In their recent acclaimed The Global Cities Reader Brenner and Keil (2006) start their selection with an excerpt from what they term ‘Peter Hall’s classic work on world cities’ (p. 21). Although there is always debate on what is and what is not included in a scholarly ‘reader’, in this case where to begin must have been an easy decision for the editors: certainly most people knowledgeable of this research field would have expected Hall’s (1966) The World Cities to appear prominently up-front despite a 40 year gap between the two publications. But this should not be interpreted as implying a simple continuity of research efforts over these decades. This is hinted at by the inclusion encompassing only the first couple of pages of Hall’s book. The reason for such brief acknowledgement is because, according to Brenner and Keil (2006, 20):
They are able to link Hall to the national urban systems school through his description of ‘the cosmopolitan character’ of world cities ‘as an expression of their host states’ geopolitical power’ (p. 7). I will, at least partially, dispute this interpretation and evaluation. I argue that Hall’s book is simultaneously before its time and of its time and it is the purpose of this essay to try and unravel these two elements in what I agree is a ‘classic’ of urban studies.
The World Cities was first published in 1966 with a second edition in 1977 and a third in 1984. A simple basic structure was maintained across editions consisting of a first chapter defining and contextualizing world cities, followed by a series of chapters on selected world cities describing their economic and spatial structures and the planning policies devised to meet their special metropolitan problems, and concluding with a glimpse into metropolitan futures and possible planning responses. In the first two editions there are seven world city case studies: London, Paris, Randstad Holland, Rhine-Ruhr, Moscow, New York, and Tokyo; in the third edition the list is increased to eight by adding Hong Kong and Mexico City and omitting Rhine-Ruhr. The latter decision is not due to substantive concerns but is simply pragmatic ‘for reasons of space’ (Hall 1984, 3). Revisions across editions of the two general chapters differ in degree. Changes to the first chapter are largely updating information and context with, importantly, the definition of world cities remaining constant. In the final chapter, the changing context identified earlier is deemed to require rather more substantial changes to discussion of metropolitan futures. Although the case studies are fascinating in their own right as succinct statements of city successes generating problems and stimulating planning coping policies, in this short essay I focus on the two general chapters largely from the first edition but with reference to changes in the third edition as and when necessary. My main concern is to understand why a book of the 1960s remains an important marker on the route towards contemporary research on cities in globalization.
All research and writing has to be of its time; you can only know previous and existing subjects and contexts, and the literatures that describe and attempt to explain those subjects and contexts. In this case Hall (1966) draws on information and ideas about large cities and changing employment structures (white collar jobs replacing blue collar, office replacing factory) that appear to favour the growth of those cities – chapter 1 is entitled ‘The metropolitan explosion’. At the heart of this work there is an interplay between the empirical and the normative: how to accommodate the additional population resulting from the demographic growth of cities, which affects their spatial structures resulting in a need for city and regional planning to create satisfactory outcomes. But the key point is that he is sensitive to temporal dynamics in his evaluations and interpretations. Thus London’s green belt, perhaps the city’s most influential contribution to postwar metropolitan planning, is understood as a prewar conception of a city not expected to grow and therefore limiting London arbitrarily to its late 1930s extent. Hence Hall favours green wedges (Randstad, Paris) to green belts (London, Moscow) as the more suitable planning response in a time of growth. The first edition of the book appears towards the end of the postwar boom and by the third edition the world economy is in the midst of a new downturn. The new context is one of demographically declining world cities with growth now represented by his new selections Hong Kong and Mexico City.
