I

If it recedes one day, leaving behind its works and signs on the shores of our civilisation, the structuralist invasion might become a question for the historian of ideas, or perhaps even an object. But the historian would be deceived if he came to this pass: by the very act of considering the structuralist invasion as an object he would forget its meaning and forget that what is at stake, first of all, is an adventure of vision, a conversion of the way of putting questions to any object posed before us, to historical objects - his own - in particular.

(Derrida 1978, 3)

Nearly a decade has passed since the symbolic and structural archaeology conference announced a coherent challenge to to the existing paradig­matic and epistemological structure of archaeology. These ten years have been momentous ones for the discipline. Not only has its practice been transformed with the introduction of new conceptual frameworks (such as the notion of the archaeological record - material culture - as text) but also whole issues, questions and debates that were previously unthinkable have become relatively commonplace - debates about the social implications and embeddedness of scientific and academic study, about the politics of archaeological production, about the context of archae­ology and the social sciences in late capitalist society, about the construction of gender relations in the past, and so on. The framework provided by the New Archaeology had provided little, if any, space for consideration of these issues, and although the last three to four years have seen a general mellowing of the initial opposition to structuralism mobilised by the positivists (as elements of structuralist theory are diluted and wash over the international scene), for many processualists it is not only still possible to distance archaeology from these issues, from the present, from society, from contemporary systems of representation, but absolutely vital that we continue to do so (see, for example, Schiffer 1988, discussed by Tilley in Chapter 5).

The strength of the reaction against the innovative and experimental work emanating from Cambridge in the late 1970s and early 1980s is nowhere more clearly demonstrated than in the confrontational and belligerent style adopted by processualists like Binford. It is, of course, to be explained by the very real threat that Hodder and his students offered to the foundations of the new orthodoxy, but also, one suspects, by the fact that it transformed processualism into that orthodoxy. There had been structuralists before of course - Leroi-Gourhan in France or Jari Nordbladh in Sweden - but neither had any widespread effect even within their own institutional systems, and on an international scale they worked largely in isolation. In the late 1970s, however, these diverse theoretical interests coalesced into a major force which has fulfilled Edmund Leach's much quoted prophecy about the future fortunes of the discipline. Almost overnight, the bright young things of the New Archaeology - these Binfords, these Renfrews, these Schiffers - were transformed from the avant-garde into the old guard, a metamorphosis from which their self image has yet to recover. The New Archaeology was, suddenly, the not so new, and further found itself criticised for not even being particularly new in the first place. And, for the last decade, it is this formation which has been on the defensive while the initiative has been taken up by a new 'new' heresy. Despite the opposition, a structuralist based archaeology has, then, taken root and established itself - at its most widespread in the United States, but at its most intense in Europe, especially in Britain and Scandinavia.

The strengths of the 'movement', which, in the typology of supposed archaeological paradigms, has since become known by the generic term 'post-processualism', is shown most clearly in a number of volumes that have emerged in the Cambridge New Directions series. The symbolic and structural conference volume (Hodder 1982) communicates clearly the sense of excitement and experimentation, and the papers it contains remain the clearest and most coherent statement of the structuralist challenge. Since then the initial unity - always perhaps largely illusory -has given way to increasingly divergent postures as the various participants have followed their interests along widely different courses. Some have sought to engage with Marxism and Critical Theory more closely (Miller and Tilley 1984; Miller 1987; Shanks and Tilley 1987a;

1987b), while others, most notably lan Hodder, have kept Marxism more at a distance, and have forged links with a hermeneutic tradition associated with British liberal historians such as Collingwood, or the theories of agency developed by Anthony Giddens and Pierre Bourdieu. Associated with this latter development has been the emergence of a gender archaeology, the strategic aims of which at present seem rather limited, but which is perhaps set to have a more general influence on archaeology as a whole.

It is thus possible to say, ten years on, that post-processualism is here to stay even if, at the same time, the identity and coherence of the movement is becoming increasingly unclear. This book is about working through and beyond the traditions in which post-processual archaeology has immersed itself and of forging, on the basis of the groundwork it has taken a decade to complete, new directions for archaeology - new questions, new ways of conceiving both what it is archaeologists do and why they do it, new ways of reading and writing the past. In attempting this advance, the contributors to this volume have looked towards those critics who are generally seen to belong to that movement in philosophy and literature which is known as 'post-structuralism' - although the applicability of this label is open to doubt and seems to fly in the face of the wide diversity and often contradictory aims of the writers'in question.

In a sense, of course, archaeology is already post-structuralist - it has absorbed a version of contemporary theory that, in the work of Giddens and Bourdieu, could be described as leading beyond some of the positions articulated by Levi-Strauss. And, indeed, one of the problems We faced when compiling this volume was that of deciding who and what should be put into this category, a problem especially pressing given that many of the ideas discussed below will not be familiar in an archaeological context. Anyone reading the papers we have selected will be struck by the variety of ideas represented - the stock and standard names are there to be found, Roland Barthes, Gilles Deleuze, Jacques Derrida, Michel Foucault, Julia Kristeva, Jacques Lacan, but alongside them are not only a number of less well-known figures (who are none the less important for being that), but also a number of unusual or unexpected names that seem to carry off whatever meaning this categorisation of 'post-structuralism' may have had in time and space. Thus there are extensive references to Freud, Marcuse and Nietzsche, to Marxists like Louis Althusser and Pierre Macherey, to anthropologists like Gregory Bateson, to novelists like Jorge Luis Borges, Angela Carter or Franz Kafka. All this emphasises not simply the problematics of categorisation, but more importantly that none of the contributors to this book would countenance the ideas with which they are concerned as the universal solution to all the problems that face the construction of the much mooted critically and socially 'aware' archaeology. Post-structuralism is not available on prescription, as the cure to the carefully diagnosed complicities of existing frameworks. Following Derrida or Foucault brings its own problems, and highlights the fundamentally permeable and porous nature of post-structuralism, which at no point forms a complete or enclosed body of thought. The necessity is always to articulate these ideas within their existing traditions in order to address any particular problem - the problem of the subject and its socio-cultural identity, as discussed by Bagnal and Nordbladh and Yates below, which produces a chain of references that runs through structuralism and post-structuralism into feminism and psychoanalysis. The papers articulate with structuralism and with radical traditions within structural anthropology (Burr), with Marxism (Tilley; Walsh; Waterman), with Foucauldian analytics (Burr; Bapty), or with Nietzsche (Bapty). A host of possibilities is opened up by these ideas. For Moran and Hides (chapter 7) the question has become that of detaching an understanding of the issues raised by Hegel and Derrida from their institutionalised discursive contexts, fraught as they are with problems of privilege and closure/inaccessibility, in order to experiment with new styles of writing and means of expression which do not return to reliance on the (relatively) esoteric. They countenance a note of caution. Whether or not they are successful in attempting to develop a subversive and reflexively self-conscious style is not as important as the fact that these questions are raised and discussed, and that the writing of the past, its forms and features, becomes the subject for debate. Walsh (chapter 10) similarly raises questions for post-structuralism in relation to the development, during the 1980s, of the commercial, commodified past, and is sceptical about the validity of the work of Baudrillard for mounting a critical debate around this subject. And this is an area in which our vigilance and critical faculties will need to be particularly sharp.

The notion of difference that Saussure brought to the study of linguistics, and which forms the basis of structuralism as a movement by its insistence on the referentiality of signs, was bound to have wide ranging effects upon archaeology once the effects of structural anthropology filtered down through the disciplinary and institutional frameworks (frameworks, the only function of which is to limit the horizon, control the range of questions and research, retard mutual influence and so, ultimately, protect each discipline from its neighbours). Saussure, as we know, divided the sign into two elements, the signifier and the signified as the sensible and the intelligible. Thus the signifier 'LEMMING' is connected to a signified

- a mental image that is represented (supposedly) by the signifier, and with which it is possible to signify this particular group and order of letters. But, according to Saussure, there was nothing that necessitated this precise combination of letters having this signified, no reason why this creature should have been called this name. The image could have been represented by any number of alternative combinations - donkey, filing cabinet, Munich, Professor of Archaeology - any would have done, and in this sense the signifier is arbitrary. It is only convention that establishes this connection between the signifier and the signified. The only criterion that the signifier need fulfil is that it be different from the other signifiers of the system, and it was in this insistence that Saussure departed from the body of traditional linguistics which had eschewed such an approach in favour of the assumption - or at any rate isolation - of linguistic forms which can then be compared to other forms and their development over time studied. Saussure insisted on the priority of the synchronic over the diachronic, and thus went against the grain of orthodox practice. Because the signifier was articulated differentially, it could only be understood as it was related and referent to all the other signifiers at any one moment in time. The sign only made sense if it was studied against the background of a system.

