Wittgensteinian and ethnomethodological investigations

Dedicated to

Royal Marine commandos everywhere

Very Special Thanks also to the crew at Manchester University and Manchester Metropolitan. And others who like to hang out there when they're not in the States. Cheers for putting me up Margaret!

From the visual arts I would like to thank Jeff Woodhead, Paul Hatton, Rob Williams and Sue Platt.

A Very Special Thanks to Irene Selway of Portsmouth University for getting me back to school.

Special Thanks must also go to 'The Team' for favours and stuff: Jon O'Brien, Mark Rouncefield, and new kid on the block Andy Crabtree. Also included are numerous friends: Lucy Warren, Caroline Schwaller, Ulrik Petersen, Luna, Rui, Melissa, and Barry Sanderson who particularly and emphatically said that he did not want his name associated with this thesis.

Here at Lancaster I must say thank you to all members of The Department of Sociology. Maeve is retired but not forgotten. Thanks Chris, Karen and Cath for this 'n that. But special thanks to Balihar Sanghera for being a cool dude these last three days. Likewise I would like to acknowledge the support I recieved from Janet Clements, Emily Brady, Ceris Reed, Anne Maguire and the porters of Furness College here at Lancaster University.

And how could I forget all those who undertook 'Sociology 318: Ethnomethodology'. Good luck to you all, whatever you're doing.

Last but not least, drinks and hugs to Cal Giles, Jenny Ball, Val King and Jason Khan.

Targeted by Formal Analytic (FA) technologies, lived immortal Society escapes seen-but-unnoticed. Unable to administer FA’s technologies Harold Garfinkel’s ethnomethodology finds order any place it cares to look. Taking this precept seriously with respect to the plastic arts. the thesis wants to avoid addressing the ‘big picture’ but, instead, remain content with investigating the small, minutiae of art-ful practise out of which the ‘big picture’ is achieved.

The thesis takes the form of three ‘forays’ into the aspects of the practises of the plastic arts. In particular, those to do with use of colour. The first section sets the scene and an opportunity for ethnomethodological exploration of recurrent themes in the plastic arts. This is followed by an extended effort to bring colour back into the public from its imprisonment in the ‘inner’ and its cognates by using the notion of ‘geometry of colour’ derived from Wittgenstein’s work. Building on from this, the final substantive ‘essay’, using real time study of artistic activity, discovers phenomena hitherto unavailable to the Formal Analytic Technologies of the sociology of art. Finally, some remarks are offered addressed to the ways in which colour use is used by sociologists as the basis for ‘pointing things out’.

Dedicated to 1

Acknowledgements 2

Art and the arts are difficult and complex subjects of study but the problems which beset the sociology of art have nothing to do with that, rather, its troubles revolve around an overwhelming concern with internal debate as to who has the best theory, the deficiencies of other theories, what other theoreticians should do, and the possibility or otherwise of all coming together to produce 'the big picture', even in the face of well known irreconcilable difficulties and differences.

The aim of this thesis is simple, how to get there is not. The aim is to think small in the face of sociologies of everything. To think small is not a reference to 'micro sociology' it refers to 'running with something in order to see what turns up'. And it has been my 'luck' to run with 'the use of colour'.

There is 'a gap' in the literature. 'Getting colours to work' is a 'renegade topic' whose observably achieved phenomena is not to be found in the accounts of physicists, or sociologists. Hence my interest is not in artistic activity as process. I am interested in artistic activity as activity. My focus is on the haecceities of artistic work and how artistic work specifically the production of the dynamics of colour is displayed in and testified to by the ordering of its work.

The first chapter is an attempt to open up a space for study at the same time as it speaks past those who cannot understand what it is doing there.

The second chapter belongs to Ludwig Wittgenstein because I've taken the words right out of his mouth. Though how they appear on paper is not his fault at all. In brief, it is an attempt to rescue colour from being anyway but in the public domain.

The third chapter is mine. And it shows. It is an attempt to explicate some features of colours, colour words, and 'the use of colour' in and as artful art practices. Apologies to Professor Harold Garfinkel.

Jeff Woodhead is a professional artist living and working in the North West and is active in organising exhibitions of regional artists for which he is paid a small stipend. To supplement his income he will often attend local colleges giving demonstrations usually on the use of colour.

The fifteen or so minutes videotape shows Jeff demonstrating the 'use of colour'. His chosen method is a simple collage. The technique is as follows. A large (8' x 6') plastic sheet is laid on the floor and large sheets of cheap paper are painted in various colours which are then torn into strips of various lengths and sizes. The strips are place colour side down on the sheet and stuck on by pasting over them with a flour based paste. The pasted backings in turn enable various other strips to be partly overlaid. Once the surface of the plastic sheet is covered with overlapping strips of colour a backing consisting of white rolls of paper is placed on top of the by now heavily pasted area. Once done the whole sheet is turned over and then carefully peeled away. The result is a simple collage of colour and constitutes 'the finished piece'.

The lesson:

Strong claims indeed.

What am I being instructed in?

Bowler's lesson is this: Situating the arts in their social, political and historical contexts the sociology of art (hereafter SA) prides itself upon having problematized at least the following: a) the artist as genius, b) 'great works', c) stylistic change, and d) readings or the consumption of art. And the way SA has problematized all these things is largely through talking about art in relation to issues of ideology, power, domination and so and so forth. Hereafter art history is The Dictionary of Art Titles and Artists

ART: ABSTRACT EXPRESSIONISM; ART DECO; ART NOUVEAU; ART TROUVE; ASHCAN SCHOOL; ASSEMBLAGE ART; AVANTE-GARDE; AURA; AUTOMATISM; BAUHAUS; BIOMORPHIC; BLAUE REITER; DER BRUCKE; COLLAGE ......ETC.

There exists in the bibliographies making up Dictionary of Art Titles and Artists (DATA) a corpus which would take great exception to their characterisation as peddlers in myth. A cursory glance will suffice:

It is Mark Cyzyk's (1991) view that the ahistorical and asocial humanist is a straw man which takes no account of Arthur Danto's 'The Artworld' (1978), George Dickie's Art and the Aesthetic: An Institutional Analysis (1974) and several other acclaimed works.

By the middle of the 1930s Max Raphael had argued against a vulgar Marxism which conceptualised art as simply reflecting the class interests or ideologies of history. And whilst calling for a new art history which grasped "all of social life in terms of a single method" he added that it should leave "relatively autonomous the specificity of each domain". And if Raphael hadn't already pre-figured much within SA Arnold Hauser certainly did. Although his two volume history of the Western tradition has been criticised for its class reductionism there is no doubt his The Philosophy of Art History, (1959) pre-figures many of the concerns currently fashionable in the sociology of art in particular and sociology and cultural studies in general. The writer and curator John Roberts has written of Hauser:

T. J. Clarke (1973) carried on the Marxist obsession with production but placed it within an analyses of the ways in which the artist was involved in and constrained by the political and social concerns of the day. Today however there is the beginning of a backlash: John Roberts argues that the fear of formalism has gone to far and calls for 'a sophistication of analysis that gives back to historical materialism the socially and historically specifiable agency of arts making' (1994:33).

Erwin Panofsky (1957), Tim Clark (1973) and Michael Baxendall (1972) have argued that artists' visual languages are linked to the society in which they live. For example, between 1915 and 1932 Erwin Panofsky had already characterised the work of van Eyck as an attempt at reconciling the demands of symbolism with perceptual-realism (cf. Early Netherlandish Painting). And in Gothic Architecture and Scholasticism he posited the theory that there was a link or cultural coherence between the development of Gothic architecture and the development of scholasticism. And in Beschreibung und Inhaltsdeutng (1932) he posited a concern with what Michael Podro describes as the "unconscious attitude which a work betrays" rather than the "meaning which it displays". Adding: "It is a meaning of which the artist himself cannot be aware, it lies outside the limits of his possible self-consciousness" (Podro 1982:202).