Although Hall’s work embodies big picture dynamics this leads him to become curiously out of sync with the narrowing dynamics of his academic discipline geography. In his description of what defines a world city he provides a place-based synthesis of attributes and functions that is a tribute to geography’s tradition of regional synthesis. This is his amalgam of characteristics that blend together to constitute a world city (Hall 1966, pp. 7-8):
World cities are clearly very special places fusing the highest levels of urban work and living. But Hall’s synthetic thinking was a far cry from the traditional regional geography school that sought out the bounding of national rural landscapes to extol ‘man-land’ relations and thereby write the work of cities out of the script (Wrigley 1965, Taylor 2003). For traditional geographical synthesizers, Hall was focusing on the wrong type of region: metropolitan, inherently uninteresting, and best ignored. In any case by the time Hall was writing of world cities, regional geography was on the wane with ‘systematic’ geography coming to the fore to promote specialization in human geography. In the newly emerging urban geography, a theory of rural marketing called central place theory was converted into description of ‘national urban systems’ to accommodate large cities (Berry 1964, Berry and Horton 1970, Bourne 1975, Bourne and Simmons 1978). This work neatly arrayed cities into national urban hierarchies thus writing non-local trade (‘horizontal’ movements of commodities and services in their various forms) largely out of urban geography (Taylor et al 2010). Hence Hall was out of sync in urban geography not just through his synthetic view of cities, but also through his comparative approach treating major cities outside restrictive central place vertical spatial relations.
As I have previously observed (Taylor 2004, 21) it is in this paradigmatic context that Hall’s world cities are sometimes noted but were never able to become an antidote to ‘parochial national thinking’ and lead an alternative ‘world’ agenda for urban geography. This had to wait for the rise of economic globalization that made thinking in terms of a world of multiple and separate national urban systems ‘anachronistic and frankly irrelevant’. Thus Brenner and Keil (2006, 8-9) have concluded that
It is true that the two foundation papers by Friedmann (1986, Friedmann and Wolff 1982) do not refer to Hall’s world cities but Sassen’s (1991) seminal The Global City does cite the first edition. However Friedmann’s and Sassen’s contributions are quite different. Friedmann essentially provides a pattern of world cities (a city hierarchy, an up-scaling of central place theory) whereas Sassen outlines a process of city development (global city formation) with detailed consideration of the agents (firms in particular sectors) who create global cities through their work. Thus Hall’s conception of world cities is likely to be less relevant to Friedmann’s project but can be an input to Sassen’s focus on process.
It is instructive that Sassen (1991) refers to the first edition of The World Cities and not to the third edition, which came out only seven years before her book. This suggests that in his revising of ideas on world cities Hall does not feed into the new research agenda that replaces the national urban systems school. The new ‘world’ research school derived from writings on multinational corporations in the 1970s that appeared to be transcending national economies. A key work was that of Hymer (1972) that implicated large firms as multinational corporations in worldwide uneven development. This led to identification of a new international division of labour (Fröbel et al 1980) whereby routine industrial production was being relocated to poor countries. It is this new global dispersal that creates the organizational need for a new local concentration of economic strategic functions in Sassen’s global cities and atop Friedmann’s world city hierarchy. These are headquarters and advanced producer services that facilitate transnational corporate transactions. They are executive, financial, professional and creative services listed as characteristics of his world cities by Hall (categories 1., 3. and 4. above) and which figure importantly in his case studies. But for Hall, these are treated at a relatively general level; they are part of the reason for the rise of white-collar jobs and the proliferation of offices in metropolitan regions. The firms – multinational corporations - doing this new work just do not feature as critical agents in Hall’s world cities. In his third edition Hall recognizes the changes happening to worldwide economic geography but he uses the concept of ‘Newly Industrializing Countries’ (NICs) rather than the more radical notion of New International Division of Labour, a restructuring of contemporary capitalism. In other words, Hall’s ideas remain wedded to an international economy rather than a world economy becoming increasingly transnational in nature. This is a fateful choice for understanding world cities and accounts in large measure for the lack of research continuity from Hall’s The World Cities to the world/global cities literature that emerged in the 1980s, soon to become part of the globalization discourse (Knight and Gappert 1989, King 1990, Sassen 1991), and to dominate the study of large cities thereafter (Brenner and Keil 2006, Derudder et al 2012).