In language, therefore, Saussure was able to assert that the value of the signifier is purely negative and differential. In writing, it is possible to write the letter d in countless different ways and in many different forms, but all that is necessary for it to signify is that it be different from the other letters of the system - that the letter d is distinguishable from the letter c or the letter b and so on. Thus, Saussure wrote, in language there are only differences without positive terms. Attention to the system of language forces us to look closely at the underlying structural rules that govern the generation and correct combination of signs. Thus any particular speech or writing act - parole - is seen to be conditioned by a system of grammar - langue - comprising regulations that are never, of course, present all at once, but exist only as they are activated by parole. A certain reciprocity therefore exists between langue and parole, which are always articulated in a relation of mutual dependency.

This brief - and inevitably reductive - account of Saussure's linguistics is to set out the basis around which the main strand of structuralism, as a means of analysing cultural phenomena, has grown up, and it is also, therefore, necessary to begin to understand what difference the 'post' of post-structuralism makes. The search for underlying rules between phenomena - the 'high' structuralism of Levi-Strauss' cross-cultural based anthropology for example - can be understood through Saussure's basic insight, and Levi-Strauss' emphasis on the synchronic is quite clear. In order to examine a structure it must be totalised, it must be stopped dead. This, of course was not a problem when dealing with the relatively static time frame of the unproblematic 'present' conventionally enclosing ethnographic studies, where areas of analysis were the apparent stability of the kinship system, or the anti-historical time of the myth. But it did mean that history became something bracketed off, returned to only subsequently as an afterthought, an annoying loose end. The problem was how to return to this later. And it can be said that post-structuralism attempts this return, this reopening of time and history, by rethinking the diachronic within the synchronic. And this move, crucial as it is, is achieved by rethinking the whole basis of Saussurian difference and the system of thought that holds it in place.

Thus 'post'-structuralism is, in fact, heavily dependent upon struc­turalism, and it is not so much a move beyond as a move through its logic. Derrida, whose texts are most closely associated with this shift, puts the point succinctly, and offers us a convenient description of the emergence of this movement and its denning force in the 1960s, especially around 1968:

At that time structuralism was dominant. 'Deconstruction' seemed to be going in the same direction since the word signified a certain attention to structures (which themselves are neither simply ideas, nor forms, nor syntheses, nor systems). To deconstruct was also a structuralist gesture or in any case one that announced a certain need for a structuralist problematic. But it was also an anti-structuralist gesture, and its fortune rests in part on this ambiguity. Structures were to be undone, decomposed, desedimented.

(Derrida 1988, 2)

Properly this description should apply only to Derrida's own work, involving a very detailed analysis and critique of Saussure's original insights. Of course, Lacan's seminars had been examining the linguistic analogy for unconsciousness since the 1940s and 1950s, developing a body of thought that was to have a seminal impact on the debates of the 1960s, particularly following the publication, in 1966, of a collection of Lacan's papers titled Ecrits. But nonetheless, it is with Derrida that any introduction attempting to frame post-structuralism finds, if not its actual beginning, at least its most convenient one.

Structures, which structuralism had been labouring to construct, define and understand, were to be undone, decomposed, desedimented. The laborious constructs of the intellectual movement, and many others besides, were to be taken apart, were to be deconstructed. Saussure, Derrida maintained, had consistently resisted the most important insight of his work. Firstly by retaining the concept of the sign ('a two sided unity') which left open 'the possibility of thinking a concept signified in and of itself, a concept simply present to thought, independent of a relationship to a system of signifiers' (Derrida 1981a, 19) - leaving open a journey back to a sign conceived an outside of difference. And secondly by privileging speech over writing, which collapses the sign wholly into meaning and consciousness, such that in speech 'the signifier seems to ease itself in, to become transparent . . . the exteriority of the signifier seems reduced' (Derrida 1981a, 22). What Derrida does is to insist, as Lacan had also done, that between the signifier and the signified is driven a bar resisting unification and signification, a bar forced through the middle of the sign by difference. For Derrida, the course to this position is charted through the metaphysical opposition of speech to writing, for writing had always been debased on the grounds that it represents a fall from the presence of meaning, because ambiguity cannot be cleared up by reflection and discussion. Writing had been charged with only re­presenting speech, which is conceived to signify via a direct connection between signifier and signified. But, of course, the signifier in speech is the same as the signifier in writing - there is nothing to distinguish the two, and therefore what characterises writing must also be regarded as characterising speech. 'Writing', in Derrida's work, comes to name not empirical writing, but the system of difference that articulates the signifier. Articulated in the same way, speech is already writing, and is, therefore, already in difference, and the pure self-presence that had been thought to underlie the phonetic sign is torn open at the seams.

What are crucial here are the implications of difference for the nature of the sign. In an interview with Julia Kristeva, Derrida describes the point succinctly and clearly:

The play of difference supposes, in effect, syntheses and referrals which forbid, at any moment or in any sense, that a single element be present in and of itself, referring only to itself. Whether in the order of spoken or written discourse, no element can function as a sign without referring to another element which is itself not simply present. This interweaving results in each 'element' - phoneme or grapheme - being constituted on the basis of the trace within it of the other elements of the chain or system. . . . Nothing, neither among the elements nor within the system, is anywhere ever simply present or absent. There are only, everywhere, differences of differences and traces of traces.

(Derrida 1981a, 26)

The sign, articulated by difference, is not present, let alone 'present to itself, it is not a fixed and stable entity, it is inherently unstable, constantly open and constantly changing. The 'presence' of the signifier is deferred onto another signifier which is in turn not fixed but itself caught up in a relation of difference - and so in a long chain of signifiers and differences. The name Derrida gives to this movement and force -differance - attempts to capture its two senses by the substitution of an a for the second e in the French, signifying both to differ and defer, both the passive and the active. Nothing can be outside of these chains and movements, for they would be no more than new marks, at once differed and deferred from other marks. Presence and absence are therefore no longer available as alternatives - they are divided from themselves by the trace within them of difference, because within this system elements differ from themselves, cutting off the possibility of anything ever being present to itself: 'We will watch it [presence] infinitely announce itself and endlessly vanish through concealed doorways that shine like mirrors and open onto the labyrinth' (Derrida 1981b, 128).

Set to work within structure, the effects of difference are seen fairly quickly. Every limit that seemed apparently to govern and to guarantee the identity of structure is overrun. Thus, for instance, philosophy, which Derrida describes under the blanket term metaphysics, has attempted to found itself upon positions of pure presence - speech in Plato's Phaedrus, the living present in Husserl's Investigations and Cartesian Meditations, absent being in Heidegger and so on. But differance passes through these conceptual frameworks, and offers us a point of entry from which the whole structure can be taken apart. Thus, for example, the distinction made by Kant between ergon (primary, essence) undparergon (secondary, addition) is taken apart by Derrida (1987). The structure of the parergon - the frame of a painting - is always absolutely essential to the ergon, and without it, it cannot exist. The parergon is shown to be, in fact, more essential than the ergon. Like the structure of the supplement, which Derrida takes from Rousseau's Confessions, it has an undecidable sense which positions it outside of presence/absence and in differance. Where the supplement means (in French) both addition and replacement, so also 'parerga have a thickness, a surface which separates them not only . . . from the integral inside, from the body of the ergon, but also from the outside, from the wall on which the painting is hung . . . from the whole field of historical, economical, political inscription . . .' (Derrida 1987, 60-1). The structure of 'parergonality' both 'joins and divides the difference between essence and accident, primary and secondary, outside and inside, intrinsic and extrinsic.

Deconstruction differs from a simple critique. It is not, Derrida insists, a method, or even a critical activity. 'It is not an analysis in particular, because the dismantling of structure is not a regression towards a simple element, towards an indissoluble origin. These values, like that of analysis, are also philosophemes subject to deconstruction' (Derrida 1988, 3). 'The movements of differance cannot be controlled or introduced by the critic, they can only be followed or brought out - it is not the function of a subject, a critical genre. It is more a property of texts, of structuring, of bringing out "a relation of the work to itself ' (Derrida 1984, 124).

And yet the question is raised - and Derrida does not address it - of how meanings are actually produced, of how texts and the system of which they are a part avoid being constantly torn open and deconstructed. If Derrida has not addressed the issue of ideology, then Roland Barthes has. In his study of modern bourgeois myth Mythologies (Barthes 1973), Barthes drew the distinction between a first order of language, which contains the ordinary or manifest content, what he calls denotation, and a second order connotation, contained within the first, which would be the ideological. An ideological discourse operates at the second level but through the first - myth would be the latent plane of connotation beneath the manifest plane of denotation, And Barthes' essays in Mythologies are, of course, the classic putting to work of an ideological analysis in these terms.

But, as Foucault at least nominally rejected the problematics that ideology involved (see Waterman, chapter 3), so Barthes was also led to reject this distinction between the levels of meaning production and signification. In his later work, in the late 1960s and the 1970s, ideology is no longer seen as a secondary text that takes over a primary and original text, since the separation between denotation and connotation is no longer possible. In S/Z (Barthes 1974), in which Barthes rejects not only his earlier interpretive frames but also the notion of a latent structure, he writes:

. . . denotation is not the first sense, but pretends to be; under this illusion, it is ultimately only the last of connotation (that which seems at once to found and close the reading), the superior myth by which the text pretends to return to the nature of language, the language of nature.

