Globalization is, as Michael Storper (1997a, 31) has recently pointed out, inherently geographical in nature. But which geography? There are many geographies in globalization which, for the initial purposes of my argument here, I will reduce to two basic positions, one emphasising geographical scale and the other focusing upon spatial flows. These are not, of course, discrete positions but they do tend to represent alternative tendencies in the way globalization is conceptualised. It is important to identify these two geographies because of their very different potentials for promoting radical restructuring of the social sciences. Emphasising rescaling can be a means for preserving a statist agenda with its conventional geographical mosaic of territories, the space of flows implies a new metageography.
This essay has been stimulated by reading the outstanding new text on globalization by David Held and his colleagues (1999), Global Transformations (hereafter GT).1 Their comprehensive treatment of the subject has forced me to revisit my earlier argument about globalization as a critical challenge to embedded statism in the social sciences (Taylor, 1996). I had argued that the mix of processes making up globalization were beyond coherent and rigorous treatment using conventional social science approaches. This was because the social sciences were interpreted as the creations, and therefore the intellectual creatures, of states, and it was surmised that the consequent state-centrism could not adequately handle the trans-state processes at the core of globalization. To my surprise, in GT the authors do manage to produce a coherent and rigorous account of globalization from a conventional social science perspective. Below I will argue that they achieve this by emphasising geographical scale and neglecting spatial flows. To illustrate this other geography in globalization I use Zygmunt Bauman's (1998) Globalization: the Human Consequences (hereafter GHC) where mobility, both physical and virtual, is the cornerstone of the argument.
The gist of my argument, which proceeds in four steps, is as follows. First, I investigate how Held et al. (1999) actually create a viable globalization thesis from a conventional social science position - I argue that they raise political science to a pre-eminent position, realising a potential always latent in a state-centric metadiscourse. Second, I focus on the implications of having geographical scale at the heart of the analysis - I identify the promotion of history as an unexpected consequence of this geography in globalization. Third, the manner in which spaces of flows are integrated into the analysis is explored with particular reference to migration - I use this to construct a geographical typology of globalization approaches. Fourth, I recast the argument in terms of metageography as a means for finding an alternative locale for where globalization takes place: I identify the world city network, conspicuous by its absence in both GT and GHC, as an alternative metageography.
A POLITICAL ENFRAMING OF SOCIAL SCIENCE
Given my original argument about the limitations of social science for studying globalization, it is important to emphasise from the beginning just how conventional GT is in its approach. This is expressed in four distinctive dimensions. First, GT is specifically disciplinary in its approach: the subtitle of the book, Politics, Economics and Culture, tells us as much. In fact the team of authors consist of two political scientists, an economist and a sociologist and the text has seven substantive chapters which divide as three political (states, war and environment), three economic (trade, finance, production) and two sociological (migration, culture). Second, GT is specifically statist in its approach: as we might expect the political science, economics and sociology are dominated by information about states and their relations to other institutions. In fact, this empirically-rich text has only two maps (on migrations) and two tables (on culture) which are not about states or accumulations of data from states out of 89 empirical figures, maps and tables. Third, GT is very historical: despite numerous references to 'spatio-temporal' the latter in this hyphenated pair certainly dominates the text. This is, of course, what Soja (1989) has identified as a central modernist feature of social science. Fourth, despite reference to the uneven nature of globalization GT has an overt bias towards rich countries (called SIACS, states in advanced capitalist societies): this is the model of statehood used for testing the effects of globalization. More explicitly, USA, UK, France, Germany, Sweden and Japan are selected for regular empirical evaluation thus eluding the extreme polarising effects of economic globalization which are so important to Bauman (1998, 70-1). This is undoubtedly a disciplinary/ statist/ historical/ first-world, that is to say conventional, social science perspective on globalization.
