Chapter two: Identity: a performative understanding

Identity in the archaeological literature is often treated either as irrelevant (where the true purpose of archaeology is regarded as being to recover and understand long-term processes) or unrecoverable, as the top level of Hawkesí ladder of inference (1954). Both these attitudes are misplaced. Firstly they thrust modern conceptions of identity back into the past, and thereby legitimate modern inequalities, such as those between men and women for example. Secondly both are based on a view that sees the archaeological material as a record. This point of view has been problematised above. Identity is certainly not irrelevant, it is central to any understanding of the past, and it is certainly not unrecoverable. Just as Barrett has argued with reference to agency, I would argue that the archaeological material is already infused with identity, with gender, because it was produced in, and helped produce a world in which identity and gender were central. Those that argue that identity is unrecoverable, if I may twist the Barrett quote, are disarmed simply because the study of identity does not adhere for its validation to such a theory of representation (cf. Barrett 2000b, 63). The material conditions in which people lived, that which archaeologists recover, are the material conditions through which concepts of identity were created, performed, sustained and undermined. These conceptions were central to past peopleís lives and thus must be central to our investigations of the past.

Today identity is conceived as being the combination of multiple factors, age, gender, sexuality, status and personhood. Each of these made up of categories, which can be taken up by the subject in day-to-day life. Young or old, male or female, straight or gay etc. these are the choices available to us, in our society, these are the positions that make us who we are. Often these notions are perceived and presented as unchanging, eternal, and universal. This assumes an essentialist position with regard to human nature; we are all the same across space and time. It is the same supra-historical position that Barrett (1994) and Giddens (1984) offer on agency. Yet Judith Butler has clearly demonstrated how all the above categories are in actuality merely regulatory ideals (1993, 1). We shall return to regulatory ideals later, but first we need to begin to think more creatively about how identity is created so that unwarranted, essentialist positions will not be inserted onto the past insidiously and thus legitimate modern inequalities.

Michel Foucault

One author who has thought more stringently than others about the nature of identity is the French philosopher Michel Foucault (1977; 1987; 1988). Foucault argues that the conception of agency prevalent in the social sciences "places its own point at the origin of all historicity Ė which in short leads to a transcendental consciousness" (1970, xiv). Instead what was required, he argued, was "not a theory of the knowing subject but rather a theory of discursive practice" (1970, xiv). It is here that identity becomes important, in the attempt by Foucault, as Stuart Hall has argued, to "rearticulate the relationship between subject and discursive practices" (Hall 2000, 16). In other words, identity emerges in an examination of how discursive practices can create differing notions of the subject and subjectivity. This begins to form the basis for an argument that rejects an essentialist view of human identity. Foucault views the subject as constituted entirely within the discursive practices of disciplinary regimes and the repeated performances of "technologies of the self" (Redman 2000, 10). This point is essential. If subjects, and therefore identities, are created through discourse, then they must be produced through historically constituted acts of performance; through conditions, and at moments, that are unique (Hall 2000, 17). This makes identity a historically constituted creation and different across both time and space. This conception of identity is thus far from an essentialist one that sees it as trans-historical human nature. Instead identity is now a process, always under construction, a strategic and positional concept (Hall 2000, 17). Foucaultís ideas around identity, gender and sexuality are intimately caught up in knowledge and power. He himself points out that sexuality for example is a

"historical construct Ö a great surface network in which the stimulation of bodies, the intensification of pleasures, the incitement to discourse, the formation of special knowledges, the strengthening of controls and resistances, are linked to one another, in accordance with a few major strategies of knowledge and power" (Foucault 1978, 105-6).