(Barthes 1974, 9)

The level of denotation is, therefore, the point at which connotation seeks to naturalise and so hide itself. It is present and active from the very beginning, from the first moment of meaning. In place of a structuralism founded on the distinction between surfaces and depths, S/Z mediates a very different conceptualisation of the text:

The text, in its mass, is comparable to a sky, at once flat and smooth, deep, without edges and without landmarks; like the soothsayer drawing on it with the tip of his staff an imaginary rectangle wherein to consult, according to certain principles, the flight of birds, the commentator traces through the text certain zones of reading, in order to observe therein the migration of meanings, the outcropping of codes, the passage of citations.

(Barthes 1974, 14)

The text, to the Barthes writing after the middle of the 1960s, is without an origin, and the reader is freed from the tyranny of the Author: 'To give the text an Author is to impose a limit on that text, to furnish it with a final signified, to close the writing. . . . When the author has been found, the text is "explained" - victory to the critic' (Barthes 1977a, 147).

The dethroning of the Author opened up the text to multiple readings -'by refusing to assign a "secret", an ultimate meaning, to the text (and to the world as text)' - the emphasis of criticism is shifted and set free of these limits - 'a text's unity lies not in its origin but in its destination . . . the birth of the reader must be at the cost of the death of the author' (1977a, 148). And it is here that a distinction is made between the 'readerly' (lisible) text, which encourages the reader to take part in the production of meaning only as a passive consumer, and the 'writerly' (scriptable) text, in which the reader is forced to adopt a more active role and to 'write' or 'complete' the text as s/he reads. The Utopian vision would be, for Barthes, that of 'a mobile, plural reader, who nimbly inserts and removes the quotation marks, who begins to write with me' (Barthes 1977b, 161).

If the origin of the text is to be problematised, then it is in order to release

the potential of the text to mean. According to Frank Lentricchia (1983, 173), the point of Derrida's working of and through the concept of differance is not to destroy all grounds and remove from the critic the responsibility of examining the contexts of the production, but to pose new questions for an historical labour. The point is to open texts up to differance. Derrida himself puts the point fairly succinctly on a number of occasions, and insists that breaking the two sided unity of the Saussurian sign does not destroy the question of production and the attachment of signifier to signified: 'For us the rupture of the "natural attachment" puts in question the idea of naturalness rather than that of attachment' (Derrida 1976, 46).

The strategic conceptual role of differance is to force into the light all those processes that appeared to guard against the radicalism of history -questions of closure, convention, system, ideology and power become of enormous importance. 'Where are the borders of a text? How do they come about?' (Derrida 1979, 85).

Thus deconstruction does not practise a radical 'death of the author' despite the problematisation of the origin (the arche of archaeology for instance) through notions like the trace. Within differance, history has to attempt to comprehend 'the becoming space of time and the becoming time of space' (Derrida 1973, 136), the movement captured in the silent a of differance - of differing as both a spatialisation and a temporalisation. But it is also important to understand the extent to which Derrida's readings of philosophy and literature make use of fairly orthodox concepts of time and place. Thus, for example, behind his well known criticisms of Foucault's Madness and Civilization offered in an article entitled 'Cogito and the History of Madness' (Derrida 1978, 31-63) are assumptions which Derrida does not question - assumptions about Descartes, his Meditations, his period and genre. The point is to isolate those aspects of the text which have permitted it to be used and read historically in a particular way, the supplement for instance:

It is certainly a production, because I do not simply duplicate what Rousseau thought of this relationship. The concept of the supple­ment is a sort of blind spot in Rousseau's text, the not-seen that opens and limits visibility. But the production if it attempts to make the not-seen accessible to sight, does not leave the text ... it is contained in the transformation of the language it designates. . . . We know that these exchanges can only take place by way of the language and the text, in the infrastructural sense that we may give to that word, and what we call production is necessarily a text, the system of writing and reading which we know is ordered around its own blind spot.

(Derrida 1976, 163-4)

Deconstruction is therefore a force of immanent critique - though doubtless Derrida would find that description far from unproblematic. It works within formations that already exist - the tradition of western metaphysics which, because it extends, in Terry Eagleton's words, all the way from Plato to Nato, it is fairly easy to accept as already determined. Thus, for instance, while the techniques and processes of a deconstructive reading may be 'well attuned to the exposure of hidden contradictions within ideology, "a revelation of its hidden aporias" ' (Butler 1984, 114), it remains problematic to the extent that Derrida is unable to account for how meanings actually do take place. Such an account would, of course, involve him in a departure from the 'internal' and 'immanent' nature of his work, where concepts are drawn from the interior of a text -differance from Saussure, supplement from Rousseau, pharmakon from Plato, hymen from Mallarme, and so on, and are turned against the text - and account for a position outside of that particular production, outside of the text in question. And it would involve him in specifying the particular conditions of writing in terms open to closer definition than those of 'trans-historical' metaphysics, which, because of its ubiquity, seems even to stand outside of history. The kind of articulation of deconstruction attempted by, for instance, Michael Ryan (1982) with Marxism, or the more guarded and selective response offered by Frederic Jameson (1983) or Terry Eagleton may be taken as indicating that differance in itself is insufficient for fulfilling all the purposes for which we would wish to use it. As Dews (1987, 37ff.) points out, Derrida is not concerned with challenging or transforming structures so much as with laying them bare, revealing their veiled functions and hidden activity. But to a great extent, Derrida himself appears to be aware of these problems, and since the later part of the last decade he has become increasingly emphatic about the political implications of deconstruction. Some of his more recent material is important in offering a guide as to reading towards history as text and the 'social text', and is discussed by Maley below, who finds that Derrida's potential contribution to an archaeology may, in one sense, be greater than that of Michel Foucault, whose value to the discipline and its development of critique has in some ways appeared more immediate than that of deconstruction.

Whether Foucault's 'histories' are really histories in the orthodox sense is a question about which much ink could be spilt - he certainly uses apparently 'historical' data in his analyses, where Derrida's work appears to be a more fundamentally 'textual' practice. Indeed Foucault has charged deconstruction with this limitation, and accused Derrida of 'a pedagogy that teaches the pupil that there is nothing outside of the text' (Foucault 1979a, 27) despite the fact that his own discussions of the place of the author (Foucault 1977a), or of the all-embracing extension of language and materiality (Foucault 1977b), might seem to align him with just such a position. To a certain extent, this web of ambiguity which Foucault weaves - the ambiguity equally inherent in the paradox of both Foucault's 'archaeology' as an analysis of the past supposedly beyond 'history', and of his 'genealogy' as an analysis of the past where 'history' is rehabilitated only through the new paradox of situating it within a post-structuralist frame - is precisely the method of his iconoclastic project, precisely invoking a consistent irony which internally maintains the trace of his externally fraught relationship to thinkers such as Derrida or Barthes. The scheme of his 'historical' analysis is, then, almost deliberately to draw implicitly the criticism of other contemporary critics with whose work his own might seem theoretically and epistemologically aligned, as much as it is to distance explicitly the orthodox tenets of empiricist history, or the accepted framework of the history of ideas. It is out of the always latent instability of this rhetorical network that he builds, like Nietzsche, a history substantial and 'real', and yet one conceived more as the missing myth of the modern age, than as the reconstruction of an abstract, reified, 'true' past feigning to point away from the conditions of its own production.

It was the 'archaeological' method Foucault elaborated through his works of the 1960s where this rebellion against the traditional framework of history was most immediately and ambitiously manifest. For Foucault, all the apparent innovation of twentieth-century historiography - such as, for example, Collingwoodian idealism or the Annales school's concern to construct a many-stranded history evading and exploding the limited domain of the 'event' - had not been sufficient to escape the immediate legacy of late nineteenth-century positivism, and the more insidious and underlying acceptance of 'history' as a particular density known and understood; it was how to frame the understanding of what happened within history, not history itself which was the problem for the western historiographic tradition. According to Foucault, the answer could only be to go beyond this maligned space occupied by the signifier 'history' altogether. Thus, 'Archaeology' rejected 'history' as implicitly associated with the study of the pre-ordained, unproblematic, order of a teleological, horizontal progress, powered by the necessity of some immanent, silently unfolding rationality, and focused instead on what Foucault refers to as the 'constant verticality' behind that facade. Archaeology was to be juxtaposed to the 'history of ideas or science', while aiming 'to uncover the regularity of a discursive practice, the basis upon which theory and knowledge become possible' (Foucault 1977c, 44).

Foucault's archaeology, then, sought to recover the irrationality of past discursive practices to the side of reason, revealing in the same movement the irrationality of the conventional historical reason which rejected these practices as the debased beginnings of itself. In the same way that archaeology is conventionally understood to be the study of past material culture, of artefacts, so for Foucault 'archaeology' is the study of the past, not in terms of outcomes and events with their origin in the machinations and ideas of the individual subject, but in the study of the contingencies and conjunctures running through the material, object world of discourse, where the subject is formed only as one more effect, one more provisional construct of a particular set of prevailing discursive relation­ships. The outcome of the 'history' of madness offered in Madness and Civilization (Foucault 1967) is not so much to see the rise of humanitarian concern for the insane as the final emergence and triumph of some naturally embedded reason or truth, as it is rather to see madness in the late eighteenth/early nineteenth century now allocated and regimented through a different economy of discourse where madness had become, in the emergence of the post-Enlightenment world view, a condition lurking within reason, a new danger which must be submitted to a very different technology of control and, ultimately, repression.

