GaWC Research Bulletin 29

GaWC logo
  Gateways into GaWC


A Metageographical Argument on Modernities and Social Science

P.J. Taylor


  1. A metageography is the collective geographical imagination of a society, the spatial framework through which people order their knowledge of the world. It provides the geographical structures that constitute unexamined discourses pervading all social interpretation.1 Well-known examples of such basic spatial ordering are the traditional Chinese Sino-centric view of the world around their 'middle kingdom', the Moslem division between the House of Islam and the House of War as set out in the sharia, the medieval Christian 'T and O map' of the world with Jerusalem at the centre, and the modern society's division of humanity into a mosaic of 'nation-states'.
  2. Like all geographies, a metageography has three components: pattern, content and meaning. Of course, these are intimately intertwined in the construction and reproduction of metageographies. Relatively simple spatial patterns are given highly selective social content that is often imbued with quite sophisticated meaning. This is how they are able to straddle the boundary between myth and reality providing a grounding for both the necessary information and the necessary belief within a society. For instance, the map of nation-states provides the basic spatial ordering of 'facts and values' in contemporary social interpretation of the world.
  3. Modern society has separated the 'geo' from the 'cosmo'. In traditional societies the metageography is inseparable from a broader cosmology: Earth and universe are a single system. From the age of the great explorations, through the scientific revolution, through secularisation of politics and economics, to the contemporary materialism underpinning globalization, new metageographies have grown continuously and conspicuously independent of other-worldly ideas.2 In the modern world-system which is a capitalist world-economy, metageographies are essentially materialist products.
  4. Modern society is not a monolithic social entity, the nature of modernity varies both historically and geographically. At the most general level this is expressed through three long cycles interacting with a core-periphery structure of the modern world-system which came into being in the 'long sixteenth century' (c. 1450-1650).3 There are three 'prime' modernities associated with each historical cycle, constructed in the core and imposed on, or diffused to, the periphery.4 These are, in order of construction, mercantile, industrial, and consumer modernity. Mercantile modernity is largely the product of Dutch hegemony in the seventeenth century: trade invigorates society based upon navigation as the key knowledge. Industrial modernity is largely a product of British hegemony in the late eighteenth and nineteenth centuries: production invigorates society based upon engineering as the key knowledge. Consumer modernity is largely a product of American hegemony in the twentieth century: consumption invigorates society based upon media/marketing as the key knowledge. Each prime modernity has its own metageography.
  5. Modernity defines a society in perpetual change, a world of constructive destruction.5 Metageography is no exception to this rule. A metageographical moment marks the destruction of an existing metageography and the beginning of the construction of a new metageography. The archetypal case is the European discovery of the Americas. This destroyed the idea that the world consisted of three continents, an idea taken by European Christendom from classical Greece, and forced a fourth continent into the European geographical imagination of the world.6 Subsequent European oceanic explorations opened up seaways that eventually formed the basis of a new and modern metageography. However, unlike other historical systems, the modern world-system's cyclical reproductions have resulted in three modern metageographies each associated with a prime modernity.
  6. The three modern metageographies are named for their basic spatial structures.7 In mercantile modernity there is a topographic metageography with the oceans forming the linkages. In industrial modernity there is a centripetal metageography with the north Atlantic forming the centre. In consumer modernity there is a mosaic metageography based upon 'nation-states'. The two metageographical moments separating these metageographies are the 'industrial revolution' and decolonisation both of which fundamentally changed the relationship between Europe, where the modern world-system began, and the rest of the world.
  7. The topographic metageography of mercantilism describes a world of nodes (ports, trading stations, plantations, mines and new settlements) linked together by the state and merchant ships of first Spain and Portugal, followed by France, England and the Netherlands. The space ranged from the Philippines (via Mexico) in the 'west' to the Spice Islands, in contemporary Indonesia, (via India) in the 'east' and centred on the famous 'triangular trades' of the north Atlantic. By the mid-seventeenth century the pattern was dominated by the Dutch and epitomised in the staggered annual arrival of the four great fleets into Amsterdam harbour: the Baltic, Levant, East Indies and West Indies fleets. For these first modern Europeans, the rest of the world exists as a cornucopia for increasing their wealth, an interpretation that implies no European superiority.
  8. The centripetal metageography of industrialism describes a simple two-zone space of industrial production (initially Europe) and of industrial raw materials (the rest of the world). The industrial revolution greatly increased the needs of Europeans for new mass commodities for its industry (raw materials for production and cheap food for workers) and at the same time provided the means for satisfying it: the military balance of power between Europe and the rest of the world was also fundamentally altered. The result was an economic restructuring of the rest of the world to act as suppliers culminating in the new imperialism. In effect the world was constructed as a single great functional region centred on the north Atlantic industrial core. The subtlety of this very simple metageography was to be found in its connection to history: this was when Europe and Europeans were placed at the top of humanity's hierarchy through combining this 'colonizer's model of the world' with the Whig theory of history neatly combining Eurocentrism and progress with imperialism and racism.8
  9. The mosaic metageography of consumerism describes a world of multiple nations each bounded by their own state. Nationalism arose originally in Europe in response to the social upheaval of industrialisation and urbanisation, and then beyond Europe in response to the social upheaval of imperialism. Nations, by generating nation-states', convert subjects into citizens which presumes a political egalitarianism at first, the individual level (democracy) and, second, at the state level (decolonisation). The latter led to the replacement of progress for the favoured few by development for all. In the centipetal metageography progress was attached to just one civilisation, European, with all others identified as stagnant (i.e. not progressive). Now all could expect the fruits of modernization as long as their state took the appropriate development path. Originally this involved 'iron and steel works all round' as 'underdeveloped countries' attempted to replicate nineteenth century European development. But Americanization gradually converted political citizenship into an economic citizenship or 'consumptionship' as US multinationals redesigned the world as a mosaic of national markets for American business.9