Michael Baxendall has argued that a distinctive social experience can be ascertained through an analysis of pictorial style. In his study of 'seeing' in 15th Century Italian art we find the following little gem:

Equally interesting is Svetlana Alpers (1983) analysis of scientific 'vision' and the work of Jan Vermeer (1632-85) and which was so important to Bruno Latour's notion of the 'immutable mobile'. Equally important to Latour was the work of William Ivins (1953, 1973).

Latour (1990:27) writes that Ivins "saw it all in a few seminal pages". What Ivins saw was that sight was social in character. But more than that Ivins saw the importance of the invention of 'perspective'. In short Ivins and Alpers described how a culture sees the world and makes it visible. Sociologists such as Robert Witkin (1995) would come to denounce them for not being properly sociological but all his sociology would entail is an insistence that their work needs supplementing with his 'New Age' sociology of self-reflexive 'intra-actional' relations and 'sensuous-identity' making.

In the literary field the New Historicism has attempted to situate works 'not only in relationship to other genres and modes of discourse but also in relationship to contemporaneous social institutions and non-discursive practices' (Johnson 1993). For example, Edward Said (1983:174) has conceptualised the relationship between literature/art and its socio-historical context in terms of 'affiliations' defined as an

These affiliations 'fix' the artist and her work in a complex of social relationships including the

Nor have the humanities been silent in matters of self-reflection and 'deconstruction'. For example, Svetlana Alpers (1977) has taken art history to task for being narrow minded in privileging oil and canvas over collaborative works, montage and other media. Frietag (1979) and Fawcett (1982) have shown how the recent emergence of art history as a discipline was unavoidably embroiled in the dispute over visual depiction in photography. And in the visual arts the artist and critic Victor Burgin (1986) has analysed the problematic relationship between classification and 'meanings'. Donald Preziosi (1989) has examined the role of rhetoric in the construction of art history and the use language for the purpose of historical interpretation. Norman Bryson (1992) has criticised art history for not acknowledging the role of the historian in producing 'historical evidence'. And following Derrida's contention that linguistic signs are arbitrary constructs whose meaning is unstable K. Moxey (1992:39) has argued that art history has no privileged position from which to claim a 'correct' interpretation of art history at all!

The sociology of art is driven by an enforced distinction between the sociologist and the humanist (aesthetician and art historian). The sociologist will concede that art historians often incorporate social, political and theological matters into their art histories adding that their analyses are inadequate in one way or another.

On the whole sociologists display a wilful arrogance as they continue to characterise the art historian and aesthetician as a peddler in myth. The general accusation from SA is that the humanist conceives of the artist as a unique genius, unaffected by social surroundings and generally disallowing 'external' factors such as politics, economics, or society to figure in an account of artistic creation, dissemination, and evaluation (Cyzk, 1991:192).

It is apparent that Art History, New Historicism and the New Art History have all attempted in one way or another to demonstrate that "formal and historical concerns are inseparable, that human consciousness and thought are socially constituted, and that possibilities of action are socially situated and defined" (Johnson 1993:19). Younger and less practised in the mysterious analytic crafts of sociology admittedly but 'traditional' art history and literature studies is a rare beast indeed.

It remains the great achievement of SA to have demonstrated that the work of art and arts work are socially produced, shaped, influenced and constrained in some way by socio-historical forces and processes. SA's (hereafter 'Formal Analytical Studies of Art' (FASA)) achievements are well known within sociology. FASA's analytic procedures are those of its parent, 'Formal Analytic technologies' and are understood world-wide: 'Almost unanimously for the armies of social analysts, in endless analytic arts and sciences of practical action, formal analytic procedures assure good work and are accorded the status of good work' (Garfinkel, 1996:5). FASA has targeted innumerable phenomena for analysis and that the achievements of FASA are beyond dispute is attested to by its bibliographies. It has targeted areas for analysis by 'describing', 'specifying', 'testifying', 'showing', 'demonstrating', and generally supplying adequate grounds of further inference. By its practices a phenomena of order are made instructably observable. Such phenomena consist in 'recurrent achievements of practical action'. These range from opinion forming to sexual oppression.

Ethnomethodology does not dispute these achievements. Ethnomethodology asks 'what are we being instructed in'.

Janet Wolff’s achievement has been to demonstrate: 'Art is a social product' (1993a:1). By which she means that artistic practice "is situated practice, the mediation of aesthetic codes, the 'cultural unconscious, and ideological, social and material processes and institutions" (1993a:137). Subsequently, her sociology of art is the study of 'the practices and institutions of artistic production'. And her work 'discloses the ways in which these practices are embedded in and informed by broader social and political processes and institutions, with economic forces historically playing a particularly important role' (1993a:139).

Vera Zolberg (1990:54-5) has produced instructions towards the study of

For sociologists, more important than art creation is the social process of status creation, to whose end certain art works are selected for inclusion in the canon of elite art. The sociologist is more interested in the symbolic use of art, rather than the work itself (Zolberg 1990:56).

Howard Becker has amassed a study of group processes and institutional constraints. Demonstrating the essentially collaborative nature of art works Becker writes:

Becker's remarks pre-figure his Art Worlds (1982) which re-emphases that the actual creation of the art object is often a product of collaboration rather than the work of an individual artist.

Though much admired for his work, FASA technologies are able to incorporate Becker's observations on the sub-cultures of 'artworlds' within more 'macro' concerns:

Common to FASA is the notion that art is situated and that any simple base-superstructure model of the relationship between art and society is inadequate. It is argued, therefore, that the simple base-superstructure model of art and society be replaced. Wolff's remarks are typical of this concern:

But what am I being instructed in?

FASA's 'relative autonomy' thesis is premised on a particular relationship between art and society, namely, 'art' is not reducible to it. Janet Wolff again:

But what is it we are being instructed in?

FASA has begun to amass a body of knowledge which demonstrates that artistic creativity is labour:

And:

Although creativity and social structure are perceived to be in a relation of mutual interdependence creative practice is

FASA has thus shown that artists' creative acts are not the 'free' transformative acts of non-alienated labour rather they are 'distorted by capitalist relations of production and work'.

Again, what am I being instructed in?

Employing the technologies of FASA a growing band of practitioners have noticed a gap in the literature. Subsequently they have begun the quest to make visible a phenomenal field of order hitherto ignored: gender. Tucked away in libraries the world over or out in the field, FASA practitioners have demonstrated and discovered that representations of women are but one site of social control in capitalist culture. After scrutinising the field in greater detail they have delineated a mechanism of masculinist logic i.e., the systematic subordination of women through the signs of sexual difference (the so-called sexual ‘visibility’ and social ‘invisibility’ of men and women).

A small but growing band have targeted and made visible a phenomena of order hidden within the arts. For example, Griselda Pollock (1988 and 1992) has argued that male artists established their 'modernity' on the backs of women often portrayed on their backs. Terry Lovell (1978:95) argues that aesthetic sensibilities are 'class- and sex-linked', and that the "politics of aesthetic pleasure will depend on the particular ways in which that sensibility has been appropriated and developed along lines of sex and class". Helen Hills (1993:379) writes that the modernist canon marginalizes women and supports "gender inequalities". Sara Mills (ed.) (1994:1) asserts that "texts gender their readership".Linda Nead (1992) sees the history of the nude in art as an element of a "cultural continuum" within which "sexual difference and subordination" is played out.And Janet Wolff (1990:138) has declared:

Demonstrating that FSA is not a distinct corpus from other bibliographies Griselda Pollock (1988:30) has begun to analyse the interdependence of gender, race and class:

So what am I being instructed in?

FASA's technological wizardry is attested to at the very moment it is interrogated by a corpus showing hostility towards its descriptions. For example, criticising Pollock's 'open dialectics' and its 'reciprocal relations' between class, gender, and race, on the grounds that it downplays 'the asymmetric causal relations' between base and superstructure, John Roberts (1994:16-17) writes:

But what am I being instructed in?