In the twentieth century conventional social science has developed as a trilogy of disciplines - economics, sociology and political science - which purport, between them, to cover all social behaviour (Taylor, 1997). Of the three, political science has been the weakest, in the sense of least influence within and beyond academe: for instance, 'does politics matter?' has been a question of some concern (Sharpe, 1981). Given the state-centric nature of the social science project and the fact that the state is a political institution, this lowest position in the pecking order at first seems surprising. However, it is easily explained by the taken-for-granted nature of the state in social science. Embedded statism meant that whether studying the economy, society or government, the theory of the state was only ever implicit and therefore very rudimentary - in one critique Macpherson (1977) famously asked 'do we need a theory of the state?'. Interestingly, this prioritisation of economics and sociology is to be found in the globalization literature, in both academic and more popular accounts descriptions of economic and cultural globalizations have dominated. This is recognised in GT in the opening sentences to the chapters on finance with its 'pervasive images of globalization' (p. 189), on production with, aside from the above, 'the commonest image' (p. 236), and on cultural globalization which is 'so visible, widespread and pervasive' (p. 327). But these statements are all a long way into this text; for its treatment of globalization GT turns the social science pecking order upside down so that it is politics which dominates.
There is an irony in the influence of globalization on conventional social science: it has forced the state out into the open. Quite simply, if contemporary global processes are interpreted in terms of degree of threat they pose to the state, then the latter can hardly be left as only implicit. And this is how politics comes to dominate GT despite its multidimensional view. The authors begin by identifying different schools of thought in globalization studies in terms of the degree and nature of state autonomy that is allowed. This starting point sets the structure of the rest of the argument. For instance, the authors choose to begin their book with a chapter on the territorial state followed by a chapter on warfare. In additional all substantive chapters end with a political assessment of the impact of globalization on state autonomy. What GT is fundamentally about is not global transformations but state transformations as a response to globalization; it is the state which turns out to be the real object of study. This is never more obvious than at the very end of the book: the penultimate section deals with changes in state power, the final section with limits to democratic politics.
Thus the book begins and ends with politics. That the authors can successfully carry off a book on globalization through thoroughly state-centric lenses is indeed remarkable. It is made possible, in part at least, by them heeding the lessons of Ruggie (1993) and Mann (1988): territoriality is taken seriously as Ruggie requires and war is integrated into social science as Mann requires. The end-result is a political enframing of social science in order to study processes of social change at large geographical scales.
GEOGRAPHICAL SCALE: WHICH HISTORY?
Good conventional social science is rigorous in its concepts and carefully analytical in its methods. GT meets these criteria in the way the authors systematically introduce globalization as their subject matter (pp.14-27). Here is the definition they give for globalization:
'a process (or set of processes) which embodies a transformation in the spatial organisation of social relations and transactions - assessed in terms of their extensity, intensity, velocity and impact - generating transcontinental or interregional flows and networks of activity, interaction, and the exercise of power' (p. 16).
At first reading this appears to be unexceptional; certainly the references to relations, transactions, flows and networks suggest an approach where spaces of flows will feature prominently in the transformation of spatial organisation. But this is not how the book turns out. The key words in the definition for understanding the authors' approach are transcontinental and interregional.2 Globalization is treated primarily as a geographical scale of activity: it lies 'on a continuum with the local, national and regional' and it 'implies, first and foremost, a stretching of social, political and economic activities across frontiers' (p. 15, emphasis in the original). Scale is 'first and foremost' because the authors see their definition as a means 'to differentiate globalization from more spatially delimited processes - what we can call "localization", "nationalization", "regionalization" and "internationalization"' (p.16). This formulation follows rejection of contemporary time-space compression, with its explicit linkage to enabling electronic technologies (and thus spaces of flows), as a definition of globalization (p. 15). The consequence of this geographical scale approach for the contents of the text is quite unexpected.