The status of archaeology as a kind of anti-history was always problematic however, founded on a contradiction increasingly evident as Foucault extended his archaeological studies to medicine, linguistics, economics and the natural sciences (Foucault 1974; 1975). The paradox was, that as a new totalising system of knowledge, claiming to outflank history entirely, archaeology found itself resting on precisely the same foundations, the same claim to 'truth', as that of orthodox history. The historical a priori of the episteme around which the archaeological analysis of the past was organised must ultimately swallow Foucault as well, rendering him and his work as one further symptom of the pattern of development he had set out to dissemble and reconstitute.

Foucault's response to this dilemma was to dispense with the archaeological framework and much of the carefully described and elaborated analytical schemes of classification which went with it, and to adopt instead the paraphernalia of a revised Nietzschean genealogy. In this guise, the aim is no longer to displace history altogether. Rather, within 'history' it is to enable a different kind of analysis based on an emphasis on what in a sense becomes, for Foucault, a new transcendental signified, an emphasis on power - albeit a constantly self-effacing 'transcendental signified' occupying a space in Foucault's thought corresponding to that where Derrida had opened differance. To understand Foucault's departure here, it is important to see the full significance of a relationship to Nietzsche which some have been inclined to see more as a superficial, than a real, shift from the premises of archaeology; that involves comprehending what it is that the genealogical Foucault means by power as something more than mere obedience to the almost obligatory referents of the 'left' thinker. Thus what is not implied, even despite the nominal appearance and subject matter of a work like Discipline and Punish (Foucault 1977c) is power as fundamentally a force of appropriation and repression, that power, for example, which in an orthodox Marxist position controls the dislocation of forces and relations of production in the class structure of capitalist society. Indeed, for Foucault such a characterisation of power is exactly one of the means by which the recent disciplinary ordering of society has been facilitated, is one of the traces on which modernity rests (see Foucault's difficult relationship to Marxism (Smart 1983)). Even when turned to reveal the social asymmetries of society, the repressive model of power still constitutes power as a closed density, as a limited phenomenon associated with certain kinds of structure rather than others, in a way that conceptually restricts rather than enables the possibility to examine and resist the interplay of discourse and institution, that underestimates from the start the multi-dimensional channels through which the network of what Foucault calls 'Juridico-Discursive' forces operate. The repressive view of power, even when connected to a critical intent, is based therefore on the same metaphysics of limit and identity, of a power/non-power relationship which is one more extension of the hegemony of the Cartesian presence/absence duality. It is within a Nietzschean concept of power as expansive and creating, as a fundamental force of life (that is, a basically active force - see Bapty, chapter 9) not only before, but even as it takes the form of restriction, tyranny and repression, where Foucault paradoxically finds the means to break the apparent indivisibility of the Cartesian union. In a way analogous to deconstruction, Foucault's conceptualisation of power thus involves not simply asserting it as an active over a passive agency, of something all pervasive as opposed to something constrained, but rather of rewriting entirely the word 'power' into a meaning that sits outside such polarity of reference, that is all inclusive exactly in the sense that it evades determination by an exclusive/inclusive bracket.

However, this is not to say that Foucault is reproducing a post-structuralism directly compatible with that of Derrida, and this is where the debt to Nietzsche is more than simply one of terminology. Foucault's interest in a concept of power, however rhetorically shaped, is to use that concept to reveal a certain truth, to grasp a new basis for the understanding of discourse, and indeed, exactly to make meaning of concepts such as history. From the start, Foucault's genealogical project, his analysis of power/knowledge relations, is fundamentally reconstructive , rather than (in a Derridean or conventional sense) deconstructive, it points away from the search for originary unity, but away too from even the suggestion of a retreat instead into nihilism. So, following Nietzsche, Foucault does not condemn history either as mode of repressive power, or as a presence inevitably flawed in the all-embracing conspiracy of western metaphysics, rather he seeks to transform the ground on which history operates and equally to dispose of the idea that history can ever be escaped (as archaeology had boldly claimed). Power is not just everywhere, it is everywhere as in the first instance a process of begetting and becoming, as the drive towards Nietzsche's basic 'will to freedom'. While none of this is to deny the harsh reality of power as repression, it is to suggest that the way to truly undermine the locus of that power is to investigate how such repression is seen perversely to serve life. The 'analytics of power' Foucault proposes should not 'concentrate the study of the punitive mechanisms on their "repressive" effects alone . . . but situate them in the whole series of their possible positive effects, even if these seem marginal at first' (Foucault 1977c, 23). Moreover, with the realisation of power as a procreative force is the justification to dissolve reliance on the logic of metaphysical polarity; this is facilitated in the movement set in motion as power covertly identifies itself as the silent origin of Foucault's analysis through the very rhetorical object of that analysis to identify power elsewhere. And it is in this sense that Foucault now seeks to produce through the medium of history, not instead of it, a genealogy.

Reference to Nietzsche helps to clarify this fundamental point. As for Nietzsche, the final legitimation of what he called the 'eternal return' marked by the becoming of the 'superman' was to be found at the juncture where the value of truth itself was questioned, where, in Nietzsche's terms, the 'will to power' was at last divorced from the 'will to nihilism', so for Foucault it is here that a critical use of the subjective is to be found - evading the pitfalls of an anarchistic nihilism on the one hand, and the attractions of the misguided will to truth on the other. It is here that genealogy is lodged, suspended in the tension between the Nietzschean concept of power, and a continued concern with the old archaeological problems of knowledge and the constitution of the subject. Out of this tension Foucault draws the rhetorical figures around which his works of the 1970s are based, such as, for example, the distinction between the 'bio-power' power of the body and that power exerted over the body elaborated in Discipline and Punish (Foucault 1977c); within genealogy such a polarity is a space of continual self-displacement and opening, not the claim to reveal a new presence, a new knowledge, as it might have been in archaeology.

Yet that said, from a Derridean perspective, it might seem that genealogy is still open to equivalent criticisms to those offered for archaeology. Genealogy may well share with deconstruction 'the rejection of the metahistorical deployment of ideal significations and indefinite teleologies' while opposing 'itself to the search for origins' (Foucault 1984, 77). Yet in the same breath it is also 'gray, meticulous, and patiently documentary' aiming to 'isolate the different scenes where they [events and discourses] engaged in different roles' only by a process of analysis that 'requires patience and a knowledge of details, and . . . depends on a vast accumulation of source material' (Foucault 1984, 76-7). Talk of meticulous documentary, patiently observed detail, the capacity to isolate scenes and structures in the past, might all seem explicitly to reintroduce the spectres of identity and the origin, indeed re-announce the claim of a metaphysical truth, even as all that is denied. One might indeed be inclined to argue that genealogy is flawed by a greater contradiction than archaeology had been; at least the latter had been consistent within its own limits, where genealogy, certainly from the perspective of other post-structuralisms, must seem internally inconsistent from the very beginning.

To go beyond this apparent impasse it is necessary again to trace the link with Nietzsche, this time in the more oblique sense of seeing the embeddedness of Foucault's, like Nietzsche's project, in the actual process of reading; genealogy is constituted in the active moment of understanding and following an analysis, not merely through such judgement as might be gained on the basis of those relatively few (but to the commentator attractive) programmatic and polemical outlines of his work Foucault gives. So the very process of commentary in a resume such as this, that process, to paraphrase Foucault, of saying for the first time what has already been said, and tirelessly repeating what was never said, almost has the unavoidable effect of casting Foucault in a mistaken light, of making solid in his work what was never meant to be solid. That vast accumulation of detail which forms the 'cyclopean monument' of genealogy is not so much to privilege the natural identity of the particular interpretation, to locate meaning in the scholarly recovery, cross-referencing, and study of detail as an end for itself, as it is exactly to create a maze of dislocation and repetition in which meaning as truth is dispersed and lost and meaning as genealogy is found in the becoming moment of interpreting that mass. Genealogy combats the ploys of metaphysics exactly in the excess with which it seems to pander to them.

Foucault's genealogical texts are to be read more at the level of a sequence of Nietzschean aphorisms, than they are even a stylised or parodic version of an orthodox piece of historical writing. In so far, then, as this discussion of Foucault has not been based around more detailed exegesis of works such as Discipline and Punish (Foucault 1977c) or The History of Sexuality (Foucault 1979b; 1986; 1988) that is because the understanding of those books can only be fundamentally misrepresented in brief summaries of the explicit arguments they nominally present; it is in the process of reading now perceived as rather a creative writing, working through recognition of the inability to close a final metaphysical meaning - that is, in a sense, of the reader participating in a discourse of confession, such as that which for Foucault has been central to the development of western concepts of sexuality, but where admission of guilt is now seen as productive rather than repressive - where Foucault's genealogical meaning is opened, where the meaning of The History of Sexuality is to be found. Foucault is forever trying to avoid the ploys of 'simple' meaning, of direct sense; rather he wishes to find himself 'Surrounded by words, taken up and carried beyond any possible beginning' (Foucault cited in Sheridan 1980, 121) - and, indeed, any possible end.

