Creating urban communities at Kilwa Kisiwani, Tanzania, AD 800-1300.

Creating urban communities at Kilwa Kisiwani, Tanzania, AD 800-1300. Introduction The archaeology of the eastern coast of Africa, and of what isknown as the 'Swahili' culture, has long been a topic ofacademic enquiry. Until recently, debate has focused on the'origins' of this specifically coastal culture and theethnicity of the inhabitants. This debate has been rightlysuperseded--in an archaeological literature increasingly unwilling tosubscribe to Verb 1. subscribe to - receive or obtain regularly; "We take the Times every day"subscribe, takebuy, purchase - obtain by purchase; acquire by means of a financial transaction; "The family purchased a new car"; "The conglomerate acquired a new company"; narrow conceptions of ethnicity--by a focus on process, andit is now necessary to reconsider the material culture associated withthese towns in the light of recent theoretical debate. Rather thanviewing the various artefacts as passive reflections of particularethnicities or social groups, they may be seen as active features of adynamic society. This paper examines coastal material culture in thelight of such insights--concentrating in particular on the role ofmaterial culture in the creation and consolidation of certain types ofauthority. The archaeology of the 'Swahili coast' The elites of the East African Adj. 1. East African - of or relating to or located in East Africa coast inhabited the stonetowns (sonamed for the coral-built architecture they contained) that grew therefrom the tenth century AD onwards, building their local political andeconomic dominance on the flourishing trade network that connected themto the wider Indian Ocean world. Towns such as Kilwa Kisiwani Kilwa Kisiwani is an Islamic community on an island off the coast of East Africa, in present day Tanzania.In the 9th century it was sold to a trader Ali bin Al-Hasan and over the following centuries it grew to be a major city and trading centre along that coast, and inland , on thesouthern coast of Tanzania (Figure 1), enjoyed a privileged andprestigious position in that trade, as the middle-men between the richesof the African continent and the exotic luxuries of the Arabianpeninsula Arabian Peninsulaor ArabiaPeninsular region, southwest Asia. With its offshore islands, it covers about 1 million sq mi (2.6 million sq km). Constituent countries are Bahrain, Kuwait, Oman, Qatar, United Arab Emirates, Yemen, and, the largest, Saudi Arabia. , the Indian subcontinent sub��con��ti��nent?n.1. A large landmass, such as India, that is part of a continent but is considered either geographically or politically as an independent entity.2. and the Far East. Gold, ivory,ambergris ambergris(ăm`bərgrēs), waxlike substance originating as a morbid concretion in the intestine of the sperm whale. Lighter than water, it is found floating on tropical seas or cast up on the shore in yellow, gray, black, or variegated , iron, timber and slaves passed through the towns in largequantities, to be exchanged for luxury imports such as fine silks andfabrics or glazed and decorated ceramics. Ideas also travelled the tradenetworks, most obviously visible in the spread of Islam This article is about followers of the Islamic faith. For territories under Muslim rule, see Muslim conquests. The spread of Islam began shortly after Muhammad's death in 632. to the EastAfrican towns, some of which contained a Muslim population from theninth century (Horton 1996) and all of which became Islamic within thesucceeding centuries, a tradition that persists to this day. The modernoccupants of coastal Kenya and Tanzania continue to define themselves bythis religion, which sets them apart from their compatriots furtherinland. Although the coast has since been subject to Portuguese, Omaniand British domination, a distinctive coastal culture remains. In fact,the recent centuries have cemented coastal identities into what is todayknown as the 'Swahili' culture, a nomenclature that is oftenextended into the past, to describe the people who previously inhabitedthis stretch of land and traded in the waters of the Indian Ocean. [FIGURE 1 OMITTED] The specifically coastal distribution of the sites associated withthis culture is striking, as they cling to Verb 1. cling to - hold firmly, usually with one's hands; "She clutched my arm when she got scared"hold close, hold tight, clutchhold, take hold - have or hold in one's hands or grip; "Hold this bowl for a moment, please"; "A crazy idea took hold of the East African littoral littoral/lit��to��ral/ (lit��ah-r'l) pertaining to the shore of a large body of water. littoralpertaining to the shore. andoff-shore islands. This geography has affected the way that wider EastAfrican environments are understood by coastal archaeologists andhistorians, as they developed certain conventions for describingdifferent areas (see for example Horton & Middleton 2000: 8). Thus,the term 'hinterland' is normally used to refer to the regionimmediately inland from the towns, extending perhaps 50km from theocean. Recent archaeologies have tended to view this area as part of thewider society associated with the towns (see below). In contrast, theterm 'interior' has come to refer to the wider East Africanregion, with which the towns were trading, but were probably not indaily contact. Archaeological research into the origins, development andsocio-linguistic composition of the stonetowns has traditionallysubscribed to one of two modes of thought--one locating the origins ofthe urban population in the Arabian world, and the other emphasising theAfrican roots of the Swahili. In general, the early claims for externalorigins are countered by the more nationalist archaeologies of recentdecades. Each view has emphasised different forms of evidence, with thefew indigenous histories that relate to the coast in the first half ofthe second millennium AD providing much of the inspiration for theclaims of Arab immigration immigration,entrance of a person (an alien) into a new country for the purpose of establishing permanent residence. Motives for immigration, like those for migration generally, are often economic, although religious or political factors may be very important. (Kirkman Kirk´mann. 1. A clergyman or officer in a kirk.2. A member of the Church of Scotland, as distinguished from a member of another communion. 