Tuesday, 30 July 2013

[Guest View] Scientism in Western Culture and the Special Case of Economics -- by Gavin Kitching

A Review Essay on Rupert Read, Wittgenstein among the Sciences: Wittgensteinian Investigations into the ‘Scientific Method’

Gavin Kitching

As my subtitle indicates, the thoughts that follow emerge from my engagement with Rupert Read's recent book, Wittgenstein Among the Sciences. Central to that book is Read's contention that western culture is saturated with what he calls ‘scientism'. Scientism is here to be understood as the view that only science can provide valid knowledge of the world, and Read argues that it is also present in philosophy itself. It is sometimes (not always) combined with the view that there is a single scientific ‘method’ that provides such knowledge. (For a brilliant discussion of scientism in philosophy and some of its consequences, see Michael Hymers, Philosophy and Its Epistemic Neuroses)

I think that this argument is not only powerful and important, I think it is true, and I wish to devote this review article to saying why it is true in the field of economics, a subject on which Read touches in section 2.4 of the book (in a very good critique of Milton Friedman’s influential paper ‘The Methodology of Positive Economics’), but a subject about whose scientism I think there is a lot more to be said than Read offers in that section. However, in trying to formulate at least some of that ‘more’ I have found myself drawn beyond economics, and back into the broader terrain of Wittgenstein Among the Sciences. For in considering the enormously varied, sometimes contradictory, but cumulatively and massively ‘overdetermining’ reasons why economists (and others) want what they do to be seen as a science, I have found myself considering questions of politics and of public policy more generally, questions on which Read touches, especially in his concluding interview with Summers, but does little to develop. Also, in anatomising the pursuit of a spurious  ‘normal science’ of economics as a defence mechanism against the very recognition of its endemically unstable and revolutionary (i.e.unscientific) nature, I have found myself drawn back to Kuhn, and to the importance of the Read/Kuhn insistence that pursuing unanimous, uncontentious. ‘puzzle solving’ within a paradigm, and doing so for long periods of time, is a hallmark of all true sciences. And this itself explains, I think, both why many economists try to maintain that pursuit in the face of endemic paradigm challenge, and why their attempts always fail.

Above all, however, in considering the deep and multi-layered scientism of economics, I have been led to the politics of Wittgensteinian philosophy itself. Rupert Read, like myself, belongs to that group of revisionist Wittgensteinians who insist that Wittgenstein’s philosophy need not commit one either to political quietism or conservatism, indeed can do much to deepen one’s radicalism through a frank recognition that it relies as much or more on one’s will and morality as on the facts of the world. However, even though I take that view I have always been troubled by the insistence of D.Z. Phillips and others that philosophy cannot underwrite or justify any set of political ideals or actions. That while philosophers may, as individual citizens, pursue environmentalist causes, or criticise capitalist greed, or fight for (or against) censorship of the internet, they cannot do so any better (or any worse for that matter) than any other citizen on the basis of their knowledge of philosophy.

I now think this view is wrong, and it is wrong because its conception of politics is too narrow, relies too much in fact on a one-sided diet of examples. It is true that a Wittgensteinian understanding of language does not equip one to be any better a campaigner for (or against) nuclear energy, or for (or against) a financial transactions tax, or for (or against) cuts in social services, than any other citizen. But it does equip one to critique the conceptions of meaning and action that are at play in standard economics discourse. And it does equip one to demonstrate that the distinction between the humanities and the sciences is not, as Read says, a distinction of ‘subject matter’ but a distinction of ‘subjectivities’, i.e. a distinction between ways of knowing, a distinction between how people act (and thus know) in different contexts. In fact Wittgensteinian philosophy can provide a deep philosophical defence of the humanities as a specific type of intellectual practice and (thus)  encourage and reinforce the efforts of those who wish to throw back the ever more  insistent attempts to ‘social scientise’ humanities disciplines. In a word, philosophy, and specifically Wittgensteinian philosophy, does equip one to engage in certain forms of academic politics. Indeed I would go so far as to claim that, insofar as it is true, it obligates one to so engage. If it is true that the vocabulary of emotions is fundamentally misunderstood if taken purely as a description of measurable ‘mental’  or ‘psychological’ ‘states’, if it is true that the vocabulary of material wants and desires is fundamentally misunderstood if taken purely as the expression of measurable individual ‘preferences’, if it is true that our vocabulary of ‘fundamental human rights’ is fundamentally misunderstood if taken purely as descriptive of certain intellectual convictions, then these truths must affect one’s attitude to the curricula and pedagogy of Departments of (respectively) Psychology, Economics and Politics.