  1. Social science as a major segment of the academy is a product of modernity. Of course, the questions asked by social scientists - on the different types and forms of relations between individuals and between them and society - are common to the cosmology of all civilisations but their particular codification as the intellectual province, 'the social sciences', is an invention of the modern world-system. However, unlike the materialist metageographies which emerged from the demise of traditional cosmology, social science did not appear with the emergence of the modern world-system.10 In fact, it is only in the twentieth century that the social sciences have blossomed in the 'frontier' between the more established disciplines of the natural sciences and the humanities. Social sciences, as their name suggests, were created to apply 'scientific methods' to the traditional subject matter of the humanities.
  2. Before the twentieth century rise of the social sciences, political, economic and social theories were developed under various auspices, borrowing pre-modern discourses such as jurisprudence and political philosophy and inventing new ones such as universal (progressive) history and political economy. The idea of a social science became popular in the nineteenth century as part of the reaction to the social upheavals of industrialisation. Reform and improvement became the watchwords in a wide range of areas: housing, health, town planning, social insurance, working conditions, unemployment, education, local government, suffrage, constitution, law. Dealing with these concerns invariably required collecting new data for official 'inquiries' which generated new social knowledge. However, recognition of the social sciences as respectable disciplines, rather than practical information, developed fully in the Anglo-American academy only after the revival of European (largely from German state universities) nineteenth century social theory within the twentieth century US universities which created the modern discipline of sociology. This invention allowed the social sciences to develop as a trinity: sociology from the earlier 'social science', economics from political economy, and political science from political philosophy.11 In each case a more specialised discipline replaced a broader field of knowledge.
  3. This imposing of disciplinary boundaries on knowledge of human activities is in keeping with the social sciences as a product of modernity. Boundaries are a common strategy for taming the incessant upheaval of the modern, like modern states disciplines mark out their domains within a threatening, changing world. Boundary -drawing presumes a mutuality. A key point about this trinity is that between them they claimed to encompass the whole modern social experience: sociology dealt with socio-cultural relations, economics with economic relations and political science with political relations. There was a shared interest in this arrangement even though each discipline would regularly 'invade' other's domains in cross-disciplinary work. With no academic space left, other disciplines which studied social change - e.g. anthropology and geography - struggled to maintain a place in the modern academy.
  4. With their roots in industrial modernity and their blossoming in consumer modernity, the trinity became imbued with 'national' thinking. This culminated in an embedded statism whereby the mosaic metageography of states assumed the role of unexamined discourse.12 This is marked by a quite remarkable yet unremarked spatial congruence in their practices: the 'society' that sociologists study, the economy that economists study and the polity that political scientists study were assumed to begin and end at exactly the same boundaries, those of the state. Hence, whether structural-functionalist or radical sociology, whether Keynsian or monetarist economics, whether pluralist or Marxist political science, for all their fundamental theoretical differences the one thing these discourses have in common is a spatial grounding in the mosaic metageography. And this is compounded by empirical macro-analyses (of whole societies, economies or polities) which rely on 'official' sources of information, appropriately called statistics (I prefer 'state-istics'). Reams of data are available for states and these are conveniently converted to quantitative models of 'society', 'economy' and 'polity'. Partly creations of the state, partly creatures of the state, the social sciences have been inherently state-centric.