I have attempted to demonstrate in quite a severe manner that Ethnomethodology has a problem with the sociology of art. Ethnomethodology wonders if the sociology of art’s phenomena of interest has disappeared or is in danger of disappearing. That is, Ethnomethodology wonders if an emphasis on art as socially produced, shaped and constrained in some way stands the risk of hiding the phenomena sociology wishes to address. For example, instead of examining what it is about human activity and human interaction that make the plastic arts the recognisably distinct phenomena they are understood to be by those who work in them, an analysis of the forces which allegedly shape art can end up taking precedence and the art and the arts become yet another incidental area in which to observe such forces and processes at work.

In the jargon, ethnomethodology asks if there is a 'missing what' known and demanded of by practitioners artful in the ways of FASA. Ethnomethodology takes it that FASA's achievements demand the existence of practices already known, recognisable, unavoidable, and without remedy or alternatives. They are the practices which are indispensable to 'the armies of social analysts' engaged in 'endless analytic arts and sciences of practical action' (Garfinkel 1996). Ethnomethodology asks 'what are we being instructed in'. And I have attempted to show that FASA's technology answers with a literature of things familiar. Ethnomethodology or anybody else is invited to read a literature with a tradition displaying lines of studies which are singled out, perhaps for renewal, perhaps for attaching to a study in progress, or perhaps for an attempt at synthesis with others. even radically different others. And what the ethnomethodological question attempts to do in return is to show that despite the well known problems which exist within sociology it nonetheless is able to do that which it sets out to do unavoidably. And it does so unavoidably because its 'technologies' ensure its phenomenon. And that is sociology's great achievement.

Perhaps I can make it a little clearer by returning to the sociology of art and its treatment of artistic practices. I will then attempt to illuminate ethnomethodology’s concern with the 'missing what' through a detailed discussion of its major ideas.

Pierre Bourdieu (1984) has insisted that a sociology which reduces art and creativity to nothing more than 'production' ignores an 'internal history' which needs to be incorporated into sociological analysis. For Bourdieu artistic practices like any other exist within a 'cultural field' of 'possibilities' where kudos is 'measured' not so much in terms of economic capital as cultural and symbolic capital. With respect to the question, What makes a work a work of art? the answer is not to be found in the formal properties of a work of art itself since the so-called formal properties of any work of art are themselves the product of specific socio-historical conditions. What makes a work a work of art is belief:

'Internal' explication of the kind allegedly found in art history and art criticism ignores the complex social and institutional framework "which authorises and sustains art" and artistic practice. In other words the 'aesthetic formalisms' of art works must be inserted into the universe of belief, which supports them.

But in this world of fields and universes of belief no one appears to actually create anything?

But what is this 'creative project'?

All familiar stuff, but what about the issue of creative activity?

Is one any further forward on the issue of creative activity?

The problem with Bourdieu is that he is a determinist in the last instance. Having once described himself as a 'constructive structuralist' Bourdieu went on to say:

So when it comes right down to it Bourdieu isn't saying anything that is in any way unfamiliar to sociologists the world over. Indeed it is positively Durkheimian; I will return to the issue of objective structures shortly but for now one need only note that if practices are possible at all in Bourdieu it is by virtue of the 'generative basis' of practices, namely, the less than conscious 'dispositions' of a 'habitus' which is imprinted and encoded onto the mind and body in a socialising or learning process. Frankly, I think Bourdieu is attempting to both have his cake and eat it; he wants to say that the artist is both 'carried by' and 'carries' the culture. And it is this which he is attempting to express with his notion of the less than conscious dispositions of habitus: The ascription "to the cultural unconscious the attitudes, aptitudes, knowledge, themes and problems, in short, the whole system of categories of perception and thought" inculcated through teaching and experience. And it is the "unconscious borrowings and imitations" which are the "most obvious expression" of the "cultural unconsciousness of an age, of that general sense which makes possible the particular sense in which it finds expression" (Bourdieu 1971:182). This cultural-mentalistic strain in Bourdieu is seldom criticised, rather, it tends to be seized upon as an example of how Bourdieu has escaped the agency/structure debate. It would be the contention of ethnomethodology however that Bourdieu displays the kind of Cartesian-constructionist turn rife in cultural studies. This Cartesian strain has been trenchantly described by Graham Button and Wes Sharrock (1992:3) and deserves quoting in full:

And so the words used by Bourdieu are once again familiar, and depressingly so because far from knowledge and belief being the wellsprings of action it is the latter which opens up the space for the possibility of 'knowledge' and 'beliefs' in the first place, that is, it is actions first and then 'knowledge' and 'beliefs'. Bourdieu’s notion of 'unconscious cultural categories' far from directing us towards real world activities directs us instead to a culture working in back or actual activities. The folly of this position is well known in ethnomethodology since its concern is with the features of any setting:

In ethnomethodological terms then Bourdieu's analysis is nothing less than a theoretical construct standing proxy for real people. Between the 'embodiment of habituation' and 'less than conscious dispositions' we have no idea how one or the other actually generates actual practices, save inferring that they do from their putative effects. So regardless of Bourdieu's instruction that we incorporate the 'internal histories' of the arts into sociological analysis a more radical i.e. ethnomethodological agenda would state that what is required is not incorporation but study.

To the uninitiated The Marriage appears to be perceptual-realist, the semiotically necessary style of the Renaissance but in fact it is a compromise of perceptual-realism with an older invocational mode of symbolising usually associated with Archaic co-actional social relations. The explanation is that the mode of symbolising displayed in The Marriage "is appropriate to a society that is still seigneurial, one in which the interactional mode of social relationship and the political forms associated with it are still at an early stage of their development" (p.154). In Cezanne's Still Life with Peppermint Bottle, c. 1894 on the other hand we begin to detect a shift from a concern with the object to a concern with 'significant form' and Witkin equates this move as mirroring arts autonomy from the institutional complex constituting the structure of the modern state and self-reflexive forms of social relations.

So once again what started looking like a sociological concern with art practice and artistic work has ended up being used to flag the more general sociological concerns to do with socially determined 'ways of seeing', social change, the social construction of identity, and the relative autonomy of art. In other words, it is all very well situating 'art' within wider social and historical processes but what is the relationship between them? From what has been said so far one might be forgiven for thinking that the art object and the work taken to create it are largely irrelevant, merely a handmaiden to some Benjamite task of reducing the 'aesthetic aura' of any art form by contextualising and intellectualising the art object and its creation in terms of familiar sociological categories i.e. 'capitalist relations', 'patriarchy', 'status', 'alienated labour', 'symbolic usage', 'cultural production and consumption', 'consumption as re-production', and the rest.

Within the sociology of art there is absolutely no serious mention or description of what 'art' e.g. painting, actually consists of as a practical endeavour. This is unsurprising since to all intents and purposes SA is merely a) about art, and b) what other social scientists and historians have written about art. But there remains no actual artistic practice in SA. And indeed were one to remove all references to actual art and artistic practice from SA one would be hard pushed to lose more than a couple of pages. Art is a difficult and complex subject but the problems SA attempt to tackle revolve around an overwhelming concern with internal debate as to who has the best theory, the deficiencies of other theories, what other theoreticians should do, and the possibility or otherwise of all coming together to produce 'the big picture', even in the face of well known irreconcilable difficulties and differences. But that there is a connection between art, the aesthetic and socio-political economic factors is, in the eyes of SA, not to be doubted; the question is how to theorise it. Given the fluidity and dynamics of 'art' and 'artistic practice' talk of the artist as acting within existing aesthetic codes and conventions really tells us nothing at all. To say as much is not a criticism since that would imply that there are more 'correct' or more complete definitions of art and artistic codes and conventions to be had. As I shall argue below, accounting for artistic activity in terms of codes and conventions as though they were definitions and rules is not only insufficient and foundationalist it is the very stuff of the constructivist-analytic theorising to be found in the sociology of art.