And in the case of economics at the moment, critique of the intellectual discourse can have a far wider political and policy resonance. For in the midst of the most severe crisis of capitalism since the 1930s, economics as a discipline has moved from its normal state of subdued paradigm disputation to intense conceptual and policy ‘warfare’. At the present time critics accuse proponents of ‘economics orthodoxy’ of perpetrating a ‘Zombie’ discipline of ‘dead ideas’ that have led directly to the current crisis and the suffering it is causing, while its defenders accuse their opponents of being hostile, not just to the discipline of economics but to ‘the market economy’ itself and to the ‘freedom’ and ‘prosperity’ it brings. This is hot, heavy, ideological stuff, the kind of stuff which finds many academics (including many academic economists) keeping their heads down. But it is also the kind of stuff, and the kind of moment, which encourages the deepest conceptual ferment and openness. At this moment a few economists (a very few!) are even prepared to debate the question of if, whether, and how, economics could, or should, be considered a ‘science’ at all. And if that is not a moment for Wittgensteinian intellectual/political intervention, I do not know what is.

In that spirit I present below a detailed examination of the role of scientism in economics, an examination which aims to assess its consequences for economics as a discipline. This examination owes something to the explicit views on economics in Wittgenstein among the Sciences, but more importantly it aims to instantiate the Wittgensteinian method for the identification and study of scientism which Read lays out in the book, and which is, I believe, its most significant contribution.

In applying Read’s method to economics in a more comprehensive way than he himself does I hope to show how a practice-focused Wittgensteinian approach to this issue helps one to avoid dogmatic critique of economists’ scientism – can even allow that it serves some legitimate purposes – and can thus facilitate and encourage the efforts of those economists who wish to ‘open’ their discipline methodologically. For dogmatic critique tends to provoke equally dogmatic defences of the status quo (and indeed currently heated debates within the economics profession are, I believe, doing just that). And it is just in these situations that well-informed outsiders have a productive role to play.

Scientism in Economics

My aim is to assess the consequences for economics as a discipline (for practices within that discipline) of the claim that economics is a science. My approach will be broadly ‘consequentialist’, I will describe a range of intellectual practices engaged in by economists, practices which they take to exemplify and embody ‘the science of economics’ as they understand it. I will also describe the ways in which such practices are methodologically and epistemologically justified as scientific, within the profession. I will begin with those practices which seem to me to most persuasively vindicate economic scientism and then proceed to those which seem notably less persuasive. Finally I shall consider certain practices of non-economists (notably those of politicians and other state policy makers) which tend to reinforce economic scientism externally.

Practice 1: Focus on Mass or Aggregated Human Action

This is the aspect of economists’ activity most consistently missed or underestimated by outsiders to the discipline, and for a very good reason. Many introductory textbooks begin (as innumerable critics note) with postulated ‘economically rational’ individuals buying and selling in a market. But epistemologically and methodologically speaking these ‘individuals’ are nullities. They are not actually ‘individuated’ in any way (physically, psychologically, sociologically, culturally) and their specifically economic activity is always mass activity – ‘producing’, ‘consuming’ ‘buying’, ‘selling’, ‘saving’, ‘investing’ etc. The only genuinely individual behaviour postulated for these entities is response to price signals, and even this is conceptualised and measured en masse. I begin by noting this ‘aggregative’ intellectual practice in economics because it is one of the most persuasive, and pervasive, supports to scientism within the profession. Economists, (like sociologists, and many historians) are concerned with the consequences of mass human activity, and consequences conceived as radically removed from individual intentions, desires, preferences, precisely because of their mass or aggregated nature. Indeed economists generally take it that one of the roles of their ‘science’ is to explain how the consequences of mass economic activity may diverge from the intentions of (in different cases) some/many/all of the ‘economic actors’ involved. This does not mean that there are no descriptive or explanatory costs for economics in its exclusive focus on the aggregated activity of spurious (i.e. non-individuated) individuals. As we shall see, ignoring the psychological, sociological, cultural identities of actual economic actors seriously prejudices the predictive accuracy of economics, while any attempt to seriously incorporating such variables makes actual economic behaviour radically indeterminate. But grasping the substantively (substantively, not formally) ‘aggregative’ focus of economics does help one better understand why economists (economists specifically) have so little methodological objection to analogising the economic effects or consequences of human  activity to the physical effects or consequences of natural forces. A change in (say) the magnitude of investment consequent upon an increase in the savings ratio can be analogised to (say) a change in the volume of a gas consequent upon an increase in its temperature, because the former seems to have ‘almost’ as little to do with individual (or even small group) human volition as the latter does. Along with many other people I would argue that appearances are deceptive here, and that the ‘almost’ above conceals a radical epistemological difference. But however that may be, it is important to understand that the kind of critique that Wittgensteinians routinely make of scientism in psychology or cognitive science cannot be extended to economics in any direct way, because, initial appearances to the contrary, economists simply have no interest in what individual human beings think, feel, or indeed do.

