This piece had its origins in a pair of posts written by Don Arthur over at Club Troppo, and followed up by Andrew Norton, Andrew Leigh, Will Wilkinson and Backroom Girl. The idea that libertarians and ‘progressives’ could hammer out some of their differences and reach a compromise far more workable than that between conservatives and libertarians is an interesting one. Parts of the debate are, I think, misconceived. Some of Rawls’ ideas about liberty – even the ones I find superficially very attractive – are pretty sketchy. That said, I do think ‘progressive fusionism‘ has potential, and I devoted part of my Oxford jurisprudence assessment to exploring the idea.
At bottom, of course, are the links between liberty and equality, and to what extent (if any) the two conflict. True to my training in analytical philosophy, I’m concerned to pin down concepts like ‘equality’ and ‘liberty’, at least as they’re used by Rawls and Hayek. Like Don, I think that Rawls and Hayek are closer together than they appear. Unlike Don, I don’t think Rawls’ normative framework is as persuasive as it appears, and I don’t think it’s clearer than Hayek’s. I found this when I came to consider one of the ideas at the heart of his ‘principles of justice’ – the idea that liberty is lexically prior to equality.
I think that liberty should have priority over equality, but that lexical priority is too rigorous a requirement. This is an admission I make reluctantly – I am, after all, a libertarian. It is nonetheless a necessary admission.
In this piece, I argue that to give liberty lexical priority over equality – particularly if one accepts some or all of John Rawls’ system as set out in A Theory of Justice – not only excludes equality (however conceptualised). It also excludes other values that are essential to good governance, as well as endorsing a thin conception of liberty. In short, according liberty strict lexical priority over equality is too demanding. That said, I also find Rawls’ suggestion – incipient in Theory, but later made explicit in Political Liberalism – that ‘basic wants’ must be satisfied before liberty can take its place at the top of the ‘values heap’ troubling.
First, I outline what Rawls means by ‘liberty’s lexical priority’. Second, I consider his conceptualisation of liberty, comparing it with that outlined by Isaiah Berlin and F.A. Hayek. Following Hayek, I import free markets into Berlin’s account of ‘negative liberty’ in order to thicken Rawls’ conception of the word. I also discuss Robert Nozick’s insight that restraining people from market participation involves continual interference with their freedom.
I then turn to ‘equality’. I argue for a non-maximal version of equality, incorporating key elements from Hayek’s account of formal equality and Amartya Sen’s notion of ‘capabilities’. Due to the latter’s concern with agency and well-being, I consider his ideas in light of Berlin’s ‘positive liberty’, and ask whether they share the potentially destructive characteristics of which Berlin warns. I also consider whether positive liberty as Sen describes it is coterminous with Rawls’ basic wants.
While I argue for the importance of this equality, I also maintain that according basic wants lexical priority over liberty may be harmful. This is not because I think that people – particularly in the developing world – should continue to live in dire poverty, but because trying to satisfy peoples’ basic needs before setting them free may have deleterious effects. To that end, I consider the large body of empirical research indicating a strong correlative link between elements of negative liberty and satisfaction of Sen’s Human Development Index.
Rawls and liberty’s lexical priority
The lexical priority of liberty is an integral part of Rawls’ schema: liberty can only be restricted for the sake of liberty. This argument emerges after Rawls develops his principles of justice. His first principle requires equal liberty for all, while his second holds that inequality – of liberty and other values – is justifiable (a) if it benefits people overall and (b) is to the ‘greatest benefit’ of the disadvantaged. Any restriction on liberty must either (a) contribute to the liberties shared by all or (b) be acceptable to the parties who end up with less liberty. This places liberty above not only equality, but also justice, welfare and efficiency.
By any criterion, this is a demanding requirement. That is not to say one cannot imagine practical applications. Consider, for example, the presence of a minority who during time of war constitute a fifth column. One can conceive of a public debate that – leaving security to one side – raises the idea that any restrictions on liberty (identity cards, employment controls, profiling) should either enhance the liberty of society as a whole or be acceptable to the group in question; perhaps both.
