goffman’s early work was expedient if piecemeal. Technical refinements would prevent unintended communication and, it was assumed, what then would be produced were the participant’s honest behaviors. In an extensive review of the “volunteer subject,” Rosenthal and Rosnow (1975) distinguished the characteristics of the “willing” or volunteer subject from those “situational determinants” of volunteering. To remedy the biases arising with volunteering, Rosenthal proposed a range of technical changes to the social situation of the experiment and even recommended new relations between experimenter and participant. More “reciprocal” and “humane” arrangements would essentially set an ideal stage of impression management: the participant half of the dyadic system would perform appropriately if offered nonthreatening appeals, rewards for participation, greetings by high status recruiters, and invitations to participate made by someone with personal relations to him. However, the notably Goffmanesque face work recommended for recruiting and employing the right participants was deemed to be insufficient for managing experimenter expectancy effects, those biases experimenters inadvertently communicate to participants. Increasing experimenter’s awareness of her own expectations, mechanizing the observation process, and employing multiple experimenters all provided good but insufficient tools for “controlling” the biases of observer’s expectancies. The face work of laboratory data collection, performed non-consciously to manage a scene for particular ends, posed special challenges as it takes place in the private spaces of the experimental chamber, outside the public scrutiny of science. Here Rosenthal detected how the social situation is susceptible to corruption because “we are too often more interested in demonstrating what we already ‘know’ how nature works than in trying to learn how, in fact, she does work” (1966, p. 403). The social situation of the experiment is further vulnerable to the complexities of the “experimenter–subject interaction system,” which involves not only the experimenter’s experimental instructions but also nonlinguistic communications wherein “people ‘talk’ to one another without ‘speaking’” (p. 403). Although he listed a handful of techniques for controlling these communications through more precise control of the experimenter, Rosenthal seemed unconvinced about their efficacy. To fully remedy the experimenter problem, he proposed that psychological research employ a new kind of expert, namely “professional experimenters” who are specially trained to conduct research for which they have no scientific or personal investment. Only such scientifically disinterested experimenters could escape the tendencies to transmit their scientific desires and to speak without talking, tendencies that can influence participants.

Informing Rosenthal’s analysis of experimenter effects was Riecken’s “phenomenology of the subject” that enumerated participants’ multiple, complicated aims to detect the experimenter’s objectives and decode the social situation. But unlike Riecken’s assertion that he was not claiming any correspondence between appearance and reality, Rosenthal’s program moved toward making precisely this claim. The scientifically well-managed impression is equated with psychological reality. Although technically sophisticated and meticulously executed, his empirical research and especially his extensive lists of methodological correctives for controlling the experimenter–participant dyad take (well managed) appearances to be reality. Modifying the experimental situation enables, according to his model, productively managed participants who would behave as natural selves. The matter of managing the experimenter seems to have remained more complicated for even after a decade of research on experimenter expectancies, Rosenthal (1966) advocated the use of experimenter alter egos, those professional experimenters who would stand in for the invested experimenter.

The real versus the staged

Another project relying upon Goffman’s early writings pried not impressions but the underside of ritual interactions, the hidden self, and that project intimated how the entire social world is not much more than stimulus props for the realization of (often invisible) individual “intents.” Social psychologist Edward (Ned) Jones took Goffman’s theory of the social self as the starting point for examining “ingratiation” and its component processes. Initially defining ingratiation as “those episodes of social behavior that are designed to increase the attractiveness of the actor to the target,” Jones acknowledged the influence of the essay “On Face-Work,” notably its focus on maintaining and saving face in such episodes. He then distinguished his theory project from Goffman’s by insisting that ingratiation goes beyond routine face-work. Jones introduced new terms that describe ingratiation as a “contract violation”: it is an instance when the actor “violates the face-work contract while seeming to validate it” (1964, p. 4). Ingratiation, by this definition, is “non-normative behavior under a normative guise” that involves cognitive processes such as “autistic distortion” (p. 22). In other words, it is “the illegitimate and seamy side of interpersonal communication” (p. 3).

Ingratiation theory significantly departed from Goffman’s, but Jones’ repeated differentiating of his project from Goffman’s is occasionally belied in his own writing. Although purporting to depart from what he claimed to be Goffman’s overemphasis on the “silent conspiracy” of face-work, Jones over and again characterized ingratiation as subversive, manipulative, and deceitful, even self-deceitful. And while aiming to draw a decisive line between the normative and non-normative social interactions and suggesting how that division distinguished his work from Goffman’s, the line sometimes was blurred. For instance, Jones acknowledged that normative as well as non-normative social interactions could wreak havoc in everyday life, observing, “Ingratiation and related activities of face-work generate a considerable amount of mischievous noise which masks and distorts feelings and judgments as they are conveyed across the interpersonal chasm” (1964, p. 19). The emphatic distancing of his own theory from Goffman’s appears to be at least partially a rhetorical gesture.