Practice 2: Conceptualising Consequences or Outcomes as ‘Things’ or Objects

This practice follows directly from 1 above. In economics only mass human activity matters and it issues in consequences or outcomes – ‘production’, ‘consumption’, ‘saving’, ‘investment’ etc - which, once they have taken these noun forms, are readily conceptualised as ‘abstract’ (albeit ‘material’) ‘things’. So much is obvious. Rather less obvious is that the attraction of this form of ‘objectivist’ conceptualisation is considerably increased for economists (for economists specifically) by the fact that each and every one of these outcomes or consequences is simultaneously a monetary magnitude. Thus ‘investment’ can not only be theoretically postulated as a ‘thing’ (in algebraic equations say), in ‘real economies’ it can be empirically measured (as a sum of money) to a high degree of accuracy. In short all the ‘things’ that economists are concerned with are monetarily measured or measurable ‘things’ (though not all the ‘forces’ they are concerned with are monetarily measured or measurable ‘forces’ – see below). But to understand  that economists deal exclusively not just in the consequences of aggregated human activity, but in a particular sub-class of such consequences that can be monetarily ‘valued’ and ‘measured’, is to better understand the appeal of scientism in the discipline. Once conceive the total of investment and the savings ratio as 'things' whose compositions and magnitudes have little or nothing to do with individual human volition then why not conceive a fall (or rise) in the latter as a ‘force’ acting ‘causally’ on the former? Why not conceive it in a manner analogous to the way in which (say) the rise and fall of air pressure in a balloon acts ‘causally’ on its altitude?

There are many answers to these ‘why not?’ questions, and perhaps the most significant of them, as Rupert Read indicates, is to do with the way in which descriptions of economic behaviour may (in certain crucial situations) affect that behaviour, a way which immediately problematises any analogy of economic ‘causality’ to physical causality. But that is not the point I wish to stress here. My aim rather is that readers understand economics well enough to feel the force, and appeal, of these scientistic analogies within the discipline, and their apparent ‘reasonableness’.

Practice 3: Axiomatic Deduction and Mathematical Modelling

Once the consequences of mass economic activity are conceptualised as ‘things’ with measurable monetary magnitudes, it becomes possible to project them into logical space and to examine their ‘relations’ under various assumptions, a process much facilitated by symbolising both the ‘things’ themselves and their ‘relations’ algebraically. From the late nineteenth century onwards, ‘normal science’ in economics, was, as Read correctly intimates, not so much empirical investigation of the economic world, as mathematically symbolised logical explorations of radically simplified ‘economic relations’ under varying initial assumptions. With a two crop agricultural ‘economy’ using varying inputs of ‘labour’, ‘water’ and ‘fertiliser’, and assuming certain technical input/output relations in the crop production functions, what is the optimum yield of both crops obtainable from the least cost combination of the three inputs? Technical papers solving such puzzles dominate the professional economic journals now as they have done for over a century, and to that degree economics appears to pass the Kuhnian ‘normal science’ test. More germanely, and whether the economists in questions know about Kuhn or not, the widespread use of moderately advanced mathematics throughout the discipline, and an equally widespread disciplinary consensus on what constitutes good, routine ‘analytical’ work in economics, all further strengthen economists’ conviction that they are engaged in some meaningfully scientific enterprise.

More recently (i.e. in the last forty or so years) the development of high speed computers with massive data storage and processing capacity, has allowed economists to build economic models which attempt to simulate ‘real economies’, i.e. to construct computer models in which the number and range of economic variables and relations is supposed to at least approximate the complexity of real economies and to which real economic magnitudes can be fitted. The ultimate point of such model building is to attain the holy grail of positivist economics – prediction. It is hoped that with sufficiently complex models fitted with sufficiently exact and robust data it may be possible to predict the future trajectory of economies under a range of realistic conditions. To date however achievement has not matched hope. This is not so much because computer models cannot produce accurate predictions (along with many inaccurate ones) of the trend of some particular variable. (The rise in house prices say, or in land values). It is because all such models seem radically unable to predict the ‘thing’ most desired – i.e. significant systemic malfunctions and crises. And it turns out ‘of course’ that the failure of even the most complex and powerful models to predict what policy-makers most want predicted derives, at bottom, from the fact that, the explanation of mass economic activity and its consequences cannot proceed in total abstraction from the attributes, (intellectual, psychological, emotional) of the individual human beings (the real ones) who compose that mass. Or, to be more precise, explanations that proceed on this basis only work, even in a limited way, when the behaviour of economic actors is ‘stable’ in ways that make it (limitedly) predictable. But of course times of crises are precisely times when such stability disappears or is much reduced. However, in this matter too I am less concerned that readers should feel the weaknesses of economic scientism (however real and profound) than its strengths. Epistemologically defensible, and therefore ‘reasonable’ objectification, exact measurement of the objects so identified, mathematisation, computerisation, a certain form of ‘normal science’ maintained for over a century. Economists, I want to emphasize here, have grounds for believing that if theirs is not a ‘science’ of society then (truly) nothing is.

