The Fate of Social Science
Peter Abell

Editors Introduction Anthony GiddensIn the article "The Future of Social Science" sociologist Anthony Giddens set out his vision for the future of the social sciences, writing that "social sciences must deliver enlightenment. There has to be a role for engaged theoretical reflection." It is precisely on this front that Professor Peter Abell (right), challenges him in this feature.

Abell expresses his concern about the popularisation of a "social theory" favoured by academics of Giddens's ilk. He warns that the convenient application of user-friendly labels could engender an easy disregard for real "sociological theory" and for a proper understanding of social, cultural and political mechanisms.

I share with many other social scientists the view that much of what is currently termed social science--particularly what is usually called "social theory"--does not pass muster. So entering a debate about the future of social science is, for me, quite tricky. It is difficult to broach the issues without running the danger of offering offence--something I dislike doing. Notwithstanding, these issues are of such great significance that they need a public airing. Indeed, I am giving voice to views which are often expressed in private but are rarely committed to paper. I hope in so doing I am contributing to an open and constructive debate.

A reply to Anthony Giddens
In the article "The Future of Social Science" Anthony Giddens highlights what he terms "the four great debates"--globalisation, technology, everyday life and economic redistribution and justice. He claims that one cannot "really be a social scientist ... without situating yourself in some way in these great problems." I have considerable sympathy with his message, but I daresay that others would provide a different list of great debates; for example, demographic changes, environmental issues and the socio-bio-cognitive understanding of the brain. But my argument is not so much to do with whether his list is a good one or not but, rather, with the intellectual orientation he brings to it. This, I hasten to add, is not so much a matter of Giddens's personal orientation but of the broader perspective of "social theory," of which he is a major proponent, and which inevitably colours his remarks.

What I mean by "social theory" will, I trust, become evident as I proceed; I want to draw a sharp distinction between this sort of "theory" and sociological theory proper.

There is no doubt that "social theory" attracts all the trappings of conspicuous success both within sections of academia and also in the wider world. One need look no further than the remarkable career of Anthony Giddens himself. But, to name just a few others, Manuel Castells, whom Giddens quotes with such approval, has with his book The Rise of the Network Society attracted a great deal of favourable attention, as have many of the French social theorists, like Pierre Bourdieu, who is equally celebrated.

For social scientists like myself, this is a puzzle, because the sort of theory they proffer does not to any significant degree contribute to the progress of either empirical or theoretical social science. Indeed, I would go so far as to say that because of its ever-seductive terminology, and in spite of its lack of rigour, it offers a net impediment. It diverts attention from serious matters. If so, how can this be and whence the applause? In fact, we need a sociological theory of social theory which provides some explanation of the demand for, and thus supply of, the genre, despite its tangential relationship to rigorous social science. But before all that I need to show how it is that "social theory" fails to engage with the social sciences as I, and those who think the same way I do, understand them.

The great globalisation debate
Let us start with Giddens's comments about the social sciences. The social sciences he describes "deliver enlightenment" (sure, yes, if possible); "engage in theoretical reflection" (yes, but of what sort?); "bring scholarly disciplines into popular and public consciousness" (yes, but making sure we sharply distinguish between the two); and "embrace the norms of scientific rigour" (yes, but what are they?).

Perhaps the easiest way to come to grips with the nature of "social theory" embodied in his paper is to consider just one of his great debates--globalisation. So what does Giddens tell us in his many works about this issue? It is a complex concept that is not just confined to "economic or financial markets" but also concerns "institutional transformation" and is driven by "new systems of communication rather than economic exchange." Furthermore, it may be reshaping the nation-state. One may perhaps cavil at some of this, but no doubt a little thought would reveal many more things which might in some loose sense be associated with the concept. (I recently gave a group of students at the University of Chicago five minutes to come up with some of the likely socioeconomic consequences of changes in information technology. They delivered more than 30.) The point being that most of us as reasonably well informed observers of contemporary society are capable of doing this sort of thing. No doubt some are better at it than others, but as Giddens opines, the newspapers are full of it. What remains unclear is precisely what "scholarly discipline" is revealed by putting such commonplace observations through the mill of social theory. To paraphrase Giddens, it is all "public consciousness" and little "scholarly discipline."