Thus far I have presented a rather balanced discussion of Brenner and Keil’s respectful critique of Hall’s The World City with which I began this essay: Hall saw his world cities as distinct from national urban systems but they remained as integral to an international economy. I will now take my argument in a direction away from that of Brenner and Keil to fulfill my initial promise to partially detach from it. There is more to be said. The first question is that if Hall is so nationally-focused as Brenner and Keil state, why does he call his case studies world cities? Why this adjective rather than ‘great cities’, or ‘leading cities’ or ‘major cities’ or simply ‘big cities’? It is because he is offering more than ‘comparative metropolitan studies’. In fact he uses the phase ‘great cities’ in the very first sentence of his book but only to emphasize that he is concerned with just ‘certain great cities’ – world cities are identified as a special subset of great cities where ‘a quite disproportionate part of the world’s most important business is conducted’ (Hall 1966, 1). Taking his cue from Geddes (1915) these particular great cities are therefore labeled ‘world cities’. The key phase ‘the world’s most important business’ is what is elaborated in the list of characteristics of world cities previously described. Here business means political as well as economic plus additional related metropolitan functions. And his discussion treats them as more than simple attributes: they are important because of their relational properties. For instance ‘major centers of political power’ implies projection of power (geopolitics), there is the import and export of ‘great ports’ plus ‘great international airports’, because of ‘great universities’ ‘students and teachers are drawn to the world cities’, and also ‘world cities have become the places where information is gathered and disseminated’ (Hall 1966, 1-2). What is missing is reference to multi-locational firms, ironically pioneered by Pred (1977) from the national urban systems school, but essential for understanding cities as the places where globalization is organized. This is the price paid for the lack of focus on multinational corporations. Nevertheless right from the start there is a strong relational aspect to Hall’s world cities.
It is their extensive relationality, sometimes explicit always implicit, that makes these cities ‘world’ rather than simply ‘great’ and which is substantiated when Hall considers cases that are not world cities in comparison with his selection of cities. Although he uses a list of the largest cities in the world by population size as the framework for his identification of world cities he is at pains to point out immediately that
In fact in the third edition he includes two cities that are not in his list: Randstad Holland and Hong Kong ‘conversely’ do not appear in his revised list of large cities for 1980 (Hall 1984, 3-4). In addition, although Hall makes only brief references to historical world cities, relational thinking is shown to be the definitive criterion for identification: the relatively small but important international trading centre of medieval Bruges is identified as a world city (Hall 1966, 8) but late eighteenth century Edo (Tokyo) is ‘almost certainly the biggest city in the world in terms of population … yet … could not claim to be a world city for the simple reason that the city was almost completely shut off from the world’ (p, 218). The latter example clearly confirms that national urban primacy, whilst analyzed by Hall for his world cities (p. 21), is insufficient in itself – relations are not extensive enough - for great cities to qualify as world cities.
In the paradigmatic world of academic disciplines it is probably unreasonable to expect writings from the 1960s to have a legacy through simple continuity of research effort and ideas. With revolutions trumping evolutionary change, particularly by the geography knowledge makers, what we had better look for is academic works that have a strong resonance for contemporary researches. This is the basic reason Brenner and Keil (2006) cannot ignore Hall’s early contribution to their subject matter. From my way of thinking there are three key geographical contributions from the 1960s that have this quality to reverberate through urban studies across the decades: Jean Gottmann’s (1961) Megalopolis, Peter Hall’s (1966) The World Cities, and Jane Jacobs’ (1969) The Economy of Cities. Thus Hall is in very good company! The one basic common feature they all share is not thinking you can understand large cities through up-scaling a simple model of rural marketing. They are all going very much against the intellectual tide of the times and the strength of that tide should not be under-estimated – the influential Theoretical Geography by William Bunge (1966) referred to Christaller’s central place theory as ‘geography’s finest intellectual product’, so important because it proved geography to be ‘a basic science’, a producer of ‘new theory’ (p. 133)1. But the multi-nodal spatial concentration from Boston to Washington that is megalopolis is the very antithesis of central place spatial organization. Hence despite the empirical veracity of Gottmann’s research on megalopolis, we find it dismissed as ‘lacking precision as well as generality’ in the prime urban geography textbook of the national urban systems school Geographic Perspectives on Urban Systems (Berry and Horton 1970, 54). Textbooks are key markers of disciplinary substance because they define the content of contemporary wisdom on a subject; in this case the brief dismissal of Gottmann is better than Hall and Jacobs do since neither of these authors even get a mention. Of course, Hall’s identification of ‘polycentric’ world cities – Randstad Holland and Rhine-Ruhr – is similarly an implicit rejection of central place thinking, as is his concern for primate city distributions previously referred to. But it is Jacobs (1969, 35) who is most explicit in her rejection of the central place way of thinking in urban studies - as she baldly states: ‘A city does not grow by trading only with a rural hinterland’. And this is despite her not usually identified as a geographer - she is the classic paradigm maker and breaker - here I follow Harris (2011, 80) and posthumously treat her as ‘a theoretical geographer’. As I see it Jacobs is the antidote to Christaller. Unlike Gottmann and Hall she is concerned for understanding cities generically but she does link with the two other authors in understanding that large cities require specific attention as shown by her sole chapter to feature a distinct category of cities being ‘How Large Cities Generate Exports’ (pp. 180-202). The key point is that Gottmann’s megalopolis, Hall’s world cities, and Jacobs’ urban growth theory are all part of contemporary urban discourse - see Nelson and Lang (2011), Brenner and Keil (2006) and Glaeser (2011) respectively - while Geographic Perspectives on Urban Systems has been long dispatched to disciplinary history relevance only.