  5. Embedded statism has created three key prioritisations in the practice of the social sciences. First, there is, inevitably, a geographical scale prioritisation. Apart from some minority interests in cities and regions (usually by geographers and planners) and on 'international regions' (nearly always by international relations scholars), social knowledge has been created at the scale of the state. There are, of course, good reasons to focus on this scale given the nationalisation of the state in the last two centuries. State national institutions have been created to make policy, implement it and promote it in order to construct the 'imagined community' which is the nation in its state. Hence, this might well be the prime scale for most social activities under conditions of late industrial and consumer modernities but it should not simply be assumed for all activities. In any critical discourse the spatial congruence should be treated as a hypothesis whose strength will vary under different conditions at different times.13 The social sciences have been part of the process of creating the national imagined communities instead of being critical observers of the process. This does, of course, make the social sciences particularly vulnerable to a change in modernity and its metageography.
  6. The second prioritisation is of analysis over synthesis. This has several roots. In the development of the modern university system in nineteenth century Germany, disciplines evolved by hiving off from broader areas of knowledge thus creating more and more specialisation. Within specialised disciplines, research became associated with 'solving' particular problems with only a very few 'leaders' taking care of the overall frameworks of disciplinary knowledge. It is this model which is emulated in US universities including their creation of the social science trinity. This specialisation matched the processes in the wider society where the reform movements targeted particular issues and generated practical and narrow social knowledges. Gradually, as the twentieth century proceeded, the grand syntheses of universal histories and political economies, about civilisations and cities, are sidelined as analysis of social, economic, and political patterns and processes within states becomes what social scientists do.
  7. The third prioritisation is of attributes over relations. Quantitative data comes in two forms, descriptions of objects which produce attribute measures and descriptions of relations between objects which produce relational measures. Looking at any social theory across the trilogy it would seem that empirical analysis needs to employ more relational measures than attributional ones: by definition social science is all about relations. In practice this has not been the case. States use their statistics to account for the content of their territories: they are the great modern counters. State-istics, such as in census volumes and UN yearbooks, are totally dominated by attribute data. The techniques of mathematical statistics (inferential and co-variance methods) have been developed to model relationships between variables but this is no substitute for the dearth of actual information on relations between objects. The social science paradox of developing theories about relations which are empirically evaluated using information on attributes is a classic effect of embedded statism.
  8. Imagined communities are constructed to eliminate real communities concretely expressed as human settlements. Cities have been the great victims of embedded statism and the mosaic metageography. Before the development of the social sciences cities figured prominently in studies of human relations. In the modern world-system they have been doubly erased. First, they were erased from the political map as states centralised their functions and eliminated alternative sources of power. Stripping cities of their autonomy has been a key feature of the European rise of modernity. Second, they were erased from the intellectual map by the rise of the social science trilogy. The mosaic metageography is a reflection of the modern obsession with boundaries. But boundaries are anathema to cities.14 Cities are the crossroads of society, where flows of people commodities and information meet and interact. Boundaries interrupt flows so that viewing cities as only parts of a country is to truncate them. This is clearly reflected in studies of urban hierarchies that have invariably dealt with 'national systems of cities' as if New York or London are merely the apex of their respective country's urban system. Trying to understand New York as only part of the USA, or London as only part of the UK, is to totally misunderstand the nature of cities.
  9. To understand cities it is necessary to study relations between them. The same is also true of states but such study is rare; it is as if the mosaic metageography is set in marble. Ironically, the only discipline which does deal with relations between states has defined itself explicitly as being outside the social sciences. International relations (IR) in its dominant 'realist' school argues that states are beyond normal social constraint (i.e. they exist in a condition of international anarchy). Its basic 'billiard ball' model is totally asocial: states are represented as balls (with no internal differentiation) which career inevitably into one another (producing war). Within the social sciences it is comparative analyses rather than relational studies which feature in multi-state research (i.e. comparative sociology, comparative politics and development economics). The exception that proves the rule is trade theory in economics which has been a minor part of the discipline and seems often to be more political than economic in its repetition of the free trade mantra.