As mentioned in the previous section GT is historical in approach, but there is much more to this than putting contemporary globalization in a genetic context: GT is a study of comparative globalizations. Once the focus is upon geographical scale, the whole of history opens up in a search for past transcontinental or interregional patterns of activity. For instance, starting points for the empirical parts of the substantive chapters includes early imperial systems (chapter 1, p. 33), early Mesopotania (chapter 3, p. 152), Mongol-Chinese migrations (chapter 6, p.287), and the spread of world religions (chapter 7, p. 332). The argument of each of the eight substantive chapters is summarised at the chapter's end by 'grids' that compare the 'historical globalizations' which have been uncovered. In the conclusion of the whole text the historical narratives are brought together as globalizations in four periods: pre-modern, early modern , modern and contemporary. These do not exactly coincide with the periodicity of the previous chapter grids (different timings of processes is in keeping with the authors' separation of the economic, political and cultural3) but the interesting feature is that contemporary globalization is dated from 1945. This is important to the overall argument since social changes after 1945 are deemed not to be the critical determining features defining contemporary globalization. In fact, in the previous chapters there are only two grids which depart from this dating, in the financial and production globalization chapters the 1970s are given as the starting point of contemporary globalization. An implication of these timing differences is that political and cultural globalization led the way to our contemporary situation with economic globalizations following at a later date. Economic processes as the laggard in contemporary globalization is indeed a surprising finding of this book.
We produce a very different interpretation if globalization is viewed from a space of flows perspective where the distinction between inter-state and trans-state processes becomes crucial for dating contemporary globalization (Taylor, 1995). However, through beginning with 'political globalization', GT integrates traditional international relations topics - UN, Cold War - into globalization. Obvious these topics qualify by geographical scale criterion but they are primarily inter-state by nature. GT does distinguish between inter-state and trans-state (termed international and transnational) as, for instance, when they contrast international law with cosmopolitan law. But this is not taken forward into their periodization except where the trans-state processes have become of overwhelming importance (i.e. in finance and production). The point is that the inter-state processes described in GT reinforced states and are best seen as the continuation of 'internationalization' rather than contemporary globalization. It is in finance and production where trans-state processes have dominated, as indeed the authors themselves make clear. Therefore, rather than lagging in the creation of contemporary globalization, these are the key instigators. Hence, it is the 1970s, with enabling technologies coming on hand to greatly enhance trans-state organisational potentials (Castells, 1996), which marks the beginning of contemporary globalization and is the focus in GHC.
As always, periodizations are not simply intellectual quibbles, they express social theory and as such can have profound influences on interpretation of social change. In this case the effect of treating post-1945 as the period of contemporary globalization is to miss what is arguably its most important feature. In the era since the 1970s we have witnessed an historical shift in the distribution of income and wealth. After a century of narrowing economic polarisation, since the mid-1970s polarisation has been increasing once more. And to dramatic effect: Bauman (GHC, p.70) quotes the UN Human Development Report which finds that 358 'global billionaires' have the same combined wealth as the income of the 2.3 billion poorest people in the world. In short, the new enabling technology has provided the potential for a new redistribution upwards to those who command 'global wealth' or who can demand 'global wages' (GHC, p.71). And new right political projects (abetted by 'new left' governments as and when necessary, see Taylor (1992)) have ensured the very wealthy elite can indeed make more money, more quickly, than ever before. This is indeed a truly critical 'global transformation' but because the focus of the GT analysis is the state, distributional effects of globalization are treated in debates over the welfare state (GT. p.183, 232) and de-industrialisation (GT, p.185, 278) in the wealthier countries - the global picture and the associated historic shift in power and wealth are simply not part of their argument.
INCORPORATING FLOWS: WHICH GEOGRAPHY?
As stated previously, the definition of globalization in GT does not neglect transactions and flows, the issue is how they are incorporated into the text. In part, the problem stems from the serious data deficiencies which confront all social scientists when trying to describe and analyse transactions and flows. Most conceptualisations depict flows as constituting systems or networks - Castells' (1996) 'space of flows' is the spatial structure of his 'network society' - but information on flows rarely allows even a glimpse at such complexes (Taylor, 1999b). Most data only depict a fragment of the whole flow process, usually measured as an attribute, evidence of variations in some input or output factor. The classic case is foreign direct investment (FDI); it is widely used in GT including to describe 'the global financial system' (GT, pp. 210-2). Although depicting 'private' capital flows such data are aggregated to become attributes of states (or groups of states), either 'sending' or 'receiving', with little or no empirical sense of the system behind the flows. FDI data, in GT and very many other texts, provide invaluable information on trends but very little on networks.