Foucaultís work has been criticised on a number of levels. In much of his work there is little resistance offered by the subjects (Hall 2000, 25). The power through which disciplinary regimes construct the subject becomes unchallenged and unchanging, the subjects themselves forming "docile bodies" (McNay 1994, 104). Lynn Meskell has accused archaeology of being "seduced by Foucault" (2000, 16). She argues that his conceptions of power and discourse strip the body of any corporeality, leaving it an empty template onto which anything can be written. This however is recognised in Foucaultís later work, particularly The use of pleasure where he begins to discuss how subjects "practice on themselves and others a hermeneutics of desire" (Foucault 1987, 5). In other words he begins to acknowledge how power and discourse can be resisted through the subjectís understanding of self. Furthermore, there have been recent calls to increase the influence Foucault has in archaeology, in order to combine narratives of power, identity and corporeality (Hamilakis et al. 2002; cf. Meskell 2000). Further examinations based on Foucaultís work have also turned directly to the materiality of the body, and these have offered insights that develop his work and also build notions of performativity that blend well with the insights gleaned through an archaeology of practice. Judith Butler (1990; 1993) has carried out the most notable of these works and it is to her work that we will now turn.

Judith Butler: regulatory ideals and performativity

Judith Butler draws on an eclectic range of sources including Foucault (1978), Derrida (1981), Lacan (1977), and Iragary (1985) to problematise all our modern notions around identity. Building on Foucault, Butler describes modern notions of identity as being made up regulatory ideals (Butler 1993, 1). These regulatory ideals, or regulatory fictions as Donna Haraway has described them (1991, 135), provide idealised and reified norms which people are expected to live up to. Thus categories such as male and female, straight or gay, young or old are not biological facts, but categories which we create and recite through performance. As Chris Fowler has pointed out "the ways that subjects attempt to recite subvert or reiterate fictions of identity mark them as a specific type of person" (Fowler 2001, 148 emphasis in original).

These types of regulatory ideals are thus created, sustained or undermined through performance, or to be more exact performativity. Performativity as defined by Butler is not a singular act, "for it is always the reiteration of a norm or a set of norms, and to the extent that it acquires an act-like status in the present, it conceals or dissimulates the conventions of which it is a repetition" (1993, 12). This reiteration is crucial in understanding performativity. It is through repeated action that these norms are created and lived up to. In relation to discourse, Butler argues that performative acts are statements which also produce that which they say. Her classic example is that of the midwife cry of "itís a girl" which is not merely a reflection of a biological given but a performative act, binding a gender onto the body (Butler 1993). In other words it produces that which it names (Butler 1993, 7). Thus performative acts are the one domain in which "discourse acts as power" (Butler 1993, 225). Thus concepts of male and female, of gender are historically and culturally unique. These regulatory ideals are indeed fictions (Haraway 1991). The baby girl is not a girl until the midwife declares her so, and thus curtails the possibility of other genders being created and explored. But we cannot assume that the same categories existed in the past. As Butler herself points out "these regulatory schemas are not timeless structures, but historically revisable criteria of intelligibility which produce and vanquish bodies that matter" (1993, 14). Indeed there is much evidence anthropologically that shows that other genders are possible, further weakening any position which sees the categories of "man" and "woman" as universals. Societies may have "multiple and fluid gender categories" which refuse to be categorised along the simplistic male/female bifurcation of the modern west (Rautman and Talalay 2000, 2). We will now turn briefly to two anthropological examples of alternative gender categories, in order to enhance this argument.

Multiple genders in Southern India and America

A classic account of an alternative to the simple male/female bifurcation of western gender comes from Celia Busbyís (1997) and Serena Nandaís (1990) work in Southern India, particularly their description of a third gender, the hijra. The hijra are predominantly male, in the western view of gender, and have usually gone through a process of ritual castration (Busby 1997, 265). They dress and act as women, though particularly unfeminine women, and actively engage in prostitution (Gilchrist 1999, 60). They also officiate at ceremonies, such as marriages and the birth of male children (Gilchrist 1999, 60). The hijra also include hermaphrodites and women who are unable to reproduce (Gilchrist 1999, 60). They are defined negatively rather than positively, thus they are not part man and part woman, but neither male nor female (Nanda 1990). The crucial factor in becoming hijra is the inability to reproduce, thus impotent men often become hijra. It is not the actuality of reproduction that defines them, because homosexual men may never reproduce, but maintain the potentiality for reproduction, and thus are men not hijra (Nanda 1990, 14). The hijra thus form a third gender within Indian society, performing in their own way, and defined through the inability to reproduce, rather than the absence of sexual organs.