SA's epistemology is largely one or another version of realism. Broadly speaking they begin from the position that the social world is an objective world consisting of an inter-related complex of individual people, observable events and tangible things. Behind this observable world of appearances lies a material reality not accessible to direct observation but which explains what can be observed. And therein lies the rub: the social world consists of an observable reality of practice (including actors' practices of giving explanations for their own practices) but behind it there lies the 'real' explanatory, objective and invisible social reality of social and cultural forces and processes which operate behind the backs of actors. The epistemological assumption from which all else follows is simple: social reality is constituted as an objective reality, but between this objective reality and practice there lies the equivalent of a 'black box'.

And the 'missing what' is also evident when Witkin argues that the rational-techno machine aesthetic of late-modern organisations constructs a particular kind of identity. In brief his analysis consists in asserting that just as the organising form of social relations under the Modern is self-reflexivity so the modernist aesthetic is a move from mimesis to self-referentiality. Just as identities are constructed from within any social encounter so art refers to nothing but itself. Transported the analysis of organisations in late-modernity Witkin’s' socio-aesthetic construction-of-identity-thesis can be summarised in the words of Francis Duffy: "We make the buildings then the buildings make us". But what is missing is any account of just how they do it. How, in specifics, in the details, does a late-modern organisational aesthetic actually construct anyones identity? At the risk of repeating myself; we are not told. It is asserted that an organisations aesthetic does construct identities but we are left in the dark as to how it is done save perhaps from a general idea that it is something to do with wearing a uniform. The kind of detail being requested is of the order given by Harvey Sacks when he suggested that organisational identities can turn on the conversationally generated identities of answering a telephone. And if that should seem obvious just consider who gets to answer the phone when a number of people are in the same room!

The sorts of dilemmas faced by SA are self imposed ones arising out of the methodological choice to attempt to give explanatory accounts of social life. They set themselves up to settle explanatory questions and in so doing they are not so much involved with actually explaining anything but are more involved in questions concerned with the form of explanation. For example, Wolff, Zolberg, Bourdieu, Witkin, Bowler, Pollock, and others claim to take the actors point of view into account but the confusion and uncertainty that arises when they even attempt to account for the artists point of view are a function of their orientation to the social object studied because what they consistently fail to realise is that their conceptions of the social world are worlds which could not work. All the sociologies of art which I have described have attempted a synthesis of institutional and interpretivist models in one way or another but they have yet to come up with an adequate theory able to come to terms with the ways in which the social world is comprised of unique circumstances which are nevertheless recognisable as instances of general types. They have yet to produce a model of the social world that is "neither so inflexible and rigid that it lacks any sensitivity to the infinite range of contextual variation in the real world, nor so flexible that nothing at all is held to be general across different contexts" (Atkinson 1990:455).

There are very real problems within SA and sociology in general which stem directly from pressure to 'do the right thing'. The phenomena is cogently summarised by Max Atkinson:

And oddly enough Vera Zolberg had the nail in front of her but failed to hit it on the head when she remarked that the art historian is required by her peers to pay attention to such things as the artistic details of a work or the life of its artist. Likewise the sociologist of art has to be seen to be doing 'proper sociology'.

As any sociologist knows, the disputes that divide sociologists revolve around the familiar ones of inter alia the tension between objective appearances and subjective experiences, between structure and agency, between researcher and researched, between the particular and the general, and so on. Conflict theorists, Symbolic Interactionists, phenomenologists, feminists, and the rest, all claim to have identified the most appropriate model of social order and social action, as well as the most suitable research methods, even though there are no widely accepted independent procedures for deciding between the competing versions. Nonetheless they provide yet more data for yet more speculation aimed at producing some new synthesis or reconciliation between alternatives. And the enabling mechanism for doing that is the fairly widespread agreement that:

1 Despite all the evidence to the contrary sociology, or the favoured approach to it, "is capable of producing descriptions and explanations of social phenomena that accurately represent or correspond with the actual events in the world to which the descriptions and explanations refer."

2 There is a general agreement that sociological accounts of the social world are of a superior order "to the kinds of descriptions and explanations that are available to and routinely used by ordinary people in making sense of their everyday lives.

3 Finally, there is the view that "everyday methods of practical reasoning are in some sense 'flawed', and hence are to be avoided, or repaired, modified or otherwise cleaned up for the purposes of doing professional sociology" (Atkinson 1990:452-3).

To which we can add

4 To reject any of the above is to deny the possibility of pursuing any kind of sociological research at all. Instead what remains are the writings of journalistic reports, fiction, or political propaganda.

In common with much of 'The Sociology of X' the practical solution to fragmentation and internal debate appears to be what Lynch (1993) calls "fictive consensus"; that is, one or another framework is accepted and the researcher then 'cracks on' with the empirical work. But, the analytically prior problem, as Lynch conceives it to be, is not so much a problem of deciding which sociological framework we should plump for, but why we should think "any particular 'framework' is appropriate in the first place" (ibid.:30). The reality, of course, is that sociologists will often adopt currently fashionable frameworks of the kind, for example, espoused by Giddens and endorsed by both Wolff and Zolberg; the theory might need a bit of 'tinkering with' but it suits their general purposes. However, the problem with adopting the sort of theory-of-everything envisaged by Giddens (and numerous others) is that in the sort of research practices envisaged

Already then the researcher has placed a restriction on what the data might reveal. One surplus detail that is likely to drop out of sight in the domain relevant here is what-is-happening-now or the actual and see-able business of doing art. In those few instances where there is an attempt to consider actual artistic practice it is again made from within familiar sociological debates: Are artists born or made? Is 'creativity' just another form of alienated labour and thus not really creative? What are the effects of wider social and historical factors and contexts? How much is the biographical or the psychological a reason for, or a cause of, what artists actually do, and so on and so forth. To put it bluntly, artistic practice becomes not something to be investigated in its own right but just another tool for illuminating ones favoured sociological theory for the purposes of downgrading or pointing out the shortcomings of one or another favoured sociological theory.

Allow me to push the point further: when if ever the artist is taken into account it is in terms of a conception of the artist as 'cultural dope'. Complementing 'structure' with 'agency', or 'agency' with culture, gender analysis with Marxism, discourses with [unspecified] degrees of artistic autonomy or whatever, makes no real difference to the actual outcome: SA get the last word and can only impute self-deception to actors who do not recognise or acknowledge the analytical explanation. Put simply, SA takes an extra-mundane position, the social 'scientific' stance, in order to claim theoretical status for its statements with the upshot that real worldly activity becomes what Lynch calls "a docile matrix for exercising a theoretical will" whereby the theoretical will provides the framework into which any 'real world' phenomenon will 'fit'. The purpose of taking such a position of course is most often done for the purposes of engaging in the politics of empowerment. So although an authoritative critique of power is both tempting to do and an admirable endeavour Lynch points out that it has serious consequences for any notion of human agency.

Many of SA's problems emanate from sociology's problems in general. And sociology is a discipline which operates discursively in 'natural language' and that fact, according to Button, Hughes and Sharrock, lies at the bottom of many of sociology's troubles. This is not to say that 'ordinary' language is flawed, requiring repair in order to conduct a supposedly scientific sociology, rather, it points to the implications for sociological usage since sociology unavoidably trades upon and deals in 'what anyone knows'. For example, consider the claim that what makes a work an art work is something to do with a struggle over meaning. Anybody knows that what counts as art is a difficult question. Indeed, hardly a month goes by without the ‘quality’ or tabloid press railing against what they see as the excesses found in contemporary art. in other words though sometimes bemusing uncertainty over the definition of artworks is familiar to any competent member of society. Following the strategy of Button, Hughes and Sharrock the reason I draw attention to this is because of the issue of the 'social fact'.

Button et al begin their discussion by noting that according to Durkheim social reality had an objective reality existing independent of and external to individual consciousness. In one stroke, they argue, Durkheim "delineated an autonomous order of phenomena as an object of study for the young science known as sociology". But although the belief in social facts is widespread in sociology they argue that Durkheim was not clear on what he meant by talk of 'independence' and 'externality'. However, they also note that Durkheim did give examples of the objective existence of social facts.