If ingratiation theory did, contra to Jones’ claims, show substantial indebtedness to Goffman, other aspects of this reformulation of ingratiation did not. Jones endeavored to “cut through the evaluative atmosphere” that surrounds our usual conception of ingratiation by “wrenching a term from its everyday context and assigning it status as a psychological concept” (1964, pp. 1–2). The social situation and the meanings it prompted were rendered unnecessary to understanding ingratiation. Detaching ingratiation from its everyday context and the flow of social interactions meant that it could be placed in the laboratory. Once relocated there, the scientist could “gain insight” through precise comparisons that are unlike those of the participant observer who cannot always penetrate a person’s deceptions and self-justifications (their “autistic distortions”, p. 12). Yet in Ingratiation, Jones admitted that scientists could be duped, noting too that detached objectivity is sometimes difficult to realize. Ingratiation theory discarded the contingent and intersubjective dynamics of impression management and face-work, factors associated with “the imprecise but intuitively cogent dramaturgic approach of Erving Goffman” (p. 17). Instead, Jones restricted observations to individual actors. By observing not social interactions or situations but individual intentions (cognitions, motivations, perceptions), the analyst’s attention shifts from the social to the individual and from the publicly observable to the invisible. Ingratiation thereby was rendered an intrapsychic phenomenon or, stated more precisely, a set of “perceptual and cognitive achievements” that could be measured by controlling and observing laboratory behaviors (Jones, 1964, p. 17). Roles were taken as the mirrors for an inner self who can perceive and use the reflections (see also, Gergen, 1965). And although periodically inserting the language of face-work, impression management, and drama, Jones introduced rhetoric of “contract” and economic exchange that seemed better suited to an individual, intrapsychic perspective. Ingratiation theory thus married Enlightenment notions of contract and commerce with those of subversion and deception.

Finally, while purportedly eschewing evaluative claims (in accord with Goffman’s attitude of moral detachment), Jones rehearsed the dangers of ingratiation for productive exchanges and sound citizenship. He repeatedly stated the need to detect, dissect, and protect against ingratiation, claiming that although “an ubiquitous social phenomenon,” undetected ingratiation opens the way for bribing and other deceits. It is harmful for individuals for although “social interaction between human beings is a strange and wonderful achievement,” individuals need to veridically perceive and sort information in order to take “effective action” (1964, pp. 17–18). Simply put, ingratiation masks and distorts accurate psychological information; it consists of surface interactions that misrepresent depth. Yet again, Jones acknowledged that ingratiation could have positive outcomes, adding that even trust “can be an outcome of hypocrisy” (p. 20). Implicit in these judgments is a valuing of objective awareness and self-awareness—a scientific awareness—over the deceptive trickery, the subversiveness, of social interactions. Veridical perception, achieved only when phenomena are “wrenched” from the social situation, stood as desiderata, and this ultimate if tacit twinning of ideal citizen and ideal scientist helps explain why Jones, unlike Riecken and Rosenthal, expressed no apprehension about the experimenter’s intentions or capacities.

Extricated from ritual interactions and the mutual dynamics of intersubjectivity, ingratiation is defined as intrapsychic, having the potential to derail effective cognitive processes as well as social actions. It is decipherable primarily in terms of perceptions and cognitions, not through analysis of social interactions. Understood as invisible, labile, and potentially harmful, ingratiation’s moral dimension can be appraised in terms of its outcomes for individual actors and groups. Departing from Goffman’s extensive if mostly tacit social critique, Jones’ open moral appraisal enabled readers to better discriminate between the black and white, the bad and the good, of individual acts. And contra Goffman’s appreciation of deception as productive in social interactions, ingratiation theory posited that spotting deception is a first and necessary step toward accurately observing the real matter of the self. In other words, analyzing deception and self-deception in acts of ingratiation makes it possible for experimenters to distinguish between real and apparent, the good and the bad intentions as well as the “real” (cognitive) self that underlies self-presentation.