A second example can be drawn from the literature on Native American culture. The existence of a third gender in many Native American tribes was reported in ethnographic literature up until the middle of the 19th Century (Gilchrist 1999, 61). Known as "two-spirit" the individual would take on the clothing, dress and manner of a member of the opposite gender (Gilchrist 1999, 61). Although this was usually a man taking on womenís clothing, in many tribes the reverse also occurred. Due to the practices demise in the mid-19th century their exact status is unclear but they appear to have been both revered and stigmatised by different tribes (Gilchrist 1999, 61). Originally male two-spirits would take up domestic tasks including basket making, whilst female two-spirit would take on the more male dominated areas of hunting and warfare. Although many two-spirits would engage in same-sex relations, they were not defined by this, as others would not (Fulton and Andersen 1992, 608). Here again, performance, dress and action were more important in defining gender than sexuality. Like the hijra, two-spirits also played important roles at certain ceremonies: burial and mourning rituals, assisting wounded, dancing at the return of warriors (Gilchrist 1999, 62). It is clear therefore that the two-spirit also occupied a liminal zone, between man and woman, human and god, living and dead (Gilchrist 1999, 62). These two examples have begun to show the range of genders possible for human beings to take up. There is no limit to the number of different gendered positions available to societies in the past. Multiple genders may have had important roles in the rituals of the Neolithic, just as the hijra do in India and the two-spirit did in Native American culture.

Constructed gender, constructed sex?

The presentation of gender above has offered an account that takes a particular western view; that whilst gender is constructed, sex is a biological constant. It was this that allowed me to discuss male and female hijra, because from this point of view their gender changed when they became hijra but their sex remained the same. Judith Butler has now robustly challenged this point of view, through her work on the body, performativity, gender and difference (1990; 1993). Sex, Butler argues, is just another regulatory ideal, like gender, that we are required to live up to by society (Butler 1993, 1). Far from being a biological given, "sex is an idealised construct that is forcibly materialised through time" (Butler 1993, 1). The maintenance of a simplistic sex=natural: gender=cultural dichotomy, is simply a continuation of the problematic Cartesian duality between nature and culture that has bedevilled archaeology and other social sciences for years (Ingold 2000). Thus Butler wishes to rephrase the question away from "how is gender constituted as and through a certain interpretation of sex?" to "through what regulatory norms is sex itself materialised? And how is it that treating the materiality of sex as a given presupposes and consolidates the normative conditions of its own emergence?" (Butler 1993, 10). The argument that sex is far from a simplistic, natural bifurcation is even supported by biology itself which has shown there are 11 different chromosomal categories of sex, with the traditional man XY and woman XX simply the extremes on both sides (Gilchrist 1999, 57). There have also been cases of sex reversal where a person with XY chromosomes has a female phenotype or XX have a male phenotype (Gilchrist 1999, 57). So even from the simplistic view of the natural sciences, sex cannot be presented as an unthinking universal bifurcation.

Categories of sex and gender are constructed against an outside, against non-viable choices, that secure the boundaries of sex (Butler 1993, 8). The construction of these categories is through exclusion, through abjection, through making some bodies unthinkable (Butler 1993, 188). In the heterosexual hegemony of modern western society, these abjected bodied maybe homosexual, but in other societies, different bodies may lie outside the regimes of power and discourse constructed through performativity.