  1. The latest promotion of free trade is part of the neo-liberal international agenda closely associated with globalization. But the latter is much more than economic globalization, it is cultural (global music), political (global governance), social (global classes) and environmental (global warming). Its apogee is often considered to be the financial markets which, since the demise of Bretton Woods in the early 1970s, have grown immensely to totally dwarf national bank reserves. Based upon a few major international financial centres, this has been referred to as 'the revenge of the cities'. There has been much debate on the importance of globalization with opinions ranging from its dismissal as hyped-up rhetoric to its promotion as a new global civilisation.15 I will approach globalization from a metageographical perspective: does globalization represent a new metageographical moment?
  2. The most controversial element in the globalization debate has surrounded the changing role of the state. Clearly key features of globalization such as the international financial markets impinge directly on the concept of sovereignty which legitimates the authority of the modern state. Hence the processes of globalization can be viewed as eroding the state or even heralding its demise. However this is too simplistic. For instance, it is states which have promoted the neo-liberalism in which globalization has prospered. And states remain the sole source of ultimate legitimacy within the modern world-system. Thus rather than being a portent of the end of the state, the alternative view is that globalization merely changes the mixture of activities which constitute the state. As many times in the past, today states are adapting to changing circumstances in the world economy: they are all 'competition states' now. Nevertheless the actions of states are severely curtailed under globalization, witness the concerns over the future of social welfare policies or the inability to control the internet. There seems to be a widespread belief that state actions alone can no longer solve many of the key problems facing citizens. This is reflected in the widespread adoption of the concept of governance replacing government (of the state) as the contemporary form of organised authority. My interpretation of what is happening is not necessarily an erosion of the state but it is most definitely an erosion of the mosaic metageography of states.
  3. If contemporary globalization is a metageographical moment it means that the fundamental spatial framework of our thinking is being dismantled. In other words, the demise of embedded statism has arrived. Evidence for this 'disembedding' is becoming commonplace in both popular and social science writings. Perhaps the starting point was the impact of the photographs of the Earth from space where, as one of the astronauts exclaimed, a world without boundaries was on display. In social science writing the main theme has been the replacement of the simplicity of states by a new complexity, sometimes referred to as the 'new medievalism'. Geographically, what has happened is that authority has moved from a single scale of practice to multiple scales with power moving upwards above the state and also downwards to units below the state. These two related transfers of power have many implications, here I concentrate on the movement upwards.
  4. There are two distinctive interpretations of new authority and new practices existing 'above the state': supra-national and trans-national. The former focuses upon the transfer of authority to new institutions which transcend the state. The archetypal case is the European Union wherein states have given up important parts of their sovereign powers to the larger whole.16 This may be a forerunner of a future world of several supra-national regional entities. However, notice that although such reorganisation changes the salience of boundaries within its realm, it remains a bounded territorial entity. Although not embedded statism, this maintains 'territorialist' thinking (of which statism has been the prime example). This supra-national world is still a world of boundaries, only the scale has changed. The same is true of the revival of a multiple-civilisations model of the world.17 In this 'clash of civilisations', each civilisation has its own exclusion territory in another 'boundary-obsessed' argument. It seems that even with the demise of the states metageography we cannot get rid of boundaries in our thinking, the mosaic just has larger pieces.