1964; Chittick 1975), and a growingknowledge of the local archaeological sequence The archaeological sequence or sequence for short, on a specific archaeological site can be defined on two levels of rigour. Normally it is adequate to equate it to archaeological record. However, the two terms are not exactly interchangeable. providing thecounter-claim. The histories, known as 'chronicles', provide foundationcharters for the stonetowns and genealogies for their rulingelites--often written long after the events to which they claim torelate. The best known, the Kilwa Chronicle, tells the story of aPersian prince, Ali bin al-Hasan, who bought Kilwa island from a localking and settled there (Freeman-Grenville 1962: 34-49, 89-93, 220-26).The subsequent rulers of Kilwa Kisiwani drew upon this'Shirazi' heritage as an important legitimisation of theirauthority (Sutton 1998: 124). The story also emphasises the splitbetween the town and its hinterland, aligning the Shirazi population ofKilwa island more closely with comparable settlers in other coastaltowns or with their foreign homeland; this claim to a separate coastalidentity was a notable feature of the various town chronicles (Pouwels1987: 30), reinforced by the urban acceptance of Islam and reflected inthe rhetoric recorded by visitors such as Ibn Battuta Abu Abdullah Muhammad Ibn Battuta (Arabic: أبو عبد الله محمد ابن بطوطةin 1331(Freeman-Grenville 1962: 27-32). Yet, as the widespread similarities between aspects of thestonetown assemblages and those of their hinterlands began to besystematically described, a new generation of scholars recognised thestonetowns as a development of the African continent (Horton 1984; 1996;Abungu 1989; Chami 1994) and began to regard the histories assemi-mythical manifestos created by later rulers. These researchersexplicitly sought alternative voices in the archaeological record The archaeological record is a term used in archaeology to denote all archaeological evidence, including the physical remains of past human activities which archaeologists seek out and record in an attempt to analyze and reconstruct the past. : thoseof the African majority that had always existed in and around thedeveloping centres. They formed part of a wider movement that recognisedthe Bantu roots of the Swahili language Swahili language,member of the Bantu group of African languages (see African languages and Bantu languages). Swahili is spoken by 30 million people, chiefly in Tanzania, Kenya, Congo (Kinshasa), Burundi, and Uganda, and serves as a lingua franca for additional (Nurse & Spear 1985) and thesimilarities between the contemporary Swahili and hinterland groups suchas the Mijikenda, who occupy the hinterland of the Kenya coast (Allen1981; 1993). Archaeologically, arguments for African ethnicity rested onthe record of locally produced ceramics and a class of pottery known asTana Ware (Phillipson 1979; Chami 1994; Horton 1996), found across theearly layers of the coastal stonetowns and also in many contemporarysites of the hinterland and interior. These ceramics indicated the linksbetween the urban households and their continental neighbours, tyingthem into a wider material tradition. Thus, the towns began to beunderstood as an African phenomenon; with clear Arabic influences, butwith Africa providing the roots and the majority of the population. This brief outline does not do justice to the subtleties of eitherargument and to some extent revisits debates long since settled.However, an appreciation of the history of scholarship here is key tothe current situation, in which researchers stress the interplay oflocal and foreign influences in the creation of a distinctive coastalculture (e.g. Horton 1987). It explains the persistence of argumentssurrounding external influences on the stonetowns, as well as thereluctance of contemporary researchers to acknowledge tooenthusiastically the 'foreign' elements of coastal society.Recent scholarship has stressed 'process', and has beendeliberately oriented towards the inclusion of the African majority inthe Swahili world, rather than exclusively studying the'patricians' who created and used the elements of materialculture previously thought to typify the society. Contemporary coastal scholarship thus recognises a complex web ofintentional and unintentional causation for the observable distributionof material culture, both locally made and imported. There remains atendency to think about coastal culture in terms of dichotomies:foreign/local has been replaced by elite/non-elite or town/country, andcertain types of artefact See artifact. are thought representative of the differentgroups (although see Fleisher 2004 for a consideration of the activerole of material culture). However, a recent survey in the region ofKilwa Kisiwani--a