I will conclude with a brief comparison between concepts of world cities and global cities as created by Hall (1966) and Sassen (2001) respectively. Focusing on their similarities and differences illustrates the former’s prescient thinking. The similarity is obviously that both scholars are defining a class of large cities as special places with exceptional influence beyond their immediate environs. Such research practice has led to two key debates, one geographical and the other historical.
The first debate is about whether the world/global cities should be treated as exceptional, and if they are, what this means for understanding other cities. By ‘downgrading’ these other cities does research neglect them thus creating a very unbalanced urban studies? Both Hall and Sassen are careful to differentiate their subject matters in terms of contemporary processes. For Hall (1966, 25-8) this is the change in capitalist process from factories in industrial regions to offices in metropolitan regions, with those cities at the forefront deemed world cities. This may have derived in part from extrapolating Rasmussen’s (1960) well-known treatment of London as a ‘unique city’, albeit he denoted the uniqueness just in comparison with ‘continental (European) cities’. For Sassen (1991, 3-5), the uniqueness of London, with New York and Tokyo and just a small number of unspecified other cities as global cities, is explicit. This uniqueness derives from the unprecedented nature of globalization: the global reach of large transnational corporations has created ‘a new type of city’ as a strategic place for extensive spatial organization. Finance is an important function in both arguments but world/global cities are treated as much more than international financial centres. However, neither author precisely enumerate the membership of his or her class of world or global cities – given that there are deemed to be relatively few such cities, this would not appear to have been an onerous task. Clearly Hall’s European bias (five of his original seven world cities) and his explicit omission of Chicago and Los Angeles as world cities will not have gone down well in the USA, especially later with the so-called ‘LA School’ attempting to position their city at the centre of urban theory (Scott and Soja 1996; Soja 2000; Dear 2002): reference to Hall’s world cities here is again conspicuous by its absence. In the case of global cities, empirically it has been found difficult to discern either a demographic gap (Gugler 2004) or a functional gap (Taylor 2004) separating a few leading cities from the rest. Alternative approaches that treat globalization as an all-pervasive urban influence have developed more inclusive concepts such as ‘globalizing cities’ (Marcuse and van Kempen 2000) and ‘cities in globalization’ (Taylor et al 2006) both of which transcend Sassen’s exclusivity. The most trenchant critic has been Robinson (2002) who accuses global city researchers of leaving most cities ‘off the map’ as ‘the voids of world and global city approaches’ (p. 540). The claim of ‘bad geography’ (Short et al 2000, 317) is a particularly profound critique since it indicts global city research for being in opposition to its own intellectual origins from Hymer (1972). Both Hall and Sassen’s work illustrate the need for world/global city researchers to confront the political implications of the geographical bias in so much of their research2.