Although it may serve some useful purpose to coin the word "globalisation" in order to more effectively focus our attention upon a number of socioeconomic processes or mechanisms which conceivably may be connected in rather intricate ways, the word does no more intellectual work than to alert us to this possibility. Certainly, in the absence of any analysis of these mechanisms it makes no sense, from a social scientist's viewpoint, to speak of the "consequences of globalisation." To do so is to engage in little more than social commentary which at best has only a tangential relation to any recognised social science. If we are to "deliver enlightenment," "engage in theoretical reflection" and "embrace the norms of scientific rigour," we have to completely change gear and engage in serious analysis of the interrelated mechanisms.

Although I believe it is essential that we should put the founding fathers of social theory firmly behind us (or treat them as economists treat Adam Smith), a historical parallel might help to make the point. Consider Marx's now largely discredited (but perhaps more accurate in his own time) theory of capitalism. Marx certainly wrote of the consequences of capitalism (and for the present purposes we might see the word "globalisation" as playing an equivalent role to that played by the word "capitalism"). He did so, however, on the back of a detailed (and for his time highly technical) exposition of the complex mechanisms he supposed were driving capitalism. Furthermore, the legacy of his analysis still controversially motivates rigorous and technically sophisticated socioeconomic theory today; for instance, John Roemer's The General Theory of Exploitation and Class.

It is this rigorous social-scientific analysis of mechanisms that is entirely lacking from (specifically) treatments of globalisation such as David Held's A Globalising World and Castells' The Rise of the Network Society and more generally from "social theory." Why this sorry state of affairs should have arisen, I shall return to later.

Labels and concepts
The future of the social sciences cannot rely upon "social theory." The coining of terms and listing of trends is no substitute for the technically demanding job of engaging in rigorous theoretical formulation. Much is, of course, going on in this respect concerning international trade, capital (debt and equity) flows, the market for corporate control, normative and institutional change, faltering national jurisdiction and many other things. Each of these areas of study develops theories in its own right, designed to reveal the inner workings of complex mechanisms. It might be thought to be helpful to bring them together in terms of a "portmanteau" concept, like globalisation, describing the evolution of a global world (cf Marx). This seems in some sense to be what Giddens is driving at and, certainly, what Castells is after.

The intellectual challenges in achieving this objective in a rigorous manner are, however, immense, and they fall way beyond the style of listing trends which might or might not reverse and dreaming up possible consequences. Further, things are not advanced by an infusion of eye-catching but ultimately unintelligible words and phrases like "nonlinear history." Indeed, one of the most disconcerting aspects of the writing of social theorists is their easy resort to ill-specified, technically sounding terminology, in what I can only assume is an attempt to add to the scientific standing of their prose. Castells, for instance, is replete with such terms and phrases.

Giving a name to complex interconnected sociohistoric mechanisms is, of course, a temptation which can befall us all, but building associations without a deep understanding (empirical and theoretical) of the subprocesses at work rarely helps. As we all know, that can lead us to classify whales as fish. I am not necessarily arguing for secure microfoundations--however attractive this might be. Notoriously, for example, Darwin hit upon the macro theoretical mechanism of natural selection whilst embracing the incorrect micro (genetic) theory--transmission through blending rather than through particulate Mendelian mechanisms. It may be possible to establish sui generis macro mechanisms (the evolution of global society?) that are consistent with a wide range of alternative micro assumptions. The skills required to theoretically model these are no doubt many, as they invite an understanding of the intercalation of many submechanisms which are of a strategic nature. But "social theory" as currently practised can at most exert only a marginal impact upon this endeavour.

The need for sociological rigour
Let me try to suggest in the briefest outline what would need to be achieved. We might, for instance, taking nation-states and global cooperations as strategically interacting players, wish to understand the impact of trade flows, debt and equity flows upon the changing regulative jurisdictions of states (i.e., changing institutions). These selfsame changes will then feed back upon the flows. If so, where might we be going? This is where theory proper starts. Theorists will have differing views about how to begin to model these issues, but to do so cogently will involve, in my view, a technical grasp of network theory (something that Castells, incidentally, has no real grasp of, despite his title), both cooperative and non-cooperative game theory, and system dynamics (the latter with apologies to economist readers). The theoretical model grows out of a technical grasp of these sorts of frameworks. I may have the wrong ones, or there may be others, but the point is that one cannot make a serious contribution to "social science" without something of this sort.