In this section I will focus on the movement of people and how this is interpreted as part of contemporary globalization. Here we find a major difference between GT and GHC. This is the only topic in GT where there are maps (for early-modern/modern and contemporary eras) showing global flows; they contribute to a very good succinct narrative on the history of major migrations. In terms of contemporary globalization (post-1945) the authors are very careful to delineate many regional and global flows of labour (GT, pp.300-2) and refugees/ asylum seekers. However, despite concern for stratification in migration, there are only two brief references to elite migrations (GT, p.302, 304). In the chapter's review of impacts, border controls, surveillance and citizenship issues arise with one other brief reference to skilled professional migration (GT, p.325). But the latter mobility is particularly important for globalization because it marks the formation of an enlarged trans-state group of cadres organising and controlling capital, perhaps what Leslie Sklair (1995) identifies as a 'transnational capitalist class'. The relative neglect of such migration in GT suggests its state-centrism precludes joining in such globalization debates on the meaning of such migrations. In contrast, Bauman identifies a 'new polarisation' of society (GHC, p.18) based upon mobility as 'the most powerful and most coveted stratifying factor' (GHC, p 9). In his 'global hierarchy of mobility' (GHC, p.69) we are all on the move but there is a basic bifurcation occurring between those with freedom to move (both physical and virtual) and those increasingly being forced to move while restricted in their destinations - in his colourful language Bauman calls the two strata 'tourists and vagabonds'. In GHC, this is central to globalization whose cultural face is consumer society where desire, not satisfaction of desire, is the guiding force: all want to choose but only the tourists can. There is no doubt that GHC provides a better sense of a new world of flows but, as a caveat, it must be added that, although some empirical evidence is offered in evidence in GHC, it does not compare to the more comprehensive and systematic presentation of information in GT.
What can we make of the dissimilarities between the two books? Here I focus on the consequences of their respective geographies in globalization. It is in this one chapter where GT is directly and wholly concerned with flows that its conventional social science approach to globalization shows particular signs of weakness. If there are indeed 'global transformations' going on, then it is not clear that careful historical comparisons are sufficient to understand them. Showing how the contemporary situation differs from previous patterns of large-scale migration will provide a list of particular characteristics of present mobility (GT, grids 6.1 and 6.2, pp.306-10) but to follow this by considering their policy impacts on states prevents taking the argument a stage further in terms of large-scale spaces of flows. The authors do identify 'a hierarchy of ease of movement' (GT, grid 6.2, p. 309) but this is not translated into the different networks through which the hierarchy operates. States are important to this hierarchical-forming process, of course, but their border surveillance is only one part of the story. In short, they describe spaces of flows through the lenses of a space of places.
The analysis in GHC, of course, is immersed in a space of flows, but describing spaces of places through the lenses of a space of flows can be equally problematic. Bauman extrapolates the old modernist theme of rapid change eroding local community to contemporary globalization (GHC, p.14) with no consideration of the reconstitution of localities under conditions of transformation, global or otherwise, as Agnew (1987) has taught us to look out for. There has probably never been more place promotion - the selling of localities using the advertising techniques of consumer society - than today ranging from global cities such as Singapore to restructured industrial regions such as the former mining towns of northern England. We can agree with Bauman that the global elites are extra-territorial in the sense of being relatively free from the incursions of sovereign territoriality (witness the Australian American, Rupert Murdoch as Britain's leading media mogul) but not in the sense of being free of geography per se. Contra Bauman, in Castells' space of flows one of the levels he identifies is 'the spatial organisation of the dominant, managerial elites' - the cosmopolitan locales of their work, rest and play (Castells, 1996, 415). To take another well-rehearsed example, the end of geography has not come even for the most virtual of world markets, global finance. Here the networks of capital flows are 'grounded' in certain privileged locations, the international financial centres of the world (as global cities, part of another layer of Castells' space of flows (p. 413)). The key point is that networks have centres as well as flows.