The second debate is about a different exceptionalism, whether world or global cities are solely contemporary phenomena or whether there can have been such cities in previous eras (Brenner and Keil 2006, 21). Hall and Sassen take different positions on this question. As discussed previously Hall recognizes ‘historical world cities’ on the grounds that particular past cities have had the necessary extensive relations to qualify for such worldliness. Gottmann (1989, 62) agrees with Hall and identifies world cities before the twentieth century. The term has also found its way into the historical literature: Braudel (1984) and Arrighi (1994) identify ‘world-cities’ in early modern Italy, and Robinson and Inglis (2006) identify classical Rome as a world city. In contrast, Sassen’s (2001) ‘new class of cities’ is not just new conceptually; it is intrinsically new as contemporary globalization. She explicitly argues that global cities are unlike previous ‘centers of banking and trade’ (p. 4) because their ‘key dynamic’ is
Taken up by Castells (1996, 415), her global cities are anchoring nodes in a new network society that is replacing modern industrial society. According to Brenner and Keil (2006, 21) although these two positions have ‘quite divergent research agendas’ they suggest that ‘these approaches can be compatible’. This is perhaps best achieved by interpreting ‘world’ not to mean ‘worldwide’ but rather ‘social world’ - what Wallerstein (2004) calls world-system - in which a city is embedded. Cities strategically influential in their ‘world’, such as early modern Genoa or classical Rome, can be reasonably designated ‘world city’ (Taylor 2011). However, the modern world-system has grown to become worldwide so that its strategically influential cities can be reasonably designated ‘global’. Nevertheless, I don’t think this solution fully encompasses Sassen’s and Castells’ position, that highlights technological innovations in communication (Sassen 2002, Castells 2009) and which means contemporary extensive influence and control are applied at historically unprecedented levels of intensity.
These two debates indicate an ongoing dynamic literature in which initial ideas and positions Hall took in the The World Cities nearly half a century ago still resonate today. This is a remarkable legacy.
Arrighi, G (1994) The Long Twentieth Century. London: Verso
Berry, B J L (1964) ‘Cities as systems within systems of cities’. Papers of the Regional Science Association13, 147-63
Berry B J L and Horton, F E (1970) Geographic Perspectives on Urban Systems. Englewood Cliffs, NJ: Prentice Hall
Bourne, L S (1975) Urban Systems: Strategies for Regulation. Oxford: Clarendon Press
Bourne, L S and Simmons J W (eds) (1978) Systems of Cities. New York: Oxford University Press
Braudel, F (1984) Perspective on the World. London: Collins
Brenner, N and Keil, R (eds) (2006) The Global City Reader. London: Routledge
Bunge, W (1966) Theoretical Geography. Lund: Gleerup
Castells, M (1996) The Rise of Network Society. London: Routledge
Castells, M (2009) Communication Power. Oxford: Oxford University Press
Dear, M J (ed) (2002) From Chicago to LA: Making Sense of Urban Theory. Thousand Oaks, CA: Sage
Derudder, B, Hoyler, M, Taylor P J and Witlox, F (eds) (2012) International Handbook of Globalization and World Cities. Cheltenham, UK: Edward Elgar
Friedmann J (1986) ‘The world city hypothesis’, Development and Change 17, 69-83
Friedmann, J and Wolff G (1982) ‘World city formation: an agenda for research and action’, International Journal of Regional and Urban Research 3, 309-44
Fröbel, F, Heinrichs, J and Kreye O (1980) The New International Division of Labor. New York: Cambridge University Press
Galpin, C J (1915) ‘The social anatomy of an agricultural community’ University of Wisconsin Agriculture Experimental Research Bulletin, No. 34
Geddes, P (1915) Cities in Evolution. London: Williams and Norgate
Glaeser, E (2011) Triumph of the City. London: Macmillan
Gottmann, J (1961) Megalopolis. New York: Twentieth Century Fund
Gottman, J (1989) ‘What are cities becoming centres of? Sorting out the possibilities’ in R V Knight and G Gappert (eds) Cities in a Global Society. Newbury Park, CA: Sage
