GaWC Research Bulletin 279

GaWC logo
  Gateways into GaWC

This Research Bulletin has been translated into Italian and published as '(In)disciplina' in E. dell'Agnese (ed) (2009) Geo-grafia. Strumenti e parole Milano: Unicopli.

Please refer to the published version when quoting the paper.



P.J. Taylor*


This essay is about the need for indiscipline in creating knowledges of human activities. Since the late nineteenth century such knowledges have metamorphosed from broad traditions of scholarship (e.g. political economy, political arithmetic, ethnography) into strict modern disciplines (e.g. economics, sociology, political science) in the process of claiming intellectual space as chairs/departments in the emerging research-knowledge world of universities. This disciplining of social knowledges lasted through most of the twentieth century but is being seen as ever more problematic when confronted with contemporary globalization. I will argue that the late nineteenth century ‘social science' disciplines have finally had their day: hence the need for us to be indisciplined.

I develop this argument from a geographical perspective in two senses. First, the disciplining of social knowledges is itself a social activity that has a geography; social knowledges reflect this geography, and they have been instrumental in reproducing this geography. Second, as a geographer I am in many ways a social outsider in the research-knowledge that has come to be called the social sciences. Put simply, geography never really fitted into the new disciplining although it has survived against the odds in the form of academic departments in many universities in many countries across the world. I begin with a brief summary of this strange interloper among social sciences in order to indicate my particular take on geography. This is then used to make sense of the first geography above: where were the social science disciplines made? Next I trace the gradual dissolution of this disciplining in the last third of the twentieth century which leads into my argument that if we take contemporary globalization as a serious phase of social change, then we need to ditch the old disciplines. Finally I speculate about whether this indiscipline will provide new intellectual spaces for geographical knowledges.


Long before it became a university discipline, geography was associated with explorations and was organised into Geographical Societies. Its success in moving into universities in the late nineteenth and early twentieth centuries was based upon its practical utility in an age of nationalisms and imperialisms. But what was to be the content of these national and imperial geographies? The broad geographical knowledges of the Geographical Societies were only of limited use in defining geography as a modern discipline: describing the world is an insufficient definition for a credible discipline. Combining both physical and human changes on the Earth's surface, two distinctive disciplinary solutions were evolved. First, there was an attempt to create geography as a ‘science' through study of relations between the physical and human worlds leading to debates about geographical (mainly climatic) determinism. Second, there was an attempt to create geography as a ‘humanity' through the art of regional synthesis of the physical and the human in regional geography. Both solutions were credible enough in some places and times for geography to become and continue as a university discipline. But this modest success should not blind us to the strangeness of this early discipline geography.

The modern disciplining of knowledge is a product of the German-speaking university system in the nineteenth century. Starting in the natural sciences, professors became research-knowledge makers and their ‘chairs' represented disciplining knowledge into university departments. Further disciplines were generated by the creation of new professorial chairs whereby the new incumbent ‘hived off' his knowledge niche from existing departments. Thus disciplines became more and more specialised: specialization is the key process of disciplining. This is the opposite of geography's two solutions: in order to adequately research either the science or humanity geography disciplines required a very broad rather than narrow knowledge base. Geographers appeared to aspire to be the last Renaissance men! Ultimately such justifications for geography in modern academia were found to be wanting. There were two possible outcomes and both occurred: closure of geography departments and revolutionary change to specialisation. There were some high profile examples of the former but not enough to undermine the geography disciplinary revolution, which came in two phases. First, a specialism had to be identified: it was found in geography as ‘spatial science'. But this emphasis on pattern soon gave way to concern for process leading to, second, a collection of reinvigorated ‘systematic geographies'. These were geographies with adjectives describing the process they concentrated on. For instance, in human geography specialisms blossomed from the 1960s as economic geography, social geography and political geography. In other words, human geography became a social science with adjectival links to economics, sociology and political science.