Practice 4: Analogising Mass Human Activity to a Natural Force or Forces

I have already touched on this above, but now wish to consider it in detail. We have seen that by conceptualising certain  consequences of mass action as ‘things’ or ‘objects’, economists are then enabled to trace ‘causal relations’ between and among them. (In considering Practice 2 above, we gave the example of treating the ‘relationship’ between the savings ratio and the volume of investment as logically analogous to the ‘relationship’ between the air pressure in a balloon and its altitude). I also noted then that, while all the objects so conceptualised can be measured as monetary magnitudes, the scientistic analogy of the mass human action that produces such magnitudes to some form of natural force is much more problematic for economists. What is problematic is not that the increase or decrease of such magnitudes cannot be measured monetarily. Of course they can, and routinely are, - usually as a ‘rate’ or percentage. What I mean is that when one thinks through how a given volume of investment is ‘produced’, how a given savings rate is ‘achieved’, how a certain volume of savings is ‘converted’ into a certain volume of investment, how a certain level of output ‘increases’ (or ‘decreases’), it immediately becomes apparent that all of these happenings are the result of mass human activity, and would not occur without that activity.  But the effect of using subjectless, mechanistic scientistic vocabulary like ‘produced’, ‘achieved’, ‘converted’ , ‘increases’, ‘decreases’ (the scare quotes above were genuinely scare quotes!) is to occlude or obscure the point that, we are not dealing here with physical or natural phenomena at all here, but with the results of human thoughts, feelings, intentions and actions. At this point in fact, the analogy between an economic system and any kind of physical or natural system breaks down. This is not in itself a problem. All analogies break down somewhere. If they didn’t they wouldn’t be analogies. But what is important is (a) to recognise that an analogy is just an analogy and is therefore ‘doomed’ to breakdown and (b) to understand all the implications of such a breakdown – what the breakdown reveals about the analogy as a whole.

In this particular case what the break down reveals is that the objects and magnitudes that economists measure (the ‘total’ of investment at time x, the ‘savings ratio’ at time y, the’ level of output’ at time z) are artefacts of economists’ conceptualisations. There is a certain sense in which they ‘do not exist’ at all. In what sense? In the sense that mass human activity (including the activity we call ‘economic’) is ceaseless. Across the world as a whole human beings are working every hour of every day, and many of them are also saving ceaselessly, investing ceaselessly, consuming ceaselessly and (indeed) disinvesting and reducing their savings ceaselessly. In short, the world of real economic activity is endlessly dynamic, endlessly changing, and that ceaselessness, that continuous motion or dynamism, is only ‘arrested’ at’ time x’ or ‘time y’ - as a given ‘volume’ of investment or a given ‘magnitude’ of saving -  in the conceptualisations and equations of economists, not in the real world. When, at 11 am on 3rd Jan 2011, economists in some country’s Department of Treasury report that their country’s GDP ‘is’ $274.6 billion it is already, in reality, not $274.6 billion, but something more (or less). And that ‘something’ is ‘forever’ unknowable, at least in the present-tense. (Because, of course, by the time ‘it’ is known ‘it’ will again be something else). This logical point, much emphasized by the Austrian School of economists in particular, has many implications, one of the most important of which is that ‘economic equilibrium’ is almost entirely an artefact of economist’s conceptualisations and equations and not a characteristic of the real world. [Varoufakis and Halevi have a very good discussion of the Austrian school, a school of economists unusual in combining a strong cult of the market with a deep, and philosophical informed, hostility to any attempt to model economics on natural science. Friedrich Hayek is probably the best-known member of this school.] For only postulatedly fixed magnitudes can be in equilibrium (or not), and such magnitudes are only sustainedly present in economist’s equations, not in real economies. But we must say ‘almost entirely’ because we cannot say that we know that real economies are never in equilibrium. What we can say, however, is that by the time we knew they were they would not be. And we can also say that the formal requirements of equilibrium or so demanding that real economies are certainly not  in equilibrium the vast, vast majority of the time, if ever.

Practice 5: Recognising that Economics Discourse Affects Real World Economic Activity only in a ‘Fitful’ and ‘Incomplete’ Way

Another implication of the breakdown of the scientistic analogy between the economic world and the physical world is that the human beings whose activity is creating economic ‘magnitudes’ and ‘relations’ are thinking, emoting entities whose economic activity (like all their activity) is affected by what they know (or think they know), by what they hope and fear, by what they expect or are seeking to avoid. And this means of course, that in the contemporary world, and indeed for the last 200 or so years, human economic activity has been affected by, among other things, what economists tell other human beings, other ‘economic actors’, about ‘what is happening’ in the economy. And economists affect ‘real world’ economic activity not only by providing knowledge, - and information - to economic actors, knowledge or information which may influence what they do or abstain from doing. Economists also affect economic actors emotionally, because the analyses they present, the projections they proffer, affect people’s hopes and fears, their expectations about how the world will be or could be.

Now all economists, even those of the most determinedly scientistic and positivistic bent, know that economics as a discourse affects economics as a real world activity. Given the explanatory emphasis that they place on the role of ‘confidence’ in the creation and solution of all economic crises, they could scarcely not be aware of this. But the more scientistic the bent of an economist the more ‘fitful’ and ‘incomplete’ is this recognition. For of course what Read, (and many others), call the ‘hermeneutic’ dimension of all social science (including economics), the ‘feedback loop’ it creates between the human beings being studied and the human beings studying them, is what most threatens to explode the entire analogy between economics and any form of natural science. And what explodes it is not that social scientists affect the object or objects they are studying while natural scientists do not. (For the experimental activities of quantum physicists affect the activity of the ‘quanta’ they are studying, and, more prosaically, every natural science can affect the physical world when its results are ‘applied’ to interventions in that world). What explodes the analogy between economics and any form of natural science is that the economic actors whose activity economists are studying understand the knowledge and information that economists produce. They know what it means in short, and that knowledge can, and often does, affect what such actors do. In short it is the way – the ‘hermeneutic’ way – that economics affects real world economic activity, that differentiates it completely -  by which I mean epistemologically and ontologically – from any form of natural science.