Gugler, J (2004) ‘Introduction’ in J Gugler (ed) World Cities Beyond the West. Cambridge: Cambridge University Press
Hall, P (1966) The World Cities. London: Weidenfeld and Nicolson
Hall, P (1984) The World Cities. London: Weidenfeld and Nicolson (Third edition)
Harris, R (2011) ‘The magpie and the bee: Jane Jacobs magnificent obsession’ in M Page and T Mennel (eds) Reconsidering Jane Jacobs. Chicago: Planners Press
Hymer, S (1972) ‘The multinational corporation and the law of uneven development’ in J Bhagwati (ed) Economics and World Order from the 1970s to the 1990s. London: Collier-Macmillan
Jacobs, J (1969) The Economy of Cities. New York: Vintage
King, A D (1990) Global Cities. London: Routledge
Knight, R V and Gappert, G (eds) Cities in a Global Society. Newbury Park, CA: Sage
Marcuse, P and van Kempen, R (eds) Globalizing Cities: a New Spatial Order?. Oxford: Blackwell
McKenzie, R D (1933) The Metropolitan Community. New York: McGraw Hill
Nelson, A C and Lang, R E (2011) Megapolitan America: a New Vision for Understanding America’s Metropolitan Geography. Chicago: Planners Press
Pred, A (1977) City-Systems in Advanced Economies. London: Hutchinson
Rasmussen, S E (1960) London: the Unique City. London: Penguin
Robertson, R and Inglis, D (2006) ‘The global animus’ in B K Gills and W R Thompson (eds) Globalization and Global History. London: Routledge
Robinson, J (2002) ‘Global and world cities: a view from off the map’, International Journal of Urban and Regional Research 26, 531-54
Sassen, S (1991) The Global City: New York, London, Tokyo. Princeton, NJ: Princeton University Press
Sassen, S (ed) (2002) Global Networks, Linked Cities. London: Routledge
Scott, A J and Soja, E W (eds) The City: Los Angeles and Urban Theory at the End of the Twentieth Century. Berkeley: University of California Press
Short, J R, Breitbach, C, Buckman, S and Essex, J (2000) ‘From world cities to gateway cities’, City 4, 317-40
Soja, E W (2000) Postmetropolis: Critical Studies of Cities and Regions. Oxford: Blackwell
Taylor, P J (2003) ‘Global, national and local’ in R Johnston and M Williams (eds) A Century of British Geography. Oxford: Oxford University Press
Taylor, P J (2004) World City Network: a Global Urban Analysis. London: Routledge
Taylor, P J (2012) ‘Historical world city networks’ in B Derudder et al (eds) International Handbook of Globalization and World Cities. Cheltenham, UK: Edward Elgar
Taylor, P J, Derudder, B, Saey, P and Witlox, F (eds) (2006) Cities in Globalization. London: Routledge
Taylor P J, Hoyler, M and Verbruggen, R (2010) ‘External urban relation process: introducing central flow theory to complement central place theory’, Urban Studies 47, 2803-18
Wallerstein, I (2004) World-Systems Analysis: an Introduction. Durham, NC: Duke University Press
Wrigley, E A (1965) “Changes in the philosophy of geography’ in R J Chorley and P Haggett (eds) Frontiers in Geographical Teaching. London: Methuen
1. Of course this is a very geographical predicament; it was not faced by urban sociologists where rural marketing and metropolitan studies were separate areas of research (e.g. Galpin (1915) and McKenzie (1933)).
2. In an intriguing personal communication (16/06/12), Saskia Sassen points out that critics of her global city concept have missed the point that ‘I do not write as an urbanist but as someone interested in political economy and the social (inequality)’. She continues ‘mine was not an urban analysis’ and therefore focus on ‘a limited number of places’ is justified because ‘it is not about “cities” as such, but about strategic space of power and a site of production of very advanced complex capabilities that global firms need’. I mention this here because a very similar point can be made concerning Peter Hall’s book. As Ron Johnston has pointed out (personal communication, 11/05/12): ‘the bulk of the material in the case studies reflects Peter's real interest - planning. He is less concerned with the nature of the cities than with their organisation’. Thus we have the ironic situation that two immensely influential figures in urban studies - the authors that have bequeathed the concepts of ‘world city’ and ‘global city’ – are not themselves ‘urbanists’, their fecund concepts derived indirectly as outcomes from other related thinking.