Button et al argue that at this point sociologists will either agree in the existence of social facts, existing 'independently' of the individuals consciousness or they will deny their existence and argue that 'social reality' is nothing other than individual consciousness. However, Button et al caution against an either/or choice based upon what they see as vague abstractions arguing choosing instead look at what the apparent choice is in substantive terms. And the conclusion they come too is that Durkheim is not saying anything like as radical as one might think. They argue that Durkheim is saying little more than that "as we go about our daily lives we cannot just do as we wish". After all we are born into a society which existed before us and we are trained in the use of that society's language. But as they also note, telling us such stuff "hardly constitutes a scientific sociology". Why not? Well for the simple but crucial reason that Durkheim’s claims are pre-scientific ones. His descriptions of social facts are not derived from, dependent upon, or justified by scientific investigation; rather, and to restate the point Durkheim trades upon what anybody knows. Not only that, he demands that such things are common knowledge. Durkheim has not gone out, done a scientific study , returned and given his results. Not at all! He is speaking to us as a fully enfranchised member of society. And it is the recognition of that which is Alfred Schutz’s’ great contribution.

At lenght:

Schutz goes on:

In sum

As Button et al once again observe, the point at (g) affirms, Durkheim’s view that individuals encounter a 'world already there'. In other words it is a world handed down from others to the extent that any social situation is 'only to a small extent purely created by me'. However, they also make the crucial point that Schutz's claims differ from Durkheim's in that they are descriptive of the orientation under the 'natural attitude' whereby the givenness of the social world is taken-for-granted.

The claim that sociology 'trades upon' what anybody knows is a cause for concern within sociology’s ranks not least because it is treated as having epistemological significance. Briefly stated: sociologists think sociology provides scientific demonstrations which supersede commonsense understandings and so for them commonsense is not incorrigible it is flawed or otherwise distorted. In ethnomethodological terminology they are now engaged in an ironic sociology. To engage in an ironic sociology is to impose a standard exogenous to activities-in-a-setting (see below). Ethnomethodology on the other hand refuses to talk about commonsense in those terms. For ethnomethodology commonsense refers to the way in which, under the auspices of the natural attitude members of society, sociologists included, treat matters as 'obvious'. In other words, in contrast to an ironic sociology ethnomethodology treats the intelligibility of any setting as endogenously produced.

The point Button et al wish to stress is that in order to explain his notion of a 'social fact' Durkheim invokes matters presumed to be known by his readers: that one's language existed before one's birth, for example. In other words the plausibility of Durkheim’s argument is largely dependent upon commonsense understandings of what anybody knows. Saying such things of Durkheim is not a criticism but a description of a plain fact of social life, namely, that there is a shared reliance upon commonsense knowledge of social structures.

Consider now, then, the following remarks from Janet Wolff and which are typical of the oeuvre:

What are we being instructed in? The story presupposes familiarity with notions like 'art', 'politics', 'work', history', 'economic forces' and 'society'. Most of all it presupposes that who counts as a great artist, who can be an artist, and what is to count as art and what counts as a great work of art, is something to do with all the aforementioned notions. In that respect her contention that 'art' is a social product is redundant since who would suppose otherwise? What would un-social 'art' look like? The intelligibility of Wolff’s remarks come from our prior understandings of art and from much more besides. For example, Wolff’s narrative informs its readers that artists live in particular socio-historical periods which will have something to do with who can be an artist and what is to count as 'art'; and those sorts of concerns and disputes are 'what anybody knows' as the frequent indignation of the press attests whenever, in its view, something out of the ordinary is brought to the attention of the public. One has only to note the reactions which greeted Claude Monet and the Impressionists, Henri Mattise and the 'Fauves', Pablo Picasso, Jackson Pollock, Andy Warhol, Carl Andre, Judy Chicago, and Damien Hirst.

Of course, what Wolff is really doing is asserting that what the artist does is a function of her social position which carries at least the implication that those practices are, at best, partial and interested distortions. But the important point is this: in order to make the kinds of assertions she does, Wolff depends upon the fact that she does not have to tell her readers everything. That Wolff's sociology of art depends for its intelligibility and plausibility upon unexplicated 'commonsense' is not a criticism. Attempting to avoid populating art with what Alfred Schutz called 'homunculi' or puppets Wolff claims that art and aesthetics should not be reduced to political and ideological structures and argues for a variation of 'structuration' theory and the 'relative autonomy' of art thesis. This is the point where things can become a little sticky because it revolves around a dispute within ethnomethodology itself.

Until recently the ethnomethodological move would have been to launch straight into an attack once popular with ethnomethodology. The attack would begin by pointing out that SA consists in two sociological theories: The 'proper' one of explicit theorising (the sociology of art must be "properly theorised" says Wolff) and the 'obvious' one drawn from the world of commonsense. Under the onslaught of Zimmerman and Pollner (1971) SA would be criticised for treating as a resource that which should be made a topic. I would reject that line preferring to follow Michael Lynch (1993) when he argues that the topic/resource distinction is emblematic of what he has termed "proto-ethnomethodology". At issue is the following passage from Garfinkel and Sacks:

The lesson which Garfinkel and Sacks are attempting to teach is this: To say that sociology unavoidably trades upon commonsense knowledge of social structures is not to invoke some kind of methodological horror. Once one accepts the lesson that one cannot step outside of the fields of language and practical actions the distinction between topic and resource dissolves. As far as proto-ethnomethodology is concerned Wolff's commonsense knowledge is the hidden ingredient of her 'proper theory' and should be spelled out. But Garfinkel could not agree with that, surely, after all, didn't his own 'experiments' show that attempts to spell out even a small stretch of conversation simply 'multiplied the task'.

At issue is the distinction between professional accounts and lay accounts and the distinction between the sociologist qua sociologist and sociologist qua member of society. Button et al argue that as far as ethnomethodology is concerned orthodox sociology offers competing accounts of members lay accounts by pointing out that lay accounts are flawed versions of 'what is really going on' (ironic sociology). Ethnomethodology on the other hand does not see any real distinction between orthodox sociology and members accounts because the sociologist is a member of society first and professional sociologist second which means that in order to talk about things she unavoidably relies upon and trades in the ways of talking and commonsense understandings of ordinary members. That is, her activities and her technical language trades upon the fact that the social world is an already interpreted intersubjective world known in common.

As a non-ironic sociology, ethnomethodology takes seriously the great questions of sociology: How do actions recur and reproduce themselves? How is it that interaction displays properties of patterning, stability, orderliness? How does social life get organised? For ethnomethodology this orderliness must be seen as arising from within activities due to the work done by parties to those activities. For the non-ethnomethodological sociologies on the other hand order is to be found in sociology's 'plenum' (Garfinkel 1991). What Garfinkel means by the term 'plenum' is that all non-ethnomethodological sociology, without exception, are a variation of Parson's sociology since they all make a distinction between 'action concrete' and 'action analytic'. Parsons made this distinction and it remains with us today.

What Parsons did, actually did, was to demonstrate with his distinction - concrete action/analytic action - that there is no orderliness in concrete activities. In short, real, actual, immortal society is not to be found in concrete action. This distinction of his, a distinction that provides for and constitutes the deficiency of concrete action, is Parsons' plenum. He needs this concrete action, that is, the one that is constituted by the distinction between concrete action and action analytic. Why does he need it? He needs it so that he can argue that it is deficient. By making his distinction Parsons was arguing that real and immortal society could only be found in the results of administering the policies and methods of formal constructivist analysis. The claim is that only the methods of constructivist analysis could provide for any and every orderliness whatsoever. Order then, is not to be found in concrete action; it is not an accomplishment of societies members.