This disciplinary route of geographical social knowledges from science through humanities to social science is unique and remarkable. One important effect is that geography has been essentially shallow in its knowledge production. It has no ‘founding fathers' because what was founded a century ago lost its relevance in the disciplinary revolution to specialization. This contrasts with the social sciences of economics, sociology and political science created to understand modern social processes whose founding fathers are still relevant, for instance, the ideas of Marshall, Weber and Mills are still widely debated. Thus there is a sense that their process-based knowledge is so much deeper with intellectual traditions that are a century or so old. Of course, these social science founding fathers are more likely to be referred to by late twentieth century adjectival human geographers than any geographers from a century ago, who are fascinating historically, but appear as being of little relevance to later scholarship.

My personal solution to the shallowness of geography has been to draw my social-theoretical ideas from Immanuel Wallerstein's world-systems analysis. This is not the place to describe this approach in detail but it is a late twentieth century development of ideas that are basically economic, are very political, and derive mainly from sociology. In other words this approach to social knowledge is trans-disciplinary. It is relevant to my work because it provides a materialist time-space structure that is the modern world-system. This historical system became global in extent about a century ago. Within the system times are treated as cyclical and spaces are treated as core/periphery. This framework is exceptionally useful for understanding where, when and how the social sciences came to dominate knowledges of human activities.


The concept of ‘social science' is a very powerful nineteenth century idea. With the emergence of a strong natural science body of knowledge and a contrary humanities tradition (also called arts or letters) that were to become entrenched in universities as distinctive new faculties of arts and science, social science emerged ‘in between' these established knowledges. Put most simply, the idea of social science was to use the methods of science to study the subject matter of the humanities (i.e. human practices). As we have seen, this new scientific activity came to consist of three disciplines: economics, sociology and political science, the latter explicitly including ‘science' in its name to differentiate it from the long standing political philosophy tradition ensconced in the humanities. I call them the ‘trinity' not just because they are three but because they aspire to cover all human activities between them and among them. First, all human activities are covered by one or other of these disciplines – there appears to be no intellectual space for additional disciplines – but, second, at the same time every human activity can be studied by all three disciplines. For instance, markets are the subject matter of economics but they also have social and political aspects to be studied; families are part of sociology's subject matter but there is an economics and politics of the family; and governments are clearly part of the remit of political science but who can doubt their relevance to economics and sociology? Thus the social sciences as a trinity are both divided and symbiotic. This has been, of course, a major source of their intellectual and academic power in universities.

The trinity derive from the needs of the great nineteenth century reform movements. The industrial revolution had turned, or was in the process of turning, national worlds upside down and this impinged upon all aspects of human activity. In particular, states became active in response to the social upheavals and embarked on reforming their societies. Different dimensions of the modern state were generated through reform movements: economic reforms led to the recognition of the ‘national economy' and the need to manage it; social reforms led to extending citizen rights and the welfare state; and political reforms created constitutional and suffrage changes producing democratic states. The self-ascribed rational modern reformers required new evidential and theoretical knowledges that would make sense of three distinctive spheres of human activity, the economic, the social, and the political. The new disciplines of economics, sociology and political science emerged in the late nineteenth century/early twentieth century in the new growing universities to meet these needs. Although, the industrial revolution was the basis of Britain's hegemonic power, this hegemonic cycle was in decline by the late nineteenth century and Germany and the USA were vying for the hegemonic succession: it is the latter two states that dominate the rise of the social sciences. Germany with its major universities provides most ‘founding fathers' and the USA continues the German idea of the research university including the social sciences into the mid-twentieth century. By the confirmation of US hegemony in 1945, the trinity was in control of social knowledges throughout US universities: it had become axiomatic to have economics, sociology and political science departments on campus. This arrangement was imitated by universities outside the USA through the third quarter of the twentieth century: US hegemonic power included the Americanization of university departments around the world as US scholars now led research agendas in economics, sociology and political science.