prominent stonetown of the tenth to sixteenthcenturies--suggests that objects did not just passively reflectdifferent peoples or activities, but at the time of their use were alsoconstitutive constitutive/con��sti��tu��tive/ (kon-stich��u-tiv) produced constantly or in fixed amounts, regardless of environmental conditions or demand. of various social relationships. Of particular interest inthis regard is the creation of an elite group in coastal society, whofor many were the Swahili (Chittick 1975; Donley 1982; Middleton 1992),and in the creation of which material culture appears to have beenespecially important. The importance of material symbols in the creationof status is also a topic of contemporary concern for archaeologistsmore generally. Town and hinterland The growth of the coastal towns from the ninth century onwards iswell-documented (see Wright 1993 for a useful summary) and relativelythoroughly understood. Various stonetown sites have been excavated andtheir developmental histories plotted; those known in particular detailinclude Kilwa Kisiwani itself (Chittick 1974), Ungwana (Kirkman 1966;Abungu 1988), Man& (Chittick 1984) and Shanga (Horton 1996) on theKenya coast and Unguja Ukuu in the Zanzibar archipelago The Zanzibar Archipelago consists of several islands lying off the east coast of Africa in the Indian Ocean:[1] Unguja Island Pemba Island Prison Island Bawe Island Chapwani Island Chumbe Island Mnemba Island Misali Island (Juma 2004).Numerous other sites are known and have been studied, but thosementioned here are the earlier sites for which thorough stratigraphic stra��tig��ra��phy?n.The study of rock strata, especially the distribution, deposition, and age of sedimentary rocks.strat information is available. In general, these towns grew dramatically insize between the ninth and thirteenth centuries (often from smallvillage sites of earlier centuries), with increasing evidence foroverseas trade accompanying the elaboration of urban architecture toinclude coral-built structures. Kilwa Kisiwani grew to coverapproximately 1[km.sup.2] over the course of these centuries, and thefirst coral building was the Great Mosque (Figure 2), apparently begunin the eleventh century although later greatly extended (Chittick 1974:61-99). Various coral buildings followed, mainly of a monumental nature,including the ambitious palace of Husuni Kubwa The Palace of Husuni Kubwa is a ruined structure on the island of Kilwa Kisiwani, in Tanzania. The majority of the palace was erected in the 14th century by Sultan al-Hasan ibn Sulaiman, who also built an extension to the nearby Great Mosque of Kilwa, although portions may date and the eventual use ofcoral in the construction of private residences--both these developmentsdate to around the fourteenth century. In the more northerly towns, suchas Shanga, a greater proportion of the population seem to have beenhoused in stone, although in both areas the majority would always havebeen living in houses built of wattle-and-daub. [FIGURE 2 OMITTED] Although these intra-site developments remain of interest, theurban hinterland is increasingly forming the subject of investigation,following the practices of settlement archaeology developed elsewhere(among others, Willey 1953), as it is only by understanding theexistence of a site within its surroundings that its history anddevelopment can be fully appreciated. Thus the survey reported here wasconceived as a contextual analysis of the developing urban centre ofKilwa Kisiwani, based on an understanding of changing settlement in awider region. The Kilwa Kisiwani hinterland survey Survey was conducted across the hinterland of Kilwa Kisiwani, inthis case defined as an area of 30 x 30km, forming the immediatesurroundings of the town (Figure 3). Foot survey and surface collectionwere accompanied by the excavation of several test pits to assesssub-surface remains and conditions of preservation. Overall, 66 siteswere located across the region, on a sample of around 5 per cent of thetotal area (see Wynne-Jones 2005b for a full description of methodologyand results). Among these sites, the most striking feature observed wasthe long-term continuity across the region. The distribution of sitesremained remarkably constant, with a general spread of sites acrosscoastal and inland locations, exploiting a range of environments. Thenumbers of sites across all periods were fairly similar, with no obviouschanges accompanying the growth of the town. Settlement type also seemsto have been stable, as almost all sites were associated with theremains of wattle-and-daub architecture, which formed a relativelyephemeral imprint upon the landscape. This impermanence im��per��ma��nent?adj.Not lasting or durable; not permanent.im��perma��nence, im��per was an importantpart of the settlement type recorded. Shallow deposits at all sitestestified to the brevity of their occupation, which may have beendependent on shifting agricultural practices, such as are seen in theregion today (Wynne-Jones 2005a). [FIGURE 3 OMITTED] Thus it appears that the process of urbanisation in the Kilwahinterland did not entail a dramatic change in the settlement of theregion. Indeed, even a few kilometres outside the town the regionalpopulation appear to have occupied the landscape much as they hadalready done for some centuries. Kilwa Kisiwani began as one of theseephemeral settlements; the early layers of the town excavations containthe remains of a small settlement based upon impermanent im��per��ma��nent?adj.Not lasting or durable; not permanent.im��perma��nence, im��per architecture(Chittick 