CityReads│Agriculture First vs. Cities First: Debates Continue
This is a rebuttal to a critique of Jane Jacobs ‘cities first ’ thesis from a geographer Taylor Peter (Reply 3 on WeChat to get the orignal paper by Smith et al.). The dispute centres on the archaeological record for city origins. Substituting a process definition of cities—city-ness—for a ‘thing’ definition, this reply opens up pre-Mesopotamian possibilities for city networks while conceding the difficulty in empirically obtaining evidence in earlier periods. Thus Jacobs’ thesis cannot be absolutely refuted, and an exciting agenda for urban research emerges for archaeologists and social scientists.
Taylor, Peter. 2015. “Post-Childe, Post-Wirth: Response to Smith, Ur and Feinman.” International Journal of Urban and Regional Research 39(1):168–171.
Source:http://onlinelibrary.wiley.com/doi/10.1111/1468-2427.12181/epdf
Cover picture:https://s-media-cache-ak0.pinimg.com/736x/dc/09/fe/dc09fe2a29ccc25bbd5233716030151d.jpg
IThe Main Arguments and Evidences of "Agriculture Proceeds city" by Smith et al
Smith et al.’s argument is both very basic and very simple: the archaeological evidence rules out Jacobs’ ‘cities first’ thesis. Therefore Jacobs is ‘contrary to fact’ resulting in her position being ‘erroneous’ and ‘faulty’. Smith and his colleagues largely concentrates on the evidence for the timing of the origins of agriculture. They review methods for dating and then report the archaeological record that these have produced. The results are summarized in Figure 1 that shows the beginning of farming to be 10,000 years old for both plants (column 2) and animals (colunmn 3).
IIPeter Talyor's Rebuttal
Accepting this archaeological record for agriculture, the dispute centres on the archaeological record for city origins. Substituting a process definition of cities—city-ness—for a ‘thing’ definition, this reply opens up pre-Mesopotamian possibilities for city networks while conceding the difficulty in empirically obtaining evidence in earlier periods. Thus Jacobs’ thesis cannot be absolutely refuted, and an exciting agenda for urban research emerges for archaeologists and social scientists.
The differences relate to the column which is headed ‘Settlement’. Peter Taylor believes the complexity of cities makes them inherently different from all other settlements. Following Jacobs and Castells, he treat cities as a process (economic development) that operates through inter-city relations (networks of cities). This process of ‘city-ness’ is especially conducive to innovation and its diffusion, due to the communication potential within and between cities that totally dwarfs levels of human Communication in other types of settlement. Thus ‘urban worlds’ become prime candidates as sources of world-changing practices such as agriculture and state. Smith et al. do not really engage with this argument, merely pointing out that ‘Taylor is free to define terms as he pleases’ and referring to ‘his vague and archaeologically unusable ‘‘cityness’’ criteria’.
First it is unclear what archaeologically unusable means: archaeologists have long used central place theory with its focus on local hinterlands, and there appears to be no reason why the same should not be the case for ‘central flow theory’ derived from city-ness and focusing on non-local trading relations.
Second it is the nature of cities that is at issue. City-ness is built upon a relational approach to understanding cities; archaeologists as represented by Smith and his colleagues use ‘thing theory’-- definition by content rather than process. The critique refers to Wirth’s three criteria of size, density and heterogeneity. Particularly in the archaeological argument, the ‘things’ emphasized are what are found in early Mesopotamian cities——reference to other sites that ‘have not revealed any monument architecture’, so that settlements not containing these things are deemed not to be cities, ipso facto Mesopotamia has a head start in claiming the first cities.
Jacobs originally drew her ideas for ‘cities first’ from Mellaart’s (1965) excavation of Ctatalhoyuk and his initial claims to have found a new ‘first city’. Subsequently excavated further by Hodder, I engaged with Hodder’s new interpretation of Catalhoyuk as having a ‘domestic mode of production’. This enabled me to combine Jacobs’ ideas with Sahlins’ (1972) ‘stone age economics’ to support the notion of urban worlds as world-changing places.