The demanding nature of lexical priority strictu sensu is most notable when one considers its impact on the implementation of Rawls’ second principle of justice. If liberty has lexical priority, then one cannot consider any question of equality, efficiency, justice, welfare, security or anything else unless liberty’s claims are fully satisfied. Rawls justifies restraints on liberty only as a means to protect like liberty for others. This exclusion even seems to take in the harm principle, except in forms that prevent harm by preferring one form of liberty to another. He gives the example of debates, where rules must limit participants’ liberty to speak whenever they wish in order to facilitate overall freedom of speech. Effectively, all restraints in a situation of maximum equal liberty are necessary for the protection of that maximum. This strikes me as a recipe for institutional paralysis.
Later in Theory, Rawls retreats from this maximal position. He comments that ‘until the basic wants of individuals are fulfilled, the relative urgency of their interest in liberty cannot be firmly decided in advance’. The argument seems to be that a given state needs to satisfy ‘basic wants’ up to a certain point. Thereafter, liberty is preferred to any further increase in other goods. This argument links to Rawls’ distinction between ‘general’ and ‘specific’ conceptions of justice. Although not always clear, my understanding is that the special conception – where liberty is lexically prior – governs societies that have developed such that they can satisfy citizens’ basic wants. Under the general conception, by contrast – where these favourable conditions do not obtain – the state may deny equal liberty.
In Political Liberalism he goes a step further, arguing that ‘the equal basic rights and liberties may easily be preceded by a lexically prior principle requiring that citizens’ basic needs be met’. These basic needs, Rawls argues, facilitate citizens’ comprehension and exercise of their rights and liberties.
Interestingly, Isaiah Berlin makes a similar point, fleshing out ‘basic needs’:
[T]o offer political rights, or safeguards against intervention by the state, to men who are half-naked, illiterate, underfed and diseased is to mock their condition; they need medical help or education before they can understand, or make use of, an increase in their freedom.
Slippage between the lexical priority of liberty strictu sensu and lexical priority + ‘basic needs’ raises two questions. First, how does Rawls conceive of liberty? And second (because he does not give them content), how are we to conceive of basic needs?
Rawls, negative liberty and markets
Rawls seeks to navigate through the shoals surrounding liberty’s definition by arguing that much of the dispute is not definitional at all, but about weighing the value of different liberties when they conflict. Despite collapsing positive and negative liberty into each other in an ingenious conception that refers to ‘agents who are free, the restrictions or limitations from which they are free, and what it is that they are free to do or not to do’, he argues for ‘freedom of thought and liberty of conscience, freedom of the person and civil liberties’. He thinks that these ‘ought not to be sacrificed to political liberty’ in the sense of participation in political affairs. This is close to Berlin’s negative liberty, which is conceptualised in terms of freedom from deliberate interference by other human beings and distinguished from participatory democracy.
There is, however, a crucial difference between Rawls’ liberty and that present in Berlin and – more notably – Hayek. H.L.A. Hart draws attention to the distinction between the very broad scope of Rawls’ equal liberty principle and his later restriction of liberty’s content to that of ‘basic liberties’, noting in the process a major elision when it comes to economic liberty. Hart observes Rawls’ ‘careful and repeated explanation that, though the right to hold property is for him a “liberty”, the choice between private capitalism and state ownership of the means of production is left quite open by the principles of justice.’ Limiting private ownership to personalty would clearly result in less liberty than would be the case if a given society permitted private ownership more widely. It’s also unclear whether Rawls includes freedom of contract among his basic liberties. Rawls’ liberty is thus substantively narrow but procedurally broad.