1974: 27-28). Yet between AD 800 and 1300 this one settlementdeveloped into the principal town on the coast. What were the changesthat signalled this transition? The advent of building in coral (seeabove) is the most immediately obvious difference that occurred in thetown, and one that separates it from the sites of the hinterland.However, the majority of the coral architecture dates to the fourteenthcentury and later, by which time Kilwa's status as an importanttown was already established. In the preceding centuries, while the townwas developing, buildings were of wattle-and-daub, or later of coral rag Coral rag is a rubbly limestone composed of ancient coral reef material. The term also refers to the building blocks quarried from these strata which are an important local building material in areas like the east African coast. bound with mud mortar. It is therefore difficult to identify the process of urbanisationwith major change in the settlement of the region, as is often dictatedby models of urban development and, indeed, as established elsewhere onthe coast (Fleisher 2003). The hinterland distribution changed littleand the stone architecture largely post-dated the development of'town'. It seems that the main changes occurring in the Kilwaregion between the ninth and the fourteenth centuries were social ratherthan spatial. The development of the trading elite was the decisivefactor Noun 1. decisive factor - a point or fact or remark that settles something conclusivelyclinchercausal factor, determinant, determining factor, determinative, determiner - a determining or causal element or factor; "education is an important determinant of in these changes, and is also the main process that can be seenarchaeologically, through the evidence of portable material culture. The ceramic record The ceramic record is particularly useful, as this is by far themost plentiful class of artefact and the one in which the most obviouspatterning is apparent. The pottery can be divided into two groups withparticular distributions, either regional or coastal. It will be notedthat the coastal group includes both locally produced andforeign-derived ceramics: it is felt that in this instance thedistribution is more pertinent than the derivation. All categories of ceramic that can be said to have a regionaldistribution (i.e. to be found in assemblages across the Kilwa region)are locally produced. Such production is likely to have occurred withinthe region, although great similarity is evident with other assemblagesalong the length of the coast. Nonetheless, particularities of localclay types, which are pinkish in colour, imply a restricteddistribution. All types recognised in the hinterland were also found in the urbanlevels at Kilwa Kisiwani, where they were identified as 'KitchenWares' by Chittick (1974: 317-94), who referred to their undoubted un��doubt��ed?adj.Accepted as beyond question; undisputed. See Synonyms at authentic.un��doubted��ly adv. use as cooking vessels. Analogies with modern vessels suggest, however,that such vessels would also be used for serving and communalconsumption, so they should not be thought of as restricted to thekitchen area. The development of forms between the ninth to thirteenthcentury all belongs to the broad Tana Ware grouping (Figure 4) withincised incised/in��cised/ (in-sizd��) cut; made by cutting. cross-hatched and triangular decoration, which was also the maincategory to which urban examples belonged. The changes in this ceramictype over time that are seen at Kilwa Kisiwani are also mirrored in thehinterland, with a growth in the numbers of bowl forms and a decrease inthe decorated cooking vessels. Throughout the development of the town,the similarity with ceramics from the wider region is notable. It isdifficult to compare numbers, as the locally produced assemblage fromKilwa Kisiwani is not quantified, having been of less interest to theexcavator ex��ca��va��torn.An instrument, such as a sharp spoon or curette, used in scraping out pathological tissue.excavator (eks´k than the imported wares, but they seem to have constituted thevast majority of the collection in the town, as in the hinterland.Chittick reports a number of the order of one million (1974:317). [FIGURE 4 OMITTED] Yet it also seems that there may have been some proportionaldifferences. The ceramics identified by Chittick as 'HusuniModelled Ware' and dated to the fourteenth century, due to theirassociation with the palace of Husuni Kubwa, can be said to have aregional distribution, and are not known from any other coastallocations. They are decorated with moulded projections, among which bowlforms predominate (Figure 4). Although they are found across thehinterland, numbers are very restricted and they are significantlyoutnumbered by necked jars of incised decoration which appear to be acontinuation of earlier types. This contrasts with the town, where bothtypes were present, but with the opposite proportions, with HusuniModelled Ware outnumbering the incised wares. Nevertheless, many of the locally produced ceramics have adecidedly regional distribution, without any vast differences observablebetween town and country. This has implications regarding aspects ofshared identity as reflected in production and in consumption practices.This is particularly true for these types of ceramics, associated withthe rituals of daily life and with domestic cooking and consumption. Anumber of authors have stressed the link between the practices ofconsumption and group identity, as it is created and recreated throughthe rituals of daily life (Goody 