Is this, however, an argument for killing social theory off? Why not allow it to flourish alongside attempts to establish what, for the want of a more general term, I shall call sociological theory? Indeed, for many years I took precisely this view, believing that eventually competitive forces would drive social theory to the wall. So far, I was wrong--for, if anything, the evolving story goes the other way round. Social theory has not only come to dominate sociology but is now making incursions into other disciplines. So lawyers, geographers and others seem to be at ease with it. Why should this be so, if it does not reach any recognised standards of "scientific rigour"? Thirty years or so ago, when I first entered sociology, there was a distinct promise, particularly in the US, that rigorous sociological theory might be about to gain a firm foothold and displace the established sterile textual debates about the classical authors. This would prove a false dawn, as nonrigorous social theory progressively swept the field. At that time the rapid expansion of sociology disproportionately attracted those who were referred to as possessing an "arts orientation" (you may recall the two cultures debate raged at this time). The burgeoning population of sociologists demanded, and were increasingly supplied with, a nontechnical approach to the discipline.

Sociologists were "trained" with minimal exposure to either statistical or formal reasoning, and social theory expanded in importance to complement this intellectual deficit. Indeed, if one inspects examination questions set under the heading of "Sociological Theory," there is little evidence of the need for sustained reasoning at all; rather, students are invited to critically compare different "theorists." This usually amounts merely to knowing what two or more social theorists have opined. The paramount problem with social theory rests with its failure to provide a systematic framework within which one can reason and solve problems about complex mechanisms. And this has arisen because of the wilful evasion of the technical rigour which is inevitably involved in any serious theorising.

Thomas Kuhn's concept of paradigm has been put to effective ideological use by many social theorists in an attempt to justify their evasion. They avidly embraced the idea whereby the social sciences (notably sociology) are inherently multi-paradigmatic and thus, apparently legitimising their activity as but one among many, all equally demanding of our attention. So social theory was in demand, supply was forthcoming and legitimised, students became teachers, publishers colluded and, to use one of their own terms, social theory was "reproduced." Supply and demand came to happy equilibrium, driving out what I have termed sociological theory proper. But why doesn't the latter disturb the equilibrium?

Surely, if genuine sociological theory attempts to rigorously elucidate the nature of complex social mechanisms, it should eventually displace social theory in the rough-and-tumble of competitive evolution. The reasons this has not happened are twofold. Firstly, advances in sociological theory are slow and laborious, demanding attention to the inherent technical problems. Secondly, and more importantly, social theory in fact shifts the objectives of social inquiry (and, thus, criteria of evolutionary success) away from the causal explication of complex mechanisms to a form of social commentary and speculation wrapped in a technically sounding vocabulary ("nonlinear history," "reflexivity," etc.). Students like this, they want this. They ask for no more, and pack lecture theatres.

We might, however, note that the apparent legitimisation of social theory as a paradigm is in fact misleading. Paradigms as Kuhn envisaged them do compete, and eventually one displaces another. Allegiance to alternative paradigms is, however, usually based upon mutual comprehension--proponents of the alternatives, whilst rejecting each other's standpoint, fully comprehending the opposing position. Social theory is not like this. Because it is formulated in a manner which is intended to pass around the implications of technically rigorous social science, it is not in Kuhn's sense a paradigm. It is, rather, a fugitive framework.

I can bring this point out by considering Giddens's mention of "three principles" to be observed in the treatment of the future: the future or "history will surprise us," it is "non-linear" and our predictions might be "reflexive." Take the last first. This is a well-established idea whereby the conceptualisation and/or predictions about social phenomena, proffered by the social scientist, may have a causal impact upon the behaviour of those towards whom it is directed. Giddens quotes Robert Merton (by now largely a historical curiosity) in this respect and gives us a couple of homilies of which, again, most of us are aware. I don't think anybody who is really versed in the literature on these matters (e.g., the theory of responsive sample spaces and the game theoretic interpretation of the effects of predictions) could offer a treatment of the issues in these terms, particularly to an audience of social scientists in a leading social-science institution. To put it succinctly, Giddens's treatment is not a "reflexive" response to the technically rigorous literature which impinges upon these matters. And this, unfortunately, is true of social theory in general.

Conclusion: A future of surprises?
The observation suggesting the future will surprise us is hardly surprising. Who denies this in contemporary social science? It is generally recognised that any predictions we may make are innovation-vulnerable--some more than others--and there is considerable technical literature about this. Indeed, as many social commentators have observed, nobody predicted the precise timing of the Soviet demise, though a number of scholars did indicate its vulnerability. It is, of course, an intriguing question as to whether such rare large-scale complex events are ever, in principle, predictable. Again, there is a technical literature on these matters which has to be mastered to enable one to advance ideas in a cogent way. As to the nonlinear nature of history, it is, I am afraid, like many terms that "social theorists" deploy, impossible to be clear about what is implied. Things will have to change.

Copyright The London School of Economics and Political Science.