Nozick – in his famous description of liberty’s disruption of patterns – makes a powerful case for the proposition that excluding economic liberty from any account of negative liberty impoverishes the latter’s content. To be fair, Nozick wants to defeat arguments in favour of wealth redistribution. In the process, however, he outlines how any regime preventing basketballer Wilt Chamberlain from making a contractual agreement for extra remuneration interferes in Chamberlain’s life to such an extent that it ‘would have to forbid capitalist acts between consenting adults’. This assumes, of course, that Chamberlain’s basketball team is in private hands. What if – recall Hart’s account of Rawls’ schema – the basketball team were state owned?
Hayek’s account of economic liberty focuses on ‘spontaneous order’. That is, markets – like languages – emerge thanks to the interaction of free individuals acting in a non-patterned, un-coerced manner. A free market merely co-ordinates the aims and purposes of countless actors; these actors cannot know the aims and purposes of more than a handful of their fellow-citizens. It does this through the mechanism of prices. Changes in the price of a commodity are simply signals, feeding information back into the system, enabling participants to produce spontaneous co-ordination ‘automatically’. This can appear to be the product of an omniscient mind, but is actually its antithesis.
Markets, then, are a signature example of negative liberty. Hayek makes the point explicit:
Much of the opposition to a system of freedom under general laws arises from the inability to conceive of an effective co-ordination of human activities without deliberate organization by a commanding intelligence. One of the achievements of economic theory has been to explain how such a mutual adjustment of the spontaneous activities of individuals is brought about by the market, provided that there is a known delimitation of the sphere of control of each individual.
Although I find Rawls’ argument for liberty’s lexical priority unpersuasive, the thinness of his ‘liberty’ gives one pause even when arguing for a less demanding priority. I propose that negative liberty is only plausibly prior to equality – or other values – if it takes in freedom from economic coercion: the liberty to form contracts, the liberty to engage in the human ‘propensity to truck, barter and exchange one thing for another’.
Equality and capabilities – or, what are ‘basic wants’?
If ‘liberty’ is a large concept, then ‘equality’ is the omnibus term par excellence. While what I propose here is close to libertarian formal equality, I explicitly invoke Hayek’s more nuanced account of formal equality, not Nozick’s. Hayek describes a minimum of equality in terms that amount to a mild version of Berlin’s positive liberty, although compassion moves him, not a desire for equality.
While both Berlin and Hayek link positive liberty with self-mastery, they counsel caution with respect to it for different reasons. Berlin is chiefly concerned with the political system-builder who thinks he has lit upon a method for ensuring self-mastery and self-realisation, only to find that he (or his followers) is coercing others in the name of achieving self-mastery. Berlin cites as exemplum Nikolai Bukharin:
Proletarian coercion, in all its forms, from executions to forced labour, is, paradoxical as it may sound, the method of moulding communist humanity out of the human material of the capitalist period.
Hayek, by contrast, argues that confusing liberty with power leads inevitably to the ‘identification of liberty with wealth’. He points out – somewhat dryly – that while freedom and wealth are both good things, they remain different. While Hayek rejects redistribution aimed at correcting inequality, he explicitly countenances the ‘assurance of a given minimum of sustenance for all’, to be funded by taxation. With this in mind, I argue that people should have enough for ‘capabilities to achieve functionings’ (Sen).
This is in addition to the traditional basket of formal ‘negative’ rights. These rights are not only valuable because they acknowledge the impartial and universal nature of moral judgments. They are also valuable because unless citizens have equality of status along at least one key axis, economic and social inequalities may re-emerge as primary determinants of moral worth.
This explicit selection of a non-maximal account of equality strikes me as sensible for two reasons. First, few people argue for ‘simple equality’ or true economic egalitarianism. Harry Frankfurt elegantly defines the latter as ‘the doctrine that it is desirable for everyone to have the same amounts of income and of wealth (for short, “money”)’, noting that it is easy to defeat argumentatively. However, Frankfurt’s contention that what is morally important is not that ‘everyone should have the same, but that each should have enough’ is extremely persuasive. He calls this a ‘doctrine of sufficiency’.