1982; McKee 1987; Emberling 1997). Bothat the level of production and consumption, there were similar practicesoccurringacross the region, indicating shared norms and accepted practice withregard to style and function, as well as shared tastes, between urbanand rural populations. There were, however, types of ceramics associated with the town ofKilwa Kisiwani which were not part of the broader regional distributionand these are of interest in the context of what was clearly a societythat shared a regionally understood material language. Among the locallyproduced ceramics it has already been noted that the modelled bowl formswere rare in the rural areas. Another form that was more associated withthe town than with the countryside was a group of red-burnished types,occurring from the ninth century levels onwards, of which again bowlswere the most common vessel form. Chittick regarded these as the'most characteristic ceramic for the early period for the wholecoast south of Lamu' (1974: 323), but he was referring of coursespecifically to stonetown sites. Although his judgement does not seem tohave been based on an overall majority of these ceramics (the'Kitchen Wares' remained the most numerous), there wereclearly significant numbers of red-burnished wares recovered from theexcavations. They are also widespread from other coastal excavations,and are particularly common on Pemba (Fleisher 2003: 248-50). Incontrast, the Kilwa hinterland assemblages contained very few burnished bur��nish?tr.v. bur��nished, bur��nish��ing, bur��nish��es1. To make smooth or glossy by or as if by rubbing; polish.2. To rub with a tool that serves especially to smooth or polish.n. ceramics, with only six sites yielding any sherds at all and theproportion hovering around 1 per cent. It seems, then, thatred-burnished wares had a different type of distribution to the unglazedcooking vessels--a coastal rather than a regional pattern. Coastal distribution patterns are even more marked for the classesof imported wares known from Kilwa Kisiwani, which included many typesfrom the Arabian peninsula and China, with sgraffiato and Islamicmonochrome ceramics in the majority. Regional assemblages had an almostcomplete lack of this type of ceramic. Only a few sherds of importedtypes, adding up to no more than ten in total, were recovered from thesites of the hinterland. Those that were found were from the largersites--often the same that contained the red-burnished and modelledwares. Thus in contrast with the broad similarities noted in the locallyproduced assemblages, the imported goods seem to have been deliberatelyrestricted to the town. Here, their importance was largely symbolic. The use of imported materials for display purposes, in order tocreate and signify status, is a practice recorded among recent coastalpopulations (Donley 1982). In the Swahili house, imported ceramicsfulfil a number of important symbolic roles and are frequently used fordisplay, inhabiting wall niches designed for this purpose (Donley-Reid1990). In the past, also, there is some evidence that imports may havebeen displayed. For example, Ibn Battuta mentions the fine fabricsimported by the merchants of Mogadishu and displayed prominently there(Freeman-Grenville 1962: 29-30). At Kilwa, he comments on the differencebetween the imported silk robes of the elite and those of the othertownspeople (Freeman-Grenville 1962: 32). By the fifteenth century, theuse of imported ceramics as decoration around the mihrab mihrabArabic mihrabSemicircular prayer niche in the qiblah wall (the wall facing Mecca) of a mosque, reserved for the prayer leader (imam). The mihrab originated in the reign of the Umayyad caliph al-Walid I (705–715), when the famous mosques at of coral-builtmosques was common practice along the coast (Garlake 1966: 62-3). Theniches in the walls of earlier buildings may have held ceramics for asimilar purpose. Imported ceramics may also have had a role in feastingor conspicuous consumption conspicuous consumptionn.The acquisition and display of expensive items to attract attention to one's wealth or to suggest that one is wealthy.Noun 1. (Kusimba 1999; Fleisher 2003)--a point thatis explored below. Certainly, the very fact that coastal tradersimported quantities of 'prestige' luxury goods, exchangingthem for commodities of quantifiable value such as gold or ivory,implies that the imports had a culturally determined value in coastalsociety. This was obviously sufficient to ensure that the goods were nottraded onwards, allowing them to become part of the hinterlandassemblages, but were retained within the town. This contradicts thepicture of shared material culture and continual town-countryinteraction painted by the regional wares. Discussion: the creation of an elite at Kilwa The patterns outlined above suggest that the most fundamentalchange involved in the creation of a town at Kilwa Kisiwani was thedevelopment of an elite during the centuries between AD 800 and 1300. Asdiscussed, the settlement of the region remained relatively stable, withthe majority of the population continuing to live in impermanentarchitecture, on sites that were only occupied for short periods. Itseems likely that this pattern of occupation was based upon shiftingagricultural practices, as it is today. The locally produced ceramicrecord ties the town of Kilwa Kisiwani into the occupation of the regionas a whole, despite the differences that existed between town andcountryside. It implies that people were producing and using the samearticles of daily existence, and that consumption practices were similaracross the urban/rural divide. Use of these ceramics was also notrestricted to the non-elite sections of the urban population, as thelocally produced ceramics are found in association with imports, andcome from palaces and from wattle-and-daub houses alike. Thus, the maindifference between town and country that is visible archaeologically isto be found in the differential access to the imported, prestige goodsthat seem to have become a marker of status. This has long beenrecognised, as part of the development of an elite that wielded powerthrough a network strategy based on differential access to wealth(Blanton et a/. 