Despite these differences, Jones’ theory of ingratiation as well as the productive research it spawned is significantly indebted to Goffman’s theory of social interaction and the self. Yet the payoff of that debt was ironic in presupposing that researchers need to get behind the social to discover the true self. Deception thus was rendered a methodological challenge to be corralled through well-designed experiments which themselves often entailed deception on the part of experimenters. Continuing the experimental search for the real self in subsequent research, Jones, together with colleague Harold Sigall, considered this quest to be common among psychologists, conjecturing, “Many psychologists for many years, beset with the vexing difficulties associated with inferring true feelings from behavior, must have had fantasies about discovering a direct pipeline to the soul (or some nearby location)” (Jones & Sigall, 1971, p. 349). In so much as fantasies involve magical thinking, the “magic” of Jones and Sigall’s fantasy ironically resembles Goffman’s appreciation of the social dynamics of self. Their fantasy inspired a solution for reaching beyond the social situation. It was a “bogus pipeline,” a scientific apparatus designed according to the assumption “that if people can be made to believe that there are devices reflecting their true inner attitudes or feelings, the measurement possibilities are almost limitless” (p. 354). The bogus pipeline consists of a dummy instrument (either an assessment scale or physical device) that is described to the submitting participants as being able to accurately discern participants’ inner, true states. The bogus pipeline’s reliance on appearances ironically illustrates Goffman’s original account of social interactions as productive events. And in order to function, the technical “pipeline to the soul” exploited the very dynamics of social interactions, especially the productive performances of deception that Jones aimed to dissect from his borrowing of Goffman’s theory.

Conclusion

While the postwar decades saw tremendous expansion of scientific studies of social life (Herman, 1996; Solovey, 2004), it also fostered other human science projects, notably theories inspired by the uncertainty of world events as well as by phenomenology, existentialism, psychoanalysis, social criticism, avant-garde literature, and expressionist art. Goffman undertook his work in this dissonant if not entirely “rudderless” cultural atmosphere, and his alchemy pried several binaries of human nature that were entrenched in conventional scientific epistemology. He demonstrated ingenious awareness of the constitutive and reflexive powers of social life, and in this regard he trumped positivists in making the invisible visible and replacing linear causality with feedback loops and probablism.

Within postwar psychology where, to many investigators, theory seemed to entail an unhappy choice between behaviorism and psychoanalysis, the early adopters of Goffman’s work appreciated its creative reformulation of the social self. Henry Riecken and Robert Rosenthal utilized that theory of the self to interrogate a vital ritual of scientific psychology, the experiment, although their efforts to substantively rethink experimentation were curbed by commitments to preserving psychology’s signature method of inquiry. By contrast, Goffman unswervingly disregarded such scientific desires for objective language, moral neutrality, detached observers, and unambiguous distinctions between the individual and the world, between real and apparent (Burns, 1992; Geertz, 1983). Despite their commitment to experimentation, Rosenthal’s programmatic methodological reforms retain vestiges of Goffman’s ideas about performing the self, and Riecken’s narrative resurfaces as a still vibrant account of participants’ subversive underlives.

Selective appropriation of Goffman’s theory of self is most evident in psychology projects that selectively utilize his early work through painstakingly teasing the inner self from the role-taking self, the real from the apparent, the authentic from the artificial. In fact, the renaissance of self-studies began in the late 1960s is notable for distinguishing between persons’ “self-image management” and their “intrapsychic” selves—for distinguishing between apparent and real (Hales, 1985; Tetlock & Manstead, 1985). This ontology of inner and public self stands as a significant, enduring translation of Goffman’s early theorizing. Jones’ theory exemplifies how ritual interactions and managed impressions came to be taken as the observational markers of hidden cognitions, perceptions, and affect. Necessary to this discriminating usage of Goffman were notions of deception and self-deception, although these came to be understood not as productive processes or performances but rather as problems of method or good citizenship. Harnessing deception via the “psychologist’s advantage” (Scheibe, 1978) enabled psychologists to gaze through surface appearances into the cognitive depths of self, an ironic use of a productive catalyst in Goffman’s alchemy.

Appropriations of Goffman’s early work show the portability and mutability of theory, illustrating as well the dynamic relations between theory and method. The cases examined here illustrate the vivacity of theory and the importance of examining the socio-political and intellectual contexts in which a theory is born and assessed. These cases show how theory can travel and be re-imagined to serve as a tool for projects whose overarching model of human nature differs from that of the originating theorist’s. Importantly, histories of theory contribute to genealogies of human nature. Social scientific theories, by definition, contain architectonic assumptions about human nature. Chronicling their appearance, travel, interpretations, and applications uncovers cross-disciplinary trades as well as the ontological tenets about human nature that circulate, mutate, and coalesce in a given era. The history of theory, then, reveals much about our desires and fears about humankind; it informs us about visions not taken as well as those we hold. Within psychology, Goffman’s theory guided ontologies of a bifurcated human, one whose reasonable or rational poise covers less than reasonable intentions and irrational ones. While this tenet pervades contemporary psychology, quite different ontological outcomes of Goffman’s work have been realized elsewhere. Performance studies, also emerging from postwar thinking, gained from Goffman an account of how human nature is the effect of inextricable forces between actors, spaces, actions, and representation. Performance theories examine how the self is brought into existence through these conditions and thus offers an underappreciated yet important approach to psychology’s self theories. Considering the vivacious and complex lives of theories such as Goffman’s can enrich our present theory building.