Archaeology and Islamic Identities in Bahrain
Insoll, Timothy, Antiquity
The discussion of identity within Islamic archaeology is rare, and it is often treated as a given, as if Islam was a monolithic category, when in reality it is composed of a variety of identity variables, such as those revolving around sectarian affiliation, ethnicity or gender for example. Ideally, Islam should be structured by 'emphatic ideological denial of the primacy of ethnic identity (and rank)' (Barth 1969: 38), but in reality divisions based on class, caste, ethnicity, occupation and the like exist. These types of identities can be manifest internally within the overall Islamic community, but Islamic identities are also evident in relation to external identity groups as well, most notably those of other religions. Such a dichotomy of labels broadly corresponds to what Jenkins (1994: 200) defines as the difference between 'groups' and 'categories'--the former 'rooted in processes of internal definition' and the latter 'externally defined'.
Essentially the problem can be framed as one of categorisation and classification. The 'rigidity of Western taxonomizing' has recently been critiqued from an archaeological perspective by Meskell (2001: 185) drawing upon similar criticisms made by Foucault (1977), and it could be argued that the label 'Islam' is sometimes used too inflexibly by archaeologists. Which in turn relates to the very notion of what Islam is--what constitutes Islam? This is a question which has been well considered by anthropologists (Asad 1986; Geertz 1968; Gilsenan 1982), but rarely by archaeologists (Insoll 1999). Yet the archaeological record--material culture--can allow the evaluation of the notion of diversity within Islam, the exploration of the creation and maintenance of identities, the extent to which diversity renders the term 'Islams' preferable to 'Islam', and the interplay between 'contextualism' and 'Levi-Straussian universalism' (Asad 1986: 2).
This paper works from the proposition that of course Islam as a recognisable category exists, but within the supra-structure there is great diversity, what has been referred to elsewhere as 'structuring principles' and 'regional diversity' (Insoll 1999:11). With regard to Islamic identities on Bahrain, the overall framework of Muslim belief and practice might offer little room for manoeuvre, but within this structure internal boundaries are constantly being adapted and altered in minor ways during the creation and maintenance of identity relations, some of which might be archaeologically visible. Yet this said, more serious transgression, what might be termed 'heresies' from an orthodox perspective, can also develop, and again potentially these can be explored within the archaeological record. In order to construct a more complete Islamic archaeology, it is perhaps necessary to attempt to include, as far as possible, all the categories of actors, in both their communal and individual manifestations.
Archaeology and identity
Obviously, the archaeology of identity in its various forms is increasingly being considered (see for example Shennan 1989; Meskell 2001; Gilchrist 1999; Casella & Fowler 2005), but rarely is religious identity the focus of attention. Instead the predominance of ethnicity and gender within archaeological discourse on identity is probably more a reflection of the priorities of the scholars themselves rather than necessarily an approximation of past reality (Insoll 2004). Therefore the neglect of identity within Islamic archaeology is far from unique (Insoll 2001). Equally, in archaeological investigations of identity in general the so-called single-issue questions (Meskell 2001: 187) have tended to be the focus of research. The multiple strands of identity are rarely considered together.
These multiple strands can take many forms as can be seen from contemporary observation, for numerous permutations of identity are possible in Bahrain and the wider Arabian Gulf today based upon region, nationality, tribe, clan, religion, gender, profession, class, ethnicity, language and other variables (Fuller & Francke 1999: 9). …