RC21｜世界體系怎麼了？

最近，RC21的群組信有一波關於「世界體系」與「都市研究」之間理論關係的小討論，很是精采。起因是Ray Hutchison問了以下這個問題：

Whatever happened to world systems theory? My sense is that this was overwhelmed by the literature on globalization and world cities and the like. A colleague in history department who worked in this area (in the past, he emphasizes) suggested that the approach became institutionalized and narrowed and has had some recent scholarship within history, but I am curious about present linkages with urban studies and urban theory. Is there any recent article that discusses this shift of focus across the field?

這問題一出現，引起許多人的回應，有的是緬懷自己過去讀世界體系時的澎湃熱情，有的是從自己現有研究暢談心得。在眾多回應者中不乏如John Logan、Sharon Zukin這些知名學者，不過這當中又以Neil Brenner的回應最讓人折服。

他說（粗體字為我個人標示）：

To my mind, the legacy of world system theory in urban studies is alive and well, above all on the literature on what used to be termed "world cities" and what is now usually referred to as "globalized urbanization."

But the legacy of world systems theory in that particular literature has to be understood in relation to the evolution of the conceptual foundations of world city theory itself.

As Peter Taylor and others have argued, early approaches to world cities (R.B. Cohen, Friedmann, Sassen) tended to focus on the "attributes" of so-called world cities--e.g. numbers of TNC headquarters, indices of producer / financial services activity. In that context, there was a direct articulation between world system theory and world city theory. World cities were understood to be the command and control centers of the world economy, located mainly but not exclusively in the "core" zones of the world system. There is a vast and to my mind very informative literature that maps out the contours of the world urban system using this type of approach (the work of Michael Timberlake, David Smith and others comes to mind). The link between the capitalist world system and the world city system also hinged upon the notion of the new international division of labor: as John Friedmann suggested, it was precisely due to the industrialization of previously peripheral or semi-peripheral zones that new command and control centers emerged in the cities of the core (this point was later elaborated at length by Sassen and many others).

More recently, however, Taylor and others have emphasized more relational approaches to the study of globalized urbanization. Here, the question is not "attributes" of particular cities but forms, degrees and modes of interconnectivity among cities.（註：參看GaWC） Taylor's excellent book, WORLD CITY NETWORK, along with his collaborative work at GaWC / Loughborough, elaborates the conceptual and empirical implications of this approach with great detail and sophistication.

To what extent does this relational approach to world city formation, focused on interconnectivity, build upon, revise or transcend the world-systems foundations of earlier, attribute-based approaches? This is a tricky question and opens up a number of methodological and empirical issues. I believe that Taylor still situates himself in the tradition of world systems research, but he also increasingly invokes writers such as Jane Jacobs (her later work, that is) to develop his relational urbanism.

I personally believe that a relational understanding of globalized urbanization is entirely compatible with the world-systems emphasis on large-scale, worldwide divisions of labor and structures of inequality. From my point of view, the major problematique of world systems theory is global uneven development. To my mind, this is also one of the major issues at stake in world cities theory / approaches to globalized urbanization. So, from this perspective, anyone interested in the study of globalized urbanization stands to benefit from some of the work of world-systems researchers. But there is still the broader question of how a relational conception of economic space might lead to a rethinking of the standard (territorialist) distinction between core/semi-periphery/periphery that underpins world systems research.

Another key literature, possibly linking some of the issues under discussion here, is that on global commodity chains (pioneered by Gary Gereffi and others), which likewise links world-systemic structures to individual organizational forms (in this case, capitalist firms, not cities) through relational data on commodity chains.

One final point. Some years ago I gave an AAG presentation in which I suggested that many comparative urbanists have retreated from the world-systemic concern with "encompassing comparison" (Charles Tilly) that animated "first-wave" world cities research. There are many reasons for this, but I believe that it is in part linked to an antagonism towards "convergence" arguments, i.e. the notion that globalization = homogenization. I would argue, however, that encompassing comparisons do not at all necessarily entail an embrace of convergence or homogenization arguments. Furthermore, without some "encompassing" methodological gesture, the notion of globalized urbanization becomes rather empty, if not logically incoherent. I hope to write up this paper sometime soon--I expect that many colleagues will disagree with my particular reading of the literature and the (Lefebvre-inspired) notion of "globalization" that underpins it.

最後，其實Brenner的回應值得用wiki的方式加註相關連結，只是我最近沒什麼時間。所以，有興趣的人就自己看看做參考。