1996). However, what has been inadequately examined isthe way in which that prestigious material culture was active in thecreation of that elite, rather than simply being evidence for theirascendancy. Through the use of a powerful material culture, derived from theworld of overseas trade, the trading elite of Kilwa created an urbanityfor themselves, based around material forms with a symbolic link In Unix, a file that points to another file or directory. It is used to allow a variety of sources to point to a common destination. The Windows 2000 counterpart is the "virtual directory." When URLs are redirected, it is called "URL mapping. tocivilisation elsewhere. Gosden (2004) has explained this type ofrelationship as a form of colonialism, through which material culturewith a symbolic centre in a particular place becomes associated withelites across a wider area. This, he feels, creates a form of colonialpower for the region from whence the material stems, as well asproviding a powerful material language to be drawn upon by the'colonised'. The coastal traders certainly manipulated theirstatus through the use of exotic material culture, particularly from theArab world, and elite status was tied very firmly to that world, as hasbeen discussed for the claims to Shirazi heritage. The role of materialculture seems to have been fundamental in the creation of urbanidentities, as the objects acted as 'constitutive symbols'(Renfrew 2001: 130) with an active symbolic role of their own. Thus theconcept of urbanity and the symbols of that urbanity were interlinked ina process by which both were created or 'substantialised'. Thecultural forms adopted by the elite were therefore shaped by thematerial culture that had become symbolic of their status. The ways that the imported material culture was used by theemergent elite no doubt varied considerably. Donley-Reid (1990) hasdiscussed how imported ceramics have such powerful symbolism that theyare used inside the Swahili house, particularly in private areas wherethey can safeguard the purity of the household. In such cases it isclear that the power of the objects is not linked to overt display,although their ownership confers and implies status. Yet the display ofimported ceramics in mosques was clearly intended for an audience.Likewise, it seems likely that in the past imported ceramics may havebeen part of feasting or public consumption practices. The developmentof such practices is also suggested by the changes in locally producedceramic assemblages, which from the twelfth century onwards contain apreponderance of bowl forms, a change which is seen along the coast andhas been interpreted as representative of a shift in consumptionpatterns, related to a developing elite and perhaps as part of acorporate power strategy that stressed the redistribution of goodsthrough public generosity (Kusimba 1999; Fleisher 2003). In such asituation, the prestigious objects convey status onto their users,acting as a form of redistribution. The distribution of some of the rarer forms of locally producedceramics implies that they might have had a similar role among thehinterland populations. The quantity of red-burnished and modelled bowlsis small outside the town, and they are significantly outnumbered by thecooking vessels with incised decoration, even after the latter haveceased to be the most numerous form inside the town. Perhaps theconsumption practices heralded by the new vessel forms were onlyparticipated in on a limited scale. Although the imported goods may nothave spread outside the town, similar power strategies may have beenpursued among the hinterland populations, using a different materiallanguage. This is an aspect that requires further consideration, as therole of objects and the value ascribed to them may have been differentin non-coastal contexts. The power of the imported objects was basedupon a culturally determined system of value, in which association withthe Arab/Islamic world is known to have been linked to prestige. Adifferent value system may have been in operation inland, making theimports less attractive than other goods. Conclusion This paper has attempted, drawing on data from the Kilwahinterland, to show how urban identities were created and maintainedthrough the appropriation and use of particular kinds of materialculture. It takes as its starting point Noun 1. starting point - earliest limiting pointterminus a quocommencement, get-go, offset, outset, showtime, starting time, beginning, start, kickoff, first - the time at which something is supposed to begin; "they got an early start"; "she knew from the the notion that material cultureis active in shaping and negotiating social realities, and has arguedthat the development of an elite at Kilwa was to some extent a functionof their use of exotic material culture. It also seems that thedevelopment of a system of prestige, based on the appropriation of suchpowerful symbols, was a key feature of the development