One way of making sense of these differences is to create an alternative typology to that devised in GT. As previously noted, with its focus on the state the authors identify three broad positions (GT, p.10): first, the hyperglobalists are contrasted with the sceptics (end of state versus continuing internationalization), and second, the 'transformationalists' are brought in as a middle position (globalization transforming state power). The latter is the authors' position and smacks very much of them creating a position where they can have their political cake (state power) and eat it (globalization). A geographical alternative4, which I have been moving towards throughout this essay, defines three broad positions as follows: first, the space of places group is contrasted with the space of flows group (state-centric mosaic versus extra-territorial/end of geography), and second, the 'spatial interactionists' are brought in as a middle position (spaces of flows interacting with a spaces of places). The latter is, of course, my position which allows me to have my geographical cake (spaces of places) and eat it (spaces of flows).
A QUESTION OF METAGEOGRAPHIES?
The space of places in the spatial interactionist position need not be the mosaic of states. The latter dominates our global conceptions of places because the global political map is a master framework through which we view our world spaces. In the modern world, non-local information about human differences and social changes has been generically interpreted through the spatial mosaic which is the familiar political map. In Lewis and Wigen's (1997, ix) terms it is a key metageography, a spatial structure 'through which people order their knowledge of the world'. Often unexamined and therefore used uncritically, metageographies are part of our taken-for-granted world, just the sort of conceptions we should consider ripe for radical consideration at a time of 'global transformations'. I want to use this new discourse on spatial structures to revise the way I have conceptualised the space of flows in the discussion above and in my original argument (Taylor, 1996). In a critique of the latter, Michael Mann (1996) distinguished my use of trans-state from global in a strictly planetary sense. I do not want to pursue his scale argument here but it has to be conceded that trans-state, like inter-state, can be interpreted as describing a confusing range of scales. The problem is the continuation of reference to state even when the 'trans-' is treated in the stronger sense of 'beyond' rather than simply 'across' (Taylor, 1995, 12). If our understanding of globalization is to escape from embedded statism, we require an alternative metageography, one which does not feature states.
I will continue to take my cue from GT and GHC but this time I focus on a particularly revealing similarity between their two arguments: neither make reference to the world city literature. GT refers to cities briefly as early modern entrepots, contemporary financial centres, and migration magnets but their state-centric focus allows for no elaboration of them as places; Bauman treats cities as the locus of modernist space wars and contemporary defensive segregation practices but without integrating them into his spaces of flows. In fact one of the most intriguing features of the globalization literature in general is the dearth of references to world city studies even though world city researchers think they are depicting a key spatial structure within globalization. The reason for this partial exclusion of world cities is no doubt partly due to disciplinary divisions - geography/planning versus the social science trinity. It is also a result of the oppressive power of the states-metageography, the lack of alternatives leaves us a very unsatisfactory 'all or nothing' situation. In the globalization literature this is expressed by either maintaining the state mosaic or else propounding a nihilistic metageography, the two positions represented by GT and GHC respectively. Thus, even though both books begin with reference to the concept of time-space compression which has been enabling for the rise of world/global cities, their state-centrism and end of geography, respectively, leave no subsequent room for urban places as world/global cities.
Since states have developed as all-encompassing 'containers' of social activities in the modern world, the states-metageography has seemed to make sense for social analyses. But it is a myth (Lewis and Wigen, 1997, 7); we should not assume that something as complex as a society can ever be adequately expressed in a single spatial structure. As recounted above, the states-metageography has led to an under-estimation of the importance of flows in the modern world. This is what Arrighi (1994) refers to as 'deficiencies in our perceptual habits' causing 'non-territorial spaces-of-flows' to have gone 'unnoticed alongside the national spaces-of-places throughout the history of the modern world system' (p. 81). He illustrates this in the most conspicuous contemporary space of flows, finance capital, tracing its network structures back to early modern Italian city states. This bias against the space of flows is to be found in GT where the early modern world is described in terms of absolutism and the territorial state while largely ignoring mercantilism and the great transcontinental fleets of the era. The 'age of absolutism' was also the 'age of mercantilism' with its rise of large trading cities and circuits of cosmopolitan merchants - by GT's scale criteria an historical globalization but one which is missing from their state-centric account. This is not just a problem for conventional social science. The states-metageography by its very nature, can distort interpretation of the contemporary spaces of flows, even in the work of its leading theoretician, Castells (1996). For example, he proposes his space of flows as a replacement of a traditional space of places ignoring the important spaces of flows of the modern past. Globalization is not distinguished by having a space of flows but by the particular structure it takes under contemporary conditions.
