Another one of those blog posts in which I post about ideas which I should probably be reading about instead.  Yes, it would be better to know more about what I’m talking about before posting, but I find it helpful to try to organise my thoughts as I go, and that’s what the blog is for, so here we are.  I’ll aim to refer to some of the literature that I’m not referring to. so to speak, as we go.

So – in the last post I re-articulated elements of this blog’s case against the labour theory of value.  One of the reasons that Marxists have traditionally adopted the labour theory of value (and thus one of the arguments against its rejection) is that it gives us a way to understand and elaborate the category of ‘exploitation’, which is taken to be central to Marxism’s critical standpoint.  In the simplest version, the argument goes something like: it’s labour that generates value; much of that value is appropriated by the owners of capital rather than those doing the work; the ratio of value appropriated to value produced is the rate of exploitation.  Obviously this can be elaborated in a lot of more complex ways, but that’s the basic idea.  Then the idea is: the central problem with capitalism is exploitation, and our political-economic goal is to eliminate exploitation.  From this perspective, the centrality of the category of exploitation is what distinguishes Marxist from non-Marxist political economy, where non-Marxist political economy has developed a complicated theoretical apparatus to elide the centrality of exploitation to capitalism.

Then there is a subtradition within Marxism that rejects the labour theory of value in one way or another – I think I’m right in saying that all the analytical Marxists end up in this space.  If this is where you are then the question becomes: well, what happens to the category of exploitation?  How can we still make use of the category of exploitation without the LTV?  And there are a range of different answers to this question.  John Roemer’s early work is one effort to answer this question, deriving a concept of exploitation from more ‘mainstream’ economic resources; Vrousalis’ recent, more philosophical, work is another.

My basic thought in this post is: do we really need the category of exploitation?  Or, more properly and carefully, do we really need the category of exploitation to be so central, either to our analytic framework or to our critical categories?  Obviously one response to these question would be: “if you want to be any kind of Marxist, of course you fucking do! Faux-radical idealist revisionist scum like you are the reason [etc. etc.]”, but this doesn’t seem like an actual argument to me.  On the other hand, there are serious reasons, in my view, why we might not want to accord the category of exploitation – even shorn of its LTV underpinnings – the status that it has been granted by much of the Marxist tradition.

My intent is largely to bracket the question “why might we not want to give exploitation a central categorial status?” until some hypothetical future post.  But I’ll take as my jumping-off point Vrousalis’ recent work on exploitation as domination.  I haven’t read Vrousalis’ recent book yet, so again this is all rather half-baked – but Vrousalis’ basic idea, I take it, is to say something like the following.  Analytical Marxism’s critique of traditional LTV-based theories of exploitation are bang on the money.  But analytical Marxism’s effort to produce an alternative account of exploitation – we’re basically talking about Roemer here – fails to pick out the appropriate set of political-economic phenomena.  Instead, we should understand exploitation as a sub-category of domination: domination for economic gain.  In Vrousalis’s more analytic vocabulary (the quote is from his paper ‘Exploitation, Vulnerability, and Social Domination’):

A exploits B if and only if A and B are embedded in a systematic relationship in which (a) A instrumentalizes (b) B’s vulnerability (c) to extract a net benefit from B.

(Vrousalis 2013: 132)

My view, for what it’s worth, is that this is a more promising way to understand exploitation than those proposed by much of the tradition. But my thought is: what’s doing the critical heavy lifting in this definition is the concept of domination that it’s operationalising.  Why not just build our approach on that concept of domination?  Or, perhaps more properly (and this is all more a critique of the tradition than of Vrousalis), why prioritise this specifically extractive category of domination in our political-economic critiques?

My motivating worry here is that there are categories of domination that we want to be able to talk about in critical political economy that do not in any immediate – or arguably even in any very mediated – sense involve one party extracting net economic benefit from the other.  There are some probably relatively easy cases to deal with – like scenarios in which there is, as it turns out, no net benefit to be had.  (A business fails, so there is, as it turns out, no value extraction, because there turned out to be no value creation over the relevant time period – one can probably deal with this by building in expectations in some way.)  But there are also major forms of domination that don’t have any clear value extractive component.  Much of the carceral state is in this category, in my view.  Prisons are often very expensive to run, and many of them don’t themselves generate economic value.  Clearly they are critical to maintaining the political-economic order, and in that sense they could be said to centrally contribute to macro-level value extraction – but this seems like a very indirect way of analysing what’s going on with the carceral state.  Better, in my view, at least in the first instance, to just talk about this category of domination directly.  Something related could be said for much capitalist military activity.  From a (very broadly) Marxist (or if you prefer just ‘radical’) perspective, we can understand a great deal of military activity as economically motivated in some sense – but even here, it is unclear that this economic dimension is best captured by the category of exploitation.  Are settler-colonial wars of conquest, extermination, subordination and expropriation examples of exploitation?  One could make the case, if one extends one’s categories in the right way – but I think the traditional Marxist approach has been to see this kind of violence as the ‘primitive accumulation’ that precedes and sets the stage for a properly exploitative capitalist economic relation – i.e. as falling outside the category of exploitation.  And these are far from the only phenomena of interest that the category of exploitation doesn’t intrinsically seem to latch on to.  More on all this, perhaps, some other day.

My worry, in short, is that the category of exploitation risks narrowing the focus of our critique of capitalism in a way that misses many of the most direct and appalling examples of coercion and domination we might want to analyse – even potentially including economic coercion and domination.  Alternatively, the tacit belief that we need to explain those phenomena in a way that retains the centrality of the concept of exploitation can warp our analysis.  And my thought, therefore, is that we should ground the critical categories of our political economy not on the idea of exploitation, but on the concept of domination that it (plausibly, as per Vrousalis) presupposes.

Ok.  None of what I’ve just said is actually attempting to make an argument that I would expect anyone to find convincing – again, my goal here is basically just to start to get some of my thoughts in this space in rough order.  But let’s assume we’re granting for the sake of argument the idea that domination is a better core category for critical political economy than exploitation.  What is the concept of domination?  

Here, again, I’m going to start by flagging my ignorance.  There is a large recent literature in political philosophy around the concept of domination – much of it seemingly centred on the ‘republicanism’ associated with Philip Pettit and other related figures.  I don’t know this literature.  When I look at it (and this is no criticism – it’s always the case when one looks at unfamiliar philosophical terrain) I find people drawing and debating fine-grained distinctions, the function and consequence of which escape me.  Clearly at some point I need to read this literature – but that time has not yet come.  So what follows is basically me just sounding off with my own associations, and I’ll try to come back around and engage with the actual debates in this space at some future time.

I want, then, to distinguish between four different senses in which one might think about the category of domination.  ‘Domination’ is very likely not the word I ought to be using for some – even, conceivably, all – of these ideas, but again my goal is just to crudely organise some thoughts.

  1. Straight-up murdering people.  Clearly this is an interpersonal form of violence.  Arguably it’s a limit case for coercion, and (again) arguably it doesn’t usefully fall within the frame of ‘domination’ at all – but I think it needs to be in play here as one of the major ways in which people exert force against other people.
  2. Direct coercion.  That is to say, using force to bend another person to one’s will.
  3. Differential bargaining power.  Here we’re fully into the terrain of ‘routine’ economic interactions (and standard economic theory) – most economic interactions involve some difference in bargaining power between the relevant parties, and this is the kind of thing that we now have a very extensive and developed formal apparatus to model.
  4. Impersonal domination.  Here I’m thinking of scenarios in which there is no immediate interpersonal relationship of domination between two individuals, but rather a ‘structural’ kind of domination that does not route via anyone’s direct intentions.  An example of this might be the macroeconomic dynamic whereby a community’s income is dependent upon the production of some specific commodity, to cater to a wealthier export market, and for whatever reason the demand in that market greatly reduces, leaving the exporting community impoverished.  This is an example of ‘market forces’, with no specific social actors intending this specific outcome, and yet it is an example of the exertion of (de facto brutal) power by one set of economic actors over another.  Probably there are better examples than this – but basically what I want to capture is that economics is full of dynamics that are the result of “human action but not human design”, and some of these dynamics will effectively be dynamics of domination, even when many, or even all, of the relevant actors are not engaged in direct interpersonal domination.

So we have here four different categories of domination – and I think they can be roughly ordered on a sliding scale from most to least personally violent.  Murdering someone is the limit case of violent action.  Coercion is the threat of violence – including deadly violence – to achieve gain.  Differential bargaining power will very often not include direct coercion at all, but it does involve a power differential.  And finally ‘impersonal’ domination doesn’t involve any direct interpersonal domination at all, but this doesn’t detract from its ability to produce de facto cases of domination.

Anyway, this is all I really wanted to say in this short post.  I’m not suggesting that anything I’m saying here is very noteworthy – and certainly everything here is so telegraphic in its articulation that I can’t imagine it would have the ability to persuade anyone of anything.  But my goal here, as I keep saying, is to get my own thoughts straight about the broad brush strokes of my research programme.  And this is the idea: domination should be a fundamental analytic category in critical political economy – more fundamental than, for example, exploitation.  Moreover – though of course I’ve not explored this at all on the blog to date – it seems to me that formal economic theory has significant resources to explore this problem space.  After all, there’s an entire subfield – bargaining theory – which looks specifically at differential bargaining power, and these analytic tools inform mainstream economic theory in all kinds of sophisticated ways.  So my broader idea, as I continue to work my way through economics, is to keep an eye on these themes, with the broad perspective or approach articulated in this post in mind.


A quick and not very useful blog post (edit: turns out not to be very quick either!) that aims to scout out some fairly familiar intellectual terrain using a slightly different vocabulary.  Specifically, I want to talk about different understandings of Marx’s value theory using the vocabulary of Huw Price’s distinction between “object naturalism” and “subject naturalism”.

As I’ve written about on the blog before (though probably not for quite some time now), this blog rejects the labour theory of value.  (I will sometimes shorthand ‘labour theory of value’ as ‘LTV’.)  This post isn’t really about the arguments for and against the LTV – hopefully in the longish medium term I will work through the analytical Marxists in a systematic rather than a haphazard way, and maybe I’ll discuss this issue more then, and with more abundant references.  Still, I will begin with a (very) brief recapitulation / summary of the basics of the argument.  I’ll then go on to talk about how Price’s opposition between ‘object naturalism’ and ‘subject naturalism’ can, in my view, shed some light on the stakes and implications of that debate.

So – what we can call the ‘simple’ or even ‘vulgar’ LTV runs something like this.  Question: what determines the value of commodities in the capitalist marketplace?  Answer: the amount of labour that went into producing them.  Understood in this way, the LTV is an explanatory or causal account that explains one phenomenon (the price or value of commodities) in terms of another (labour inputs in the production process).  From this starting point you can then (if you wish) make a number of further claims.  For example, you can explain the tendency of the rate of profit to fall in terms of the tendency of capitalist firms to increase the degree of automation in production: fewer labour inputs per unit produced means less value per unit, so less surplus value available for capitalists as profit.  You can also use this value theory to ground a political-ethical case against capitalism in terms of exploitation: if all value is generated by labour (and none by, say, the risk-taking associated with investment, or entrepreneurial innovation, or capitalist managerial acumen, or just the value of raw materials or of technological inputs into the production process) then workers deserve to receive the full fruits of their labour, which in turn necessitates a massive overhaul of our political-economic system.

The problem with this ‘simple’ or ‘vulgar’ labour theory of value is that it’s manifestly a bad, false theory: labour inputs cannot usefully be taken to determine prices.  This is for a whole range of reasons, including the following:

  • How do you measure labour inputs? Clearly there is no such thing as homogenous labour: there are countless different kinds of labour, and it’s unclear how to compare them.  One could, of course, compare them by analysing their productivity in terms of the value they produce – but this is very clearly question-begging if labour inputs are to serve as any kind of independent explanatory variable in determining commodities’ market value.
  • Clearly things other than labour inputs contribute to the value of commodities: machinery, raw materials, market conditions, etc.  Labour applied to the production of a desirable and rare commodity clearly results in a more valuable product than the ‘same’ (or similar) labour applied to an undesirable and abundant one.  Etc.
  • One of the features of capitalism is wild swings in the value of commodities – boom and bust cycles for firms, for industries, for national economies, and for the global economy.  Clearly these expansions and contractions of value are not driven by expansions and contractions in labour inputs – quite the reverse: swings in labour inputs (e.g. of employment and unemployment) can be explained by these boom and bust cycles.

I’m not suggesting that this list of objections to the labour theory of value is complete – but I think it’s clear that there are some very formidable hurdles for the theory to overcome if it is to be treated as a useful explanatory framework.

Now, theorists in the Marxist tradition (and, before that, in the classical political economic tradition) are not nitwits, and they are aware of these objections to the LTV.  At this point in the argument (or well before) they therefore object that my characterisation of the ‘vulgar’ LTV is an absurd straw man.  They then go on to (if you are sympathetic to the LTV) clarify the rich and sophisticated account of which this summary is a debased shadow, or (if you are unsympathetic to the LTV) add a series of gimcrack epicycles.

What are these clarifications/epicycles?  I think broadly speaking there are two key moves here.  First, LTV advocates insist that the LTV is a theory, not of price, but of value – where value as a concept is clearly not totally disconnected from price but also cannot be reduced to it.  This in turn leads to a vast set of debates about the so-called ‘transformation problem’: how are value and price related, or how can value be ‘transformed’ into price?  Second, LTV advocates insist that the LTV is interested not in some homogeneous empirical labour concept, but in “socially necessary labour time”, or “abstract labour”, or both.  In short, both “actual labour” and “price” acquire counterpart ‘shadow’ categories that are somehow related to them, and yet also non-identical with them.  The theoretical claims that seem to be clearly false when made about the relation between actual empirical labour inputs and actual empirical prices can then be said to be true of these more abstract and opaque parallel categories.

Now, I don’t think these categories (value rather than price; socially necessary labour time and/or abstract labour rather than empirical labouring activities) are empty or useless.  The problem with these categories, rather, is that they do not seem to be able to serve anything like the same explanatory role that “labour inputs” serves in the ‘vulgar’ labour theory of value.  The reason they cannot serve this function is that these categories – at least if understood correctly, by my lights – are intrinsically ‘back-constituted’ by the market processes they apparently aspire to explain.

So, for example: the “socially necessary labour time” it takes to produce a commodity is not just a fact about labour – it is a fact about the entire state of the industry and the broader world economy within which the labour in question is embedded.  If a technological innovation on the other side of the world can transform the socially necessary labour time required to make a widget in my factory (by increasing general global productivity in the widget industry, and thereby reducing socially necessary labour inputs for the industry as a whole), this ‘measure of labour’ is not simply measuring labour: it is measuring general market conditions.  This category therefore cannot serve an explanatory function as an independent variable that aspires to explain price or value within that market – because the behaviour of the broader economic system is already embedded within the category.  What may appear at first glance to be a basic explanatory category is in fact an output of the system.

Now, as I say, this isn’t necessarily a problem with the category.  One of the ways in which Marx uses this, and related, categories in Capital is to explain the impact of changes in general economic conditions on workers within specific industries.  Thus, for example, technological innovation may increase productivity within some segment of an industry, reducing the value at which its products are sold, in turn reducing the demand for a category of labour that had previously counted as ‘socially necessary’, and throwing people out of work.  Here there is a connection between labour inputs and value, but the connection comes close to reversing the explanatory poles of the ‘traditional’ or ‘vulgar’ labour theory of value.  Again, this isn’t a problem – capitalism is a complex system in which every component can in principle influence every other component.  But if this is how we are wielding our categories, we need to reject the idea of a labour theory of value in which labour is taken to be an independent explanatory variable, somehow “determining” value or prices.  In my own view, this rejection is precisely the project that Marx is engaged in in ‘Capital’ – this is one of the ways in which ‘Capital’ is a critique of political economy.  As usual, though, my main focus on the blog isn’t Marxological, and the blog’s official position with respect to Marx interpretation is to simply point at the work of N. Pepperell, which is much more sophisticated on these matters than anything I put up here.

In any case, as I see it the LTV is therefore caught on the horns of a dilemma.  Either empirical labour inputs serve as an independent explanatory variable in a theory that is manifestly false, or empirical labour inputs are replaced by an alternate category or set of categories that always already tacitly track or encode the economic information that the theory apparently aspires to explain.  In my view much of the literature in defence of the labour theory of value is in practice shuttling between variants of these two positions – using back-constituted (but non-explanatory) categories when the theory is challenged empirically, while simultaneously (or alternately) suggesting that these same categories are fundamental, foundational, and explanatory.  This won’t do – we need to choose.  And in my view the correct choice is to reject the labour theory of value as any kind of meaningful explanatory apparatus.  We need an alternative account of the determination of values or prices, and this account can then in turn contribute to the content of categories like “socially necessary labour time”.  This alternative account, at least by my lights, will almost certainly draw heavily on boringly mainstream explanatory principles like supply and demand, bargaining power, and so forth.  In other words, if these categories of the so-called ‘labour theory of value’ are understood correctly, we are drawn into the orbit of something fairly close to mainstream economic theory.

So – the above is a whirlwind summary of why I reject the labour theory of value.  But that’s not what this post is about.  This post is about using Huw Price’s distinction between subject and object naturalism to capture two different orientations to the ‘materialist’ dimension of Marx’s value theory.

Now, I’ve already talked about two reasons why people (specifically, people on the radical left) might choose to adopt the labour theory of value.  The first reason was: the LTV seems like it might be able to ground an account of the falling rate of profit, and therefore of capitalist crisis.  The second reason was: the LTV seems like it might be able to ground a moral critique of capitalism as exploitation.

In the remainder of this post, I want to talk about a third reason why people might find the LTV plausible – and that is a sense that Marx’s theoretical framework is “materialist”.  Here the basic idea, I take it, is that a distinction can be drawn between ‘idealist’ and ‘materialist’ analysis.  ‘Bourgeois economics’ falls on the bad, ‘idealist’ side of this divide, while Marxist economics falls on the good, ‘materialist’ side.  On this approach, when Marx writes about the “metaphysical subtleties and theological niceties” involved in bourgeois political economy’s analysis of the value of commodities, Marx is suggesting that bourgeois political economy is engaged in mystificatory idealist analysis.  Marx and Marxism, by contrast, bring these idealist fantasies down to earth by looking at the concrete material processes and material phenomena that idealist analysis obfuscates.  And this is part of a broader materialist approach, whereby concrete material phenomena and practices are placed at the centre of analysis, rather than free-floating ideas or obscure metaphysical principles.  Marxism, the argument goes, reveals the material bases for these mystified idealist categories.

I think a lot of Marxists would endorse something like this story – albeit no doubt rephrased in their own preferred terms.  And I think this story provides one of the grounds for endorsing the labour theory of value.  The argument here, I take it, is that this distinction between ‘idealist’ and ‘materialist’ analytic frameworks lines up with the distinction between ‘objective’ and ‘subjective’ value theories.  The dominant mainstream value theory within economics – marginalism, and its epigones – is subjective, therefore idealist, therefore bad, while the labour theory of value is objective, therefore materialist, therefore good.

Here perhaps some very telegraphic intellectual history is useful as context.  When Marx was writing ‘Capital’, the dominant value theory in political economy was the labour theory of value.  The LTV had been endorsed in some sense (though there is much thorny ongoing debate over what sense) by the great ‘classical’ political economists: Smith, Ricardo, Mill.  More or less contemporaneously with the publication of ‘Capital’, however, there was a major transformation in value theory – the shift that came to be known as the ‘marginal revolution’.  The first edition of ‘Capital’ was published in 1867.  Jevons had published his ‘General Mathematical Theory of Political Economy’ five years earlier, in 1862.  Menger’s ‘Principles of Economics’ was published in 1871.  Walras’s ‘Elements’ was published in 1874.  Between them these latter three works revolutionised political economy, by proposing a value theory that was both mathematised and ‘subjectivist’.

From one Marxist perspective, this marginal revolution represents a profound wrong turn in political economy.  There are many charges laid against marginalism, and the economic theories that derive from it.  One charge is that the mathematisation of economic theory turned its back on the real world, preferring idealised models to the study of concrete reality.  This is one version of the critique of idealism.  Another charge is that by proposing a ‘subjective’ value theory – a value theory that is built upon the subjective preferences of individual economic actors, rather than objective facts about the material world – modern economics is building a form of idealism into its very foundation stones.  It’s this latter point that I’m interested in, in this post.

Here I think it’s useful to step back a little, and ask: what exactly do we mean when we talk about ‘materialist’ and ‘idealist’ approaches to political economy (or to analysis in general)?  I think the general vibe of these concepts within Marxist discourse is reasonably clear – but pinning them down is much more difficult and contentious.  In large part that’s because – as with all concepts within diverse traditions – different people simply mean different things by them.  For the purposes of this post, I’m going to – quite tendentiously – map them onto the opposition, within the discourse of analytic philosophy, between ‘naturalism’ and ‘non-naturalism’.  I recognise that this is far from the only way that one can parse these terms, and there are many dimensions or connotations of these concepts that aren’t captured by this mapping.  Nevertheless, I hope the remainder of this post will make the case that it can be illuminating to think about this opposition in this way in this context.  So.

Let’s fiat, then, for the purposes of this post, that the Marxist concept of ‘materialism’ has at least some commonality with the analytic philosophical concept of ‘naturalism’.  What do we mean by ‘naturalism’?  

This is where the work of Huw Price that I’ve been trying to get to finally comes into the picture.  In his work on naturalism, Price draws a distinction between two different forms of naturalism, which he argues potentially have very different philosophical implications: object naturalism and subject naturalism.

For Price, object naturalism is the most prominent and popular form of philosophical naturalism.  In Price’s words, this perspective “exists in both ontological and epistemological keys” (the quotes here are from Price’s ‘Naturalism without representationalism’):

As an ontological doctrine, it is the view that in some important sense, all there is is the world studied by science.  As an epistemological doctrine, it is the view that all genuine knowledge is scientific knowledge.

(Price 2004: 5)

These doctrines seem like the kind of things that scientifically- or naturalistically-minded people might want to endorse.  From this perspective, if we are making meaningful (one possibly wants to add “non-analytic”) claims about the world, then those claims should be translatable into the language of the natural sciences.  If my claims to knowledge are legitimate, they should be rephrasable, however clumsily, as scientific claims. 

Price argues that there are significant philosophical problems posed by this doctrine.  I really can’t claim to be on top of this philosophical literature, so my summary of these arguments is going to involve some hand-waving.  Nevertheless, it seems like this doctrine prima facie runs into problems when we come to the (large) categories of statement that we intuitively want to say can count as knowledge claims, but which do not refer to anything picked out by natural science.  Norms are one such prominent category.  We do not observe norms – that is, norms themselves, rather than behaviour that takes itself to be normatively driven – anywhere in the scientific study of nature.  And yet many of us want to say that we can have normative knowledge of some kind.  How does object naturalism deal with this?  Another example of the same broad category of problem is mathematics.  How are we to understand mathematical objects?  It seems clear that mathematically objects are not observed by scientific study of nature, and yet of course one doesn’t want to say that mathematics is a non-scientific mysticism.  Price calls this kind of issue “placement problems”.

There are, of course, many ways of dealing with these kinds of ‘placement problems’ from within an object naturalist philosophical perspective.  But Price proposes an alternative approach: subject naturalism.  For the subject naturalist, in Price’s words:

philosophy needs to begin with what science tells us about ourselves.  Science tells us that we humans are natural creatures, and if the claims and ambitions of philosophy conflict with this view, then philosophy needs to give way.

(Price 2004: 5)

Price’s argument is that subject naturalism in this sense is explanatorily prior to object naturalism, and moreover that one can – and should – be a subject naturalist without being an object naturalist.  From the subject naturalist perspective, phenomena like norms and mathematical objects can be explained in basically pragmatist terms, as the result of naturalistically-analysable human social practices.  These phenomena remain, however, non-naturalist in the ‘object naturalist’ sense.  There are no objects of the physical sciences that correspond to norms or to mathematical phenomena – and yet these phenomena can be naturalistically analysed, by analysing the human practices that produce these artefacts.  We cannot, for example, pick out any scientifically-describable object that we are denoting when we talk about mathematical objects.  We can, however, describe the human practices associated with producing those mathematical objects, as the topics of our (naturalistically analysable) mathematical discourse.

Subject naturalism therefore aligns with (at least a common interpretation of) Wittgenstein’s approach to dissolving philosophical perplexities.  It may seem that there are spooky objects (like “meanings”) that are difficult to explain in either scientific or philosophical terms.  But we can dissolve this spookiness by turning our attention away from the ‘objects’ and towards the social practices that produce those ‘objects’.

Now – these are potentially fairly deep philosophical waters, but my goal in this post is not a deep philosophical one.  My goal is just to pick up this distinction between subject and object naturalism and apply it in a fairly obvious and banal way to political-economic value theory.

If we pick up and use this distinction in a fairly rough-and-ready way, then, I think the resonance with the LTV debate I very briefly summarised above is fairly clear.  We have this phenomenon that we’ll call “value”.  How are we going to explain this phenomenon in a ‘materialist’ way?  I think the LTV debate I’ve discussed above has two broad categories of answer to this question.

First, we can explain ‘value’ in a materialist way by finding a concrete material phenomenon to which ‘value talk’ really refers.  The candidate concrete phenomenon proposed by the labour theory of value is: labour.  From this perspective, we have two choices: we can either talk in a mystificatory way about this abstract phenomenon ‘value’, or we can figure out what real concrete phenomenon is in fact picked out by this talk, and thereby bring the value-talk down to materialist earth.  This is the ‘object naturalism’ approach to value theory.  It proposes translating political-economic value-talk into an alternative idiom with clear, non-mystified, empirical, scientifically-analysable referents.

Alternatively, we can take a ‘subject naturalism’ approach to value theory.  On this approach, we can explain ‘value’ not in terms of a real phenomenon to which ‘value’ refers, but rather in terms of the social practices that we need to understand in order to understand how value is produced, as a non-natural but social phenomenon.  From this perspective, asking “what is value?” is a little like asking “what is addition?”  There is no phenomenon that is denoted by addition – but we can explain what addition is by explaining what one needs to do in order to add.

This latter is the pragmatist approach to value theory.  My view is that it is clearly the right approach.  A few caveats or additional points need to be made, though, before we too incautiously apply this kind of analytic pragmatist framework.

First: contemporary analytic pragmatists are extremely preoccupied by the issue of specifically linguistic practice.  But this is not our concern, as political economists.  We are not particularly interested in what people say – or even necessarily in what people think: we are in the first place interested in what they do as economic actors.  Second, and relatedly, we are interested in the complex interactions between many different individual economic actors’ actions.  It may very well be that we cannot understand what economic actors do by looking only at individual actions – we need to look at aggregate behaviour which may be qualitatively different from individual-level behaviour: the study of emergent effects is a core part of political economy.  The phenomena we are interested in therefore may be doubly removed from economist actors’ deliberate intentions.

With all this said – what does applying this perspective suggest about the debates over the labour theory of value with which I began?  I think there are two main lessons here.

First: a ‘subject naturalist’ value theory is greatly to be preferred to an ‘object naturalist’ value theory.  My own view (which I haven’t done much to make the case for in this post) is that Marx’s own value theory can best be understood as a ‘subject naturalist’ theory in something like this sense.  That is to say, Marx is not aiming to provide an account of the “substance of value” in the sense of “what value-talk really denotes”, but rather a complex account of the social practices that, in combination, produce the aggregate social phenomenon of “value”. Whatever one thinks of this interpretive claim, I think that ‘subject naturalism’ is the correct way to apply a ‘materialist’ perspective to political-economic value theory.

Second: if we think of Marx’s value theory (or just value theory in general) in the way I’m recommending, then the opposition I discussed above between ‘materialist’ objective value theory and ‘idealist’ subjective value theory no longer makes much sense.  From this perspective ‘materialist’ no longer has to line up with ‘objective’, and ‘idealist’ no longer has to line up with ‘subjective’.  We can adopt a quote-unquote ‘subjective’ value theory, in the sense of subject naturalism.  That is to say: we can ‘deflate’ idealist categories not by finding the natural phenomena to which they really refer, but rather by analysing the social practices by which they are really produced.  From this perspective, a value theory can be both ‘subjective’ – in the sense that it denies that its categories denote natural phenomena, and looks instead at how these categories are produced by distributed social practice – and ‘materialist’ – in the sense that it studies those social practices as themselves natural phenomena. Moreover, and perhaps ironically, if we understand Marx’s value theory in something like these terms, then along at least one key dimension Marx can be seen as aligning with the ‘subjectivist’ revolution that ultimately came to be represented, in the economic mainstream, by marginalism, and against ‘objective’ value theorists, rather than the other way around. From this perspective, ‘Capital’ is not so much the last gasp of classical value theory, as an early (and non-mathematised) critique of classical value theory. Of course this is too simple, and misleading in its own way. Nevertheless, I think it at least stands up enough to serve as a reasonable first-pass challenge to the idea that Marx(ism) and ‘subjective’ value theories are obviously and profoundly incompatible.

Now, the fact that we take our value theory to be ‘subjective’ in the sense of ‘subject naturalism’ doesn’t mean that it has to be ‘subjective’ in other senses.  We don’t, for example, necessarily need to tether our value theory to beliefs – the internal (or subjective) state of mind of social actors could in principle be completely beside the point, when studying the emergent properties of their aggregate actions.  And it doesn’t mean we need to adopt the specific value-theoretic approach of mathematised marginalist economics.  It does, however, mean that one can’t (or shouldn’t) rebut ‘subject naturalist’ approaches in general on the grounds of their ‘subjectivism’, in opposition to materialism.

Now this is a pretty trivial point to have reached by a rather arduous and circuitous route.  Arguably it is worse than trivial, in the sense that it makes the case for a common-sensical and already very widely accepted position (“value in the political-economic sense is an emergent property of distributed economic action which can be scientifically studied”) in terms of a much more obscure and tendentious position (analytic neo-pragmatist anti-representationalism).  It would probably make more sense to try to justify the latter position by analogy with the former, in most contexts.  But so it goes.  My hope is that if I can just keep on assembling such arguments and associations, on the blog, eventually they’ll add up to something more useful.  We’ll see if that turns out to be the case.

Continuing in the ‘dialectics as social theory’ rather than ‘dialectics as philosophy’ thread on the blog, I want to apply the framework I summarised in the last post to three figures that I happen to have been reading this year: Rousseau, Mises, and Hayek.  I’m not going to be saying anything very dramatic or original here, and quite a lot of the content of this post has already been covered in earlier posts, but as ever I’m using the blog for working things through, however ploddingly.

In previous posts I have differentiated dialectical and foundationalist philosophical approaches.  Dialectical approaches see every moment within a complex social system as granted its normative content by other moments within the system.  There is, therefore, an in-principle-infinite chain of authority that can in principle be followed to unpack the basis of the normative content of any moment in the system.  Foundationalist approaches, by contrast, see some moments of the complex normative system as generating their own authority – their normative content is ‘self-grounding’.  I am (obviously) advocating for a ‘dialectical’ or anti-foundationalist approach.

There’s a lot more that could be said about all this – and arguably I should unpack this distinction and its implications as I understand them further before I start applying it – but I’m just going to keep going.  In this post, then, I want to present Rousseau and Mises as examples of thinkers with foundationalist metatheories, and Hayek as an example of a thinker with an anti-foundationalist metatheory.  I hope it goes without saying that a thinker’s metatheoretical commitments are only weakly associated with most of their other commitments, and criticising or endorsing a thinker’s ideas at a metatheoretical level shouldn’t be taken as implying anything very strong about criticism or endorsement of other dimensions of their work – but I still think it can be clarifying to talk metatheory.

First Rousseau, then.  As I suggested in an earlier post, Rousseau’s concept of the general will seems like an excellent candidate for a ‘self-grounding’ understanding of social authority.  In ‘The Social Contract’ Rousseau writes that:

I hold then that Sovereignty, being nothing less than the exercise of the general will, can never be alienated, and that the Sovereign, who is no less than a collective being, cannot be represented except by himself


Sovereignty, for the same reason as makes it inalienable, is indivisible


It follows from what has gone before that the general will is always right

I understand that there is a vast literature on how to interpret these and related passages, and I can’t claim to be familiar with it, but at least for the purposes of this post I’m just going to assume that these passages mean what they sound like they mean: Rousseau has a weird sort of magical concept of an unalienable, indivisible and infallible general will, which is ‘self-grounding’ and ‘self-justifying’ in the sense I’ve been criticising.  In Rousseau’s words:

The Sovereign, merely by virtue of what it is, is always what it should be.

I think this is as clear a statement as one could wish for of the kind of ‘essentialism’ or ‘self-validation’ that I’ve been talking about – in Brandomian-Hegelian terms, ‘pure independence’.  Moreover, Rousseau is actively critical of social theorists who aim to analyse the constitution of the general will in a way that presents it as an emergent (or even just as a composite) phenomenon, rather than as an ‘indivisible’ unity.  Thus, for example, Rousseau mocks political theorists who “dismember the body politic” by analysing it as made up of subcomponents with different functions:

our political theorists, unable to divide Sovereignty in principle, divide it according to its object: into force and will; into legislative power and executive power; into rights of taxation, justice and war; into internal administration and power of foreign treaty. Sometimes they confuse all these sections, and sometimes they distinguish them; they turn the Sovereign into a fantastic being composed of several connected pieces: it is as if they were making man of several bodies, one with eyes, one with arms, another with feet, and each with nothing besides. We are told that the jugglers of Japan dismember a child before the eyes of the spectators; then they throw all the members into the air one after another, and the child falls down alive and whole. The conjuring tricks of our political theorists are very like that; they first dismember the body politic by an illusion worthy of a fair, and then join it together again we know not how.

This passage, of course, isn’t really an argument – it simply aims to mock the idea of analysing a collective institution as composed of subcomponents the nature and interrelation of which can be critically analysed.  In fact, though, of course, there’s nothing analytically wrong with the approach Rousseau is mocking.

Rousseau’s approach, in short, seems to me like an exemplary ‘anti-dialectical’ understanding of social authority, and therefore philosophically very bad.  It also, separately, seems politically very bad, for the reasons I was discussing in relation to the ‘Orwellian’ critique of pragmatism.  As I’ve been arguing, I don’t think it’s fair or accurate to say that pragmatism ends up having to collapse its sense of normativity into a homogeneous understanding of a unitary ‘social’… but it sort of seems like Rousseau does.  In any case, I regard this as a bad line of thinking, and an example of the kind of thing I’m trying to avoid.

Next up, Mises.  In the first chapter of ‘Human Action’, Mises presents his theoretical orientation as follows:

Since time immemorial men have been eager to know the prime mover, the cause of all being and of all change, the ultimate substance from which everything stems and which is the cause of itself. Science is more modest. It is aware of the limits of the human mind and of the human search for knowledge. It aims at tracing back every phenomenon to its cause. But it realizes that these endeavors must necessarily strike against insurmountable walls. There are phenomena which cannot be analyzed and traced back to other phenomena. They are the ultimate given. The progress of scientific research may succeed in demonstrating that something previously considered as an ultimate given can be reduced to components. But there will always be some irreducible and unanalyzable phenomena, some ultimate given.

Mises then claims that human action is one such “ultimate given”:

Concrete value judgments and definite human actions are not open to further analysis. We may fairly assume or believe that they are absolutely dependent upon and conditioned by their causes. But as long as we do not know how external facts–physical and physiological–produce in a human mind definite thoughts and volitions resulting in concrete acts, we have to face an insurmountable methodological dualism. … Human action is one of the agencies bringing about change. It is an element of cosmic activity and becoming. Therefore it is a legitimate object of scientific investigation. As–at least under present conditions–it cannot be traced back to its causes, it must be considered as an ultimate given and must be studied as such.

Mises is here making a cluster of claims.  Characteristically, he is presenting his treatment of human action as an ‘ultimate given’ as a result of scientific epistemic humility.  He is not making a metaphysically foundational claim, Mises argues – rather, he is provisionally treating human action as an ‘ultimate given’ on the grounds that we are not in a position to explain human action further, at this point in our scientific progress.  If we take this argument at face value, Mises’ treatment of human action as foundational is an example of the kind of ‘provisional foundationalism’ I discussed – and approved of – in the previous post in this series.

This is certainly a defensible way to read Mises.  For myself, though, I am not inclined to take this methodological claim all that seriously, because I am not inclined to grant the claim about the state of our scientific knowledge on which it is based.  Mises claims:

as long as we do not know how external facts–physical and physiological–produce in a human mind definite thoughts and volitions resulting in concrete acts, we have to face an insurmountable methodological dualism…. We may or may not believe that the natural sciences will succeed one day in explaining the production of definite ideas, judgments of value, and actions in the same way in which they explain the production of a chemical compound as the necessary and unavoidable outcome of a certain combination of elements. In the meantime we are bound to acquiesce in a methodological dualism.

This seems to me to be a basically entirely unwarranted move.  We know lots about how natural phenomena produce thoughts, volitions and acts.  We have sciences of psychology, sociology, etc. that look at how social and personal environment shapes behaviour, and while I can’t myself speak to physiological science, it seems obvious that we have considerable knowledge in that domain too.  To be sure, Mises was drafting ‘Human Action’ in the 1930s and 40s, so these sciences were considerably less developed at the time he was writing than they are now – but in my view this was still a very clearly unwarranted claim at the time Mises was writing.

What’s going on here, in my view, is a case of ‘god of the gaps’ reasoning – or, alternatively, the kind of ‘quest for certainty’ that (following Dewey) I’ve been criticising throughout this series of posts.  Mises is arguing that because we don’t have complete knowledge of the relationship between social and naturalistic categories and human action, we have in effect no knowledge at all, and this warrants a methodological dualism, treating the natural world and human internality and action as fundamentally different categories of phenomenon.  This is in my view a bad argument, and personally I find it hard not to read it as driven by a kind of motivated reasoning.  Just as Rousseau wants there to be a magical collective self which is ‘self-validating’ and which cannot be reduced to its component parts, so Mises wants there to be a magical individual self which has no analytic connection to the phenomena studied by the natural sciences.  Having drawn this (basically Cartesian) dichotomy, Mises then takes this dichotomy to provide warrant (again, in Cartesian style) for a purely a prioristic deductive science, grounded in an introspective understanding of human mind and action.

In my view none of this is right.  In reality, there is no reason not to treat human action as a product of many other phenomena we can study using the tools of the natural and social sciences.  And if you throw out this dichotomy then the warrant for treating ‘praxeology’ as an aprioristic science is also undermined.  In terms of the metatheoretical categories I’ve been adopting and advocating in this series of posts, this is a ‘foundationalist’ approach which is actively suppressing the extent to which its ‘foundational’ categories can be understood as constituted by other categories – it is, again, problematically non-dialectical.

So: we have in Rousseau an example of ‘collectivist’ foundationalism, and in Mises an example of ‘individualist’ foundationalism.  I’m arguing that both of these are one-sided theoretical approaches, because they both actively refuse to analyse the way in which their ‘foundational’ categories are constituted in practice by other dimensions of the complex social system within which these categories are embedded and from which they are constituted.  For all the vast theoretical differences between Rousseau and Mises, in other words, at this very abstract metatheoretical level, their work is driven by a shared philosophical mistake.

Now on to Hayek.  I’ve already discussed Hayek in several posts on the blog before, and therefore I will try to be even briefer in my summary of his philosophical perspective.  My view, in short, is that Hayek (and here I’m talking about the late Hayek of ‘Law, Legislation and Liberty’) is not guilty of this metatheoretical mistake.  Whatever my theoretical differences with Hayek, my claim is that at a metatheoretical level, Hayek’s apparatus is good stuff, and is basically wholly compatible with the ‘empirical dialectical’ approach I’ve been outlining on the blog.  And in a way this makes sense, because although the Austrian tradition sees Hegel as an avatar of totalitarian collectivism, I’ve been arguing that the elements of Hegel I’ve been discussing basically boil down to a kind of complex systems theory, focussed on the production of normativity by social practice.  The production of normativity by a complex social system is, of course, exactly the topic of Hayek’s late work on the nature of law and liberal constitutionalism.

In this late work, then, Hayek makes a double move.  On the one hand, he is critical of forms of individualism that analyse the individual wholly independently from the society that forms them.  In this respect, Hayek’s late work represents a major break with Mises (and this is one of the reasons why the schism within the Austrian tradition between Mises-oriented and Hayek-oriented figures is as significant as it is).  Individual behaviour, Hayek argues, is shaped by societal norms, and individual freedom ultimately needs to be grounded in those norms.  On the other hand, Hayek analyses those societal norms as themselves the emergent property of a dynamic complex system.  For Hayek, the norms that find expression in common law are products of “human action but not human design”.  Moreover, because the system in question is dynamic, the norms themselves, as well as our understanding of those norms, are evolving, in a way that means the ‘correct’ norms cannot even be attributed to any specific aggregate pattern – Hayek thereby avoids the trap that Brandom calls ‘regularism’.  In both of these respects, then, Hayek’s late metatheoretical apparatus is strongly compatible with the basic Brandomian-Hegelian account of the social constitution of norms.  This even extends to the centrality of common law to their accounts of normativity.

In summary: I’ve argued that both Rousseau and Mises are, so to speak, ‘problematically non-dialectical’, each reifying a particular moment of the complex social system they are studying as ‘foundational’ and ‘self-constituting’ or ‘self-validating’.  For Rousseau, this reified moment is the ‘general will’, leading Rousseau to a dogmatic and somewhat sinister collectivism.  For Mises, this reified moment is the acting individual, leading Mises to a dogmatic aprioristic Cartesian individualism.  Finally, I argued that Hayek’s late work – and basically I’m talking here about volume 1 of ‘Law, Legislation and Liberty’ – is, by contrast, strongly compatible with the metatheoretical approach I’m advocating.  

But so if Hayek’s late work gets my metatheoretical seal of approval (not, obviously, that this means anything to anyone other than me…) am I suggesting that Hayek’s general political-economic approach is the right one?  No – I’m not.  And that’s because I have a number of disagreements with Hayek’s framework at the theoretical (rather than metatheoretical) level.  I very briefly summarised some of those disagreements in my post on volume 1 of ‘Law, Legislation and Liberty’.  In my next post I want to return to these issues – but now armed with more of the intellectual resources of Brandomian-Hegelian ‘empirical dialectics’.

I recently read Don Lavoie’s ‘Rivalry and central planning’ – an account of the ‘socialist calculation debate’ which I can’t recommend highly enough.  Lavoie is a partisan – his goal is to present a ‘revisionist’ account of the debate which makes the case for the Austrian side.  But it’s also simply an excellent piece of ‘internalist’ intellectual history.  Moreover, to my mind Lavoie’s reconstruction of the Austrian arguments is much more clearly articulated than any of the debate’s ‘primary texts’, and the book is well worth engaging with on that basis.

The core of the calculation debate is, of course, over the feasibility of socialist or communist central planning.  The debate is multi-faceted and it’s not the goal of this post to even begin to attempt to summarise it, but the postage-stamp-sized version of the Austrian argument is that central planning isn’t going to work well because of a set of ‘knowledge problems’: local knowledge, tacit knowledge, and – especially – ineradicable uncertainty mean that central planners simply don’t have the categories of information required to engage in efficient economic planning.  Therefore, the argument goes, the Marxist goal of rational central planning is a pipe dream.  By contrast, the Austrians argue, market-based ‘spontaneous order’ – or ‘catallaxy’ – can be responsive to local knowledge, tacit knowledge, and the forms of unexpected discovery associated with ineradicable uncertainty, in a way that central planning never can be.  Therefore markets are better than planning.

Along the way in making this argument, Lavoie summarises two elements of Marx’s work.  First, Lavoie argues that Marx is committed to a naive concept of planning, which hasn’t reckoned with the very serious obstacles to ‘rational’ central planning.  I think this is a very fair critique – though again I don’t want to get into this side of things in this post.  Second, Lavoie gives an excellent summary of one dimension of Marx’s account of capitalism: the centrality of uncertainty and disequilibrium to market dynamics.  

It is this second element of Lavoie’s summary of Marx’s argument that I want to focus on in this post.  Interestingly – and in my view correctly – Lavoie argues that this dimension of Marx’s argument is a point of commonality between Marx and the Austrians.  Lavoie makes this argument by contrasting this position (shared by Marx and the Austrians) with two rival understandings of market dynamics.  On the one hand, there is the position that sees capitalist markets as simply chaos (a view that Lavoie argues Hayek wrongly attributes to Marx).  This is wrong – markets are not chaos – rather they are a form of ‘spontaneous order’.  On the other hand, there is the position that capitalist markets can in principle attain perfect efficiency (a view expressed by, for example, the first fundamental theorem of welfare economics).  This is also wrong, because of the ‘knowledge problem’ associated with ineradicable uncertainty.  It is impossible for markets to attain perfect efficiency, even in principle, because one of the functions of markets is to discover information that is not and cannot be known to any of the market participants when those market participants take the actions that will, ultimately, lead to the discovery of the information.  In Hayek’s phrase, market competition is a ‘discovery procedure’ – and precisely because it is a discovery procedure market actors can in principle never possess the kind of perfect knowledge required for fully efficient coordination.  For the Austrians, in other words, a large part of the value of markets lies in their failures of coordination, in their constantly renewed moments of disequilibrium, because such moments of disequilibrium are a necessary precondition of the production of the knowledge that can then be disseminated through the price system.  (Indeed, the workings of the price mechanism are themselves one of the mechanisms via which such knowledge is discovered.) I’m being very telegraphic here – Lavoie spells all of this out in much greater detail and one day I would like to too – but that’s the general idea.

One strand of Lavoie’s argument in the second chapter of ‘Rivalry and central planning’, then, is that Marx: a) understands this dimension of capitalism very well; b) regards this dimension of capitalism as a flaw rather than as a virtue of the system; c) believes that this feature of capitalist markets can feasibly be replaced with a form of central planning that would not exhibit disequilibrium dynamics; and d) is wrong about this.

Now, I think Lavoie is right to say that Marx understands this element of capitalism very well.  Ironically, this element of Marx’s argument is frequently missed by both Marx’s critics and his defenders.  Marx’s critics frequently don’t understand how sophisticated and developed Marx’s understanding of market dynamics is.  Of course, Marx doesn’t use the Austrian term of art ‘catallactics’, but Marx in ‘Capital’ is definitely giving an account of a complex spontaneous order that operates via constantly renewed disequilibrium.  Indeed, this element of Marx’s account of capitalism deeply informs Schumpeter’s concept of ‘creative destruction’.  At the same time, many Marxists are also indifferent to this element of Marx’s argument.  Perhaps a bit provocatively, I think you could give a reasonable first-pass typology of a range of strands of recent Marxist theory in terms of what they miss about this element of ‘Capital’.  On the one hand, there are forms of ‘political Marxism’ which see Marx’s contribution as an emphasis on class conflict – either at the micro level of the site of production, or at the national level of ruling capitalist class versus the proletariat, or at the international level of core versus periphery.  Of course all of these forms of class conflict are indeed essential to Marx’s account of capitalism – but it is easy for accounts of Marx’s argument that emphasise these issues to miss the elements of Marx’s analysis that do not focus on any of these forms of conflict, but rather on the ‘spontaneous order’ that emerges from dispersed social action.  On the other hand, there are forms of Marxism that make central use of categories which are presented in Marx’s analysis as ‘emergent’ phenomena – and yet ‘reify’ such categories in a way that severs them from the explanatory apparatus developed in ‘Capital’.  In my view quite a lot of recent Hegelian and ‘value form’ Marxism can be understood in this way – that is, as treating what are for Marx emergent categories in ways that render them analytically opaque by losing track of the mechanisms of their emergence.  In this sense, I think it’s important to see that ‘Capital’ is a ‘microfounded’ account of large-scale emergent phenomena, and looking exclusively at either ‘side’ of that dichotomy (either just the microfoundations or just the large-scale categories) will give an unrepresentative account of what Marx is doing.  (As always, I need to flag here how much of my own understanding of Marx’s argument is informed by N. Pepperell’s work.)

On this side of things, then, I think Lavoie is exactly right in his re-presentation of Marx’s analysis of capitalism.  Moreover, I think Lavoie is right to say that Marx thinks the disequilibrium dynamics he analyses are bad qua disequilibrium dynamics.  For example, one of the many dimensions of capitalism that Marx is interested in is large-scale economic crisis – financial system meltdown, the forms of underemployment of resources associated with recession and depression, boom and bust cycles, etc. – and of course these elements of capitalist dynamics are intimately connected to the constantly renewed disequilibrium and speculative uncertainty that both Marx and the Austrians agree are core to capitalism as a system.  For Lavoie, Marx thinks that this kind of disequilibrium process can be contrasted with rational central planning which would not exhibit these attributes, and rational central planning is to be preferred on this basis.

So, I think Lavoie is right about all of this.  I also think Lavoie is right that Marx is severely underestimating the difficulties associated with ‘rational’ planning.  So what’s the problem?  Isn’t Lavoie – and, by extension, the broader Austrian critique of Marx – just right, full stop, by my lights?

Well, in a sense, yes.  But here’s the issue.  In Lavoie’s discussion of Marx, he foregrounds Marx’s dissatisfaction with the disequilibrium dimensions of capitalism, and their consequences.  As I say, I think this is indeed a central element of Marx’s critique.  But I think it’s important to remember that there are other dimensions of Marx’s critique of capitalism.  In particular, one of the most central dimensions of Marx’s critique of capitalism is that it is oppressive.  And I think it is important to distinguish two different senses in which capitalism is oppressive, for Marx.

Here I am shifting from discussing Lavoie’s book – which is quite narrowly focused on the specifics of the calculation debate – to discussing the broader debate between the Austrians and Marx.  That is, I’m giving myself permission to paint with broader brush strokes.  Broadly speaking, then, I think there are two key elements of the Austrian enthusiasm for market catallactics which Marx’s analysis challenges, quite distinct from the critique of disequilibrium dynamics in general.

First, there is the problem of power and direct oppression.  I think it’s fair to say that one of the reasons that Austrian economists (and many other economists more broadly) often value markets as freedom-enhancing is the idea that market exchanges are voluntary, and in this respect differ from commands laid down by a violence-monopolising state.  One of the things that Marx relentlessly emphasises in ‘Capital’ is the power relations that run fractally through the apparently ‘voluntary’ exchanges of the market – in particular, but by no means limited to, the exchanges associated with labour.  For Marx, the pro-capitalist emphasis on voluntary exchange is frequently simply engaged in denialism about the many forms of coercion that can be and are encountered in ‘market society’.  This is the element of Marx that contemporary ‘political Marxism’ emphasises.

So that’s one problem with the Austrian position: are we sure that things are so free out there, really?  This problem is, as it were, at the micro level – it concerns the kinds of interactions that are going on all the time in capitalist society.

The second problem is at the macro level – that is, at the level of emergent spontaneous order, or ‘catallaxy’.  Here’s the problem: granted that market society is a catallaxy in which a spontaneous order emerges as the product of human action but not human design, what actually is the spontaneous order?  Is it good?  Or is it oppressive?

This, in my view, is where the large-scale elements of Marx’s argument in ‘Capital’ start to bite – because Marx in that work has an extremely involved account of the spontaneous order of capitalist society, built up from a micro-level analysis of a very large number of different social practices and institutions that comprise capitalist society.  Again, my goal here isn’t to summarise ‘Capital’, but Marx’s overarching claim is that, yes, capitalism represents a ‘catallaxy’, but it is a bad catallaxy: the emergent patterns of capitalist society have an oppressive, not a liberatory, impact on a large proportion of capitalism’s inhabitants.  The invisible hand of the market is not beneficent, even metaphorically; it is drenched with blood.

Now, this obviously isn’t to say that we need to accept this element of Marx’s argument (which, again, I am not even attempting to summarise here – I am simply gesturing at Marx’s conclusions).  But it is important to recognise that this element of Marx’s argument is not a rejection of ‘catallactics’ qua catallactics.  The claim is not simply that spontaneous order is bad because it is spontaneous, and because it exhibits the forms of inefficiency that are intrinsically associated with any ongoingly-evolving spontaneous order.  Rather, Marx’s critique is directed at a specific spontaneous order – the spontaneous order of the capitalist society he is examining.  Marx’s claim in ‘Capital’ is that this spontaneous order is bad – and this argument cannot be rebutted simply by making a case for catallactics against planning.  We need to engage with the specifics.  

This, I think, is where the critique I made a few posts ago about Austrian economics’ inconsistent application of the principle of epistemic limits comes into play.  My view is that Austrian economics wants to make two qualitatively different categories of argument.  One is that the actually-existing spontaneous order of capitalism is good not bad, that the invisible hand is more to our benefit than to our detriment, etc.  This is a normative social-scientific claim, which relies on us being able to analyse and understand in quite some detail the structure and dynamics of both capitalism and alternative political-economic systems.  The other category of claim Austrian economics wants to make is that our knowledge, understanding, and ability to act in ways that have the consequences we desire, are all so limited that we had better leave well alone in the face of spontaneous order – essentially ‘Chesterton’s fence’ at the level of large-scale political economy.  But it seems to me that these two positions are in significant tension.  If our understanding is so limited, how are we in a position to adequately assess the virtues of capitalist spontaneous order?  On the other hand, if we claim to understand capitalist catallactics well enough to make a dispositive case for capitalism’s virtues, what has become of our epistemic humility?  It seems to me that Austrian economics moves back and forth between these two epistemological poles of its argument, and that there is a tension – even, perhaps, at times, a convenient double-standard – in this movement.

So, let’s say we grant (as I think we should) that Marx is being naive in his imagining of a form of rational economic planning that could replace the spontaneous order of capitalism.  The claim I’m making is that this doesn’t in itself dispose of Marx’s broader arguments, because there are two further scenarios that need to be considered.  First: the possibility that the specific spontaneous order of capitalist society is sufficiently awful that forms of planning – even with all their inefficiencies and oppressions – are nevertheless an improvement.  Second: the possibility that other spontaneous orders are available – that we can transform our practices in ways that replace bad catallactics with good catallactics, or at least worse with better.  Moreover, there is arguably significant overlap between these scenarios, because one of the surprising claims underlying the Austrian critique of planning is that the institutions of a planned economy must themselves in practice be a complex system which does not lie under any individual’s control, if they are to function in the ways they often do in practice.  Thus there is a startling moment late in Lavoie’s book in which he argues that the USSR under Stalin is a good example of catallaxy:

although the Stalinist economy ‘professes to be planned,’ to use Hayek’s phrase, it in fact relies on the outcome of the clash among rivalrous, decentralised decision-makers – that is, it is anarchically rather than consciously organized. (155)

I think this view has a lot to recommend it – but in my view it also risks wreaking havoc with a lot of other Austrian arguments.  For if Stalin’s USSR and US market society are both examples of catallaxy, then it’s unclear the extent to which the categories ‘catallaxy versus planning’ can get a purchase on the relevant comparative institutional question.

Perhaps this seems like a facile debating point – and perhaps it is.  At the very least, this issue merits a lot more time and care than I’m giving it here. But I think this problem nevertheless captures something, which is that there are countless possible spontaneous orders.  Almost all of those spontaneous orders contain some degree of planning.  It’s unclear to me, then, how a general emphasis on catallaxy can guide us in choosing which actions we wish to take, in order to influence, in whatever ways, the specific nature of the spontaneous order we inhabit.  One response to this problem is, of course, full stoicism, or quietism.  But if we reject that route – as all participants in the calculation debate have, to some extent – then I see no real alternative to wrestling with concrete social-scientific questions of political economy.  The level of abstraction at which the socialist calculation debate is carried out cannot in itself be an adequate guide to political-economic action.  Which, in fairness, I think all the participants, on both sides, already knew – but that’s all I’ve got to say for now.

Fully in the ‘thinking out loud to externalise things’ space here still – I just started reading Frank Knight’s ‘Risk, Uncertainty and Profit’, and I wanted to externalise a few thoughts prompted by the book’s framing.

Knight kicks things off with a carefully caveated comparison of economics to physics.  He argues that the marginalist analysis of perfect competition is analogous to the ‘core’ laws of nature in physics, in the sense that it represents a ‘tendency’ that is always present within the behaviour of the free enterprise system.  However, this tendency does not determine the behaviour of the system on its own – there are other, countervailing tendencies which also need to be analysed.  If you naively consider only the tendency described by the laws of perfect competition in your analysis, then you will draw badly faulty conclusions about the real dynamics of the economic system.  So Knight advocates a two-step analytic approach.  First, you understand the tendency of the system associated with the assumption of perfect competition, then you analyse all the ways in which the system in fact departs from this tendency – and the latter, in practice, is where much of the work of economic analysis lies.

This seems to me to be an extremely common approach to economics – start with the idealised perfect competition model, then modify it with departures from perfect competition until you get something that seems like a reasonable approximation of the actual behaviour of whatever slice of economic life you’re studying.  I’m not saying this is necessarily the right way to approach things – in general I think that methodological pluralism is a virtue.  But let’s say that we accept this two-step approach.

The question then is: what are the important departures from perfect competition?  I think it may be useful as a first pass to distinguish two broad categories of departure (not, of course, that these are exhaustive): first, those associated with knowledge, and second those associated with power.  

Knight is primarily interested in the first category, knowledge.  Specifically, Knight is interested in the distinction between risk and uncertainty, where ‘risk’ denotes quantifiable uncertainty, and ‘uncertainty’ denotes unquantifiable uncertainty: known unknowns and unknown unknowns, as it were.  Knight discusses John Bate Clark’s account of the importance of change in generating profit.  Because the tendency of perfect competition is towards zero profits, we need an account of the source of profits that lies outside the perfect competition model.  Knight quotes Clark arguing that change provides that source, because change upsets the equilibrium of the economic system.  Knight in turn argues that Clark’s categories aren’t adequate to this insight, because if future changes – or even the probabilistically analysed possibility of future changes – are properly understood by economic actors, then they do not really count as a relevant departure from complete knowledge, and therefore do not in fact upset the equilibrium.  What generates the dynamic artefacts that Clark attributes to ‘change’ is really uncertainty – things that we do not know in the fundamental sense that we cannot even quantify the risk associated with them.

So that’s one kind of departure from perfect competition.  The other kind (that I’m talking about in this post) is power.  The idea in a perfect competition model is that every social actor is a price taker because ‘exit’ is trivially easy.  But of course there is a vast literature on the ways in which market power involves departures from this model, giving some economic actors more bargaining power than others.  I’m pretty daunted by this literature and haven’t even really begun to work through this space – but this is obviously a lot of what economics ‘is’.

It seems to me that ‘mainstream’ neoclassical economics can and has – at least in part – incorporated both of these categories of departure from perfect competition into its analytic resources.  There is a lot of mainstream economics on information and uncertainty, and there is a lot of mainstream economics on market power and bargaining power.  And yet, at the same time, there are figures on both of these two ‘wings’ of economics arguing that the neoclassical apparatus is not adequate to the core of each of these insights.  On the ‘uncertainty’ side there are the Austrians and Austrian-adjacent theorists, arguing that the neoclassical apparatus blinds its advocates to the centrality of our epistemic limits to economic dynamics.  And on the ‘power’ side there are the Marxists and Marxist-adjacent theorists, arguing that there are many forms of power operating in the economy that the neoclassical mainstream has failed to reckon with.

I don’t really have a view on all of this, at least not a view worth having – but I guess what I want to do in this post is lay down a marker to keep an eye on these two issues as I try to make my way through ‘mainstream’ economics.  My intuition, baseless or otherwise, is that these two ‘wings’ of heterodox economics (where, very crudely, the right wing corresponds to an emphasis on epistemics, and the left wing corresponds to an emphasis on power) have a point when they criticise neoclassical economics from the perspectives of these preoccupations.  And yet I also suspect that neoclassical economics has a lot more resources for tackling these issues than its heterodox critics often give it credit for.  Anyway, this is one (very partial) way (among many) of slicing up the intellectual terrain, and I wanted to jot it down.

Usual caveats in spades here.  First: I’m writing this quick post to get thoughts organised in my head rather than because I’m claiming any very significant knowledge.  I’ve read some Hayek and some Mises and some other odds and ends within the tradition of Austrian economics, but my knowledge of this tradition is still fairly slight, all up.  Second: of course almost every tradition is very diverse and no quick summary – even one much better informed than mine – is going to be adequate to that diversity.  With appropriate caveats, though, what are some key elements of the tradition of ‘Austrian economics’?  I think I want in the first instance to distinguish four elements of this tradition.

  1. Subjective value theory.  I discussed this a few posts back – ‘subjective value theory’ can potentially mean quite different things, but I think a lot of this strand of ‘Austrian economics’ has been absorbed into the mainstream of economics via the triumph of marginalism.
  2. An emphasis on catallactics and catallaxy, or spontaneous order.  Again, although mainstream economics may not use this terminology, economics as a discipline has been fascinated by the ‘spontaneous order’ of the market since Adam Smith.  Austrian economics I think gives this theme particular emphasis, but of course it’s hardly unique in discussing it.
  3. A strong commitment to laissez faire economics and classical liberal governance.  Again, the Austrians are hardly unique in this, but it’s a particular preoccupation of the tradition.


  1. A strong emphasis on the limits of our knowledge.  I think it’s reasonable to say that this is what’s most characteristic of the Austrian tradition.  There are a bunch of different ways in which economists working in the Austrian tradition talk about the limits of our knowledge: the straight-up overwhelming complexity of the spontaneous order we inhabit (who can credibly claim to begin to understand the detailed dynamics of a system this complex?); the unknowability of local or (especially) tacit knowledge for an outside observer; the sheer diversity of values among the countless people that inhabit our societies (how can we begin to get a handle on ‘preferences’ when individuals are so diverse?); the intrinsic unknowability of the future, and the fact that most of our economic categories and institutions are essentially time-bound and time-dependent, and therefore contain an unknowable element; and finally the fact that creative innovation – of technologies, of practices, of ideas, etc. – is a key and unpredictable element of economic dynamics, and likewise introduces an intrinsic uncertainty or unknowability into our efforts to analyse those dynamics.  Many of these arguments about the limits of our knowledge are somewhat ‘philosophical’.  That is to say: Austrian economics isn’t just worried about what we don’t know as a contingent result of limited data (for example) – it is interested in the idea that there are many aspects of our social reality that we cannot know, in a relatively strong sense, and it is interested in what the implications of these intrinsic limits to our knowledge are for political economy.

I think, then, what’s most characteristic about the Austrian tradition is its use of arguments drawn from point (4) to expand on and defend the positions in points (1) through (3).  And I think this is visible in the arguments Austrian economics has with the two main alternative economic analytic ‘paradigms’ that it spends much of its time critiquing: Marxism and neoclassical economics.  For Austrians, the Marxist desire to engage in large-scale rational planning of the economy is always and definitely going to end up shipwrecked on the rocks of the intrinsic limits of our knowledge about the dynamics of the economic order.  This (obviously) is the core of the socialist calculation debate.  But Austrian economics is also extremely sceptical about the kinds of modelling (and statistical analyses) carried out by the neoclassical mainstream.  For Austrians, the apparatus of general equilibrium analysis, etc., is also hubristic in its belief that it can usefully capture the dynamics of a system that is, for the many reasons gestured at above, well beyond our capacity to adequately understand.  For Austrians, then, the entire neoclassical apparatus – with its belief that it can model the dynamics of capitalism, and then identify the ‘market failures’ associated with those dynamics, facilitating government intervention to address those failures – is sort of Marxism lite.  For sure (most) neoclassical economists don’t aspire to a fully rationally planned economy – but neoclassical economics is still much too confident in its ability to rationally comprehend the relevant complexities, and is therefore also much too confident in its ability to intervene in the economy on the basis of such models.  (This is the sense in which Milton Friedman – though clearly a political ally of the Austrians – is emphatically not an Austrian himself either by temperament or in terms of his analytic apparatus: he is a mainstream economist who happens to also be committed to (1) through (3) above.)

Now, I have a range of criticisms of various Austrian economists, some of which I’ve already discussed on the blog, some of which I’ll hopefully eventually get to.  So I wouldn’t want the above summary to be understood as signalling my own commitment to the ideas of this tradition!  But I do think that the ‘epistemic’ element of Austrian economics is one that is worth taking seriously.  As I gesture at above, I think the ‘epistemic’ arguments in Austrian economics are in fact very varied – there are a lot of quite different arguments that can in principle be captured under general headings like ‘the knowledge problem’.  I think this can be quite confusing!  So one of the things I want to do, as I continue engaging with this tradition, is try to tease apart the different Austrian arguments about the limits of our social-scientific knowledge, and the different implications those arguments can plausibly be taken to have for political economy.

Hayek on rules and order

August 11, 2022

I just read the first volume of Hayek’s ‘Law, legislation and liberty’ – ‘Rules and order’ – and as usual, I want to put up some quick notes while it’s still fresh in my mind.  I don’t have a specific argument I want to make here, so this is again largely ‘notes to self’.

I guess the first thing to say about ‘Rules and order’ is that in my view it’s easily the best work by Hayek that I’ve read.  ‘Law, legislation and liberty’ is the culmination of Hayek’s late turn to broader social theory, away from ‘traditional’ economics or political economy.  For my money, both ‘The road to serfdom’ and ‘The constitution of liberty’ – two earlier works that are arguably precursors – are deeply flawed.  ‘The road to serfdom’ is a popular polemic that I feel isn’t really going to be persuasive to anyone who isn’t already sympathetic to its arguments.  ‘The constitution of liberty’, while a much longer and more serious work than ‘Serfdom’, is also a huge mess.  I wrote up my thoughts on ‘The constitution of liberty’ last year on the blog, but in summary: it is packed with ad hoc arguments and claims that don’t seem to fit in to any very coherent broader thread, and moreover its more ‘philosophical’ or social-theoretic claims are in my view often very muddy or confused in their expression.

‘Rules and order’ fixes both of these problems – in a sense it’s ‘The constitution of liberty’ done properly.  It develops a coherent, clearly articulated, wide-ranging argument, and that argument moreover in my view has a lot to recommend it. (The stuff that I didn’t like in ‘The constitution of liberty’ is likely still to come in volumes two and three of this work, but I’m talking about volume one here.)

So, on the substance, I’m going to very briefly note a few themes in no particular order.

  • The most central theme of the book is the distinction between planned and spontaneous social order.  A planned order is chosen and directed by some specific social actor or actors’ intention; a spontaneous order is the emergent effect of many decisions none of which are oriented to producing this order as their outcome.  This is Hayek’s central distinction, and it seems like an important and useful one.
  • Hayek then derives from this distinction an account of two different forms of rationalism – one which sees order as necessarily planned, and one that is willing to accept spontaneous order.  For the former kind of rationalism, only order that has been fully chosen, intended and constructed is legitimate order; for the latter kind of rationalism, order may emerge without the intention to produce that order, and we should not intrinsically see this lack of intention as a problem with the order.  Hayek calls the former a ‘constructivist’ rationalism, and the latter an ‘evolutionary’ rationalism.  (He also cites Popper’s articulation of the same distinction in terms of ‘naive’ versus ‘critical’ rationalism.)
  • Hayek then connects this distinction to two different ways of thinking about normativity or social rules.  From a ‘constructivist’ perspective, rules must be fully chosen if they are to be legitimate – this approach results in social contract theory as the understanding of the basis of political legitimacy.  From an ‘evolutionary’ perspective, rules or norms may have evolved over time without being specifically intended in this form by any social actor, and this should not intrinsically be seen as a problem with them.  Here the model is not social contract theory but common law.
  • This distinction also results in a difference in explanatory priority: for social contractarians the willing and choosing subject is fundamental, and rules, norms, a political order, etc. are chosen by this subject.  For the evolutionary approach, the subject is always formed by an existing tradition, and cannot get ‘outside’ of that tradition (though it may transform it).
  • Within the spontaneous order side of his dichotomy, Hayek distinguishes in turn two different ‘levels’ of spontaneous order.  One is the spontaneous order of the marketplace (or of capitalist political-economic dynamics more broadly).  This is the focus of the first (long) period of Hayek’s career, in which he is primarily working as an economist.  The second kind of spontaneous order is the emergence of rules or norms that can find expression in law.  This is the focus of the latter period of Hayek’s career, culminating in ‘Law, legislation and liberty’.  The latter (legal) order is an institutional precondition of the former (market) order, because a legal framework (and the normative practices associated with private property, exchange, etc.) is required for the spontaneous order of the market to get off the ground.
  • These distinctions then also allow Hayek to draw a distinction between ‘law’ and ‘legislation’, where ‘legislation’ is a constructed and intended legal order, and ‘law’ is an evolved set of norms that can be codified in common law but already exist prior to that codification as an emergent or evolved phenomenon.  In Hayek’s vocabulary, an emergent order is ‘cosmos’ and a constructed order is ‘taxis’; evolved law is ‘nomos’, and constructed law is ‘thesis’.
  • With some important but definitely secondary caveats, Hayek basically thinks that cosmos and nomos are good, taxis and thesis are bad.
  • This is fairly close to a kind of Burkean conservatism.  However, Hayek differs from at least the most ‘vulgar’ forms of Burkean conservatism in his account of, and at times enthusiasm for, ongoing evolution of the spontaneous legal order.  [Of course Burke also says that tradition can evolve, and this is central to his Whiggism, so we’re arguably dealing with somewhat fine distinctions here, but to me Hayek seems to be more centrally an ‘evolutionary’ thinker than Burke.]  There are two mechanisms for the transformation of the evolved legal tradition that Hayek discusses.  First, when we are confident that the tradition has got it wrong we can simply use legislation to transform it.  Second, the application of the tradition in practice may reveal internal inconsistencies within the evolved rules we are applying – in this scenario, we must make decisions about which dimensions of the tradition to apply and which to discard.  This latter, in particular, is a mechanism via which the tradition can genuinely evolve, without the intervention of direct ‘constructivism’.
  • Again to a first approximation, Hayek thinks that evolved law is legitimate, while legislation has a tendency towards despotism.  The goal of maintaining a legal order protective of the forms of economic and political liberty that facilitate the (‘lower-order’) spontaneous order of the market, therefore, in general requires constraining legislation – and certainly constraining the specifically targeted decisions of an administrative state – and deferring to common law traditions.  
  • This connects (a little clumsily, I think?) to the distinction between ‘impartial’ law and ‘the will of the powerful’ that animates the argument of ‘The constitution of liberty’.

Ok – those are some of the major themes of ‘Rules and order’.  Obviously there’s more to it than all this, but this is a reasonable first pass I think.

I don’t really want to spend much time on ‘editorial comment’, as it were, but a few quick remarks alongside this summary.  First, as I say, I think this is all a very well-developed metatheoretical apparatus.  Second, I think that meta-theoretically speaking it is largely correct.  I obviously differ from Hayek in a large number of ways at the level of theory, but the apparatus he develops here for the discussion of spontaneous order seems to me to be mostly good and right.  I’m not the person to develop this, but I think there are worthwhile connections to be made between Hayek’s discussion of common law and Brandom’s historicised normative pragmatics, for example.  I recently read Burczak’s ‘Socialism after Hayek’, one of the arguments of which is that Hayek can usefully be understood as a ‘postmodern’ theorist.  I think that argument is sometimes a bit of a stretch, but I see what Burczak is getting at.  For myself, though, I would articulate the point by saying that at a metatheoretical level, the apparatus developed in ‘Rules and order’ is really quite pragmatist (in the philosophical sense).  From my perspective this is a virtue (though Hayek would no doubt not have welcomed the comparison!)

All that said, I want to mark a few ways in which I would want to depart from, or at least significantly modify, Hayek’s apparatus.

First, as ever, there is the issue of conflict, coercion, and dissensus.  Hayek throughout his corpus is systematically under-attentive to many of the conflictual dimensions of social life.  Of course, Hayek is interested in competition, but there are many forms of conflict that are also central to social life that Hayek is basically indifferent to.  And yet conflict is part of these spontaneous orders too!  Conflict contributes to the production of these spontaneous orders!

Second, and relatedly: Hayek’s tacit model of common law here presupposes a high level of internal homogeneity of norms within the relevant society.  And yet why should we assume that such normative homogeneity exists?  What if dimensions of common law are better characterised by one subcomponent of society simply imposing ‘their’ norms on another?

Third: If we accept (as I think we should!) Hayek’s account of the evolution of norms through the need to choose between conflictual implications of our existing normative framework in new cases, and if we tether this element of Hayek’s apparatus to a more conflictual understanding of society as a whole, the implication, I believe, is that the ‘evolutionary’ process in question may necessarily at times become radically transformative: we may have no choice but to choose some dimension of ‘our’ normative tradition over another, in a way that has radically transformative implications for the entire tradition going forward. Again, I don’t think Hayek would exactly disagree with that, but I think it’s clear that his basic model of societal evolution is more ‘incrementalist’ than ‘revolutionary’ – and yet from my perspective his metatheoretical apparatus is neutral on this question.   

Fourth and finally: given that social actors are (of course) engaged in intentional action in order to produce the spontaneous orders Hayek is discussing, and given that some forms of intentional action may be quite far-reaching, where are we to draw the line between intentional action that contributes to the creation of a spontaneous order, and intentional action that aspires to override a spontaneous order?  It is not clear to me that this distinction is as clear-cut as Hayek implies.  Even if we think about large-scale purposive state action (Hayek’s paradigmatic case for hubristic constructivist interference in a spontaneous order): states are still social actors within a complex system, and their actions also both contribute to and are shaped by an emergent spontaneous order. The distinction between ‘evolution’ and ‘construction’ seems to me to be muddier than Hayek’s discussions of this issue often suggest.

All of these are points I want to return to as I continue to work through and work with Hayek.  But for now I just wanted to get these notes down.

I’m still circling here around Hayek and the calculation debate, but I want to take a quick detour into the work of Michael Polanyi on science.  Unlike Hayek, with whom I am still getting to grips, I know Polanyi’s key works well – I drew on Polanyi heavily in my Ph.D. thesis, and I think Polanyi’s 1962 article on ‘The republic of science’ is one of the best works in the social studies of science full stop.  Polanyi is also, in a fairly direct sense, engaged in a closely related intellectual project to Hayek – both Polanyi and Hayek were members of the Mont Pelerin Society and both are interested in the value and functioning of ‘spontaneous order’, in contrast to communist-style planning.  I think Polanyi’s work is of great intrinsic interest and value – but the reason I’m blogging on it now is because I want to start to draw out parallels (and perhaps points of difference) between Polanyi’s analysis of science as spontaneous order and Hayek’s arguments about spontaneous order in market society.

For Polanyi, then, it is critical to the functioning of science as an epistemic system that science be in a relatively strong sense unplanned, decentralised, polycentric, etc.  That is to say: scientific knowledge, for Polanyi, cannot and must not be understood on the model of the knowledge held by a single mind.  It is an intrinsic, not merely a contingent, feature of scientific knowledge that it be dispersed across a complex spontaneous order (‘the republic of science’).  For this reason, Polanyi argues, efforts to regiment and homogenise science are misguided at a fundamental level.

Polanyi does not, however, have one specific argument for this position – rather he has a cluster of arguments that need to be carefully distinguished.  In this post I want to highlight four different elements of Polanyi’s argument for the intrinsic polycentricity of science, and make some provisional and tentative connections to broader (Hayekian) arguments about spontaneous order.

First, and arguably most superficially, science must be ‘decentralised’ for Polanyi because there is simply too much of it for any one human mind to hold.  Perhaps there was a time when the greatest (and most well-resourced) intellectual figures could plausibly achieve mastery over all the major scientific (or pre-scientific, depending on how you periodise the intellectual history) domains, but in recent centuries it is simply impossible for anyone to gain mastery over anything more than a very small slice of our scientific knowledge.  Science is, in this very direct sense, a vast cognitive division of labour.  “We” possess vast scientific knowledge, but this “we” is a vast community – any individual within that community can of course possess only a small fragment of that knowledge.  Given that science hangs together in ways that span individuals’ knowledge, it can very plausibly be claimed that “we” know things that no individual knows.  (Even an individual scientific paper may be co-authored by a range of researchers with different skills and specialisations, such that the conclusions of the paper as a whole are not ‘known’ in the strongest sense of full comprehension by any individual author.  Think, then, how much more true this must be of broader coherent swathes of scientific knowledge.)  Scientific knowledge, then, is intrinsically dispersed simply because there’s so much of it.

Polanyi’s next argument is about tacit knowledge.  Scientific knowledge is often understood as explicit propositional content, paradigmatically published in scientific journals – but Polanyi emphasises that “knowing how” is as important to the scientific enterprise as “knowing that”.  We can, moreover, distinguish two different senses of “knowing how”.  First, there are the kinds of skills that cannot be communicated by means of explicit propositional knowledge: for example, how to operate such-and-such a piece of equipment is the kind of thing that one may have to learn in practice.  Second, there are the kinds of creativity and inspiration that for Polanyi are critical to the scientific enterprise, but that are missed by ‘positivist’ approaches that reductively overemphasise propositional knowledge.  (For what it’s worth, I’m less impressed than Polanyi is by ‘divine spark of creativity’ arguments – I think Polanyi is at times problematically mystical when discussing this sort of thing.  But of course Polanyi is right that tacit knowledge is important.)  For tacit knowledge, as for explicit knowledge, you could in principle I suppose imagine some sort of superhumanly skilled and creative individual who would render the division of labour associated with tacit knowledge redundant, but in any imaginable real-world scenario of course we are again dealing with a division of labour here, driven perhaps by slightly different factors from the division of labour associated with explicit propositional knowledge.

There’s a third argument in this broad space that isn’t really central to Polanyi, but that I think I might as well canvas since I’m itemising arguments: this is the problem of the incentives for knowledge sharing.  There’s a whole bunch of work, much of it coming out of Robert K. Merton’s sociology of science, about how the reputational economy of science incentivises publication of research findings.  But there are also many institutional incentives that push in the other direction: difficulties in publishing replications; reasons not to publish the ‘wrong’ findings; reasons to hoard data or other intermediate scientific inputs; incentives to commit fraud.  All of these problems can be (a bit crudely) grouped under the heading of “incentives not to share knowledge” – and if knowledge cannot be shared then the knowledge is intrinsically going to remain ‘decentralised’.

So these are three reasons why we would expect science to be intrinsically polycentric or decentralised.  Polanyi’s really interesting argument, though, from my perspective, is only weakly related to these.

Polanyi’s final argument is that science must remain decentralised because disagreement is central to the entire scientific project.  Science operates via (to use more Popperian/Lakatosian language) conjectures and refutations.  There is therefore a double-movement (apparently but only apparently paradoxical) to the scientific knowledge-production process.  On the one hand, would-be scientists must be taught the best existing knowledge, and their ability to demonstrate expertise in that knowledge is what grants scientists entry to the ‘club’ of recognised scientific research.  On the other hand, scientists must be willing and able to challenge existing knowledge in the service of making new discoveries and overthrowing old errors.  The first of these two elements pushes science towards a certain uniformity of belief – a shared canon is established and disseminated via scientific training, etc.  The second of these two elements, however, pushes science towards an intrinsic diversity of belief.  Scientific progress simply cannot happen if scientists don’t disagree.  More strongly – scientific knowledge itself cannot be scientific knowledge (but will rather become dogmatism) if the scientific community does not have the internal capacity to reject it.  There must therefore be not only cognitive division of labour – whereby knowledge and expertise are divided up across the scientific community by means of specialisation – but also genuine substantive pluralism (a stronger claim).  Although the preponderance of scientific opinion at any given time can reasonably be taken to represent ‘scientific knowledge’, it is intrinsic to the very concept of scientific knowledge (which must be fallibilist or it is no longer science) that any given scientific ‘consensus’ may be wrong, and that (therefore) scientists who are prepared to reject any given element of scientific ‘knowledge’ must also be part of the scientific community.

I think Polanyi is right about all of these points (caveats aside).  Moreover, I think the last point – about the necessary pluralism of the scientific endeavour – is a profound insight, which really does explain why science as a collective project must be in some sense polycentric and decentralised at its core, if it is to function as science at all.

But what about Hayek?  As I say, my goal here at the moment is not really to discuss Polanyi on science, but rather to extend my own field of competence, from the political economy of science to political economy more broadly.  So what are the analogies between Polanyi’s account of science and Hayek’s account of the market as catallaxy?

Well, I want to return to this in future posts.  Very quickly and crudely, though, I think we can see analogies with all four of these points.

  1. The division of epistemic expertise in the Polanyi account of science of course corresponds to the division of labour within market society.  This is a practical ‘decentralisation’ of a sort associated with the need for specialisation.  Of course, division of labour can exist within a planned economy – but here the second analogy is that, just as no one mind can in fact master all of scientific knowledge, so no one planner (or planning bureaucracy, or whatever) can master the information required to organise such a division of labour.  This is, as it were, the ‘weakly practical’ argument against ‘socialist calculation’.
  2. Like Polanyi, Hayek is interested in the problem of tacit knowledge.  One of the other Hayekian arguments against central planning is that tacit knowledge intrinsically cannot be communicated to the central planner, because the knowledge is not explicitated into propositional content.  This seems like it could potentially be a more ‘intrinsic’ obstacle to central planning than the difficulty of mastering the sheer quantity of ‘explicit’ information involved in economic decision-making.
  3. The problem of incentives not to share knowledge is a challenge for planners just as much as for the scientific community.  Again, one of the arguments against central planning is that there are many reasons why the information informing the plans would be faulty, due to the many incentives economic actors may have to pass on faulty information (or fail to pass on relevant information).
  4. Finally, and again I think arguably most importantly, there is the issue of intrinsic and necessary pluralism in science as a precondition of the creation of scientific knowledge (as opposed to scientistic dogmatism) at all.  Here I think the analogy is to Hayek’s arguments about the market as a discovery process.  Hayek’s argument here is not just that there are insuperable practical difficulties involved with aggregating the knowledge dispersed across the market.  His argument is, rather, that the relevant knowledge simply doesn’t exist without the pluralism of a decentralised process.  Specifically, all kinds of knowledge about what to produce and how to produce it does not exist without the experimental ‘entrepreneurial’ trial and error process of decentralised market dynamics.  Just as we need pluralism of belief for scientific knowledge to take on the positive epistemic attributes of scientific knowledge, so we need a decentralised discovery process to generate the kinds of information that could be aggregated in a planning process.  (This, I take it, is one of the key arguments of Don Lavoie’s book on the socialist calculation debate – which I recommend!)

Now, I’m not in this post trying to take a position on the strengths or weaknesses of any of these arguments – at this point I’m still trying to get clear on what the Hayekian (or, more broadly, the Austrian) arguments about planning and spontaneous order even are.  Moreover, I’m completely confident that these four points do not exhaust either Hayek’s arguments about spontaneous order, or the broader Austrian arguments in the socialist calculation debate.  (For example, I don’t think the core of Mises’ original argument about calculability is captured by any of these four points – hopefully I’ll come back to all this.)  But for myself, I find this way of breaking down some of Hayek’s arguments about ‘the knowledge problem’ of socialist planning clarifying.  Hopefully I will return to these issues, expand on these points, and supplement them with different arguments in future posts.

This blog post is still very much in the ‘working things through as I go’ space rather than the ‘I have a developed, researched argument that I’m prepared to stand behind’ space, and should be treated accordingly.  But circling again around Hayek’s positions on knowledge in political economy, some quick jotted thoughts on the socialist calculation debate.  I’ve read the major ‘primary texts’ of this debate but I still have a lot of reading to do, so as I say this is all highly provisional, with zero claims to either originality or precision.  All that said, it seems to me that one can usefully distinguish, at least in a preliminary way, between two different categories of ‘subjectivism’ on the ‘Austrian economics’ side of the debate.  (Obviously I’m using ‘Austrian’ here to denote the school of thought rather than the nationality, since for example ‘non-Austrian’ Neurath was born in Austria.)  On the one hand, you have subjectivism in the sense of the ‘subjective’ theory of value associated with the marginal revolution.  On the other hand, you have a stronger, more philosophical subjectivism that pushes back against the mathematised marginalist apparatus of general equilibrium theory.  Again, I’m aware that I simply haven’t done the reading to be asserting historical narratives with any degree of confidence, but it seems to me to be at least non-absurd to suggest that the socialist calculation debate represents a shift in the centre of gravity within ‘Austrian’ economics between the first and the second form of subjectivism.

So – start with classical economics, and ‘objective’ theories of value.  The paradigmatic value theory here is the labour theory of value, represented by Adam Smith, Ricardo, and orthodox Marxism, where value is primarily determined by the quantity of labour expended in some sense in the site of production (obviously there are a lot of epicycles around this core point, but still).  Then in the late nineteenth century you get the ‘marginal revolution’, and the eclipse of the labour theory of value in favour of a ‘subjective’ value theory, primarily driven by utility, preference and profit functions at the site of exchange.  One of the major figures associated with this ‘marginal revolution’ is Carl Menger, whose work articulating this value theory is also the foundation of the Austrian school of economics.  Here, then, we have one sense of ‘subjectivism’ – marginalist theory as opposed to the ‘objective’ value theories of classical political economy.

Now, my view is that the marginal revolution represents a significant advance in economic science.  Not to be too crude about things, but the ‘objective’ labour theory of value is nonsense – whatever flaws there may be in the marginalist apparatus, it is clearly in my view a much better starting point for thinking about economic value and price-formation than the labour theory of value.  Controversially (and unimportantly) my own view with respect to Marx is that ‘Capital’ is engaged in a complicated critique of the classical labour theory of value, and it’s a mistake to attribute the labour theory of value to Marx, at least in any straightforward sense – but this is such a heterodox and unpopular view that it really isn’t relevant to any kind of intellectual history.  For most readers of Marx – both Marxists and critics of Marx – the labour theory of value is one of the ‘central dogmas’ of Marxism.  For orthodox Marxists of this tradition, the advantages of the labour theory of value include: it provides an underpinning of an account of exploitation as objective extraction of surplus value; it locates labour – and thus the proletariat – at the ‘explanatory centre’ of political economy, which is taken to provide warrant for labour’s political centrality as subject of history (compare much Austrian economics which gives a similar dual explanatory-normative role to ‘entrepreneurs’); and because it is an ‘objective’ rather than a ‘subjective’ theory of value, it is ‘materialist’ rather than ‘idealist’.  All of these points are wrong, in my view – but that’s not important here.  The point is that as ‘bourgeois’ (or mainstream) economics shifted from ‘objective’ to ‘subjective’ theories of value, Marxism remained a hold-out (as to some extent it still is) in retaining a commitment to a classical, pre-marginalist account of value.

One of the things that’s going on in the ‘first wave’ of the socialist calculation debate, then, is a fight between ‘objective’ and ‘subjective’ theories of value in something close to the senses described above.  Now, this remark needs to be immediately qualified, because Neurath – whose writings on the economics of central planning are the immediate occasion for the debate – was not an orthodox Marxist in the sense described above, but was rather an advocate of ‘in-kind’ economic planning.  That is to say, Neurath basically thought that you could do without all of this ‘value’, ‘price’, and ‘money’ nonsense altogether, and simply work directly with quantities of commodities and of factors of production.  This position is the one that Mises is initially criticising, in his essay ‘Calculation in the socialist commonwealth’.  However, Mises understands that Neurath’s position is only one (minority) view within a broader socialist ‘objective value’ tradition, and appropriately devotes significant sections of his text to addressing efforts to articulate pro-planning positions similar to Neurath’s using labour time as the unit of account.

So, the first wave of the socialist calculation debate (basically – Mises vs. Neurath) can plausibly be seen as a fight between ‘objective’ theories of value (classical political economy in its Marxist iteration, plus advocacy of central planning via in-kind accounting) and ‘subjective’ theories of value (in the sense of marginalist economics).  I don’t think that’s all that’s going on in ‘round one’ of the calculation debate (obviously there’s lots in Mises’ essay that can’t be reduced to marginalism, at least not trivially), but it’s a start.

Then in ‘round two’ of the calculation debate (basically – Lange vs. Hayek) the terrain shifts in terms of analytic framework.  Now Oskar Lange comes along and says “Yes, ok, I accept the entire marginalist apparatus – and, more than that, the entire neoclassical general equilibrium apparatus, which, by the way, I had a significant role in formulating – and I completely reject ‘objective’ theories of value, including the labour theory of value. Now, let me nevertheless use that neoclassical marginalist apparatus to formulate an account of a form of socialism that, yes, has some ‘market characteristics’ in relation to consumption goods, but also includes complete government ownership and control of the means of production.”  Obviously there are debates to be had about how successful Lange was in doing this, but that’s the project and the claim.

Then in response to this complete 180 on the socialist side of the calculation debate, Hayek (the representative of Austrian economics in round two) also dramatically shifts the Austrian framework.  Rather than articulating his rebuttal of Lange within the idiom and framework of Menger-style marginalist theory, Hayek argues that the kind of marginalist theory adopted by Lange (and by mainstream neoclasssical economics as a whole) in fact misses what is most important in the subjectivism of the Austrian school.  Hayek argues that a much more fundamental subjectivism – a philosophical subjectivism which cannot be captured by the kind of deterministic modelling associated with neoclassical marginalism – is the core of the Austrian tradition.

I don’t have a view – because I still haven’t read widely enough in the Austrian tradition – about the accuracy of Hayek’s re-presentation of the Austrian position, but regardless, I think this can plausibly categorised as a second, stronger form of ‘subjectivism’, in contrast to the ‘subjectivisim’ of marginalism.  Plausibly, then, we can typologise the debate in terms of three broad analytic frameworks: objective value theory; marginalist value theory; philosophically subjectivist value theory.  In round one of the calculation debate the socialists are advocating an objective value theory, while the Austrians are advocating something closer to a marginalist value theory; in round two of the debate the socialists are advocating a marginalist value theory, while the Austrians are advocating something closer to a philosophically subjectivist value theory.

And now, having schematised the debate in this very crude way, I want to ask: is this even accurate?  What are the ways in which this typology breaks down?  In particular, I’m interested in the ways in which the two forms of ‘subjectivism’ that I’ve crudely sketched above do and don’t overlap.  It seems to me that Hayek is right that at least some of Mises’ points can’t easily be fully captured within the idiom of a marginalist framework.  At the same time, many of Hayek’s criticisms of the Lange position feel to me like criticisms that can be made within the idiom of neoclassical economics broadly understood – criticisms about imperfect information, incentive compatibility, etc. etc. Of course, there has been a lot of work on these kinds of issues within the neoclassical mainstream in the decades since the socialist calculation debate, so I’m not saying Hayek should have had these resources to hand at the time!  But still, it feels like the neoclassical mainstream has a lot of ability to accommodate many of the Austrian critiques of Lange.  And yet, at the same time, it seems clear that the stronger, philosophical scepticism of ‘calculation debate 2.0’ is genuinely incompatible with mathematised mainstream neoclassical economics in a range of ways.  So what are the areas of overlap between these two ‘subjectivisms’ and what are the areas of difference?  These are the kind of things I want to be thinking about as I read more in this literature.

I just read Hayek’s essay ‘Individualism: True and False’ and wanted to put up a few notes while it’s relatively fresh in the mind.  The essay, perhaps oddly, reminds me a bit of Richard Rorty’s essays, in the following way: it clearly has some worthwhile points, but it insists on making those points by drawing an extremely sweeping distinction between two purported styles of thought, and then filing everything it likes on one side of the distinction and everything it dislikes on the other.  If you squint a bit, this can look clarifying, but in my view it is on balance more confusing than not, because the bundles of commitments Hayek organises on each side of his dichotomy in fact don’t have to belong together, and it leaves the reader with the difficult job of disentangling it all.  

Anyway, I don’t want to attempt anything close to a comprehensive disentangling in this post, but I did want to post a few notes to self as usual.

Hayek’s core distinction in this paper is between two traditions of individualism.  One is grounded in predominantly English-language classical liberalism (Locke, the Scottish Enlightenment, Burke, Lord Acton, and also Tocqueville) – this is the good, true individualism.  The other is grounded in continental theorists (Descartes, Rousseau, the Encyclopedists, the physiocrats) – this is the bad, false individualism.  What does this distinction amount to?  Well, as I say, I think there would be room for more clarity on this score, but one of the core elements is the difference between what Hayek takes to be a bad rationalist philosophical approach (on the continent) and a good anti-rationalist philosophical approach (among mostly English-language classical liberals).

That is to say, one of Hayek’s core points is the distinction between rationalist versus anti-rationalist individualism.  This distinction has several dimensions.  First, Hayek opposes the egalitarian dimension of rationalism – he rejects the idea that there is a universal and invariant faculty of reason that is equally shared by all humans, and believes instead in a higher degree of social and psychological pluralism, so to speak.  Second, Hayek rejects what he sees as rationalism’s overweening arrogance about the capacities of human reason.  We should not imagine that we have the capacity to remake our world as we please – our faculties of comprehension and of action are more limited than utopian Enlightenment rationalism believes.  Relatedly, Hayek rejects the rationalist rejection of slowly evolved community norms.  For the rationalist, norms and practices that cannot be justified by the faculty of reason should be abandoned; for Hayek, we should respect the potential wisdom of institutions and practices that have developed and evolved over time, even if we cannot rationally articulate the substance of that wisdom.  Hayek endorses the Burkean rejection of revolutionary rationalist fervour on these grounds.  Moreover, the sheer diversity and complexity of social life should make rationalist planning of a new social order so overwhelmingly difficult as to be impossible without the suppression of that diversity and complexity.

There’s more to Hayek’s argument than this, but this will do for now.  Articulated like this, I always feel like I should be more sympathetic to Hayek’s philosophical position than I am.  Philosophically speaking, I’m a pragmatist.  I’m a big believer in fallibilism in epistemology – I think it’s definitely the case that our cognitive faculties are sharply limited, and that we should be cautious about too-grand claims to be able to comprehend and order the overwhelming complexity of our societies.  Moreover, I agree with Hayek that the ‘individual’ of classical individualism should be seen not as atomistically distinct from society but as embedded in and shaped by its social world.  I agree with Hayek that social pluralism is an active virtue.  Finally, I agree that freedom of the individual should – in part for these reasons – be seen as a central political-economic value and goal.

So why do I always find myself so unsympathetic to Hayek’s arguments?  Part of it is, no doubt, about political tribalism: I am on the left, Hayek is on the right, and we are always more resistant to finding value in the beliefs of our political opponents than in those of our political allies. But I think there’s more to it than this.  The political right/left distinction, crude as it is, also expresses something about social philosophy: it’s a matter, in part, of philosophical – and, of course, normative – substance, not just of tribal allegiance.  I think one of the things that rubs me the wrong way about Hayek is that he often seems to me to be differentially applying elements of his argument in ways that amount to a political thumb on the scales.  In other words, I often feel like Hayek’s ‘downstream’ political preferences are driving his ‘upstream’ political philosophy in ways that feel dubious to say the least.

For example, take Hayek’s Burkean conservative arguments about tradition.  As I discussed (too telegraphically) in my post on Burke, there are several very obvious problems with this philosophical-political approach.  For example, there’s the problem that there is in fact no reason to presuppose that traditions which have evolved over the centuries contain much in the way of wisdom.  Maybe they do – but maybe they are simply oppressive, or bigoted, or shortsighted.  There are bad traditions and good traditions – obviously (to me) we don’t just want to defer to tradition blindly, on the grounds that it’s tradition.  And Hayek agrees – he famously argued that he is not a conservative, in part for this reason!  But at times Hayek will apply Burkean arguments about deference to tradition, even as at times he rejects them, and it’s hard not to feel that this differential application of such arguments is partly driven by motivated reasoning.

I feel similarly about Hayek’s rejection of the idea that we can comprehend the complexity of our social world.  There is a lot to recommend this argument, in my view!  And it certainly should make us think twice about our capacity to engage in large-scale rationalist planning.  And yet at the same time Hayek’s arguments about the virtues of the market themselves involve substantive claims about how our political economy functions – Hayek is making social-scientific claims about which large-scale institutions will produce positive and negative results, and in my view there is only so much scepticism about our ability to understand social complexity that is compatible with such strong substantive claims about what forms of social organisation are to be preferred.  Again, I’m not saying that this circle can’t be squared – clearly Hayek thinks it can be – but I feel like often, in Hayek, there is the differential application of a sceptical argument, where Hayek is all about epistemic modesty when he is evaluating his political opponents’ claims, and yet is simultaneously willing to make quite sweeping judgements about what forms of social organisation are objectively superior when he is advocating for his own views.

So, let me try to write out a preliminary typology of some elements of Hayek’s ‘anti-rationalist’ views here, again with the assumption that I’ll try to nail all this down more carefully later.

  1. Burkean conservatism.  This is, roughly, the idea that our faculty of reason is sharply limited, while our institutions, norms and practices have evolved over the centuries, and thus contain greater wisdom than the individual human mind.  The first problem here is simply: what if the evolved institutions are actually bad?  This is a problem that, frankly, Burkean conservatism often simply waves its hands at without credibly addressing.  The second problem is that society is complex and plural, and there is never just one tradition that we can defer to – we always constitute our traditions by choosing in the present what to recognise as tradition – and such judgements must themselves be based on something.  For this reason, Burkean conservatism is often simply cloaking its own independent preferences in the cloak of tradition.

But of course Hayek is not simply a Burkean conservative!  He also endorses a second idea:

  1. Spontaneous order as a progressive force.  Here the idea is something like: spontaneous order is the mechanism by which individuals with their diverse preferences can work out the ways of life that suit them best, and out of that process a broader social ordering emerges that is more responsive to individual preferences – and therefore preferable – than would be any social order that is the product of large-scale rational planning.  This is, as it were, the future-oriented version of Burkean conservatism: something close to Whig progressivism.  Where Burkean conservatism defers to tradition by default, spontaneous order progressivism assumes that individuals freely pursuing their preferences will transform traditions in a way that is to their, and quite possibly our, benefit. If this position is something like a ‘mirror image’ of Burkean conservatism, then a similar objection applies: why should we assume that ‘spontaneous order’ is in fact good?  Might it not be that spontaneous order is extremely bad?  Shouldn’t this possibility at least be considered?  Again, it’s unclear what the mechanism would be that would reliably cause good order to emerge from this process of social evolution.  More on this in a bit, perhaps.

Then we have a set of commitments about the state:

  1. The state is the major source of unfreedom, and restricting state power is therefore the major goal for freedom-oriented politics.  This is the ‘libertarian’ dimension of Hayek’s ‘classical liberalism’.  As I’ve written about on the blog before, I think this is badly wrong.  Of course, the state is a major oppressive force!  But there are many forms of oppression and of unfreedom in society, and many of Hayek’s arguments in this area operate by simply fiating these forms of unfreedom out of the space of legitimate political consideration.  Contra Hayek, poverty is a form of unfreedom.  ‘Civil society’ oppression is very widespread, and can in fact potentially be mitigated by state action.  Etc.

But, of course, Hayek is not an anarchist because:

  1. Constitutional constraints on state power and limited government committed to the rule of universally applied impartial law is crucial for establishing the conditions under which freedom and spontaneous order can flourish in society.  In particular, a state that is committed to the protection and enforcement of property rights, while constrained from administrative overreach is the optimal form of government.

And related to this:

  1. A set of concrete social-scientific claims about the virtues of market society as catallaxy.

Anyway, I’m rambling a bit here, but I guess the point I want to try to get at in this post – hopefully to be worked through more precisely at some future point! – is that Hayek’s nexus of commitments seems to me to require some quite sharp footwork with respect to epistemological warrant.  On the one hand, Hayek’s philosophical stance puts heavy weight on the limits of human reason, and therefore defers to spontaneous order in preference to rational planning.  At the same time, Hayek is making social-scientific claims about the virtues of specific forms of spontaneous order, and the institutional structures and policy decisions that facilitate them – and these social-scientific claims seem to me to require a much stronger set of claims about our ability to know what policies and institutions will produce desirable social outcomes than Hayek is elsewhere willing to accept.  If we can in fact know these substantive things with the degree of confidence that Hayek often appears to assert, then doesn’t that seem to undermine the broader scepticism about social-scientific rationalism that undergirds Hayek’s project?  And if, by contrast, we accept that we can’t know what’s best for social order in a relatively strong sense, doesn’t that go for Hayek’s policy preferences too?

Again, I’m not saying that this apparent tension can’t or couldn’t be resolved – but it seems to me that Hayek’s work often flits quite happily up and down the ‘epistemic scepticism’ spectrum, making extremely strong and concrete political-economic claims at one moment, and then throwing up its hands at the impossibility of the kind of social-scientific knowledge that could inform rational institution-design at other moments.  I find it hard not to feel that this movement up and down the scepticism spectrum operates in ways that are extremely convenient for a right-wing political project.