Moreover, this point is of more practical importance in the case of economics than in any other social science. Because, far more than any other social science, economics has a very strong ‘popular’ presence in the world. There is a large and ‘high profile’ section of the economics profession operating as ‘analysts’ and ‘pundits’ in the mass media, in the business and financial press, in the news and current affairs departments of radio and TV, or (these days) in numerous online economics and business websites. At all times these popular, high profile economists, operating at the ‘interface’ between the discipline of economics and the real world of economic activity, are seeking to influence happenings in that world as much (at least as much) as they are reporting those happenings. (Or, to be more precise, they are seeking to influence what happens by reporting what is happening in certain ways). At times of economic crisis in particular, their activities are often overwhelmingly focussed on what the ‘confidence’ and ‘expectations’ of economic actors, often to the extent that the knowledge or information such economists provide is slanted, even consciously censored, in ways designed to’ lift’ confidence, or to at least not’ damage’ it further. At times of crises in short this high profile section of the economics profession is not only aware of the hermeneutic dimension of economics, it becomes absolutely front and centre in all its activity. In a word, at times of crises, orthodox ‘public’ economists become advocates,  or even propagandists, become intensely motivated to persuade their fellow human beings (company CEOs, pension funds managers, politicians, treasury officials, the ‘ordinary’ consumer of this or that product) to act in certain ways and to cease to act in others. But when do physicists (however high profile) become ‘intensely motivated’ to persuade atoms or molecules to change their behaviour? When do biologists or geneticists (even Richard Dawkins) become ‘intensely motivated’ to persuade genes or genomes of anything?

However,  the vast majority of ‘ordinary’ academic economists do not belong to this high profile section of the profession. They are therefore not regularly engaged in attempts to influence or persuade economic actors, indeed normally their only influential or persuasive interaction with other human beings, is with their students. Such students may be ‘economic actors’ (to the extent that they are consumers or paid workers as well as students) but their economist teachers do not address them as such actors when teaching them. It is therefore far easier for such ‘workaday’ economists to ‘recognise’ the hermeneutic dimension of their subject (especially at times of crisis) but continue to ignore it in their lectures or in the articles and textbooks that they write. What this means, specifically, is that the structural social position of workaday economists (who are, after all, the vast majority of the profession) allows them to continue to teach and research in ways explicitly or implicitly informed by the scientistic analogy between the economic world and the physical world, while treating the ‘persuasive’ role of economics and economists as a particular use that can be made of ‘economic science’ rather than as a practice which calls that science into fundamental epistemological doubt. This is what I mean by saying that the majority of economists recognise the affect of economics on economic activity only in ‘fitful’ and ‘incomplete’ ways.

Such recognition is ‘fitful’ because in their day-to-day practice there is no pressure for its continual, or even frequent, recognition, and because persuasive, even propagandistic, activity by high profile ‘colleagues’ is felt, uncomfortably, to conflict with notions of scientific ‘objectivity’ and ‘impartiality’ central to the professional identity of many economists. And this is precisely why ‘persuasion’ is presented, by mainstream economists, as simply one, vaguely embarrassing, vaguely unrespectable, ‘use’ that may be made of economic science.

Such recognition is ‘incomplete’ because the explosive implications of economic persuasion for the scientific claims of economics are simply not grasped. And by ‘not grasped’ here I mean something essentially behavioural, i.e. I mean that the scientistic analogy between the economic world and the physical world continues to structure all the teaching and research of such economists. Whereas, if they grasped the full epistemological and ontological implications of the persuasive use of economics, continuing with the analogy, and the teaching and research which assumes it, would be impossible for them (except perhaps for purposes of critique).

Practice 6: Engaging in an Unrealistic ‘Normal Science’ as a Defence Mechanism

In discussing axiomatic deduction and mathematical modelling in economics ( Practice 3 above) I noted that economists do engage in an uncontentious ‘normal science’ and have done so for more than a century. However, seen from a natural science perspective that ‘normal science’ is peculiar in at least two respects. Firstly, as Rupert Read emphasizes, it routinely uses unrealistic initial premises or assumptions and (contra Friedman) this is certainly not the norm in routine natural science theorising or experimentation. And secondly, as Read also states, such ‘normal science’ continues even when, in Kuhnian terms, significant groups of heterodox or oppositional economists are continuing more or less radical, more or less ‘revolutionary’, opposition to the ‘paradigm’ which it assumes. Indeed Read is right to emphasise that economics has never, in its entire history, been free of fundamental paradigm disputation. As I have emphasised elsewhere, virtually every cohort of graduate economics students across the world contains a larger or smaller minority of students who become critical, to a greater or lesser extent, of its founding assumptions and methodologies. Such students either leave the discipline entirely or become recruits to its ‘heterodox’ wing of teachers and researchers. In the terms made famous by Isaac Deutscher, economics has produced ‘heretics’ and ‘renegades’ for as long as it has existed. So whilst many of its ordinary practitioners may do ‘normal science’ they do not – and contrary to the Kuhnian account – do it as part of some long period oscillation between ‘normal science’ and ‘revolutionary’ ferment. Because economics, like all the other social sciences, exists in continual, paradigm-threatening, ferment.