Breaking free of the states-metageography by announcing an 'end of geography' is equally problematic. In GHC (pp. 18-9), power has become 'non-territorial' residing in a cyberspace through which the elite operate. Bauman writes of a 'power web' in which 'shifting of bodies and rearrangement of bodies in physical space is less than ever necessary to reorder meanings and relationships'. Predictably, the example of financial flows is used to illustrate 'this 'new "bodyless-ness" of power'. This does not mean that the new elite are ethereal, Bauman shows that they are very territorial in their arrangements to isolate themselves from unwanted local intrusions on their lives, but it does mean that the basis of their powers is beyond real geography, hence no metageography just the virtuality of cyberspace. But is this really the case? Such analyses tend to privilege information over knowledge. Certainly information is transmitted at unbelievable speeds in unbelievable amounts all around the world on the basis of which innumerable decisions are made. These decisions combine current information with existing knowledge to produce a given outcome but what of new knowledges, how can they be created? Creation of knowledge is not the simple aggregation of information, it requires judgements in new situations involving not a little collective lateral thinking. Put another way, it a reflexive process which requires a culture of innovation in which to blossom (Storper, 1997b). If the creation of knowledge cannot be disembodied and we live in a society which worships the new, then it follows that 'knowledge complexes' remain a necessity. The new time-and-distance-destroying technologies enable a new geographical dispersion of activities but they have also led to new geographical concentrations (Sassen, 1994), not least in terms of finance: the leading decision-making offices of financial institutions are to be found in the leading cities of the world, notably Sassen's (1991) triad of global cities, London, New York and Tokyo.
Here we get a hint for an alternative metageography. In Castells' (1996) space of flows there are several layers and different networks within layers but he identifies one in particular as the 'most direct illustration' of nodes and hubs (p. 415). This is the world city network where the cities represent place-based knowledge complexes within a system of information flows. This is a very different image of the world, an urban network instead of a political mosaic as a possible new world-city-metageography (Beaverstock et al., 2000a). Thinking in terms of inter-city relations rather than inter-state (international) relations genuinely transcends states. For example in our work on relations between world cities we are concerned not only with flows which cross boundaries (e.g. London-Paris) but also those which do not cross boundaries (e.g. Chicago-New York). Although the latter example exists within a single political jurisdiction and can therefore be identified as 'American' (in the restricted US meaning of the term), it can also be 'global' in nature. The latter occurs when a given flow is part of a corporate global strategy. A simple example will illustrate. The offices of the largest global law firm, Baker and McKenzie, are connected through 'Bakernet', an electronic communications network which operates through three regional 'hubs' in Chicago, Hong Kong and London (Beaverstock et al. 2000b). From the perspective of this firm, information flows between the Chicago and New York offices have the same status as between the London and Paris offices. This is not to say, states can be ignored (see Taylor, 2000), but the world city network does constitute a space of places5 in a space of flows, a possible alternative metageography in the making.
Seeing globalization through 'city-centric lenses' instead of state-centric lenses does produce quite different political agendas, as well as stimulating research agendas. One example will suffice to conclude this discussion. GT concludes with a fascinating paradox: just as more and more states are becoming democratic, globalization means that communities of fate are transcending states (GT, p.446). For an analysis based upon geographical scale, this spatial mismatch between political inputs and social outputs is worrying but also quite predictable. But if the world is seen as a network, simple territorial scale is no longer the problem. Democracy is ultimately about citizens, people with rights to determine the institutions within their political communities. The latter need not be states; as the word citizen suggests, they were originally cities., of course. In a new world of flows, it may be that it is the hubs within networks which provide the most sensible points of political intervention in a global system: cities for citizens once again (Douglass and Friedmann, 1998).