Foucault's genealogical project in many ways complements and intertwines with that strand of cultural criticism represented by Gilles Deleuze and Felix Guattari's two-volume work Capitalism and Schizophrenia, which remains one of the most mysterious and important texts of post-structuralism. The first volume of the project, which was published in France in 1972, Anti-Oedipus (1984), very rapidly stirred up a controversy. A frontal assault on the Oedipus complex, the values of the family, and the recovery of the ego, it ravages and pierces the repressive framework of Freudian and post-Freudian theory. The central problem is that of desire, which has been defined, since Plato, as 'a void in a subject that has been filled by the acquisition of an object' (Bogue 1989, 89), and codified by Lacan within whose conceptual architecture Guattari, as a practising psychoanalyst, had worked until the late 1960s. For Deleuze and Guattari, however, desire is something positive and productive, not negative; it is production, and the body of a self is not a lack of being 'filled' by subj edification, but a machine cut up by the various machines of 'desiring-production'. Desire, belonging to the unconscious (in which there is no such thing as negation) is the potentially subversive and disruptive force against which society and history are played out (see Nordbladh and Yates, chapter 8), and which the Oedipus complex, with which psychoanalysis is in complicity, attempts to force into territories and self-limiting circuits:

Psychoanalysis, at the most concrete level of therapy, reinforces this apparent movement with its combined forces. Psychoanalysis in itself ensures this conversion of the unconscious. In what it calls the pre-Oedipal, it sees a stage that must be surmounted in the direction of an evolutive integration (toward the repressive position under the reign of the complete object), or organised in the direction of a structural integration (toward the position of a despotic signifier under the reign of the phallus). The aptitude to conflict of which Freud spoke, the qualitative opposition between homosexuality and heterosexuality, is in fact a consequence of Oedipus: far from being an obstacle encountered from without, it is a product of oedipalisa-tion, and a counter effect of the treatment that reinforces it.

(Deleuze and Guattari 1984, 74)

Against the tyranny of analysis and of the ego, Deleuze and Guattari propose a 'schizoanalysis', designed to liberate desire from the unconscious, to subvert the current structures and precipitate change, 'promoting visions of revolutionary psychological formation and new earths to inhabit' (Leitch 1981, 213). Published eight years after the French edition of Anti-Oedipus, A Thousand Plateaus (1988) takes the critique developed in volume one further, and in a different direction. Psychoanalysis recedes from view, and the text is written around fifteen 'plateaus', dealing with different periods of time, the plateau referring to the work of Gregory Bateson (1972) who claimed to have discovered a 'non schizomogenic structuring' of society in Bali, where instead of the rising crescendo and climactic structure of occidental culture and sexuality, a pitch of intensity is reached and maintained, not dissipated. Each of the fifteen plateaus is dated and named, but there is no plateau organising them in a series, no line of evolution or teleology. Randomly organised, the component essays of Mille Plateaus disrupt all the orthodox emphases of history, and form points of open focus in a landscape without any boundaries. What 'unites' everything is a rejection of the limitations of ego and of the law of castration (the Oedipal codes) which reunite history in retrospect from that point, and an emphasis upon the Nietzschean concept of a becoming which holds everything in flux. Against this, as described in plateau 5, society attempts to hold itself in place by regimes of signs, and they emphasise that language is not designed to communicate information but to categorise, impose order and thereby control. Words form themselves into the mots d'ordre which are these regimes and which have two sides, the content level, such as the prison system in Foucault's Discipline and Punish (Foucault 1977c), and an expression level, the acts which bring about transformations of bodies, such as judgements, verdicts and classifications (Bogue 1989, 138). The mots d'ordre is the way by which desiring production is kept under control socially and historically.

Difficult as it is, it is likely to be the form of writing used by Deleuze and Guattari, writing so many strands of post-structuralist, structuralist,

psychoanalytic and Marxist thought, that will prove decisive in finally transforming the way in which we write and practise history:

It criticises structuralism, particularly the Lacanian variety. It openly deplores all hermeneutics. It aims to liberate the flow-breaks of desire. It is Barthesian. To let the libidinal schiz-flows breach all borders and barriers. It is Derridean. To unleash molecular forces so as to undo all molecular formations. It is Foucauldian. To set revolution growing.

(Leitch 1983, 220)

But it raises the question of the subject, without a consideration of which it is impossible to understand the potential impact of post-structuralism upon archaeology.

IV

Derrida writes:

Now if we once again refer to the semiological difference, what was it that Saussure in particular reminded us of? That language [which consists only of differences] is not a function of the speaking subject. This implies that the subject (self-identical or even conscious of self-identity, self-conscious) is inscribed in language, that he is a 'function' of the language. He becomes a speaking subject only by conforming his speech ... to the system of linguistic prescriptions taken as the system of differences.

(Derrida 1973, 145-6)

At the heart of the post-structuralist position (and there is probably a degree of conformity on this point which allows us to speak, provisionally, in the singular) is the attack upon the notion of the subject as the origin of meaning, as situated at the origin and in some sense partially outside the chain of signifiers it seeks to articulate. For metaphysics, which is the main target of Derrida's critique, the subject was an essence that in some sense is external to the meaning of its self-representation - thus Descartes' cogito escapes self-doubt only at the point of its own perception of itself, its self-knowledge, and it is around this certainty that the various chains of signification can be hung and so determined. In, for instance, the phenomenological rethinking executed by Edmund Husseri, two types of language can be identified, an essential language of expression, comprising the signs of thought, and a secondary language of indication, composed of signs charged with responsibility for indicating to an outside point a discourse which has already taken place on the inside. Two forms are therefore identified - presentation (expression) and representation (indication) which Husseri arranges hierarchically and axiologically. Derrida's reading (1973) undoes this priority and so collapses all meaning into a structure of indication, characterised by ambiguities and fissures of sense.

The consequence of-this argument is that it is no longer possible to consider the subject-object duality as an essential polarity, for both must be understood in relation to the system of signifiers in which they are articulated. Thus Derrida argues that the proper (i.e. personal, unique, individual) name is still part of a system of writing, and therefore of classification, because the activity that divides the proper from the objects that surround it is itself a categorisation, one that separates the human/individual from the material/object. By virtue of being a sign inscribed within a system of signs, the subject can no longer be treated as as an effect of a system of written prescriptions: '. . . the proper name was never possible except through its functioning within a system of classification and therefore within a system of differences . . .' (Derrida 1976, 109).

Subjectivity - like objectivity - is an effect of differance, an effect inscribed in a system of differance ... it confirms that the subject, and first of all the conscious and speaking subject, depends upon the system of presence, that the subject is not present, nor above all present to itself before differance, that the subject is constituted only in being divided from itself, in becoming space, in temporising, in deferral.

(Derrida 1981a, 28-9)

In contrast to the Cartesian formulation, and to those like Edmund Husserl who try to construct a system of meaning around the subject as self-presence, post-structuralism asserts that the subject cannot be self-present, and that its identity is a social and cultural production. The subject becomes through the movement of a signifying chain, and it is within this chain, which articulates the space/time of the subject, that the familiar metaphysical notions of intentionality, consciousness, and meaning are defined. Rather than being the essence of existence (rather than 'being' therefore), the sign that announces subjective identity/ ontology is itself already a common name, a classification, 'sufficiently equivocal to hold . . . the non proper as well as the proper' (Derrida 1984, 64). So the signature, about which Derrida has written a great deal, has to be iterable - repeatable - and therefore has to fill the same requirements as any graphic mark.

What would be at issue here, then, would be precisely the role attributed to the subject in the interpretive space of the text. What Derrida reminds us is that the subject is only to be read in the form of a text, that it is not located on the exterior of a text but is articulated on its inside. Thus it becomes possible to oppose the idealism manifested in the work of the Yale school inspired by deconstruction by applying more generally the notion of the text, such that analysis becomes the aim of examining how texts interpolate readers into their body. If everything is a text, then the production of meaning must be seen as a form of intertextuality - i.e. the way in which text works another text to produce a new text. The ontology of the author as a point of origin is undone -though this need not involve a total erasure of the author as a radical Barthesian 'death of the author'.

The subject is to be seen not as constitutive of ideologies - as someone who manipulates or is manipulated by ideology - but as ideologically constituted. This is clearly one of the informing ideas behind Foucault's project, and explicitly so when it emerges in 'genealogical' form, and the question becomes how such 'fundamental' categories as sexual identity are formulated out of the object relations of discourse and power as apparently natural, and scientifically analysable entities. At a rather different level, an application of this notion is also to be found in Lacan's theories by which the subject 'becomes' through its entry into 'la symbolique', the symbolic, a generic term for all the processes of signification, including language but also other forms of discourse.