  5. New territorialist spatial frameworks are a symptom of the demise of embedded statism but they are a poor indicator of its replacement. For this we need to look at transnational processes. The enabling mechanism for the rise of contemporary globalization was the combination of communication and computer technologies. These allowed new enhanced abilities to project power across space in the form of instantaneous transfer of information, knowledge and command. The end result has been the creation of a global 'network society' of which the world wide web is the archetypal expression.18 This has been famously described as the replacement of the old 'space of places' by a new 'space of flows'.19 However, since the modern world-system has always operated as a mixture of spaces of places and spaces of flows I prefer to see this change as an enhancement of the influence of the latter.20 Hence the replacement is not in spatial practices but in metageography: the rise of a network metageography in place of its mosaic form.
  6. This is, of course, bad news for traditional social science embedded as it is in the old metageography. A network metageography demands a social knowledge rich in connections, linkages, flows and relations, just what research immersed in attribute measures cannot provide. But social science has not, of course, stood still in response to these changing circumstances. A new heterodoxy abounds in which 'trans-disciplinarity' is the watchword.21 Hence the breaching of territorial boundaries is being matched by a breaching of disciplinary boundaries. Like states, the disciplines are not disappearing but their role is changing. Disciplines are still strong in old institutionalised areas such as university departments and scholarly societies and are therefore able to preserve their reproduction. Heterodoxy, on the other hand, is to be found in the myriad of new social science journals which are invariably interdisciplinary or trans-disciplinary in nature. Examples of these new fields of knowledge are world-systems analysis (which combines social science with history and geography), international political economy (which combines international relations with international economics), cultural studies (which bridge the social science-humanities divide), business studies (which introduce economics to the other social sciences), and world city studies (which have largely involved geography, planning and sociology).
  7. The earliest indication of a threat to the states metageography came with the study of multinational corporations. In the 1970s, several commentators pointed out that some of the largest firms had annual sales that were larger than the annual GNP of many countries.22 These same firms were instrumental in creating a new 'international division of labour' through their global production activities. World cities were subsequently identified as the 'command and control centres' for firms in the organisation of their global business.23 More recent research has shown that it is not the headquarters of multinational corporations which have to locate in world cities, but rather the advanced producer services which are needed to conduct trans-jurisdictional global business.24 Thus, it is a particular type of multinational firm which locates in world cities: global service firms in accountancy, law, banking/finance, advertising, management consultancy, insurance and other corporate services. They need to be in these cities because their product is wholly knowledge-based and to provide the best service they have to be in the knowledge-rich environments which world cities provide around the world.
  8. World cities constitute a global network of nodes.25 These are global service office complexes concretely represented by the great slabs of offices that typify these cities' skylines. The flows between these nodes are the information, knowledge and commands transmitted between offices both within the same firm - 'seamless' global service - and between firms. Of course, this world city network is only one of innumerable networks that make up the space of flows in global network society but it is a critically important element.26 It can be interpreted as the 'skeleton' through which globalization has grown and prospered. It most certainly provides a glimpse of what social activities might be like in a borderless world.27
  9. In a metageographical moment it is relatively easy to discern the demise of the old but identifying the new replacement is a very different type of argument. Historical interpretation gives way to futuristic speculation. World cities have featured as dystopias in some arguments about globalization28 but I think this reflects the tenacity of old territorialist thinking rather than a realistic assessment of the future role of cities. At this stage many possibilities remain open and there is no reason to assume the real communities of cities should be any better or worse than the old imagined communities of nation-states. The great irony would be if cities were to replace states in a new metageography. This is the ultimate revenge of the city: the world city network as the putative network metageography. But whatever the future holds, it might well be a sensible goal of the social sciences in their new heterodoxy to aim to be at least as city-centric as state-centric in their researches.