 It seems to me that these two distinguishing characteristics of ‘normal science’ in economics – its unrealism and its coexistence with continual paradigm challenge - are linked. To see how and why, let us revisit the prototypical example I gave of economic ‘normal science’ under ‘Practice 3’ above.

“With a two crop agricultural ‘economy’ using varying inputs of ‘labour’, ‘water’ and ‘fertiliser’, and assuming certain technical input/output relations in the crop production functions, what is the optimum yield of both crops obtainable from the least cost combination of the three inputs?”

That this technical puzzle is based on grossly unrealistic premises hardly requires emphasis. An ‘optimal’ yield for 2 crops ‘derived’ in abstraction from such factors as soil quality, climatic interventions, and varying skill levels of farm management and labour, is unlikely to be of any practical help to any real farmers. It is true that this kind of optimisation method can be developed and refined by adding additional variables, but it will still struggle for practical relevance insofar as indeterminates like climatic variation and managerial expertise cannot be meaningfully modelled. And the same limitations apply whatever ‘industry’ or ‘sector’ it is applied to (and it is applied to a great many in the standard economic literature). Indeed there are, as Read intimates, reasons for thinking that, in such mathematico-logical exercises, terms like ‘crops’, ‘water’, ‘fertiliser’, ‘production’ ‘yield’ ‘inputs’, ‘labour’, ‘management’, ‘sector’, ‘industry’ have no realistic reference at all. They are in fact  simply ordinary language terms attached to mathematical symbols and given a narrow mathematical use. They only appear – through their ordinary semantics - to have a reference beyond the mathematical.

Moreover, and for much the same reasons, such ‘science’ is unlikely to receive any significant attention or critique, even within the economics profession, let alone outside it. Such work is ‘uncontentious’ (and thus ‘normal’ within the paradigm) only in the sense that it counts as proof of professional competence, (and is thus career enhancing) while having no wider significance, meaning, or use at all.

But just for that reason such exercises are a perfect form of ‘normal science’ for a discipline under continual, and informed, paradigm challenge. Since they have no practical application whatsoever they pose no threat to the paradigm by being found explanatorily wanting or practically inefficacious in the real world. In other words, this kind of ‘normal science’ will never generate the kind of insoluble puzzles or contradictions which Kuhn sees as the harbingers of a paradigm crisis. Also, this kind of economic ‘normal science’ can also claim to be uncontentious - and thus ‘scientific’ and objective’ - insofar as it has no explicit persuasive intent or function at all. In both these respects then the mathematical-deductive form of economic ‘normal science’ is, as intimated above, a perfect ‘defence mechanism’ for those economists who wish to see themselves as pursuing an unproblematic, ‘technical’ and ‘mature’ science in a situation of continual paradigm ferment. Indeed, as a defence mechanism such a practice clearly allows its practitioners to ignore or deny the fact that there is continual paradigm ferment, and it is difficult not to think that this is at least part of its ‘professional’ appeal.

All these observations lead us to a wider and more important one, their logical converse as it were. – viz that the more relevant economics seeks to be for the real world of economic activity the more overtly 'popular' it becomes in intent, tone, and presentation. (Its mathematical content usually diminishes radically for example). And equally, the more overtly persuasive it becomes in intent, tone and presentation the more contentious and contested it becomes, both among economists and non-economists. In a word,when economists wish to influence the real economy they must, and generally do, give up practicing their normal science! This is surely an extraordinary observation to make about any putative science, whether from a Kuhnian point of view or any other.

Scientism in Economics: External Supports 

Along with Rupert Read and many other Wittgensteinians I hold that scientism in economics is itself only a part of the deep scientistic drift in western culture as a whole. Even to state why I believe this would take me well beyond the scope of this review essay, as indeed it is beyond the explicit scope of reference of  Wittgenstein among the Sciences itself. However, in line with that broader conviction, I would wish to emphasise that scientism in economics is not solely generated within, or supported by, the practices of the economics profession. It has significant external supports as well, most notably in the world of politics and public policy.