Thanks to the distinction that is made, real and immortal society, is to be found as a result of sociological analysis. Indeed, what order and society amount to can only be specified in the practices of the human and social sciences. Another way of putting it is to say that sociological methods and their champions share the belief of sociology in general: social life is patterned, regular and orderly, but, this patterning, this regularity, this orderliness is deemed to be obscured and hidden and can only be revealed by means of the correct method of investigation e.g. constructivist analytic sociology. The assumption then, is that social life is like a jigsaw - the patterning is to be found in the pieces, but it is not recognisable as 'the real social world' as such until all the pieces have been put together so as to make the whole visible. In short it is the sociological methods of constructivist analytic sociology which brings the pattern, the regularity, the orderliness of social life to view. Note at this point that the patterns that are supposed to be brought into view are deemed to be undetected by social actors. Actors have only a partial view and these views may themselves be ideological interests, views based on false consciousness, or just plain irrational.

Ethnomethodology takes great exception to the views espoused by the high priests and priestess of constructivist analytic sociology (CAS). What order and society are as issues are not to be found in the human or social sciences, they are to be found in their achievement, that is, 'order' is found 'in-an-as-of-the-workings-of-ordinary-society'. If Garfinkel is correct, we can ask why sociology continues to carry on as it does. And the answer appears to be that CAS is involved in the search for a correspondence between theoretical objects and the referents of those objects in the world 'as it really is'. And this view rests upon a distinction between the perceived object in the 'world out there' and the object itself. It is the familiar distinction between 'the appearance of an object' and 'the object in the real world': The 'world as it appears to be' and the 'world as it really is'.

The correspondence theory of reality and variations of it state that social structures may exist independently of our knowledge of them. Such a view raises the issue of the criteria for a fit between a theory and the world. The answer to the 'fit' lies in logico-empirical procedures (scientific method) which serve as rules by which any observer can obtain a more or less accurate view of real objects independent of one's particulars. Alternatively, it is the Schutzian 'congruence view of reality' that is the one taken up by Garfinkel. By this view the terms 'perceived object' and 'concrete activity', or 'object' and 'appearance' are synonymous and interchangeable. In other words the perceived object is the concrete object.

Whereas the correspondence theory or conceptual schema provides an approximation as to what is 'really out there' the congruence theory holds that however the experiencer experiences something 'out there' is 'out there' in the way she/he experiences it. If it is heard as a bump in the night then that is what it is until such time as it comes to be child-fell-out-of-bed. Doubts about a noise are experienced as 'doubted noise' for all practical purposes. The thing, the object e.g. noise stands along with its doubted character. On this view it might appear that ethnomethodology is postulating multiple subjective realities, all of which are objective in their own way. You should resist the temptation to think so because the alternative to an objective world in Schutz's sense is a world without sense; a meaningless world which an experience had failed to actualise. 'Subjective' here does not mean a perception influenced by idiosyncrasies of circumstance but a failure to intend meaning. We are all actualises of our realities. This is different to the 'fitting' found in the correspondence theory: there, the objective character of relevant objects and events is given by the operations of the theory, e.g.. determinism is not an ontological characteristic of a set of natural events; it is a property of the theory that relates the events to each other. That is, in terms of the correspondence theory of reality theory serves the function of organising the possibilities of experience, lay or professional, with one of many objective worlds.

In terms of the congruence view, however, there is no disjunctive between experience and reality and so no reason to seek a correspondence between them. Nothing lies behind or has greater reality than our experiences. So, for Schutz the problem of finding a fit between theory and reality does not exist. Rather, the task is to reflect upon our experiences of the world. For Schutz the social world is continually constituted in experience as an endless series of activities which are accounted for as rational, factual, logical or whatever attributes may be accorded them. So, the phenomenon for investigation then becomes the methods by which experience is found to be rational, just, factual or whatever. This is a rejection of Parsons' idea that the solution to the problem of order is some sort of notion of widespread cultural integration of values and normative orientations. If we suspend this presumption we find a new phenomenon for investigation.

Neo-Parsonian theories of shared expectations and a common culture, unconscious or otherwise, only work through the theoretical device of a shared culture and have nothing to do with real worldly phenomena; that is, they act as theoretical indicators of real worldly activities but they are not actual real worldly activities (Baccus 1986). Bourdieu's practices only work through the unconscious cultural categories of habitus. How understanding is achieved by actors on a day to day, moment by moment basis is not established as a problem for investigation. Garfinkel however, suspends the giveness of a common culture in order to make it visible and investigable. The influence of Schutz can now be seen as we note that ethnomethodology looks at how the world of daily life is experienced from within in order to see just how it is possible for the Durkheimian notion that the social world is experienced as an objective given. In short the ethnomethodological question is 'how might social structures be made visible in everyday life?'

Garfinkel pulls no punches; in his view theories are supposed to be a means to better and better studies of real worldly phenomenon, but in fact sociology’s studies of 'real worldly' phenomena are there to test and validate their theories. Garfinkel suggested that it ought to be possible to treat sociological theories as sets of instructions, the directions contained in these instructions should be precise and easily followed. Following the instructions should enable you to find ‘the animal in the foliage’. The reality of constructivist sociology, however, is that its sociological theories are the foliage, that is they hide the animal they are trying to locate.

Anderson and Sharrock (1986:22) have argued that the phenomena of constructivist sociology disappears from view because of the requirement of 'objective' data, measurement systems, consistency across investigators etc. What we get from constructivist sociology is the replacement of the phenomenon by artefacts of method -data runs, variables, indicators and the like. And as Anderson and Sharrock further argue, such troubles are not something discovered by ethnomethodology since they are known about by any competent sociologist! Sociology knows of these problems but they presume that the problems are theoretical ones in the sense that their 'problems' in getting a fit between theory and 'the world' lies in the sophistication (or lack of it) of their theories, when really their problems lie in the fact that they have made the phenomenon disappear. In brief what happens is that the haecceities (Garfinkel 1991), the 'things', the doings, the 'thisness and thats' that characterise ordinary activities for those engaged in them seem to disappear whenever sociological theories and methods are brought into play. For example when sociologists (as members of a second order discipline) discuss the family or work they seem to discuss things that we as ordinary people find hard to recognise.

Ethnomethodological studies do not lead to the disappearance of the phenomenon because they study and explicate the methods members themselves use as they routinely and in taken for granted ways display their knowledge of and understanding of the facticity of the social world as an organised and massively ordered world. So ethnomethodology neither accepts the long standing points of sociological consensus nor recommends the abandonment of further attempts at the systematic investigation of social phenomena. Rather, the development of ethnomethodology involved a fundamental revision of sociology's traditional views on social order, of the sorts of questions to be asked about it, and of the kind of empirical research to be done. The major response to traditional sociological conceptions of order is that they presuppose a social world that simply could not work, it is a world populated by puppets or 'cultural dopes'. As his breaching 'experiments' showed, the adoption of the disinterested scientific stance soon brought confusion to a setting. And this is Garfinkel’s sociology in a nutshell: the issue of social order is not between conflict and consensus but between intelligibility and senselessness.

The difficulties that sociology faces will remain for so long as they search for explanations of the realities underlying common-sensically available appearances of social order in preference to an examination of how such appearances are interactionally produced, managed, recognised and used as if they were 'the facts of the matter'. For ethnomethodology no theory has pride of place. We must accept that the criterion for the acceptability of a theory is its descriptive adequacy. If to secure adequacy we have to make adjustments by invoking terms not defined in the theory itself e.g. the obvious and the mundane, then the theory must be sacrificed. We can't fudge. As Anderson and Sharrock (op.cit.:19-23) insist, any theory must stand or fall on its own terms. Rather than slipping the obvious and the mundane through the theoretical backdoor Garfinkel urges us to treat the obvious and the mundane as anthropologically strange since obviousness is itself an orderly and methodic product of members interpretive competencies. And it is Garfinkel's argument that sociological analysis, and empirical research should be directed to explicating the methodic practices employed by members in the collaborative production of social order.