The subject becomes, for Lacan, by a process of splitting and identification (the 'mirror' and 'castration' phases). When I take on the mode of representation, and learn to refer to myself as 'I', I take on a completed pre-existent system of relationships (differences), in which space has been allocated to me. Like the image reflected in a mirror, it is an idealisation presented back to me, a specular identification, and therefore there is, within the sign 'I', a split between me as the subject of the sentence and the 'I' that is not represented in the sentence at all. The subjective - I, formed as I acquire language - is situated with respect to chains of signifiers, such that not every 'I' is equal - the point of entry into the symbolic and cultural system is therefore already mediated, according to culturally significant variables of sex, class, colour, race, etc., which can be understood as the attachments made to the subject as supplementary significations. Becoming a subject is therefore bound up in a process by which I become 'me', white, male, middle class etc. The subject 'is' not, cannot 'be'. It is a text to be read as a text. The symbolic order is that which gives to the subject its identity as a subject, allowing it both to signify and to become signifiable, for the signifier is first that which represents the subject for another signifier, not for another subject.

The subject, then, comes into being through its entry into the symbolic system, and this establishes the concept of the subject as onto-judicial rather than simply ontological. Man, Foucault insisted, 'is an invention of recent date', and that once its position within the modern period and discourse are removed, it will vanish 'like a face drawn in the sand at the edge of the sea' (Foucault 1974, 387). The alliance of notions of human essence and integrity can be linked directly to the emergence of social-judicial psychology, which allows legal systems to assault for crimes the body of an individual conceived as the origin of the crime. In a sense, the criminal/victim thus becomes a necessary exercise to protect society from its own complicity and guilt.

There is, however, no need to assume that the discourses into which we are 'subjected' are homogeneous and coherent, and if we take the issue of women's position and identity within contemporary western societies, it is quite clear that the discourses that are responsible for the construction of the subject are multiple, incomplete and contradictory. Catherine Belsey puts the point succinctly: women participate both in the liberal humanist discourse of freedom, self-determination and rationality and at the same time in the specifically feminine discourse offered by society of submission, relative inadequacy and irrational intuition. The attempt to locate a simple and coherent subject position within these contradictory discourses, in consequence to find a non-contradictory pattern of behaviour, can create intolerable pressures.

(Belsey 1980, 65-6)

It is here that we can begin to see more clearly the ambiguity inherent in structuration, since the stability of subjective identity/positionality depends precisely upon the ability of systems/subjects to keep the contradictory nature of their own constitution from coming to fruition -to keep in place the unacknowledged conditions of action and to keep from attention the unintended consequences of action.

The desire to found a non-contradictory mode of self-expression could, of course, be a cause and focus for the feminist movement (and Belsey maintains it is precisely because the contradictory discourses of women's position could not be stabilised that a feminist discourse has appeared). It is to be doubted to what extent a Lacanian position can be used without serious modifications. Coward and Ellis (1977) are optimistic, and their fusion of Marxism with structural/Lacanian psychoanalysis is productive. 'Marxism cannot conceive of a subject who remains outside the structure, manipulating it or acting as a mere support, if it did so it would cease to be a revolutionary philosophy' (Coward and Ellis 1977, 61). Ideology becomes, typically, 'a machine of representation, a practice to produce a specific articulation, that is, producing certain meanings and necessitating certain subjects as their supports' (Coward and Ellis 1977, 67). But the point must also be that the Lacanian position is insufficient, since it recognises only the post-subjective identity, and therefore represents the subject and the process of sub j edification as inevitable and irrevocable. Lacan's situating and decentring of the subject differs from that of Derrida to the extent that Lacan's definition of the pre-subjective as 'lack' is far closer to Heidegger's erased 'being', to a negative ontology, than is recognised by Coward and Ellis. Ultimately, Lacan's discourse must determine the subject as a species of being-present. If 'there is something missing in the Marxist [i.e. Althusserian] analysis of the subject' (Coward and Ellis 1977, 82) then this can only partially be solved by turning to ec fits and La Seminaire.

In fact, Lacan's formulation of the subject remains essentially phallocentric or phallogocentric. For Lacan, what determines the subject and assigns to it full identity and subjectivity (as opposed to the partial and specular identity of the mirror phase) is the identity of the phallic Signifier where exchange (along a chain of signifiers) is no longer possible. Thus in the Oedipal or castration complex, the subject finds its identity by forcing its desires into the framework delimited by Oedipal triangularity (dama-papa-me) and submitting to the rule of the phallus as the name-of-the-father (for it is the father who prohibits masturbation/incest and enforces the socio-cultural frames of the sexual desires). Feminists have been justified in objecting to the language that is used by Lacan, and Culler (1983a) is correct to point out that merely denying that the •phallic-signifier' refers to a penis does not amount to a vindication of Lacan's choice of concepts. What is perhaps more interesting, however, is the way in which women have begun to rework the psychoanalytic formulas of Lacanian psychoanalysis, and have been led to align themselves with the unconscious, the pre-subjective, or what Julia Kristeva calls the 'semiotic' or 'pre-thetic', all of which amount to a rejection of the 'phallocentric' conditions of identity.

It is inappropriate and moreover counterproductive to attempt to find a 'real' woman, a proper signifying space, since these are specular identifications mediated through the cultural constitution of a superego or ego-ideal, the conventionalised model of self through which the subject perceives its own identity. To talk of a 'woman' in the sense of a definitive identity is to force political and social conceptualisations to play the rules of the game, and thus to continue to conform to societies' norms and systems of reproduction. Thus for Helene Cixous 'you can't talk about a female sexuality, uniform, homogeneous, classifiable into codes, any more than you can talk about one unconscious resembling another' (Cixous 1980, 246).

Women's unity, therefore, renounces the conditions of identity and becomes the infinite, the inexhaustible, the dissolution of all forms of identity and the rejection of all classifications. To define a woman is to censure her - 'Female sexuality has always been theorised within masculine parameters' (Irigaray 1980, 99). If it is the rule of the father as the law and ontology of the phallic signifier which, for Lacan, guarantees the symbolic order through the triangularity of the Oedipus complex, then women mark their resistance and difference from this position, their distance from this order, by asserting the plurality of their sex. Where for Freud and Lacan women's sex is not a sex - not the site of the presence of a vagina, but the site of the absence of a penis - Irigaray asserts women's plurality. The woman cannot be threatened with castration for touching her genitals, because she touches herself constantly and cannot be forbidden from doing so, 'for her sex is composed of two lips which embrace continually. Thus within herself she is already two - but not divisible into ones - who stimulate each other' (Irigaray 1980, 100).

Women evade and subvert the laws of castration, escaping the polarity between active and passive sexuality which Freud saw as the precursor of puberty and the only reality of sexual difference among children. The woman would begin as the passive side of the polarity which would later become that of the feminine. 'She is neither one or two. She cannot, strictly speaking, be determined either as one person or two. She renders any definition inadequate' (Irigaray 1980, 101). A woman 'has sex organs everywhere', and the geography of her sexuality evades the (male) law of genitality established via the Oedipus complex. The woman's world becomes, therefore, in contrast to phallic finitude, 'a sort of experience in expansion for which no limits could ever be fixed and which, for all that, would not be incoherency' (Irigaray 1980, 104).

This argument could be shown to realise woman's identity in what Derrida has elaborated as the rule of the hymen in opposition to that of the phallus. Where the phallus is singular, determining, seminating, and so guaranteeing the symbolic order, hymen divides. In the French it has two meanings, both the vaginal membrane and marriage - at one and the same time, therefore, both virginity and consummation, purity and impurity, inside and outside, self and other. Hymen creates differences, keeps them apart (differentiates them) while at the same time joining them, preventing them from separating.

The debate over how to found around this notion a political strategy, and what the implication of women's plurality is, has been vigorous, and some of the criticisms levelled at, for example, Cixous are justified. By embracing women's identity as the irrational, Cixous is led into accepting an essentially masculine position. Julia Kristeva's work, which attempts to steer a middle path the two forms of feminism that have coexisted uncomfortably in France since 1968, offers the clearest statement of what is at stake:

The belief that 'one is a woman' is almost as absurd and obscurantist as the belief that 'one is a man'. I say 'almost' because there are still many goals which woman can achieve: freedom of abortion and contraception, day care centres for children, equality on the job, etc. Therefore we must use 'we are women' as an advertisement or slogan for our demands. On a deeper level, however, a woman cannot 'be': it is something that does not even belong in the order of being ... In 'woman' I see something that cannot be represented, something that is not said, something above and beyond women and ideologies.