Doing justice to these wider supports would also take far more words and space that I have available here, but suffice it to say  that just because any significant public policy will  - in any healthy democratic society – be the subject of deep political and ideological contestation, there is much political utility in any form of discourse which can be claimed to be ‘above the fray’ because ‘impartial’, ‘objective’ and (above all) ‘scientific’. Developing that argument further would take me into the familiar-enough terrain of ‘economics as ideology’. But this is terrain which I do not wish to enter here, partly because it is so well trodden (and I have little to add to those, Marxists and others, who have trodden it), and partly because doing full justice to its complexity would take a further mass of words. I will therefore make only one observation central to the theme of this review – the appeal of scientism in economics. It is this. In a world of overt and heated ideological contestation over policy it is an important, if paradoxical truth that any ‘heterodox’ branch or tradition of economics, any branch that openly parades its ‘normativity’, its ‘value judgements’, its ‘human actor centredness’ is likely to be markedly less appealing to all politicians and policy makers (whether of left, right or centre) than one which claims the imprimatur of ‘impartial, ‘exact’, ‘technical’, ‘objective’ science. And the reason for this is that mainstream politicians believe (and I think rightly) that scientistic discourse will be far more effective in assisting them to overcome their ideological opponents than any overtly normative discourse would be. In other words, it is a hallmark of a scientistic culture that ‘science’ is a far more effective political and ideological weapon (a far more effective partisan tool) than any form of discourse that honestly acknowledges or proclaims its partisanship! I should add here that I hold this to be a part – an important part - of the ‘sickness of our time’. But I will not even attempt to validate that judgement here.


This relationship of this essay to Wittgenstein among the Sciences will have been misunderstood if seen simply as an exploration of scientism in economics stimulated by, or rationalised by, passing reference to one relatively small section of Rupert Read’s book. As noted earlier its debt to the book is far deeper than that. This essay was both provoked, and enormously facilitated, by what I take to Read’s major achievement in the text – the elaboration of a distinctively Wittgensteinian method for discussion of scientism. Because the question of whether an intellectual discourse – any intellectual discourse – should be classified as a science or not can, at the very  worst, degenerate into a empty wrangle about terminology. But even when the worst does not come to the formalistic ‘word policing’ worst, and the debate is more substantially about criteria for the identification of science as a practice or family of practices, the outcome of such debates can be still be rather flatly indeterminate – i.e. some conclusion to the effect that “ well it is if one prioritises X and Y criteria, but it isn’t if one prioritises A and B criteria”.

As against this kind of approach Read, as I understand him, advocates one which asks not what science ‘is’, but what practices (both intellectual and experimental, both mental and physical) are associated with the claim that psychology, or social anthropology, or economics, are sciences. More importantly still Read emphasises, that, when closely investigated, some of these disciplines may indeed contain reasonable or defensible projections/adaptations of natural science procedures, as well as highly unreasonable and indefensible ones, and as a result there may indeed be grounds for thinking of the discipline in question as a science and grounds for thinking of it as something quite other.  But when that occurs, one does not end the matter in the indeterminate ‘criterial’ way above, but extends the investigation pragmatically ever outwards, asking about wider and wider layers of practice – professional, organisational, social, political – with which the claim to scientificity is associated. Then one evaluates them epistemologically, ethically, politically. And it is in that wider pragmatic investigation that he is able to unambiguously anatomise ‘scientism’, and its human and environmental costs, rather than in the technical practices of the social science disciplines alone.

In this review essay therefore I have tried to follow Read’s methodological procedure. Doing so has not led me to the conclusion that economics “is not a science”, for, as I hope emerged particularly from my investigation of Practices 1-3 there are respects in which it is, or at least could reasonably make the claim to be. However, in continually widening and deepening my investigation of the range of practices justified by, and justifying, the claim of economics to be a science (Practices 4-6 and the discussion of  ‘external supports’) I come to the conclusion that doing economics as if it were a science has conceptual, epistemological and indeed political consequences which do a great deal more harm than good, both to society at large and indeed to economics and economists themselves.

And to end this review essay I would like to outline, briefly, what I take to be the single most damaging of those consequences, and one which embraces questions of epistemology, ethics and politics simultaneously.

In the famous section of Volume 1 of Capital concerned with the ‘fetishism of commodities’ Marx remarks, in passing, that with the commodification of the products of human labour “a definite social relation between men... assumes in their eyes, the fantastic form of a relation between things.”

Marx’s ideas on the fetishism of commodities have been enormously debated of course, with the debate focussing most insistently on whether he held that ‘fetishised’ forms of consciousness were ‘false’ forms hiding the realities of capitalism, or whether they were in fact true and accurate, if ‘alienated’ forms, i.e. true and accurate accounts of a deeply alienated capitalist world. Translating that question into the terms of this review would mean asking whether, in describing and explaining the economic world as a world of relations between things (conceptual ‘things’ and their money magnitudes) scientistic economics gives a false account of capitalism, or whether, on the contrary, it gives a true account of a deeply alienated world. On this view of what Marx is saying, scientistic economics gives a true account of (deeply alienated) capitalism because it is itself a deeply alienated account. [For a good preliminary discussion of these issues, see David Andrews, "Fetishism as a Form of Life" in Marx & Wittgensteinedited by Nigel Pleasants and me]

In either case Marx wishes to see the end of such alienation. But on the first interpretation he might have thought that a social relation among things could ‘reassume’ the form of a definite relation among ‘men’ (people),even ‘in the eyes’ of economists, through some form of intellectual or ideological critique of their discourse. On the second interpretation however economists are just a small minority of the ‘men’ (people) living in a capitalist world. Therefore they, just like their peers (workers, capitalists, journalists etc) cannot see economic relations as human relations until they can live them as human relations, (i.e. until they are, actually, human relations) – which meant, for Marx, until capitalism is transformed into socialism.