Another recommendation by Garfinkel is to treat members as practical rule-using analysts rather than as cultural dopes who are rule governed. Rather than stipulate what rules members are really following or governed by, ethnomethodology tries to locate the 'seen but unnoticed' rules which members might be orienting to and using to recognise and produce orderliness in some setting. In other words, ethnomethodology wishes to make statements about rules and practices which could warrantably said to be oriented to by members. Thus the theoretical question is not why in principle social order is as it is but how for practical purposes are particular manifestations of social order achieved.

All members make use of commonsense knowledge of social structures and thus they are practical sociologists. The orderliness of society is highly visible. Without any special effort people can see what is going on at a glance and what they have to do. If social life wasn't this visible it is hard to see how it could be further produced or how the pattern of daily life could be sustained. Members share bodies of knowledge, lists of definitions, rules etc. but these are a 'seen but unnoticed' background of common understandings. By this strategy we come to see that the orderliness of social life is the achievement of ordinary social actors, the outcome of what they do, and their commonsense knowledge of social structures. In short social order is participant produced. Recognisable features of an activity or setting and their display are to be seen as 'locally' produced and produced in the ways the activity is being done by those engaged in it. That something is recognisable as 'this' activity or setting in the first place is a local production.

As I have said, talk of 'commonsense' is talk of an orientation to matters which members are entitled to adopt and expect of others. If you were to walk into Lancaster's art studios and observe the activities found therein you would not have to spend a great deal of time wondering how to classify the people there. Being in the art studios provides you with a way of classifying the people them as 'artists', 'painters', 'sculptors' and 'audiences'. Similarly watching the video of J.W. it's easy to spot who is doing the work and the instructing and who is being instructed. One does not watch the activities for a while and then test various categories out on whoever you're looking at. On the contrary you start out with the category's 'art studio' and 'artist' and invoke them to describe what is going on. To do otherwise would seriously call into doubt your competence as a fully enculturated member. This commonsense sociological work is not something to be improved upon with the addition of 'proper' sociological theory since such work is not done to provide any kind of empirical theory of the activities you observe. As Sharrock and Button (1991) would describe it such work is socially provided socially sanctioned and done for practical purposes, for making sense. That is, it is a feature of our culture that the categories 'artist', 'painter', and 'sculptor' are the pre-theoretical and indispensable categories which are relevant to the organisational setting 'art studio' and the organisation of activities within that place (Should you ask someone what it is that they do and you were told with a grin 'Collect the rent' you might reasonably suppose that you've mistaken the landlord for a resident artist but all we need to note here is that we sometimes make mistakes). Janet Wolff, Pierre Bourdieu and many others are prone to replace commonsense terms such as ‘artist’ and 'work of art' with 'cultural producer' and ‘cultural product’ but the categories they reject are essential to the identification of whatever phenomena she might purport to explain, indeed, those categories are essential to providing Wolff with phenomena for her theories to talk about. Wolff claims to study artistic practice but as Button et al (op. cit.) have noted, it is in the "categories and common sense understandings internal to a social setting that the identity of social actions are decided". Sociologists can respecify all the categories they wish, that's not what is at issue. The issue is simply that whatever they come up with the activities they talk about are identified in terms of the categories found within the setting themselves; they constitute in other words, an "occasioned corpus of settings features" (Zimmerman and Pollner, 1971). At the risk of repeating myself what an activity is, what it is identified as, its intelligibility in other words, is done pre-theoretically. It is decided in terms of the commonsense understandings internal to the setting and with the categories found therein. Armed with and only with the category 'cultural production' it is highly unlikely Wolff would be able to distinguish between somebody painting and somebody dancing the lambada. And with and only with the category 'cultural producer' Wolff would be unable to identify and distinguish between a 'painter' and a stand up comic.

No matter what the sociologist goes onto say about 'art', 'artists' and 'art works' the formulation of their phenomena will necessarily involve the invocation of socially prescribed categories. Of course the sociologist wants to say that the artist is not just 'painting a picture' but is in fact being partially determined by and reproducing historically specific political and ideological structures. Informing me that what an artist is really doing is reproducing particular ideological or hegemonic structures is irrelevant as far as I am concerned since whatever activities Wolff and numerous others wish to re-classify and re-explain the fact remains that they will have to be identified as 'art', 'artist', 'painting', 'writing', or whatever first since that is what they are going to talk about. It's not that Wolff is going to say that the person working at an easel over there is not in fact an artist painting a picture but is in fact Eric Cantona scoring a goal for Manchester United. I really have no idea of what it means to say that artists do not just paint they also reproduce political and ideological structures; it seems to me that it is simply asserted and repeatedly so. But no matter what is being peddled the sociologist cannot get away from the fact that it is 'art', 'artists' and 'art works' they are talking about and whether or not they think such stuff is obvious and uninteresting they have no way of identifying 'artistic practice' other than by the title of that practice: this is 'painting' and this is 'scoring a goal'. And whether or not the sociologist agrees with Hadjincolaou's (1978) claim that the pink bow of Goya's Marquesa de la Solana betrays its political and ideological position she does not actually identify it with 'cultural product,' but by what it in fact is, a painting by Goya. If upon watching the video of J.W. making a collage someone were to ask 'Who is he?' I would be loath to reply 'He is a relatively autonomous cultural producer situated within historically specific political, economic and ideological structures'. I would be loath to describe him that way but even if I did my description of J.W's activities have already involved a great deal of sociological work. I would see that J.W. is in an art studio, that J.W's utterances are directed towards others for the purposes of explaining what he is doing and how to do it, that the paper strips are of particular colours, that certain colours are laid first, an so on and so forth. In short, I would have identified J.W. demonstrating the use of colour with collage. It is only seeing the activity for what it is in the first place that makes possible any waffle about it one might wish to come up with later. Seeing that the activity is 'artistic' and not an exercise in covering the floor is indispensable to understanding and depicting the socially organised character of the activities found within. And rather than thinking that the sociological work starts from this point we should be aware of the fact that a lot of sociological work has already been done in coming up with a characterisation of J.W's activities as 'artistic' and admire the extent to which the invocation of that notion is essential to the observer looking for and finding the orderliness of 'demonstrating the use of colour'. Talk about 'natural language use' and 'commonsense understandings' no way implies an access to a 'social reality'. As far as ethnomethodology is concerned, talking about 'natural language use' and 'commonsense understandings' is the very same thing as talking about 'social reality.' In other words the socially prescribed ways of talking and looking and commonsense understandings facilitate the doing of sociology and it precisely because our descriptions of activities are socially organised in the pre-theoretically available ways that they are that sociological inquiry is possible at all.

Button et al do not seek to secure the epistemological foundations of sociological inquiry but talk instead about the practical possibility of inquiry and the practical dependence upon the presupposition of the social world as an intersubjective world. And this reminder of the intersubjective character of the social world draws attention once again to the fact that intersubjectivity is intrinsic to the phenomena being studied. One might attempt to avoid this fact but such attempts are futile. In getting done whatever it is the sociologist wishes to get done, she unavoidably acknowledges, accepts, depends upon and demands it of others that the social world be experienced in taken for granted ways.

The issue of intersubjectivity does not deny individual experience, indeed, it only makes sense to talk of, say, the perspective of the individual against the background supposition that the world is the same for me as it is for you. The idea of 'sameness' here is encapsulated by two postulates:

(i) the reciprocity of perspective and

(ii) the interchangeability of standpoints.

According to Schutz these two 'postulates' provide the 'world in common'. Biographical differences set at zero allow for a world which is the same for me and you though you are not me and vice versa. This world known in common is defined in terms of a socially provided 'stock of knowledge'. The reaction of many practitioners working in the human sciences is to begin blathering on about 'individual meanings' and 'different realities' and such like. But the point to grasp is that no matter how 'theoretically informed' such assertions are they are nothing more than assertions of one kind or another; that is, they in no way present an epistemological challenge to the social order. Put another way, talk of 'reality disjunctures', 'competing realities', and such like presuppose their accomplishment in the first place.