(Kristeva 1980, 137)

^What is at stake, indeed, in the issue of identity is the question of Strategy. Nordbladh and Yates (chapter 8) note that within archaeology and anthropology a division is made between biological sex and cultural gender, where the former is assumed as a natural, given division, and the latter as a secondary realm of cultural construction. They point out that biologically the two poles of gender are not as clear cut as has often been assumed, and that other combinations of chromozones are possible than the simple male (XY), female (XX) opposition. Semantically, they examine the notion that sexuality can be forced into this framework, and again find that a naturalised ground of sexual identity cannot be located - culture is implicated as far back as one can go, therefore nature and culture form a continuum, breaks in which are the acts of power and ideology. Their argument endorses Deleuze and Guattari's assertion that man 'does not live nature as nature, but as a process of production. There is no such thing as either man or nature now, only a process that produces the one with the other and couples the machines together' (Deleuze and Guattari 1984, 2).

V

Post-structuralist thought is marked by diversity, indeed it intrinsically produces diversity as it opens the radical potential of the structuralist emphasis on difference. To summarise the aims of post-structuralism, to attempt to pull that diversity into some systematic set of concepts, must, in a sense, run contrary to post-structuralism's very departure. Perhaps then, to maintain the post-structuralist insight, we should not so much conclude at this point, as merely set out certain fluid themes around ; which the shift towards a post-structuralist archaeology might be conceptualised; for the rest, the text of this book must be left to write itself.

Post-structuralism is a discourse on identity where structuralism was a discourse 'about' or with identities. In this development structures are to be torn apart and destroyed, not simply recovered and reconstituted. Thus post-structuralism is a discourse on form as well as content -everything is in difference, which drives a knife through presence. One mark of this process is the destruction of the space occupied by the idea of context. As identity is founded in non-identity, so context can no longer be neatly denned within parameters engendering themselves from the outside as a concrete exemplification of what they govern on the inside. The question eternally becomes: What parameters tell us how to apply the parameters, and to what context do these meta-rules themselves belong?

This assault on identity and the attendant notion of context - and the domination by those terms of the conventional possibility of an archaeology - is variously taken up in this volume. The problem of context is directly articulated by Yates (chapter 6), Moran and Hides (chapter 7) and Bapty (chapter 9), as it develops into a necessary critique of contextual archaeology, and more generally by Maley (chapter 2) in relation to his Derridean critique of Foucault. More specifically, the authors adopt a variety of rhetorical postures to outstrip the inevitable claims of context in the post-structuralist positions/archaeologies they propose. Burr (chapter 1), Yates (chapter 6) and Tilley (chapter 5) all argue for the use of the apparently insane, irrational, abnormal and irreverent as a means to this departure. Bapty (chapter 9) appeals to a Nietzschean process of 'overcoming', Moran and Hides (chapter 7) to the particularity of personal experience and knowledge set against the dominant presence of the author, and accepted rhetorical convention.

A further elaboration of this theme is in relation to the key question of the special kind of context associated with the western concept of subjectivity. As post-structuralism dispels the myth of natural identity, so it dispels the self-constituted subject prescribed by the Cartesian cogito, the subject as a founding presence, a basic unity of being, outside language, outside materiality. The papers in this volume identify this issue as a basic site on which to activate a post-structuralist critique. For Yates (chapter 6) the unity of the subject is to be replaced by the Freudian/Lacanian notion of the unconscious: as a structured set of signifiers, subjectivity is an effect, a function of the displacements between the symbols and imaginary orders operating, for example, in the rock art of Bronze Age Scandinavia. Nordbladh and Yates (chapter 8) similarly question the indivisibility of the subject which has preserved the cultural production of the male/female distinction, and suggest that a feminist archaeology should more radically seek to dissolve that duality. Bagnal (chapter 4) outlines the feminist movement in post-structuralism as it has been articulated through the work of Cixous, Kristeva, Irigarary in relation to this problem.

Post-structuralism involves a basic emphasis on writing, and on reading as itself a process of writing, that is as one more extension of, in Derrida's terminology, the play of differance, one more supplement to what was always already written. According to the rule of the supplement, writing adds as it completes, and through this process there must always be an excess, a remainder beyond any apparent statement of presence.

The post-structuralist theory of writing leads, in this volume, both to an explicit consideration of how strategies of writing the past, or of writing about cultural phenomena more generally, should be undertaken, and more implicitly to an examination of how the idea of writing as difference and supplementarity can open to new meaning the density of 'history' which archaeology occupies. There is, then, a general emphasis on reading/writing (archaeological) texts as literature in a deconstructive framework, from Burr's (chapter 1) invocation of the genre of 'magical :; realism', to Yates' (chapter 6) use of Kafka and psychoanalysis, and Maley's (chapter 2) more 'orthodox' Derridean stance. Alternatively, for Tilley (chapter 5) and to a certain extent Waterman (chapter 3) and Bapty (chapter 9), reading/writing is to be pursued in relation to a more Foucauldian theory of discourse, and the explicit attempt to write through, and therefore to identify, the play of power relations.

The more implicit analyses of the effects of the post-structuralist general theory of writing as difference are seen as wider issues, such as the notion of 'history', and the apparent absoluteness of the past/present relationship are questioned. Walsh (chapter 10), in his discussion of post-modernism and the public presentation of the past, points to the danger of a reified present, of a history conceived as a single unproblematic identity before the post-structuralist space, before, in a sense, writing. Tilley (chapter 5) in his 'history' of Anglo-American archaeology, begins to explore more directly what happens when the excess of the present which, through the action of the supplement, must always be in the past (in this case the recent past of archaeology) is acknowledged, and Yates (chapter 6) considers the radical potential of this movement in terms of an analysis of the distant past openly posited around Freudian psychoanalysis. In these ways, history, if never an absolute contingency, can now begin to be written as writing, that is as a realm where the contingent must rhetorically emphasise itself, and be used to break down the falsity of such divisions as past and present.

Materiality is conventionally invoked by western metaphysics as a limiting principle - the principle of externality. Thus, for example, in contextual archaeology, the 'material' model of material culture demarcates its limits as a purely discursive entity; the material is the 'other' of the 'discursive'. In post-structuralism, however, the invocation of the material is revealed in its paradoxical nature, a paradox continually disturbing the unity of identity and context. Material culture is a text which must be read, but now that reading is action - it is, as we have seen, always already writing. Archaeology, as the study of material culture is, in a sense, now moved to centre stage, since there can be no culture that is not material culture, which is not an effect of textuality.

The ramifications of this point, overlapping with the expansion of writing and the rupture of context, are fundamental to this book, and its intent to announce an archaeology no longer accepting its own disciplinary regulation within a limited field of study. This point is made in a number of different ways by Tilley (chapter 5), Yates (chapter 6) and Bapty (chapter 9), and implied by the freedom with which the signifier 'archaeology' runs through the nominally 'non'-archaeological texts of, for example, Maley (chapter 2) and Moran and Hides (chapter 7). At one level, and paradoxically, the post-structuralist emphasis on textuality/ materiality ironically confers through the play of non-identity the 'identity' archaeology has always craved; it is the possibility of this paradox this volume aims to explore.

Yet, of course, it would be wrong to give the impression that the papers in this book slavishly subscribe to post-structuralism without stopping to question it, or see an 'easy' emergence of a post-structuralist archaeology. On the contrary there are many notes of caution. For Maley (chapter 2) the ultimate validity of Foucault's archaeological/genealogical project is to be closely questioned, and similarly Waterman (chapter 3) is critical of post-structuralism's - and more particularly Foucault's - rejection of ideology analysis, as it seems also to reject the basis for a valid political practice. Walsh (chapter 10) espouses related doubts in his consideration of post-modernism, post-structuralism and the heritage industry, empha­sising the difficulty of combating the recent appropriation of the past in the heritage boom by anything less than an equally absolute response. He endorses Foucault's statement that there are lines of force, not sense, and like Waterman (chapter 3), calls for an articulation with Marxism as ultimately a more fruitful strategy.

Ultimately, of course, this book is about questions, not answers. New questions, not the old dressed up to look a little more rigorous/scientific/ logical, minor tinkerings and adjustments being the usual means by which archaeologists - some post-processualists included - have used radical ideas to (paraphrasing Lentricchia) increase the noise without noticeably increasing the speed or performance. We hope that this book goes deeper. Phil Kohl has written that archaeology is nothing today if not acutely self-conscious. In this book, we have disagreed: it is often little more than acutely louder. A break, an epistemological shift, is required, and it is towards this that we have worked here. To paraphrase Barthes: the birth of a critical practice can only be achieved at the cost of the death of an archaeology.