There are two related fundamental characteristics of the trinity that are important here. First, given their origins, the social sciences are not only creations of the states, they are creatures of the modern states. All macro human activities are viewed through state lenses: economics studies national economies, sociology studies national societies, and political science studies national politics. This appears all very obvious but there is a hidden geographical premise here: the nationalist assumption that nation-state boundaries bounded all critical macro human activities. This geographical congruence of economic, social and political activities (and therefore seen as all human activities) was never evaluated or tested either theoretically or empirically. It was just a given because it was what the modern nation-state was all about: containing human activities within its territory. The process of the modern state (through its agents) doing its thing, was treated by the trinity as the natural order of things. Thus the social knowledges created by economics, sociology and political science have been inherently state-centric to the detriment of all other ways of considering human activities. For instance, this embedded statism has had a severe negative effect on the macro study of cities. Thus instead of understanding cities as constituting networks in commodity chains, they have been treated as ‘national urban systems', bounded rather than networked. Great cities such London, New York and Paris are modelled as leading national cities in bounded models such as the rank size rule, their extra-national links counting for nothing in the model. Put simply, urban studies were nationally contained as if cities were constituted as a hundred plus (the number kept increasing with decolonization) separate ‘national urban systems' across the world. This territorial disciplining of urban studies is probably the most impressive example of embedded statism in social knowledges. Dynamic cities abhor boundaries.

Second, the nineteenth century reform origins of the trinity meant that initially they did not consider their purview to be universal. The concept of progress was at the heart of the reform movements – it was what they were facilitating – and progress was not seen as universal. Quite the opposite in fact: progress was the crucial distinguishing characteristic of modern western civilization. This civilization had discovered a rational scientific path to progress and the trinity was conceived as part of this special rationality. Thus the knowledges produced were restricted to the ‘west', to the core of the modern world-system in the mid-twentieth century. And this knowledge was nomothetic in nature, ‘western man' was seen as a rational being capable of being understood through formal theories and laws. But what of the rest of the world, the periphery of the modern world-system? No behavioural laws were expected here; this is the realm of idiographic study. In fact two areas of study developed to cover contrasting parts of the non-West. In the regions of former civilizations Orientalism emerged. Since these civilizations were no longer dynamic their unchanging nature could be understood through careful literary analysis of ancient texts. Where there had been no ‘civilization' anthropology emerged to study tribes through careful participant observation. And what of the European world before modernity? There is no choice, history (before the modern which is interpreted as before industrialization) has to be inherently idiographic using careful interpretation of archives. Thus by the mid-twentieth century, academic knowledges of human activities had come from six main sources: the trinity for understanding modern states, Orientalism and anthropology for making sense of the contemporary un-modern, and history to interpret the pre-modern. A final note on the misfit, geography, which ironically did not have a geography: this seventh minor source of knowledge of human activities straddled the core-periphery divide because human-environmental and regional geographies could be and were studied everywhere.

The great wave of decolonization in the third quarter of the twentieth century irretrievably changed the geography of the trinity. Decolonization created new states whose prime purpose was to modernise and develop. Self-evidently not a job for Orientalism or anthropology, the trinity was harnessed to the task. Whereas the nineteenth century concept of progress had been a civilizational property, in the mid-twentieth century such purposive societal change was renamed development and development was what states did. Enter the trinity to help what soon became known as ‘developing states'. Their nomothetic modern knowledge could now be extended and economists devised economic development models, sociologists measured modernization trends and political scientists devised democratic constitutions. By the 1960s the periphery had become collectively known as the ‘third world' and few doubted that this knowledge transfer based upon deep disciplinary brain-power would end in the promotion of these countries from poor to rich. The only dispute was the timing but most expected development to have succeeded by the end of the century.


But this optimistic knowledge transfer did not outlive the post-war boom period. In the late 1960s and early 1970s a series of upheavals - urban riots, anti-Vietnam war protests, the 1968 student rebellion, the ending of the Bretton Woods agreement by floating the dollar, the first oil price hike – ushered in a downturn in the world economy that was to undermine the trinity. Rising expectations of what the social sciences could do in the good times produced a severe reaction in the bad times. The trinity was unmasked: it did not have the answers, it was enfeebled, it lost a deal of its intellectual credibility, and a first phase of indiscipline in the study of human activities was begun.