It is important at this point to emphasise that despite the above discussion one aspect of identity, be it sex or gender cannot be privileged at the expense of another. All areas of identity be it age, race, status, sexuality, sex or gender interrelate with one another. Gender may be understood very differently depending on the age of the subject or the sexuality. We cannot assume either that all of these areas had an affect on the identity of people in the past. Foucault has demonstrated the changing concern with sexuality since classical times, for example (Foucault 1978; 1987; 1988). There may also have been alternative categories of identity that remain hidden from us. However the need for a vocabulary in order to discuss these issues means that despite difficulties surrounding them, these terms will continue to serve as short hand for the various areas of identity we wish to discuss. Despite this it is essential to remember that each area is entirely interdependent on the next, no-one area can be understood without reference to another (Butler 1993, 116).

There have been several critiques of Judith Butlerís work, one of which has come from Bryan Turner (1996). Turner takes age as a test for Butlerís theory of performativity, reiteration and interpellation, the hailing of the subject by particular discourses (cf. Althusser 1971). Turner distinguishes between the social construction of ageing, which he agrees is a variable concept, and the physical, phenomenological experience of ageing (1996, 29). In other words on the one hand, people in different societies may be expected to act differently in social situations depending on their age, and differing amounts if respect may be shown. Thus the elderly who may be treated as respected elders in one society, might be disregarded as weak or infirm in another (Turner 1996, 30). But on the other hand, the physical process of ageing affects everybody in similar ways, through loss of strength and eyesight, greying hair and failing mental capacities. Thus Turner argues we do not need to reject the phenomenological experience of life in order to accept the social construction of different categories.

Judith Butler acknowledges the temptation of such a critique, but at the same time refuses to bow to it. She points out that as soon as one acknowledged the essentialism of certain aspects of the human body, one is drawn back into a discourse in which sex is a purely biological construct (Butler 1993, 10). Furthermore, whilst one might accept the basic phenomenological facts of human existence (pain, pleasure, consumption, excretion) as irrefutable, "their irrefutability in no way implies what it might mean to affirm them and through what discursive means" (Butler 1993, xi). She also questions why these "biological necessities" are privileged over that which is constructed (Butler 1993, xi). Bodies only appear, Butler argues, within the constructions, discourses and constraints of our understanding of the world (1993, xi).

Another critique has come from Elizabeth Grosz, who has questioned the requirement for gender if sex is seen as performative (1995, 212). She argues that gender is now a redundant category, its meaning met by sex on the one hand and sexuality on the other (1995, 213). Grosz argues that Butlerís work is more powerful if one rejects the concept of gender altogether and locates all of the variability of the human body, within sex (1995, 214). I disagree with this argument. I do not agree that the performative nature of sex means that gender is now a defunct concept. The possibilities of performativity leave open scenarios in which gender and sex would not be the same. Although both are produced through the citation and reiteration of regulatory ideals so are all aspects of identity. Ultimately conflating the two reduces our available vocabulary for delineating distinctions between different subjects and between the different concepts they might have of identity. Thomas Laqueur, for example, has demonstrated how in medieval times men and women were perceived as being of the same sex, but of different gender (1990). To be more precise, women were seen as the literal inversion of the male (Turner 1996, 27). Our ability to describe this point of view is diminished if we drop the distinction of sex and gender from our vocabulary, even if it adds more power to description of bodily instability today (cf. Grosz 1995, 212).

One final difficulty with the work of Judith Butler is that it discusses the citation and performance of only one form of personhood, that is only one form of conceptualising human existence; that of the western individual. Other forms of personhood have been shown to exist in anthropological accounts and thus may also have existed in the past. It is to an examination of this that we now turn.

Personhood, individual, partible or combinable?