Most disputes about inequality are misconceived: they are not really about inequality per se. No one cares much about the difference in wealth between lawyers and bond traders, say, even though – while both have expensive educations and considerable technical skill – bond traders often earn incomes orders of magnitude larger than lawyers. What they care about is that some people live in awful conditions due to poverty, or that their poverty prevents them from functioning. Few scholars write more movingly of the plight of the poor than G. A. Cohen. However, even he – in his story of the woman who cannot afford to visit her sister in Glasgow – is not moved by her ‘inequality’, but by her lack of capacity.
Second, it is possible to give Frankfurt’s ‘doctrine of sufficiency’, Hayek’s ‘minimum of sustenance’ and Rawls’ ‘basic needs’ practical substance through Amartya Sen’s account of capabilities. That is, combining Frankfurt’s moral argument with Sen’s empirical observations make Rawls’ basic wants intelligible. I do not pretend this is simple. Frankfurt observes that ‘calculating the size of an equal share is plainly much easier than determining how much a person needs in order to have enough’. That conceded, economists could do much of the empirical spadework. While Sen’s ‘capabilities’ account is concerned with effective power to act or pursue one’s ends, and argues that poverty leads to a lack of effective power to act, it is much more modest in its claims than most egalitarian philosophies.
Of course, to the extent that Sen’s capabilities account contains within it the destructive elements characteristic of positive liberty noted by both Berlin and Hayek, I argue that liberty retains priority. Similarly, I reject Rawls’ argument for the lexical priority of ‘basic wants’ for the same reason as I reject arguments for liberty’s lexical priority: it is too demanding. It requires fully satisfying ‘basic wants’ before considering any other value. It is worth remembering that lexical priority accords the value ‘on top’ infinitely great relative value. Other values can never catch it.
Some empirical points and concluding comments
As attractive as combining Sen’s capabilities with Hayek’s nuanced description of formal equality may be, I still argue that liberty remains prior. This is, however, a more generous priority regime, one without the strictures of Rawls’ lexical priority.
My reasoning is partly empirical, and is borne out by the great body of data collated over the last 18 years by economists compiling the Economic Freedom of the World Index (since 1994) and the UN Human Development Index (since 1990). Amartya Sen developed the latter index, while Milton Friedman and Gary Becker developed the former. The available data focuses on economic liberty, rather than other (negative) liberties more broadly. Even so, it is very telling. In recent years, the EFW Index has published data correlating its indices with UNHD indices. The 2004 graph is below.
Economic freedom correlates strongly with higher average income per person, higher incomes for the poorest 10%, longer life expectancy, higher rates of literacy, lower infant mortality, greater access to clean water and less corruption. There is debate as to whether lower levels of official and private corruption is an effect of liberty or a cause. There is great (and sometimes acrimonious) debate over the link between economic liberty and political liberty. It is also now generally recognised that a secure property regime is necessary before a rise in average incomes takes place. Nonetheless, the evidence that poor countries require economic liberty is solid.
Significantly, there is no relationship between the income share of the poorest 10% and the degree of economic freedom. Inequality – as egalitarians have long argued – remains a pervasive feature of all nations, both rich and poor. However, the amount of income the poorest 10% earn rises with economic freedom. And, as other indices on the UNHDI show, the poorest in free countries often have – to use Harry Frankfurt’s word – ‘enough’. An HDI breakdown across key indices is below.
In sum, while there is a growing body of empirical evidence suggesting that people who enjoy both free minds and free markets have their ‘basic wants’ satisfied more rapidly than people labouring under other regimes, there are other necessary elements of governance that have little to do with liberty per se but undoubtedly enhance its effect. They may also help to secure basic wants. The best arguments for liberty’s priority over equality are empirical, but like all empirical accounts, are rebuttable by facts on the ground.