Over the next couple of decades there were a plethora of reactions to the way the trinity had guided knowledges of human activities. The most common title for these new approaches was postmodernism but this concept does not cover all the new approaches. For instance, the world-systems analysis being loosely applied in this essay originated in the mid-1970s as a critique of the trinity's failure in development studies. Note that this particular approach has remained materialist in nature but without being nomothetic. In fact the common currency of the reactions was the rejection of the universalism that the trinity projected. Whether the reactions were radical, historical, institutional, cultural, or any of the many post-somethings of these times, the simple idea of one set of comprehensive knowledge of human activities, as claimed by the trinity, was laid to rest. Thus whereas previously a lot of lip service was paid to inter-disciplinary and multidisciplinary studies which bolstered the trinity through emphasizing their mutualities, now there was talk of uni-disciplinarity (world-systems analysis's call for a single ‘historical social science') and, more generally, trans-disciplinarity, transcending old disciplinary boundaries. But this has not meant that the disciplines themselves have disappeared. They remain as powerful institutions in world academia. The result has been, therefore, two contrasting patterns of knowledges of human activities: traditional disciplines, notably the trinity, and new fields of knowledge commonly called ‘studies'.

The traditional disciplines remained powerful because they had two crucial power locales for the reproduction of knowledges of human activities. First, they normally control the university departments and therefore the awarding of PhDs the training certificate for academia. However trans-disciplinary a dissertation might be, the candidate will typically be trained in one discipline and be awarded her or his doctorate in that discipline. He or she comes on the labour market as a ‘sociologist' or ‘political scientist' or with some other disciplinary label and will be employed by a department of that discipline in a generations long process of reproduction. Second, the disciplines have the earliest academic journals and these scholarly tomes are usually the most prestigious places in which to publish. In this way new ideas, including trans-disciplinary ideas, are often incorporated into disciplinary reproduction. In addition the disciplines have the largest and most established academic conferences where both ideas are traded and hirings made. If nothing else we can expect the disciplines, especially the trinity, to be resilient to change.

But there is a challenge; since the 1970s ‘studies' have blossomed and multiplied: business studies, communication studies, cultural studies, media studies, organization studies, post-colonial studies, urban studies, women's studies, to name just some of the most well known. The key point is that none of them can be captured within the boundaries of a discipline. And these new knowledges of human activities are reproducing themselves in two guises like the disciplines. First, since the 1970s there has been a massive growth in new journals, usually launched by private sector publishing companies, which cater for adherents of the new studies. These new journals have soon outnumbered the traditional disciplinary on university library shelves. Second, informal seminar groups and networks, sometimes linked to the new journals, have proliferated since the 1970s and more lately have been especially facilitated by internet communication. Often transient and temporary, nevertheless it is here that the excitement of generating new knowledges is to be found, not in the old staid disciplines.


One important feature of the new studies has been that the knowledges they produce are no respecters of state territorial boundaries. The old geographical congruence in knowledges of human activities is conspicuous by its absence in the various ‘studies'. As such they have contributed to ideas in globalization that emphasize the downgrading of the state's importance. But their contribution in no way eliminates trinity thinking on globalization. The simplest way of conceptualizing globalization is as an ‘up-scaling' of human activities from the state scale of analysis to the global scale. This can lead to quite lazy and unimaginative thinking about globalization: at the state level there was nation economy, national civil society and national government and for globalization these are simply transmuted into global economy, global civil society, and global governance. The trinity is dead; long live the trinity!