1. This definition is taken from Lewis, M. W. and Wigen, K. E. (1997) The Myth of Continents (Berkeley: University of California Press) p. ix.

2. On the separation of geography from cosmology, see Livingstone, D. N. (1992) The Geographical Tradition (Oxford: Blackwell).

3. Wallerstein, I. (1979) The Capitalist World-Economy (Cambridge: Cambridge University Press).

4. Prime modernities are described in more detail in Taylor, P. J. (1999) Modernities: a Geohistorical Interpretation (Cambridge: Polity) chapter 2.

5. Berman, M. (1988) All that is Solid Melts into Air: the Experience of Modernity (New York: Penguin) remains the classic statement on this.

6. Zerubavel, E. (1992) Terra Cognita: the Mental Discovery of America (New Brunswick, NJ: Rutgers University Press) is the best source on this process.

7. These are described in more detail in Taylor, P. J. (2001a) 'Metageographical moments and globalizations' in M. A. Tetreault and R Denemark (eds) Odysseys (London: Routledge) (in press).

8. See the comprehensive discussion of this combination in Blaut, J. M. (1993) The Colonizer's Model of the World: Geographical Diffusionism and Eurocentric History (New York: Guilford).

9. Taylor (1999) op. cit. chapter 7.

10. For more details on this see Wallerstein, I. et al. (1996) Open the Social Sciences (Stanford: Stanford University Press) chapter 1.

11. Discussion of this 'trinity' is drawn from Taylor, P. J. (1997) 'The crisis of the boundaries: towards a new heterodoxy in the social sciences', Journal of Area Studies 11: 11-31.

12. Discussion of embedded statism is drawn from Taylor, P. J. (1996) 'Embedded statism and the social sciences: opening up to new spaces', Environment and Planning A 28: 1917-28.

13. Wallerstein, I. (1984) The Politics of the World-Economy (Cambridge: Cambridge University Press) chapters 1 to 3.

14. The discussion on cities and boundaries is drawn from Taylor, P. J. (2001b) 'Is there a Europe of cities?' in R. McMaster and E. Shepherd (eds.) Scale in Geographic Inquiry (London: Routledge) (in press).

15. See Held, D. et al. (1999) Global Transformations (Cambridge: Polity) 'Introduction' for an excellent summary of the debates.

16. This 'scale interpretation' of globalization is favoured by Held et al. op. cit., see Taylor, P. J. (2000) 'Embedded statism and the social sciences II: geographies (and metageographies) in globalization' Environment and Planning A 32: 1105-14.

17. Huntington, S. P. (1993) 'The clash of civilizations', Foreign Affairs 72 (3): 22-49.

18. Castells, M. (1996) The Rise of the Network Society (Oxford: Blackwell).

19. Castells, op. cit. chapter 6.

20. In the past the space of flows have been opaque precisely because of state-centrism, see Arrighi, G. (1994) The Long Twentieth Century (London: Verso), p. 84.

21. This discussion of heterodoxy is drawn from Taylor (1997) op. cit.

22. See, for example, Barnett, R. J. and Muller, R. E. (1974) Global Reach: the Power of Multinational Corporations (New York: Simon and Schuster), p. 15.

23. Friedmann, J. (1986) 'The world city hypothesis', Development and Change 17: 69-83.

24. Sassen, S (1991) Global City (Princeton, NJ: Princeton University Press).

25. For a formal specification, see Taylor, P. J. (2001c) 'Specification of the world city network', Geographical Analysis (in press).

26. This point is made by Castells op. cit. p. 415.

27. Taylor, P. J. (2001d) 'Visualising a new metageography: explorations in world-city space' in G. J. W. Dijkink (ed.) Festschrift for Herman van der Wusten (in press).

28. For instance, Petrella, R. (1995) 'A global agora vs. gated city-regions', New Perspectives Quarterly Winter: 21-2, and Giddens, A, (1998) The Third Way (Cambridge: Polity), p. 129.


Edited and posted on the web on 4th September 2000