The prevalent notion of personhood in the west has been that of free, knowledgeable individuals, entire unto themselves. This mode of understanding has often been used to understand other societies, be they current non-western societies or those of the past. This in effect has thrust the modern western notion of personhood unwarranted into the past and onto non-western societies. In archaeology, recent calls to renew the search for individuals (Hodder 1999; 2000; Meskell 2000; 2001) have been criticised from this point of view (e.g. Fowler 2000). The individual is a particular and peculiar way of approaching the world. Recognition of this began with the seminal work of Marilyn Strathern, The Gender of the Gift (1988). In this work Strathern expounds an alternative version of personhood, taken from her ethnography of Melanesia (1988). Principally Strathern argues that people in Melanesia are dividuals rather than individuals (M. Strathern 1988, 13). They are not bounded in the western sense, whole and intact, they are partible, divisible and dividual. The partible nature of people, made up of the social relations in which they reside, means that groups become homologues of the singular and vice versa (M. Strathern 1988, 13). However this is contrasted with a pair, which is not equitable with either the singular or the group (M. Strathern 1988, 14). Both men and women in Melanesia are composed of male and female parts; thus gender differences only emerge under certain conditions of social action (Gosden 1999, 133). As Strathern puts it, "being "male" or "female" emerges as a holistic unitary state under particular circumstances" (1988, 14). One such circumstance is in the pair mode, where the gender duality within each person is rejected and a singular gender emerges (M. Strathern 1988, 15). This contrasts with the group mode where "each male or female form may be regarded as containing within it a suppressed composite identity; it is activated as androgyny transformed" (M. Strathern 1988, 14). Different parts of the Melanesian body can represent different gendered substances, they are in effect a "mosaic of male and female substances" (Busby 1997, 270). This gender duality can even be seen in some objects that take on male or female associations in different contexts. The flute, for example, used in male initiation rites in the Eastern Highlands is seen as a female penis (Gosden 1999, 134). Another example is penile bleeding which can be seen as menstrual (M. Strathern 1988, 126).

Exchange is crucial to this understanding of personhood. Built up by the multiple social relations in which they are engaged, these dividuals are separable into different parts, some of which can be given away through gift exchange (Fowler 2001, 139). In this model of exchange objects do not act as synecdoche for the person, they are understood as being drawn from one person and absorbed by another (M. Strathern 1988, 178). People are enmeshed in social relations, in the flow of substances, gifts and interaction (Fowler 2001, 139). Gifts can be multiply authored and build a biography that only further entwines them in the creation of social relations. They can also be linked to narratives, counter-narratives and morality (Maschio 1998; Hagen 1999).

Strathernís model of personhood in Melanesia offers an alternative, relational account to the traditional model of western personhood, but other alternative accounts have also emerged of personhood in different parts of the world. For example Celia Busby has described how the people in Southern India have a combinable sense of personhood (1997). When men and women marry they are represented by a single body (Busby 1997, 269). More than this however, for them to be fully effective as men and as women they need to go through this combination (Busby 1997, 269). As we have seen above with the hijra, gender in Southern India is closely associated with reproduction. The hijra are defined as neither male nor female because of their inability to reproduce. Men and women thus in contrast are defined as their gender through the act of reproduction, which in turn combines their personhood into a single body (Busby 1997). Indian bodies in this context Busby argues are not partible like Melanesian bodies but permeable (1997, 275). Busby thus points out that although people from India have also been called dividuals, their notion of personhood is demonstrably different from that in Melanesia (1997, 275).

Three different models of personhood thus emerge, the western individual and the two dividual models from Melanesia and India. Chris Fowler has defined these as relations that separate (Melanesia), relations that integrate (Southern India) and relations that alienate (western individual) (2001, 140) (fig. 1). These provide three models for personhood, which can be used to help open up different understandings of the material conditions of life in the past. It should be emphasised however that these models are not mutually exclusive, nor do they present a totality of possibilities for human personhood (Fowler 2001, 140).

Figure 1: Three models of personhood (after Strathern 1988; Busby 1997).

All this seems to have taken us a long way from the un-theorised agents we were left with at the end of the last chapter; what room is there now within these concepts of regulatory ideals and personhood for agency, habitus and practice? In order to answer these questions we must first briefly examine the work of Chris Fowler (2001), an archaeologist who, whilst drawing on Butler and M. Strathern, has steered clear of social theorists like Bourdieu and Giddens. I hope that a brief analysis of a particular study of his will allow me to demonstrate how older concepts of habitus and agency can still offer us important advantages in the study of the past.

