Philip Manning's account of Erving Goffman's sociology comes hard on the heels of Tom Burns' gritty exposition (Erving Goffman, Routledge, 1992). Where Burns writes from inside Goffman's preoccupations, as if setting a difficult record straight, Manning stands back, pointing out the salient features of an oeuvre whose unusual form might be perfectly rational. Writing from the perspective of bureaucratized (mainstream) sociology, Manning thinks Goffman is best yoked with the ethnomethodologists and the symbolic interactionists to break the ground for a "general (structurationalist) theory or paradigm for sociology" (p. 179). But this is more hopeful than realistic, since many ethnomethodological Conversation Analysts (e.g. Schegloff) regard Goffman's analyses as too crude and too wild. And Manning does not offer fresh ethnomethodological reasons for reading Goffman again, even though the thoughtfulness with which Manning sums up the rules issue is unmatched in the literature. Further, no ethnomethodologist could accept the prospect of disparate empirical sociologies converging towards a Giddens-type theory that copes with both social and system integration. So Manning is a little guilty, I feel, of enlisting Goffman in a sociological project of whose legitimacy Goffman himself was no propagandist. It remains to be seen whether Robin Williams' Goffman book (Polity again) will take yet another tack, or what sort of Goffman emerges from the Sage book by Mike Hepworth. But no doubt Yves Winkin's intellectual biography of Goffman (also from Macmillan) will clarify Goffman's actual sociological intentions. (Neither Burns nor Manning, oddly enough, make anything of the fact that those intentions were left dangling by Goffman's early death. Yet the idea of Goffman's being an unfinished sociology must lie behind any Goffmanian study, and must suggest too that what might follow Goffman could be not only the empirical testing of his theory but also "more Goffman" in contemporary guises.)
Erving Goffman and Modern Sociology consists mostly of lucid resumes, each appended by brief responses to typical criticisms. The resumes are informed by some very good new ideas about Goffman. For example, Manning notes crucial differences of mutual expectation with regard to situational propriety, involvement, accessibility, and civil inattention when interactants are familiars instead of strangers. Here is a badly underdeveloped Goffman theme, of pressing relevance these days when the transition from stranger to familiar is made by more people than ever before, as is the transition the other way. (Some progress in this vein could start to elucidate the new world order of civil wars.) Also, Manning's reading of Frame Analysis as a continuation, minus the dramaturgical metaphor, of The Presentation of Self in Everyday Life is inspired, depending as it does on a vital distinction hitherto not made so explicitly. This is the distinction between trust and cynicism in interaction. Manning argues that in the 1959 additions to the 1956 Presentation of Self (Edinburgh) Goffman discovered trust `behind' the cynical dramaturgical metaphor (or remembered it from his 1953 Communication Conduct thesis). And through this argument Manning grasps Goffman's mainstay method of exploiting a metaphor in the first place to create a field of inquiry before deliberately seeking out awkward exceptions that force the narrative to "spiral" (p. 55) to new definitions. So one can see why Goffman shuttles between confidence in the ultimate explanatory power of analysis and despondency in the face of empirical complexity: both moods derive from the same method.
At this point a weakness of Manning's approach becomes apparent. Manning's detachment towards Goffman carries over into a presentation of Goffman as a detached user of clearcut methods and metaphors, as though Goffman's strong moral views about the inbuilt injustices of Anglo-American interaction are not a smouldering and sometimes incandescent presence in Goffman's best writing. A quite different Goffman, however, can be constructed if one starts from his texts as texts, concentrating on their rhetoric and composition while puffing aside the caveats of Goffman's Prefaces and Introductions (these in any case tend to parody protocols rather than confess genuine doubts. A textual Goffman is a sociologist with a mission to unmask vested orthodoxies wherever these are encountered, while a method-using Goffman is one whose sociological eye is not violently ironic. The key to Goffman's thought, then, may be "Goffman" rather than his methods in abstract, and Goffman like every serious writer is not less or more than his or her language.