Agnew, J 1987, Place and Politics (Allen and Unwin, London)
Arrighi, G 1994, The Long Twentieth Century (Verso, London)
Bauman, Z 1998, Globalization: the Human Consequences (Polity, Cambridge)
Beaverstock, J V, Smith, R G, Taylor, P J 2000a, "World city network: a new metageography?" Annals, Association of American Geographers 96 ***-***
Beaverstock, J V, Smith, R G, Taylor, P J 2000b, "Geographies of globalization: US law firms in world cities" Urban Geography
Castells, M 1996, The Rise of the Network Society (Blackwell, Oxford)
Douglass, M and Friedmann, J 1998, Cities for Citizens (Wiley, New York)
Held, D, McGrew, A, Goldblatt, D and Perraton, J 1999, Global Transformations. Politics, Economics and Culture (Polity, Cambridge)
Lewis, M W and Wigen, K E 1997, The Myth of Continents (University of California Press, Berkeley, CA)
Macpherson, C B 1977, "Do we need a theory of the state?" Archives of European Sociology 18, 223-44
Mann, M 1988, States, War and Capitalism (Blackwell, Oxford)
Mann, M 1996, "Neither nation-state nor globalism" Environment and Planning A 28, 1960-4
Ruggie, J G 1993, "Territoriality and beyond" International Organization 41
Sassen, S 1991, The Global City (Princeton University Press, Princeton, NJ)
Sassen, S 1994, Cities in a World Economy (Pine Forge Press, Thousand Oaks, CA)
Sharpe, L J 1981, "Does politics matter?" in Urban Political Economy Ed. K Newton (Pinter, London) pp 1-26
Sklair, L 1995, Sociology of the Global System (Prentice Hall, London)
Soja, E 1989, Postmodern Geographies (Verso, London)
Storper, M., 1997a, "Territories, flows and hierarchies in the global economy", in Spaces of Globalization Ed K. R. Cox (Guildford Press, New York)
Storper, M 1997b, The Regional World (Guildford Press, New York)
Taylor, P J 1992, "Changing political relations", in Policy and Change in Thatcher's Britain Ed P Cloke (Pergamon, Oxford) pp.
Taylor, P J 1995, "Beyond containers: internationality, interstateness, interterritoriality" Progress in Human Geography 19, 1-15
Taylor, P J 1996, "Embedded statism and the social sciences: opening up to new spaces" Environment and Planning A 28, 1917-28
Taylor, P J 1997, "The crisis of boundaries: towards a new heterodoxy in the social sciences" Journal of Area Studies 11, 11-31
Taylor, P J 1999a, Modernities: a Geohistorical Interpretation (Polity, Cambridge)
Taylor, P J 1999b, "'So-called world cities: the evidential structure within a literature" Environment and Planning A 31, 1901-4
Taylor, P J 1999c, "Worlds of large cities: pondering Castells' space of flows" Third World Planning Review
Taylor, P J 2000, "World cities and territorial states under conditions of contemporary globalization" Political Geography 19, 5-32
1. This essay emanates from a request by Ed Brown, co-ordinator of a course at Loughborough University called "Geographies of Global Change", to consider whether GT would be a suitable text for the course. My answer - "this is a very good unsuitable book" - required some further justification ....
2. There are problems of definition with these terms as Lewis and Wigen (1997) have made clear. The eurocentrism is clear if we compare the Roman Empire with the Chinese empires, the former qualifies as transcontinental, covering parts of three continents, whereas the latter does not.
3. I will not comment on the 'sectoral' approach to globalization employed here - for a discussion where this debate is taken up in the context of multiple sectors of modernity versus a more holistic approach, see Taylor (1999a, 20-5). The authors recognise the need to bring the different dimensions together which they attempt in the concluding chapter of GT.
4. The same caveats as the authors of GT apply to their typology is relevant here; these are broad categories defining tendencies within the literature.
5. Actually Castells treats global cities as processes rather than places - for a discussion, see Taylor (1999c)
Edited and posted on the web on 5th October 1999; last update 24th February 2000