Chris Fowler: citation of personhood in the Manx Neolithic

Chris Fowler, drawing on the work of Judith Butler and Marilyn Strathern, has recently offered an engaging and original account of personhood in the Neolithic of the Isle of Man (Fowler 2001). Fowler examines the human remains on the Isle of Man and argues they represent "actions which cite aspects of previous discourses" (2001, 149). The Neolithic on the Isle of Man saw the mixing of human bone, animal bone and material culture (Fowler 2001, 150). At Ballharra, a chambered cairn, an inhumation was found containing body parts from three different, male, skeletons (Fowler 2001, 155). Fowler argues that this deposition "may have been an expression of the regulatory fiction of selfhood; that a person should be integrated with others, because that is how to live; or it may have been a resistance to a different regulatory fiction" (2001, 155-6). Other monuments featured different citations, for example at Knocksharry the deposits placed greater emphasis on separation (Fowler 2001, 157). The activities and performances at monuments on the Isle of Man were, Fowler argues, citations of particular forms of personhood (2001). The forms of personhood cited were very different from that of the modern individual though, and were tied in with ideas around life, death, separation and exchange. Overall Fowler argues that the higher the degree of bodily separation the greater the "immersion or absorption in the material world" (Fowler 2001, 159). In other words, he interprets partibility in death as a sign of partibility in life and thus in personhood. The separation of the corpse acts as a citation of this regulatory ideal (Fowler 2001, 159).

Fowlerís account offers novel insights into the nature of personhood in the Manx Neolithic. There are several difficulties however. Principally, there seems little room for subversion or resistance within his narrative. The act of splitting up a corpse is only citation, it never acts as a creative, subversive or dangerous act. Thus there seems little way of explaining change. Despite the quote above, which includes reference to such resistance, this is downplayed by Fowler, and the subjects described by his work, although different from the knowledgeable agents of Barrett (1994) remain internally undifferentiated. It seems that Fowlerís work, like that of the early Foucault, offers little more than docile bodies to fill his account. This is a problem also identified with Butlerís work (McNay 2000), and thus it is little surprise that it finds it way into Fowlerís analysis when he draws so heavily upon Butlerís concepts of performativity, citation and reiteration (Butler 1993).

From this perspective Judith Butler, along with Foucault and others drawing on poststructural constructivism, have been criticised recently by Lois McNay, who has argued that they approach identity, and crucially agency from a negative paradigm (2000, 2). The negative moment of subjection she argues is privileged (2000, 2). Although Foucault made moves towards a more agency rich position, through his concepts of technologies of the self, this remains unexplored in detail, and remains in McNayís opinion an unsatisfactory account of agency (2000, 8-9). Agency, McNay concedes, is not totally absent from these accounts, but locates itself "mainly through the residual categories of resistance and to or dislocation of dominant norms" (McNay 2000, 3-4). Thus although the work of Foucault and Butler destabilises gender and identity, and offers a new way of conceptualising sex and sexuality, they fail to offer a generative schema for agency and practice. They therefore fail to explain the creation and subversion of categories, norms and fictions in either the past or the present. In order to correct this imbalance, we need to reincorporate an archaeology of practice, with its concomitant theories of agency and habitus (Giddens 1984; Bourdieu 1977).

Bourdieu and Butler, practice and performativity

The similarities between Bourdieu and Butler have been noted by several authors (e.g. Gilchrist 1999, 56; McNay 2000, 40). Both place emphasis on bodily materiality, the way one carries oneself, on corporeality in other words. Lois McNay has argued that the combination of the two allows each to deal with the otherís principal weakness. For Butler, this is the way agency is underplayed in her account through her reliance on a negative paradigm of subjection, and for Bourdieu his failure to deal adequately with the unstable nature of gender and the ways in which agency can emerge from the margins (McNay 2000, 46-56). Bourdieuís account of agency is generative rather than prescriptive. The concept of habitus offers a way of thinking about agency that whilst creative and temporal, is also historically and socially specific (McNay 2000). Due to her reliance on Foucault, Butler remains trapped between competing relations of dominance and resistance, whilst Bourdieu offers a more "nuanced view of political agency" (McNay 2000, 56). Butlerís proposed view of identity however resists the naturalised view of the modern west and allows, in contrast to Bourdieu, the ambiguities and dissonance of performative identity to emerge (McNay 2000, 54). What is needed is thus an understanding of both practice and performativity, of performative practice in fact.