I am going to take contemporary globalization more seriously than such unsubtle sleights of hand. From a world-systems perspective globalization is the form that the down side of the American hegemonic cycle has taken. But it is more than this because the American hegemonic cycles marks the concluding phase in the modern world-systems' ordinary reproduction processes. The twenty first century is interpreted as defining the demise stage of the historical trajectory of the system. In this situation the ways of social change will be distinctively different that what went before. Thus the ordinary societal reproduction processes, rather than being the trinity's natural order of things is actually an unusual era we might look back on as the ‘modern interlude' between times of change. This means that the trinity's deep knowledge is knowledge of a past interlude when states dominated human activities in a manner that was historically quite unprecedented. Such knowledges of human activities – economics, sociology and political science – will have little to offer in the new dynamism where networks are more important than boundaries. We will need ‘network knowledge' rather than ‘territorial knowledge' but most of all it will have to be strategic knowledge, flexible and adaptable to rapid social change. Such knowledge will perforce be relatively shallow in nature, always moving on. Indiscipline will have to be the order of the day.

And so we return to geography, in terms of both contemporary concern for spaces of flows and traditional concerns for spaces of places. With respect to the former, it would seem that in the twenty first century strategic knowledge will surely be network knowledge with cities as foci of change. With respect to the latter, we should remember that despite its disciplinary face, required to access the university, geography has always really been a field of study. This locates it closer to the new studies than the old disciplines. Thus it comes as no surprise that traditional geography, or something very much like it, can and should be another major source for transient strategic knowledges. We have come to a time when specialization of knowledge of human activities must be superseded by more synthetic thinking: overviews of myriad interactions of processes are going to be necessary knowledge formats in our new global, network, risk society. Traditional regional geography was all about synthesis, of course, and something along these lines will have to be reinvented to begin to understand the resilience of places in the twenty first century. And past environmental geographers are now allowed a quiet ‘I told you so' response to twenty first century predicaments. They used climatic determinism to try and understand the beginnings of civilization; climatic determinism might well come to fruition in the end of civilization. Clearly indiscipline will have to extend beyond knowledges of human activity: the gap between Snow's ‘two cultures' in the research-university world will finally have to be transcended.


I have written this essay as a personal memoir and therefore have not littered it with references. It is, in fact, a bringing together of some ideas I have previously published in several different places. These sources are listed below and can be referred to for further references.

Lee R, Martin W J, Sonntag H R, Taylor P J, Wallerstein I and Wieviorka M (2005) Social Science and Social Policy: from National Dilemmas to Global Opportunities Paris: UNESCO (MOST)

Taylor, P J (2004) World City Network: a Global Urban Analysis London: Routledge

Taylor P J, Watts M J and Johnson R J 2002 “Geography/globalization” in R J Johnston, P J Taylor and M J Watts (eds) Geographies of Global Change; Remapping the World for the Twentyfirst Century Oxford: Blackwell

Taylor, P J (2000) "Embedded statism and the social sciences 2: geographies (and metageographies) in globalization" Environment and Planning A 32, 1105-14

Taylor , P J (1999) Modernities: a Geohistorical Interpretation Cambridge: Polity; Minneapolis: Minnesota University Press

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 (1996) The Way the Modern World Works: World Hegemony to World Impasse London: Wiley

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 (1993) "Geography at fin de siecle - full circle or new meaning for the global? in R J Johnston (ed) Geography:Changing World, Changing Discipline London: Routledge

Taylor, P J (1989) "The error of developmentalism in human geography" in D Gregory and R Walford (eds) Horizons in Human Geography London: Macmillan

Taylor, P J (1989) "The world-systems project" in R J Johnston and P J Taylor (eds) World in Crisis? Oxford: Blackwell

Taylor, P J (1985) "The value of a geographical perspective" in R J Johnston (ed) The Future of Geography London: Methuen

Wallerstein I, Juma C, Keller E F, Kocha J, Lecourt D, Mudimbe V Y, Mushakoji K, Prigogine I, Taylor P J and Trouillot M-R (1996) Open the Social Sciences, Stanford, CA: Stanford University Press



* Peter J. Taylor, Department of Geography, Loughborough University, UK. Email:


Edited and posted on the web on 14th August 2008

Note: This Research Bulletin has been translated into Italian and published as '(In)disciplina' in E. dell'Agnese (ed) (2009) Geo-grafia. Strumenti e parole Milano: Unicopli.