of what werecognise as a 'town' at Kilwa. This contrasts with thetendency to think of urbanism as a fundamental shift in settlement form.The materialist argument could be extended to setting, followingDeMarrais (2004), and the stone architecture that developed at Kilwa, aswell as the practices of day-to-day interaction there, can be seen asconstitutive of the social order. It is hoped that by this means it ispossible to understand the process of urbanisation as the ascendancy ofan elite, created by, as much as creating, the settings and symbols ofpower. This approach to material culture allows archaeologists to breakaway from the narrow definitions of foreign and local aspects ofsociety, as it is clear that ceramics, and material culture moregenerally, cannot be equated with ethnicity or identity. Instead,attention can be shifted to an analysis of how objects are used and howthey in turn use people, as aspects of local and foreign identity aredrawn upon to create the structure of society. Acknowledgements Thanks are due to Professor David Phillipson for his help with theformulation of these ideas, and also to my colleagues David Robinson David Robinson or Dave Robinson is a name shared by the following individuals: David Robinson (philanthropist) (1904-1987), British entrepreneur, philanthropist and owner of racing stables who was knighted in 1985 andSusanne Hakenbeck for their comments on this paper. The fieldwork atKilwa was funded by the Arts and Humanities Research Board, BritishInstitute in Eastern Africa, Smuts Memorial Fund, H.M. Chadwick Fund,Ridgeway A ridgeway is a road or path that follows the highest part of the landscape. Roads and pathwaysOne of the best known ridgeways is the Ridgeway National Trail, also known as The Ridgeway Path Venn Travel Fund, Tweedie Exploration Fund, UAC of NigeriaFund, Churchill College and the Department of Archaeology, Cambridge. Received: 24 May 2005; Accepted: 27 February 2006; Revised: 15November 2006 References ABUNGU, G.H.O. 1988. Ungwana on the Tana, in P. Sinclair & S.Wandibba (ed.) Urban Origins in Eastern Africa, project proposals andworkshop summaries. Stockholm: The Swedish Central Board of Antiquities. --1989. Communities on the River Tana, Kenya: An ArchaeologicalStudy of Relations between the Delta and the River Basin AD 700-1890.Unpublished PhD thesis, University of Cambridge. ALLEN, J.D.V. 1981. Swahili culture Thousands of years ago the eastern coast of Africa was seen as highly valuable commercial land. Therefore many Muslims and Bantu speaking people migrated and settled along the African coast. Due to this combination of peoples a unique language and culture arose on the eastern coast. and the nature of East coastsettlement. International Journal of African Historical Studies 14(2):306-34. --1993. Swahili origins: Swahili culture and the Shungwayaphenomenon. London: James Currey. BLANTON, R.E., G. FEINMAN, S. KOWALEWSKI & P. PEREGRINE. 1996.A dual-processual theory for the evolution of Meso-Americancivilization. Current Anthropology Current Anthropology, published by the University of Chicago Press and sponsored by the Wenner-Gren Foundation for Anthropological Research, is a peer-reviewed journal founded in 1959 by the anthropologist Sol Tax (1907-1995). 37 (1): 1-14. CHAMI, F. 1994. The Tanzanian Coast in the Early First MillenniumAD: an archaeology of the iron-working, farming communities. Uppsala:Societas Archaeologica Uppsaliensis. CHITTICK, H.N. 1974. Kilwa: an Islamic trading city on the EastAfrican coast. Nairobi: British Institute in Eastern Africa. --1975. The Peopling of the East African coast, in H.N. Chittick& R.I. Rotberg (ed.) East Africa and the Orient: Cultural Synthesesin Pre-Colanial Times: 16-43. New York New York, state, United StatesNew York,Middle Atlantic state of the United States. It is bordered by Vermont, Massachusetts, Connecticut, and the Atlantic Ocean (E), New Jersey and Pennsylvania (S), Lakes Erie and Ontario and the Canadian province of & London: Africana PublishingCompany. --1984. Manda: Excavations at an Island Port on the Kenya Coast.Nairobi: British Institute in Eastern Africa. DEMARRAIS, E. 2004. The Materialization of Culture, in E.DeMarrais, C. Gosden & C. Renfrew (ed.) Rethinking materiality MATERIALITY. That which is important; that which is not merely of form but of substance. 2. When a bill for discovery has been filed, for example, the defendant must answer every material fact which is charged in the bill, and the test in these cases seems to theengagement of mind with the material world: 11-22. Cambridge: McDonaldInstitute for Archaeological Research The McDonald Institute for Archaeological Research is a research institute of the University of Cambridge in England. HistoryThe Institute was established in 1990 through a generous benefaction from the late Dr D. M. McDonald, a well-known and successful industrialist. . DONLEY, L. 1982. House Power; Swahili Space and Symbolic Markers,in I. Hodder (ed.) Symbolic and Structural Archaeology: 63-73.Cambridge: Cambridge University Press Cambridge University Press (known colloquially as CUP) is a publisher given a Royal Charter by Henry VIII in 1534, and one of the two privileged presses (the other being Oxford University Press). . DONLEY-REID, L.W. 1990. The Power of Swahili Porcelain, Beads andPottery. Archeological Papers of the American AnthropologicalAssociation 2 (1): 47-59. EMBERLING, G. 1997. Ethnicity in complex societies: archaeologicalperspectives. Journal of Archaeological Research 5: 295-344. FLEISHER, J. 2003. Viewing Stonetowns from the Countryside: AnArchaeological Approach to Swahili Regional Systems, AD 800-1500.Unpublished PhD thesis, University of Virginia. --2004. Behind the Sultan of Kilwa's 'RebelliousConduct': Local Perspectives on an International East African Town,in A. Reid & P. Lane (ed.) African Historical Archaeologies: 91-124.New York: Kluwer Academic/Plenum. FREEMAN-GRENVILLE, G.S.P. 1962. The East African Coast. SelectDocuments from the First to the Earlier Nineteenth Centuries. London:Clarendon Press. GARLAKE, P. 1966. The Early Islamic Architecture of the EastAfrican Coast. Nairobi & Oxford: British Institute in EasternAfrica. GOODY, J. 1982. Cooking, Cuisine and Class. A study in comparativesociology Comparative sociology generally refers to sociological analysis that involves comparison of social processes between nation-states, or across different types of society (for example capitalist and socialist). . Cambridge: Cambridge University Press. GOSDEN, C. 2004. Archaeology and colonialism: cultural contact from5000 BC to the present. Cambridge: Cambridge University Press. HORTON, M.C. 1984. The Early Settlement of the Northern KenyaCoast. Unpublished PhD thesis, University of Cambridge. --1987. The Swahili Corridor. Scientific American Scientific AmericanU.S. monthly magazine interpreting scientific developments to lay readers. It was founded in 1845 as a newspaper describing new inventions. By 1853 its circulation had reached 30,000 and it was reporting on various sciences, such as astronomy and 257(Sep): 86-93. --1996. Shanga: The archaeology of a Muslim trading community onthe coast of East Africa. Nairobi: British Institute in Eastern Africa. HORTON, M.C. & J. MIDDLETON. 2000. The Swahili: The SocialLandscape of a Mercantile Society. Oxford: Blackwell. JUMA, A. 2004. Unguja Ukuu on Zanzibar: An archaeological study ofearly urbanism. Uppsala: Societas Archaeologica Upsaliensis. KIRKMAN, J.S. 1964. Men and monuments on the East African coast.London: Lutterworth. --1966. Ungwana on the Tana. The Hague: Mouton moutonlamb pelt made to resemble seal or beaver. & Co. KUSIMBA, C.M. 1999. The Rise and Fall of Swahili States. WalnutCreek Walnut Creek,residential city (1990 pop. 60,569), Contra Costa co., W Calif., in the San Francisco Bay area; inc. 1914. It is the trade and shipping center of an extensive agricultural area where walnuts are among the major product. : Altamira. MCKEE, L. 1987. Food supply and plantation social order: anarchaeological perspective, in T.A. Singleton (ed.) I, Too, am America AM America was a morning news program that started in 1975 with Bill Beutel and Stephanie Edwards as co-hosts. Peter Jennings gave news reports throughout the program. British comedy troupe Monty Python made one of their earliest American television appearances on :218-39. Charlottesville: University Press of Virginia. MIDDLETON, J. 1992. The World of the Swahili. New Haven New Haven,city (1990 pop. 130,474), New Haven co., S Conn., a port of entry where the Quinnipiac and other small rivers enter Long Island Sound; inc. 1784. Firearms and ammunition, clocks and watches, tools, rubber and paper products, and textiles are among the many &London: Yale University Yale University,at New Haven, Conn.; coeducational. Chartered as a collegiate school for men in 1701 largely as a result of the efforts of James Pierpont, it opened at Killingworth (now Clinton) in 1702, moved (1707) to Saybrook (now Old Saybrook), and in 1716 was Press. NURSE, D. & T. SPEAR. 1985. The Swahili: reconstructing thehistory and language of an African society AD 800-1500. Philadelphia:University of Pennsylvania Press The University of Pennsylvania Press (or Penn Press) was originally incorporated with the Commonwealth of Pennsylvania on 26 March 1890, and the imprint of the University of Pennsylvania Press first appeared on publications in the closing decade of the nineteenth . PHILLIPSON, D.W. 1979. Some Iron Age sites in the Lower TanaValley. Azania 14: 155-60. POUSWELS, R.L. 1987. Horn and Crescent. Cultural change andtraditional Islam on the East African coast, 800-1900. Cambridge:Cambridge University Press. RENFREW, C. 2001. Symbol before concept: material engagement andthe early development of society, in I. Hodder (ed.) ArchaeologicalTheory Archaeological theory covers the debates over the practice of archaeology and the interpretation of archaeological results. There is no single theory of archaeology, and even definitions are disputed. Today: 98-121. Cambridge: Polity Press. SUTTON, J.E.G. 1998. Kilwa: A History of the Ancient Swahili Townwith a Guide to the Monuments of Kilwa Kisiwani and Adjacent Islands.Azania 33: 113-69. WILLEY, G. 1953. Prehistoric Settlement in the Viru Valley, Peru.Washington: United States Government Printing Office United States Government Printing Office:see Government Printing Office, United States. . WRIGHT, H.T. 1993. Trade and Politics on the eastern littoral ofAfrica, AD 800-1300, in T. Shaw, P.J.J. Sinclair, B. Andah & A.Okpoko (ed.) The Archaeology of Africa: food, metals and towns: 658-70.London: Routledge. WYNNE-JONES, S. 2005a. Scale and Temporality tem��po��ral��i��ty?n. pl. tem��po��ral��i��ties1. The condition of being temporal or bounded in time.2. temporalities Temporal possessions, especially of the Church or clergy.Noun 1. in an Urban SettlementSystem: Fieldwork in Kilwa Region, Southern Tanzania. Nyame Akuma 64(December). --2005b. Urbanisation at Kilwa, Tanzania, AD 800-1400. UnpublishedPhD thesis, University of Cambridge. Stephanie Wynne-Jones, British Institute in Eastern Africa, Kenya(Email: stephanie.wynne-jones@biea.ac.uk)