It is insufficient merely to combine these two authors however. We also require the work of Giddens. This is not to reintroduce the a-historical knowledgeable agent, but to consider how actions can have ramifications beyond the life span of a single subject. The concept of time-space distanciation allows Giddens to describe how institutions and groups can allow a single subjectís actions to permeate beyond his immediate surrounds both temporally and spatially (1984, 377). The concepts of both habitus and performativity place too much emphasis on the subject themselves, on their learned action, on their identity and success or failure at living up to different regulatory norms. The agency of subjects can affect otherís conceptions of habitus and identity. Word of mouth for example can spread and affect those who may never have met a particular individual. These consequences may be an unintended outcome, but this makes it no less worth considering. In archaeology, the concepts drawn from Giddens and Bourdieu have allowed John Barrett to offer the view of the archaeological material that I discussed earlier (2000a, 24). That far from being a record, the material is the physical conditions through which life was led. This is an essential point when we begin to consider how different archaeological deposits not only cited different regulatory norms but also created them, and created the conditions through which they were understood. We cannot stop here however. If we are to understand how identity was created in the past, indeed how people lived at all, we must consider other work I have discussed. Conceptions of personhood drawn from Marilyn Strathern and Celia Busby are also essential in preventing the modern western concept of the individual becoming a presumed essentialist universal.

Thus we return to the difficulty of essentialism. Much of what has been written about agency has been accused of being essentialist. What I hope I have indicated here is that, despite this, agency remains an important way of understanding human action in the past. In order to counterbalance accusations of essentialism we need to take on board the nuanced temporal habitus of Bourdieu, and combine it with understandings of personhood and identity drawn from Foucault, Butler and M. Strathern. Despite the essentialist nature of some of his work, Anthony Giddens still has much to offer us. It is only through the combination of multiple theoretical sources that we can deal with each oneís weakness. Too often in archaeology, simplistic polemics are offered declaring the superiority of this philosopher or that social theorist. Instead we require more complex, less mono-causal explanations for why archaeology has failed to deal adequately with the complexity of identity and agency in the past. This may make for less elegant narratives, but that I believe is a small price to pay.

Even the sources outlined above only touch on the range of considerations required. Memory and emotion are also important, as recent archaeological and anthropological sources have shown (Bradley 2002; Riviere 2000; Whittle 2003). Equally so may be concepts of shared values, often underplayed in prehistoric archaeology (Whittle 2001). We also need to consider the nature of human beings interactions and understanding of their environment, of their relations with plants and animals (Ingold 2000; Fairbairn 1999). The nature of thought also needs to be considered. The work of Maurice Bloch has shown how human thought is not linear and narrative in nature (the so called folk theory of thought) but linked through nodes of thought in complex ways (1992; 1998). Thus a consideration of what, in a society "goes without saying" is also essential (Bloch 1992). We need to consider a multiplicity of agencies at a multiplicity of scales, a plurality of genders associated with complex unstable identities. We need to think about how certain practices became unthinking, part of the habitus, but also how these may have been challenged subverted and overthrown. We need to offer temporal and contingent notions of agency, identity and performativity. These notions must be historically specific so accusations of ontological essentialism can be avoided.

The above is offered as some kind of summary for the arguments laid out in detail over the last two chapters. In what follows I shall attempt to take this abstract position and show how it can be applied to the data through an analysis of the creation of identity at the Neolithic causewayed enclosures of Windmill Hill and Etton. But first we must take a moment to paint the background to the creation of these sites by discussing Earlier Neolithic of Southern Britain.