I have no idea which of these interpretations of Marx’s views on fetishism and alienation is right, for, as on so many other matters, there seems persuasive evidence, both in Capital, and in other parts of his oeuvre, for both. (It is difficult, for example, to think he would have devoted as much time and energy as he did to the intellectual critique of classical political economy had he held to some unalloyed version of the second interpretation). It may also be that these two interpretations are not mutually exclusive, i.e. that Marx did indeed hold that classical political economy (like the scientistic economics that succeeded it) could only be rendered clearly false by transforming the reality that it accurately described, but that an important initial stage in that political struggle for that transformation was to demonstrate that the truth and accuracy of classical political economy could only be obtained by certain kinds of intellectual manoeuvres. And the point of Marx’s analysis of classical economics was to show that it could produced true and accurate accounts of capitalism only because these manoeuvres were themselves deeply alienated, and alienating. That is, they were manoeuvres which subtly and brilliantly obscure the extent to which economic activity under capitalism is still, at bottom, human activity. But they are ‘true’ and ‘accurate’ insofar as capitalism itself operates in ways which obscure the same fact. In fact classical political economy was both a description (a true description) of the means by which that obscuring is done and itself a further (intellectual) layer of that obscurity.

At any rate whether or not the above is an accurate rendition of what Marx was doing, it will certainly do as a rendition of what I  am doing in my discussion of the scientism of contemporary economics. Modern scientistic economics is, in Marx’s terms, a deeply fetishised and alienated form of discourse. But more to the point, those of its practices which are most scientifically plausible are those which describe certain economic activities that are lived as a relation between things, not only by economists but by everybody else. And conversely, those of its practices which are least scientifically plausible are those in which it continues to describe as relations between things economic activities which economists, as well as the rest of us, can see (albeit with difficulty) are relations between people – human relations. One might say that the latter practices are those in which economic’s alienated nature stands revealed, even to economists.

And this does indeed lead us to conclude that it is only when economic relations (including global economic relations) are lived as relations between people that scientism in economics will end. It will end because it will no longer be required, and (more prosaically) because perceiving economic relations as relations between things will then be simply to perceive them wrongly – empirically inaccurately – in a way which will be ‘obvious’, to economists and everybody else. But that, of course is a lot easier said than done, especially in a world of global capitalism where the ‘relations between men’  are extraordinarily complex relations between people thousands of miles apart, and of widely differing languages, cultures, nationalities and ethnicities, as well as classes. In a recent book I try to suggest how a beginning might be made to the humanisation of global capitalism through its conscious political regulation.

But until economic activity becomes transparently human activity (if it ever does) there remains the awkward pragmatic question of what is to be done, now, about scientism in economics.

In addressing this question, one point should be made clear from the beginning. Nothing said in this review would come as news to the (not inconsiderable) community of ‘heterodox’ economists around the world. Indeed I am indebted to my reading of their work for all seven of the items in the critique of scientism above. [During 20 years living and working in Australia I learned a great deal from attending the annual conferences of the Australian Society of Heterodox Economists (SHE)] As noted, economics is probably unique in the volume of ‘heretics’ and ‘renegades’ it has produced - and reproduced - in every generation, since its inception. But it is equally unique in its power either to drive such people from the profession altogether (‘renegades’) or to exclude them from institutional influence and marginalise them intellectually if they remain practitioners (‘heretics’). In fact, another practice closely associated with scientism in economics, and one that I was too intellectually ‘delicate’ to mention earlier, is a positively brutal use of power – on journal editorial boards, on promotions committees, on curricula committees, on grant awarding bodies – as a back up to, or in frank substitution for, failed or exploded intellectual argument. [(also p.13) See the discussion of the, often brutal, academic politics of economics in Quiggin]

In this situation I certainly think there would be nothing to be lost (for the heretics) and perhaps something to be gained, if economically informed Wittgensteinian (and other) philosophers were to ‘pitch in’ on their side – at conferences, in media and website debates etc. For whilst I do not think that Wittgensteinians know what is wrong with scientistic economics any better than the best of the heretics do, they are perhaps equipped to describe and anatomise the conceptual roots of scientism in a way which – through its clarity and transparency – can further strengthen the heretics in the fight. Whether, in ‘pitching in’ Wittgensteinians, and others, should demand, or suggest, that economics should stop calling itself a science, and adopt some other name - ‘Economic Studies’ ‘Political Economy’, ‘The Human Science of Economics (ugh!)’ – seems to me entirely a matter of political tactics. It would certainly be important to stress that, whatever it was called, the systematic study of economics could still make use of at least some of the statistical and mathematical tools that scientistic economics has developed. But aside from that, I do not care, Rupert Read does not care, and Wittgenstein would not care (how deeply he is still misunderstood in these ways!) what economists call what they do. We care what they do!

1 comment :

  1. Wow. That's really good stuff. Thanks, Gavin.

    [Readers may be interested to check out http://blog.talkingphilosophy.com/?p=4367 in which I expand a tiny weeny bit more on what is meant to be going on in 2.4 of the book.]