If the writing discipline of sociology is a political struggle whose victors feel entitled to scorn the social visions of the losers, it is probably sensible of Manning to portray Goffman as a complementary foil to the ethnomethodologists (Goffman ferrets out the rules implicit in "What should I expect?" while ethnomethodologists and Conversation Analysts pore over the rules implicit in "What should I do?"). But, if sociology is radically committed to imagining society as a phenomenon that might be critically explored, then it is depressing to read Manning's apologies for Goffman that say things like (of spatial units) "significant amounts of analytic and empirical research are needed" (p. 168) and (of conversational moves) "considerable amounts of detailed research" are required (p. 169). This is neither to decry detailed analyses of audio and visual recordings nor to decry the practice of exhaustively analysing such data (Birdwhistell's goal in 1945, before Goffman and before Garfinkel). As Hilbert argues (The Classical Roots of Ethnomethodology, University of North Carolina Press, 1992), ethnomethodology and CA might be performing a function analagous to that of the astronomer Tycho Brahe whose findings went on to inform Kepler's and Newton's laws (likewise Manning wonders whether sociology currently resembles biology before the discovery of DNA in 1940, and then wonders if 1840 might not be a better date of comparison). But recommendations that Goffman be tested by methods that he himself found premature, while acknowledging Goffman's intuitive genius, slight it (l) by confining intuitional practice to Goffman alone and (2) by not considering that Goffman's creative thought, which goes to the heart of our immediate social experience, may not also deeply question the possibility of understanding life without stylistic and metaphorical involvement (for the latter view see Juliet F MacCannell, "Forms of talk/figures of speech," in S H Riggins (ed), Beyond Goffman, Mouton de Gruyter, 1990; Patricia T Clough, Chapter 6, The End(s) of Ethnography, Sage, 1992; and Lawrence Hazelrigg, `Reading Goffman's Framing as provocation of a discipline,' Human Studies, 15, 1992, 239-264).
Missing from Manning's Goffman is Goffman's concern with self. Despite the prominence of self in the title of his most celebrated book, Goffman never locks on to the self as a topic in its own right. All his remarks on self are asides in studies of interaction where interaction is conceived as a miniature social system rather than a dance of people who are running up and down the same gamuts from forgetting themselves to feeling painfully self-conscious. Not only that but Goffman consistently minimizes the role that his own self might be playing in his ethnographizing. Yet when Goffman performs a field study - when whom and what you are for others is an inevitable obsession - he returns again and again to the moral perils of selfhood (indeed he says in the Introduction to Asylums: "A chief concern (of this `volume') is to develop a sociological version of the structure of the self"). My view is that the creation of selves in interaction is a neglected area of microsociological research (see my `Conversion of self in everyday life', Human Studies,15, 1992,169-238). For, if selves were the foci of studies, people would stop disappearing into their ethnomethods and frames, and, as a result, interactional shifts would begin to make the sense of collaborative biographizing that is quite impossible to see either in a frame analysis or through ethnomethodological preconceptions. But it takes a self - not necessarily an ethnomethodologist or a sociologist - to see a self. And it is not obviously scientific to write analyses that say: "I am becoming this self through seeing him and her and them as becoming those selves that are describable novelistically." Yet I would argue that it might be in the spirit of Goffman to write like that, though he, of course, was notoriously averse to self-disclosure. Further, I might suggest that, since one needs to know the whole of society to know a single self (modernist sociology), one could learn what society is by knowing the whole of a public self (as in Sartre's studies of Baudelaire, Genet, Mallarme, and Flaubert). To go even further, what if the study of society should therefore proceed as the study of its sociologists, the official ones (like Durkheim and Weber), the self-appointed (like Zola), and the ingloriously mute? Then a book about Goffman would be a book about America (as a total institution?), which Philip Manning's book, for all its virtues of economy and grace, is not. Alas, I have strayed from Erving Goffman and Modern Sociology, by wandering after Goffman. Erving Goffman and Modern Sociology is first-rate sociological exegesis but it is not Goffman. But that's the fault of Goffman's extra-sociological perception.Andrew Travers