Smith et al.do devote two paragraphs to Ctatalhoyuk to repeat that it has a ‘non-urban nature’, by arguing that it ‘does not come close to qualifying as urban’ because the settlement does not have a hinterland (central place theory) or the right institutions and activities (content). There is no mention of Sahlins so the idea of a ‘neolithic urban’ is discounted by omission.They returned back to archaeological record——Catalhoyuk ‘post-dates’ early agriculture and therefore ‘the proponents of Jacobs’ model are left without empirical support’. But she only refers to the settlement as the ‘earliest city yet found’.
The evidence archaeologists discover and interpret need carefully concerns. Archaeological data is probably unique in its degree of patchiness.How representative is it of the process being studied? In terms of spatial distributions of artifacts, ‘blank areas on the map are difficult to interpret’ because this ‘may not mean that the artifacts or sites did not occur there, but only that they have not been found there’. Smith et al. obviously appreciate this problem and include it in their discussion of the timing of early agriculture-- evidence of morphological changes in plants and animals are supplemented by consideration of ‘less archaeologically visible management strategies’ to push back the dating a whole millennia.
The visibility and survivability of early settlements is equally important for dating the origins of cities. Concern for monumentality, as in Smith et al. makes missing cities less likely. Pyburn discusses this issue for the case of the Maya lowlands that had been traditionally interpreted as ‘only marginally urban’. She argues that:
The effect of a tropical biomass on preservation and architectural visibility is profound. There is, for example, no reason to expect that wattle-and-daub structures would be as visible to archaeologists 1,000 years after abandonment as are large structures with stone foundations built on elevated platforms……. potsherds from periods not represented by houses [show that an] architectural sample must therefore be incomplete.
such frailty-- Pyburn’s ‘invisible architecture’--in the urban archaeological record opens up an exciting new research agenda for understanding cities. In fact, it is relatively easy to find city candidates in ‘wrong places’ at ‘wrong times’ and serendipity plays a role in some discoveries. Serendipity does not mean random, so there is no way of knowing the representativeness of our current knowledge of settlements. But, more importantly, there will have been myriad past settlements that will never be found. In situations where you cannot know, the best strategy for interpreting evidence is to be modest in claims and assertions.
The most disappointing feature in this debates is the framing of the debate as a disciplinary dispute. Smith and his colleagues fall back on references to their discipline’s textbooks, which is a tool for reproducing a discipline by socializing new generations of scholars and the bedrock where a discipline is built. Thus they tend to be inherently conservative in nature. What Smith and his colleagues refer to as ‘conventional understanding’-- the basic outlines of agriculture and cities in prehistory have been known since the early twentieth century’, citing the work of Gordon Childe, is imbued with nineteenth-century presumptions. As Wallerstein’s “Post-Child” call ‘to ‘‘unthink’’ nineteenth century social science’, the author believes disciplines applied to studying social change formed in universities a century or so ago are coming to the end of their useful lives. Thus, instead of the popular interdisciplinarity or multidisciplinarity, I advocate indisciplinarity, which means bringing evidence and theory to a study irrespective of their disciplinary origins. This is particularly appropriate for studying cities, since their fit into existing disciplines has always been a little awkward.
The author considers Jane Jacobs to be a founding mother of indisciplinarity. Her ‘cities first’ thesis can be viewed therefore as her being inconsistent, given she is known as a scholar for whom induction precedes deduction. Or it could reflect the ‘complex and difficult-to-model set of non-linear processes’ that Smith et al identify. In the case of understanding urban origins and the power of cities and their networks, it should not be a matter of archaeological evidence versus social science theory, or even archaeological evidence combined with social science theory; rather a genuine two-way interaction is in order with post-Childe archaeological urban debates, for example on disentangling city-making and statemaking, alternative urban forms, changing urban networks and evolution as explanation engaging with post-Wirth social science urban debates, for example on cities as agglomerations, creative arenas and connected spaces. I am a social scientist trying to make my contribution to this endeavour.