Schutz, of course, argued that members descriptions comprise "ready made, already-constituted, socially derived constructs and characterisations - not only typifications of persons and conduct, but also typified methods of describing the environment, of typical means for bringing about typical ends in typical situations, etc." (Watson, 1994:172). This stock of knowledge comprises the culture and language is the medium for the transmission and use of this stock. As Watson says: "language is the typifying medium par excellence" (ibid.). And it is this pre-given, pre-constituted and already interpreted character of the social world which provided Sacks and Garfinkel the wherewithal for formulating a non-ironic sociology.

Sacks and Garfinkel took great pains to point out to the conference the epistemological divide in classical - and current - sociology, namely ironic and non-ironic analytic approaches. The former is based upon a neo-Kantian epistemology of correspondence, a programme of and claims to literal description, and an external standard for assessing members' cultural knowledge. For example, consider Andrew Sayer's (1995:241) attempt to counter what he terms 'popular' objections to normative theory. Like countless others before him Sayer requires his plenum:

And Sayer's insistence is an exercise in propaganda. With his optimistic attitude towards certain forms of social scientific theory and methodology Sayer betrays a tendency to dogmatic closure found in epistemological thinking of the kind which declares to have reached bedrock: this is what is really going on. Sayer's assertions bring to mind both Alfred Schutz's concerns about the social scientist's use of scientific constructs in order to supersede the constructs of commonsense reasoning, and Harold Garfinkel's warning that orthodox sociology's account of 'order' is utterly reliant upon the ironic characterisation of members of society as 'cultural and judgmental dopes' pushed around, with the necessary democratic qualifiers, by external forces unbeknown to them. Sayer obviously sees himself as the 'engineer' of Levi-Strauss possessed of special tools and techniques for discovering order, but since the engineer is himself a species of bricoleur (Derrida 1970) Sayer’s characterisation is doubly ironic. Derived from formal procedures of science or worse, scientism, Sayer appeals to a standard 'outside' of the settings and courses of activity examined thus exemplifying a social scientific elitism whereby commonsense understandings are deemed "primitively deficient, superficial, misconceived, inadequate, interested, stereotypical, mystificatory - in short as an inferior, naive or degenerate version of what is yielded by (an idealised) scientific approach" (Watson 1994:173). It is plainly Sayer's view that 'Marxism' and 'feminism' offer correctives to folk knowledge of what is really going on.

Even Sayer's 'lay understandings' is ironic. Though it looks and sounds respectful, his 'lay understandings' instructs the reader to do contrastive work and find 'lay understandings' as inferior. Sayer, along with other correspondence theorists, maintains the distinction between 'perceived objects' and 'concrete objects' in an attempt to discover 'what is really going on'. However he (and they) are doing this and more: To the question 'what is really going on?' Sayer gives two 'possible answers' 'capitalism' and 'patriarchy'. No doubt there are more; the point is that possible answers in the form of description ('women are systematically oppressed because society is patriarchal', etc.) points to a further correspondence, namely that between what is observed and what is said to be actually going on or, in other words, between the description and the events or objects in the world. Involved here is the 'problem of literal description' which has been treated at great length by Sacks, Garfinkel and contemporary ethnomethodologists, but to no avail if our critics are anything to go by. The 'problem of literal description' is an insoluble problem for orthodox sociology but at the same time it highlights the distinctiveness of ethnomethodology. Sacks's methodologically-radical programme found its earliest formulation in his difficult but trenchant article 'Sociological Description'.

If one accepts the etcetera problem:

Sacks is pointing out the following disconcerting fact: for a correspondence to be arrived at between a description and some object the description would have to be complete. However, since all that can be done is incomplete description, and since any description can be 'read' as 'far from complete' or as 'near to complete' then the possibility exists that the description can be read as an irony, metaphor, a sketch, or whatever. The conclusion for those who think that it is absurd that their 'more sociological' accounts are not superior to 'lay understandings' are devastating since Sacks' paper is an outright attack on their warrant for the very way that they do their sociology and the claims which they make. A corrective for those who have a faith in the discipline producing 'better' theories and methods and 'better' and more 'accurate' descriptions of 'what is going on' it is little understood and almost universally ignored. Stated bluntly, Sacks' paper is nothing short of a reappraisal in how to investigate the social world sociologically.

From Durkheim to Weber and to contemporary sociology we find a common endeavour; sociology attempts to provide a 'principled resolution' to the above problems. Indeed, what differentiates one sociological 'perspective' (for want of a better word) from another is to be found in their 'resolution' of 'the etcetera problem' (Hak, 1992). But why any particular solution should be accepted over and above any other is unclear not least because there is no in principle reason why we must accept any of them at all in the first place! What their protagonists assume to be 'superior', 'more accurate', 'scientific', 'rigorous', and 'more sociological' descriptions are in fact mere versions and no better or worse for all that than any other answer to the question 'what is really going on?' As Watson notes, Sacks effectively trashes 'constructivist analytic' "ways of proceeding and their warrants for making the claims they make as analytic claims, thereby rendering them, to say the least, inconsequential" (op cit.:179).

As this point I might stand accused of setting up a sociological straw man so it is worth dealing with the issue of generalisation. Looking again at Sacks' formulation of 'the etcetera problem' we find that what he has to say is well known to sociologists. However, assuming sociologists are aware of the etcetera problem it is legitimate to ask why sociology carries on in the manner it does, and the answer to that lies in the assumption that generalisations provide for some sort of resolution. Unsurprisingly, Sacks pre-empted that response as can be seen in his formulation of what he called 'the problem of generalisation'.

If one buys into correspondence theory then one buys into the issue of literal description, that is, correspondence theory makes the distinction between the description proposed and the intended object and then faces the problem of correspondence as outlined above. Consequently, no description can be treated as a set of instructions for recovering the phenomena without being 'filled in' by other than that found in the description. One might get from the description various objects which appear the same and thus one appears to have a generalised description but one can never capture the particular object since any generalisation neglects the features of particular objects that make of them the particular objects that they in fact are. A simple example might throw some light on what Sacks is saying.

Suppose we observed two particular encounters. The first being an encounter between a doctor and a patient. Having observed this encounter we might try to account for what we see in some sociological way or another, by making a sociologically informed list perhaps. For example we might produce a sociological account which contains the following interrelated observations:

a) Power

b) Deference

c) Domination

d) Expertise

e) Claims to knowledge

f) etc.

I shall call the description Towards a theory of symbolic exchange: Your submission for my expertise. In terms of theoretical rigour Sharrock and Anderson (op. cit.) suggest that what we should be able to do is follow the theory as though it were a set of instructions and having followed the theory, and only the theory, we should recover the doctor-patient encounter we initially observed. Consider now the second encounter. This one I shall call 'The Lawyer-Client Relationship'. Placed to the right of the doctor-patient encounter I shall list relevant sociological observations and produce the final account which I call Towards a theory of symbolic exchange: Your submission for my expertise.

Completed my two finished sociological descriptions look thus:

a) Power a) Power

b) Deference b) Deference

c) Domination c) Domination

d) Expertise d) Expertise

In Garfinkel’s terms the animal has been hidden by sociology’s foliage. In terms of the theory and only the theory one cannot choose between the two and so it looks like we have a generalisation here, but of course that is absurd if what that means is that one could not point to an actual doctor-patient encounter or that one routinely mistakes a real life encounter between a lawyer and her client for an encounter between a doctor and her patient. To make mistakes of that magnitude consistently is to have, in one way or another, one's competence called into question. What makes the doctor-patient encounter not a lawyer-client encounter and vice versa is that the one is a medical encounter and the other a legal encounter and to 'see' that, as it were, is by virtue of something not provided by the theory, namely, one's cultural membership and commonsense knowledge without which the theory would be unintelligible. So, generalisations not only face the 'etcetera problem' and have to be brought to a close somehow, they also fail in their own terms to recover the particular since it is only by virtue of unexplicated and unacknowledged commonsense knowledge and understandings that one can find the particular in the generalisation.

Consider now that much of the normative language of SA refers to what you and I would call artistic styles. According to Witkin: