Hayek’s ‘The Constitution of Liberty’

I wanted to assemble some pretty brief thoughts on Hayek’s ‘The Constitution of Liberty’, before it all goes down the memory hole.  The book is doing a number of different things, and I wanted to typologise them as a sort of notes to self.  Specifically, I would say that much of the content of the book can be broken down into five categories.  I’m going to discuss those categories in ascending order of how seriously I take them.

One: ad hoc right-wing talking points.  Perhaps I’m being a bit ungenerous here, but I was struck by what a high proportion of the book is making right-wing arguments that aren’t really connected to any of its overarching theses.  For example: a number of arguments are presented for why vast hereditary wealth is actually good (e.g. it’s important to have a ‘cultural vanguard’ that can experiment with cultural novelty, and, unlike those experiencing ‘new wealth’, aristocrats are already used to being wealthy so they’re unlikely to have their character deformed by sudden prodigality). From a left perspective a lot of these ad hoc arguments seem pretty silly.  I’m just going to ignore that stuff.

Two: an argument in the philosophy of law about what ‘rule of law’ means.  I would say that this is the most central thesis of the book, but I also found it a little difficult to pin down – perhaps if I read ‘Law, Legislation and Liberty’ I would be able to do a better job re-presenting Hayek’s position here.  Broadly, I’d say that Hayek wants to divide the exercise of state power into two categories: state power equally applied to all citizens in a manner that follows clear, universal rules (the rule of law), and state power applied only to some subset of citizens, because of the preferences of rulers.  Hayek’s argument is that only the former – universal laws, applied universally, which because of their universality guarantee equality before the law – are a legitimate form of the exercise of coercive state power.

I think there’s clearly a point to be made in this space.  There is a very strong argument to be made that arbitrary dictats mobilising state power against political enemies or disfavoured minority groups are oppressive and illegitimate, and go against the concept of ‘rule of law’.  And presumably one could try to construct a spectrum of state action with “arbitrary ruling class whim” at one end and “universal and universally applied law” at the other end.  But Hayek seems to want this distinction to do some heavy theoretical and political lifting that it’s not clear to me that it can, at least as articulated.  This is so in two distinct respects.

First of all, Hayek seems to think that the clear illegitimacy of arbitrary state power being used to lock up ruling class actors’ political opponents, etc. also applies to many areas of actually-existing state policy: the administrative state as a whole, for Hayek, seemingly fails to exhibit ‘rule of law’, because of the way it empowers state actors to make context-specific policy decisions.  Again, there are arguments to be made in this space (hopefully one day I’ll write some more about the sinister dimensions of the administrative state), but Hayek’s argument seems to me on its face to be way too strong.  It also seems to be inconsistently applied – Hayek is happy for certain administrative state programmes to persist, in a way that does not seem to me to line up with any coherently articulated ‘rule of law’ versus ‘arbitrary state action’ distinction, but rather to line up with Hayek’s own policy preferences.

Secondly, Hayek is weirdly indifferent to the potentially oppressive dimensions of universally applied law.  Chapter 10 contains this extremely strange passage:

It is not to be denied that even general, abstract rules, equally applicable to all, may possibly constitute severe restrictions on liberty.  But when we reflect on it, we see how very unlikely this is.  The chief safeguard is that the rules must apply to those who lay them down and those who apply them – that is, to the government as well as the governed – and that nobody has the power to grant exceptions.  If all that is prohibited and enjoined is prohibited and enjoined for all without exception (unless such exception follows from another general rule) and if even authority has no special powers except that of enforcing the law, little that anybody might reasonably wish to do is likely to be prohibited. (154-155)

This seems to me to be a deeply silly and wrong-headed passage.  In the first place, Hayek does not adequately address the very significant problem of what institutional structures can enable us to institute a society in which those with power do not have the ability to make exceptions in practice for themselves.  This problem must, in my view, be central to any serious classical liberal political-economic program, but ‘The Constitution of Liberty’ largely skirts around it.  In the second place, there is an arguably even deeper problem of principle here: those making the laws may simply have different preferences from many of those to whom the laws are applied.  The idea that we can rely on the universality of law as a check against the repressiveness of law seems totally incompatible with the insight that our societies are diverse, and that different people within them want to live very different kinds of lives.  Since the importance of pluralism of preferences is supposedly one of the motivating features of the kind of classical liberalism that Hayek is here advocating, I find this dimension of Hayek’s argument to be extremely strange.

So, on Hayek’s ‘rule of law’ arguments: clearly there are important points to be made in this space, and I would like to return to these issues again in the future, but for now I’m basically not going to take this element of the work seriously, because Hayek’s arguments seem to me to require pretty radical reconstruction to get something sensible out of them.

Three: concepts of liberty.  Here I’m basically talking about the first chapter of the book, where Hayek very usefully typologises four different senses of ‘freedom’ or ‘liberty’. Hayek’s typology is: 

1) ‘Negative liberty’ in the sense of freedom from another social actor’s coercive power.  This is the sense of ‘liberty’ that Hayek is interested in, and that he believes political-economic policy and institutions should aspire to cultivate.

2) Participation in governance – “what is commonly called ‘political freedom’” (13) or ‘positive liberty’.  Like Isaiah Berlin and others, Hayek is concerned that this sense of ‘freedom’ is in practice compatible with high degrees of oppression, because a community can claim to govern itself – and thereby exercise ‘political freedom’ at a collective level – even though at an individual level such governance is despotic or oppressive.  Hayek is not interested in pursuing this sense of freedom or liberty, and he thinks it is important that it not be confused with ‘negative liberty’.

3) Metaphysical or subjective freedom – freedom of the will.  Obviously one could claim that this is a politically-economically-relevant concept, but Hayek doesn’t think it’s very relevant to the problems posed by political-economic institutions, and I agree.

4) “the power to satisfy our wishes, or the extent of the choice of alternatives open to us”.  This is Hayek’s true target in this section of the book.  He cites the institutional economist John Commons and the philosopher John Dewey as representatives of this position.  In our own time (postdating ‘The Constitution of Liberty’), Amartya Sen is I think the most distinguished advocate of a version of this sense of ‘liberty’, with his ‘capabilities approach’.  Advocates of this sense of ‘liberty’ argue that ‘negative freedom’ is too limited an understanding of freedom – that the lack of ability to act as one chooses associated with economic deprivation is itself a form of unfreedom.  One can have all the ‘negative liberty’ one wishes, but if one is starving, and with no ability to remedy one’s plight, one is experiencing not just a low degree of ‘welfare’, but also an absence of freedom.

Hayek is very critical of this understanding of freedom.  He argues that it is “a verbal trap” and “dangerous nonsense” (16).  Not only is this concept of freedom confused, Hayek argues, but it is also politically malign, because it is the mechanism via which redistributive policy finds ideological warrant, disregarding ‘negative liberty’ in the name of (in Sen’s terms) expanding social actors’ ‘capabilities’. 

Hayek’s position here is the ideological core of much ‘libertarian’ or ‘classical liberal’ political philosophy – but in my view it has at least three things wrong with it.  First, and fairly trivially, I think it is tendentious for Hayek to argue that Dewey, Commons, Sen, etc. are engaged in sophistry, rather than simply articulating a different political-normative framework from Hayek’s own.  Second, and perhaps most substantively, I just disagree with Hayek’s normative framework, and agree with Sen’s.  I see no reason not to understand deprivation as unfreedom, nor do I see any reason not to make the reduction of such unfreedom a major goal of our political-economic institutions.  Third, and a bit more subtly, Hayek is also dismissive of the idea that the forms of deprivation associated with ‘capabilities unfreedom’ make social actors vulnerable to coercive exploitation.  Hayek – like so much of the tradition of which he is a part – simply does not take seriously the idea that the intensely limited bargaining power associated with capability deprivation is any kind of threat to negative freedom.  This is, in my view, a problem with Hayek’s framework above and beyond ‘irreducible’ normative disagreements.

Four: a ‘Burkean conservative’ theory of history.  In the fourth chapter, Hayek distinguishes (rather ‘ideal typically’) between “two different traditions in the theory of liberty” – a French tradition which is a prioristic, rationalistic, and speculative, and an English tradition which is empirical, unsystematic, and historical.  For Hayek, the ‘French’ a prioristic tradition has a tendency towards utopianism – and therefore totalitarianism – because it disregards the real lessons of history in its effort to forge an ideal society based on abstract principles of reason.  By contrast, the ‘English’ tradition is grounded in the organic evolution over time of real institutions of freedom.  Hayek aligns himself with this latter tradition, and worries about the corrosive impact that idealist rationalism may have on actually-existing organic institutions of English freedom, and its successors.

This all deserves a lot more attention than I’m going to give it here, but I think there are two quick remarks worth making about this theory of history.  The first – which is just an observation, really – is that Hayek’s metatheoretical position here marks a significant break with the ‘Austrian’ tradition with which he is typically associated.  The Austrian approach – exemplified by Mises’ ‘praxeology’ – was historically extremely keen on a priori reasoning as against empirics.  Moreover, the Austrian approach was typically both methodologically and normatively individualist, while Hayek’s position in ‘The Constitution of Liberty’ – with its emphasis on the value of organically evolved community norms and structures – is at the very least flirting with a significantly more communitarian framework.  Hayek’s positions here are all perfectly defensible, but I think it’s striking how far he has moved in this work from the paradigmatic Austrian approach.

The second point concerns the selective application of a Burkean admiration for organically evolved institutions.  Again, this issue deserves much longer treatment than I can give it here, but the basic theoretical problem is that no culture, norms, or institutions are homogeneous or unitary, and so there are always choices to be made about exactly what it means to respect them.  The practical problem, as it were, is that Hayek wants to argue that (for example) the U.S. war of independence – which overthrew existing society in a violent revolutionary uprising, and instituted its own alternative institutions based on the founders’ reasoning about an ideal polity – is somehow aligned with the ‘organic’ side of his dichotomy, while much less violently transformative contemporary administrative policies that Hayek dislikes somehow align with ‘a priori utopianism’.  This often seems like a stretch, to me – the Burkeanism of this strand of the text can sound a lot like special pleading, whereby political-economic structures that Hayek likes are ‘organically evolved’, and those he doesn’t like are utopian.  Again, it’s not that there’s nothing here, and I’d like to return to this kind of metatheory, but to my mind Hayek is not very successful in deriving the conclusions he wants from his overarching framework.

Five: distributed knowledge.  Finally, we come to the argument which is, I think, most strongly associated with Hayek’s name: the idea that local knowledge distributed across the economy as a complex system requires certain categories of institution in order to successfully shape large-scale economic action.  This is the argument that Hayek famously made in his 1945 article ‘The Use of Knowledge in Society’.  I was surprised, when reading ‘The Constitution of Liberty’, to discover how little work this category of argument is doing in the book.  To my mind, Hayek’s ‘distributed knowledge’ argument – whatever conclusions one ultimately draws from it – is his most significant contribution to economic theory, and deserves to be taken very seriously.  But ‘The Constitution of Liberty’ is not a work in which Hayek significantly advances this argument.  Obviously there’s not much point complaining that a book doesn’t do what you were expecting it to – but there it is.

So – that’s my critical summary of the book, and what I take to be its core claims.  Enough for now.

Advertisement