It is not, of course, possible to give a comprehensive bibliography of all the interesting or relevant material - the literature both 'primary' and 'secondary' (a classically deconstructable distinction!) is enormous, although a large amount deals very specifically with the implications for literary theory, etc. Perhaps the most accessible general surveys are those of Culler (1983a), Leitch (1983) and Norris (1982), which deal with the main range of post-structuralist authors, though perhaps biased towards Derrida, and not always as critical as they might be. The five essays in Sturrock (1979) are more balanced and provide a good introduction to the main critics. There are volumes on Derrida, Barthes, and Foucault in the Fontana Modern Masters Series (Norris 1988; Culler 1983b; Merquior 1985), though the latter is rather unsympathetic toward its subject. Other volumes on Foucault are numerous, the best being those by Dreyfus and Rabinow (1982) and Sheridan (1980). Note should also be made of a forthcoming volume which contains essays on Barthes, Derrida and Foucault discussed in an archaeological framework (Tilley in press). On Lacan, see the survey by Coward and Ellis (1977), which owes more, perhaps, to Laplanche and Leclaire's (1972) account than to Lacan's own, but that is not necessarily limiting. MacCannel's volume (1986) is interesting, but lacks direction, while Frosh (1977) places Lacan in a wider psychoanalytic perspective. The only volume to deal with feminist post-structuralism - the work of Cixous, Irigaray and Kristeva - is by Moi (1985), who sets their work in a wider perspective than just that of post-structuralism. Deleuze and Guattari are covered in a recent volume by Bogue (1989) much of which, regrettably, remains opaque, and is heavily biased against Guattari; it contains, however, a useful bibliography for locating other commentators. Some of the figures whose impact outside France has been less marked, particularly Lyotard, are discussed by Dews (1987), who compares post-structuralism with the Critical Theory tradition.

Other useful surveys, of varying depth, spread and sympathies, and differing in the extent to which they articulate post-structuralism with other traditions, are worthy of mention: Jameson (1972; 1983) offers some stimulating discussion, particularly on the place of Marxism within post-structuralism, on which see also Ryan (1982) which should be read bearing in mind the comments made by Eagleton (1986). Lentricchia (1983) discusses the impact of the French avant-garde on American literary criticism, and surveys the American Yale 'school' of deconstruction (de Man, Miller) in a manner more critical than that of Leitch (1983) who covers the same material.

Barthes, R. (1973) Mythologies. St Albans: Granada.

Barthes, R. (1974) SIZ, trans R. Miller, New York: Hill & Wang.

Barthes, R. (1977a) Image-Music-Text, trans. S. Heath, London: Fontana.

Barthes, R. (1977b) Roland Barthes, trans. R. Howard, New York: Hill & Wang.

Bateson, G. (1972) Steps to an Ecology of Mind, London: Intertext.

Belsey, C. (1980) Critical Practice, London: Methuen.

Bogue, R. (1989) Deleuze and Guattari, London: Routledge & Kegan Paul.

Butler, C. (1984) Interpretation, Deconstruction and Ideology, Oxford: Oxford University Press.

Cixous, H. (1980) The laugh of the Medusa', trans. P. Cohen and K. Cohen, in E. Marks and I. de Courtivron (eds) New French Feminisms: An Anthology, Amherst: University of Massachusetts Press.

Clifford, J. (1988) The Predicament of Culture: Twentieth-century Ethnography, Literature and Art, London: Harvard University Press.

Coward, R. and Ellis, J. (1977) Language and Materialism, London: Routledge &Kegan Paul,

Culler, J. (1983a) On Deconstruction, London: Routledge & Kegan Paul.

Culler, J. (1983b) Roland Barthes, London: Fontana.

Deleuze, G. and Guattari, F. (1984) Anti-Oedipus: Capitalism, and Schizophrenia, London: Athlone.

Deleuze, G. and Guattari, F. (1988) A Thousand Plateaus: Capitalism and Schizophrenia, London: Athlone.

Derrida, J. (1973) Speech and Phenomena, trans. D. B. Allison, Evanston: North-Western University Press.

Derrida, J. (1976) Of Grammatology, trans. G. C. Spivak, Baltimore: Johns Hopkins University Press.

Derrida, J. (1978) Writing and Difference, trans. A. Bass, London: Routledge &Kegan Paul.

Derrida, J. (1979) 'Living on: border lines', in H. Bloom et al., Deconstruction and Criticism, New York: Seabury.

Derrida, J. (1981a) Positions, trans. A. Bass, London: Athlone.

Derrida, J. (1981b) Dissemination, trans. B. Johnson, London: Athlone.

Derrida, J. (1984) Signesponge, trans. R. Rand, New York: Columbia University Press. Derrida, J. (1987) The Truth in Painting, trans. G. Bennington and I. Macleod, Chicago: Chicago University Press.

Derrida, J. (1988) 'Letter to a Japanese friend', in D. Wood and R. Bernasconi (eds) Derrida and Difference, Evanston: North-Western University Press.

Dews, P. (1987) Logics of Disintegration: Post-Structuralism and the Claims of Critical Theory, London: Verso.

Dreyfus, H. L. and Rabinow, P. (1982) Michel Foucault: Beyond Structuralism and Hermeneutics, Brighton: Harvester.

Eagleton, T. (1986) Against the Grain, London: Verso.

Foucault, M. (1967) Madness and Civilization: A History of Insanity in the Age of Reason, trans. R. Howard, New York: Vintage.

Foucault, M. (1972) The Archaeology of Knowledge, trans. A. Sheridan, London: Tavistock. Foucault, M. (1974) The Order of Things: An Archaeology of the Human Sciences, trans. A. Sheridan, London: Tavistock.

Foucault, M. (1975) The Birth of the Clinic: An Archaeology of Medical Perception, trans. A. M. Sheridan Smith, New York: Vintage.

Foucault, M. (1977a) 'What is an author?', in M. Foucault, Language, Counter-Memory and Practice: Selected Essays and Interviews, trans. D. F. Bouchard and S. Simon, Oxford: Blackwell.

Foucault, M. (1977b) 'Language to infinity', in M. Foucault, Language, Counter-Memory, Practice: Selected Essays and Interviews, trans. D. F. Bouchard and F. Simon, Oxford: Blackwell.

Foucault, M. (1977c) Discipline and Punish: The Birth of the Prison, trans. A. Sheridan, London: Alien Lane.

Foucault, M. (1979a) 'My body, this paper, this fire', trans. G. Bennington, The Oxford Literary Review 4(1): 4-29.

Foucault, M. (1979b) The History of Sexuality, Volume 1: An Introduction, trans. R. Hurley, London: Alien Lane.

Foucault, M. (1984) 'Nietzsche, genealogy, history', in P. Rabinow (ed.) The Foucault Reader, London: Peregrine.

Foucault, M. (1986) The History of Sexuality, Volume 2: The Use of Pleasure, trans. R. Hurley, London: Viking.

Foucault, M. (1988) The History of Sexuality, Volume 3: The Care of the Self, trans. R. Hurley, London: Alien Lane.

Frosh, S. (1987) The Politics of Psychoanalysis, London: Macmillan.

Hodder, I. (ed.) (1982) Symbolic and Structural Archaeology, Cambridge: Cambridge University Press.

Hodder, I. (1986) Reading The Past: Current Approaches to Interpretation in Archaeology, Cambridge: Cambridge University Press.

Irigaray, L. (1980) 'This sex which is not one', trans. C. Reeder, in E. Marks and I. de Courtivron (eds) New French Feminisms: An Anthology, Amherst: University of Massachusetts Press.

Jameson, F. (1972) The Prison House of Language, Princeton: Princeton University Press.

Jameson, F. (1983) The Political Unconscious: Narrative as a Socially Symbolic Art, London: Methuen.

Kristeva, J. (1980) 'Woman can never be defined', trans. M. A. August, in E. Marks and I. de Courtivron (eds) New French Feminisms: An Anthology, Amherst: University of Massachusetts Press.

Laplanche, J. and Leclaire, M. (1972) 'The unconscious: a psychoanalytical study', Yale French Studies 48, 118-75.

Leitch, V. B. (1981) Deconstructive Criticism: An Advanced Introduction, London: Hutchinson.

Lentricchia, F. (1983) After The New Criticism, London: Methuen.

MacCannel, J. F. (1986) Figuring Lacan: Criticism and the Cultural Unconscious, London: Croom Helm.

Merquior, J. G. (1985) Foucault, London: Fontana.

Miller, D. (1987) Material Culture and Mass Consumption, Oxford: Blackwell.

Miller, D. and Tilley, C. (eds) (1984) Ideology, Power and Prehistory, Cambridge: Cambridge University Press.

Moi, T. (1985) Sexual/Textual Politics: Feminist Literary Theory, London: Methuen.

Norris, C. (1982) Deconstruction: Theory and Practice, London: Methuen.

Norris, C. (1988) Derrida, London: Fontana. Ryan, M. (1982) Marxism and Deconstruction: A Critical Articulation, London: Macmillan.

Schiffer, M. (1988) 'The structure of archaeological theory', American Antiquity 53: 461-85.

Shanks, M. and Tilley, C. (1987a) Re-constructing Archaeology, Cambridge: Cambridge University Press.

Shanks, M. and Tilley, C. (1987b) Social Theory and Archaeology, Cambridge: Polity.

Sheridan, A. (1980) Michel Foucault: The Will to Power, London: Tavistock.

Smart, B. (1983) Foucault, Marxism and Critique, London: Routledge & Kegan Paul.

Sturrock, J. (1979) Structuralism and Since, Oxford: Oxford University Press.

Tilley, C. (ed.) (in press) Reading Material Culture, Oxford: Blackwell.