What Money Can't Buy is that rare thing: a work of philosophy addressed to non-philosophers that is neither superficial nor condescending. Its prose is clear and elegant, free of the neologisms that clog up most modern works of political theory. Its message is simple and direct. Yet the questions it raises are deep ones. ‘Popular philosophy’ doesn't do justice to this high-minded work. What Money Can't Buy is a call to argument, not a sop to middle-brow cultural consumerism.

What Money Can't Buy has two main aims, one descriptive, the other critical. On the descriptive side, it traces the extension over the past three decades of market mechanisms into realms of life previously regarded as lying outside their orbit. We have drifted, as Sandel puts it, from having a market economy into being a market society. The bulk of What Money Can't Buy is devoted to illustrating this claim. Sandel clearly scours the press – or has researchers scour it for him – and has put together a veritable horror show of misjudged marketization. He starts with the erosion of that fine old institution, the queue, as a result of ticket touting and ‘fast-track’ options for wealthy customers. He then moves on to weightier matters: the use of economic incentives in education and social policy; the commercialisation of sport; and the invasion of advertising into schools and hospitals. (A particularly fascinating chapter explores the booming market in ‘death bonds’ – life-insurance policies sold on to speculators, who thereby acquire a rooting interest in the death of the insured.) Philosophers among Sandel's readers will no doubt be perplexed by this profusion of detail: why ten examples where one will do? Non-philosophers will enjoy a good story, well told. What Money Can't Buy is, among other things, a narrative of changing social mores in the style of Montesquieu or Tocqueville.

But of course, Sandel's purpose is not purely sociological. He wants not just to describe but to judge, sometimes to condemn. Sandel is no absolutist; he admits that in some cases the advantages of a market or incentive scheme outweigh its costs. His point is that we cannot simply assume, just because a policy is efficient in the economic sense, that it is best all things considered. Economic considerations are a subset of practical considerations, with no special precedence. Take, for instance, the way in which ‘Shakespeare in the Park’, a free theatrical project in New York Central Park, is increasingly monopolised by the rich, who secure tickets by paying others to stand in line for them. From the economic point of view, it's not clear what is wrong with this. If a rich man pays a poor man to take his place in a queue, both presumably benefit, and no one else suffers. (In the jargon, the transaction is Pareto efficient.) Yet we feel, instinctively, that that something is amiss. Sandel's aim in What Money Can't Buy is to put his finger on this ‘something’.

Extensions of the market confront two objections, says Sandel: they are a) unfair and b) corrupting. Of these, the unfairness objection is the less well developed. At least two versions of it jostle for attention in What Money Can't Buy, neither of them entirely persuasive.

The first version of the unfairness objection focusses on the fact that, as the number of things for sale in a society grows, so do the advantages of wealth and the disadvantages of poverty. Marketization exacerbates inequality. But regrettable though this tendency is, it cannot be the ultimate basis of our objection to marketization, for it might always be offset by a more steeply progressive tax regime. That it hasn't been so offset is an indictment of contemporary politics, not of marketization.

The second version of the unfairness objection appeals to the fact that not all market transactions are fully voluntary. A man in dire straits who makes a degrading sale to feed his family may properly be said to have been coerced. Sandel gives the example of poor peasants in India and China selling their kidneys to Westerners. If this trade strikes us as distasteful, that is at least partly because it is exploitative – the sellers presumably had little choice.

I agree with Sandel (and against libertarians like Hayek) that sales made to alleviate extreme poverty are not fully free. But that is an argument against extreme poverty, not against such sales. If the market in kidneys were shut down, poor peasants might sell their daughters instead. If all exploitative markets were closed to them, they might simply starve. That does not seem an improvement. The culprit, then, is not a particular class of transactions, but the conditions that make such transactions irresistible.

But these are quibbles, for Sandel's chief concern is not unfairness but corruption. Sandel is unusual among contemporary philosophers in holding that not all social evils can be reduced to inequality or coercion. A trade between two equally rich, equally free individuals may nonetheless be wrong if it corrupts the character of the object traded. What could this mean?

At the root of Sandel's corruption objection is the observation, taken from Fred Hirsch, that markets are not just neutral mechanisms for distributing goods; they change the character of the goods they distribute. ‘Markets’, as Sandel puts it, ‘leave their mark’. A classic example, briefly touched on by Sandel, is prostitution. Paying for sex doesn't just economise on time and effort; it changes one kind of act into another, psychologically cruder kind. Shakespeare in the Park offers another example. Originally intended as a civic celebration, an affirmation of shared citizenship, it has been transformed by ticket touting into an ordinary consumption item. The performances may remain the same, but their social meaning has changed.

Underlying these remarks on corruption is a certain picture of human action, whose source is Aristotle. An act, on this view, takes its nature from its intrinsic end. A friendly act is essentially one that aims to assist, comfort, and so forth. A ‘friendly act’ undertaken for pay is not really a friendly act, though it may outwardly resemble one. The point can be generalised. Supplanting the intrinsic end of any act with an extrinsic end – typically money – transforms it into an act of a different kind and thereby corrupts it. And what is true of individual acts is also true of practices and institutions.

Those who reject the metaphysics of intrinsic ends will naturally be unpersuaded by Sandel's account of corruption. To borrow Glen Newey's example, a doctor accused of prostituting his talents for pay might simply retort that he is in the business not of doctoring but schmoctoring, ‘which is just like doctoring, except that its internal goal is to earn money for the schmoctor’.Footnote ¹ But Sandel is surely right to insist, against the ‘schmoctor’, that we cannot impose any purpose we want upon our practices and institutions. The realm of human action is not made of infinitely malleable stuff. ‘Schmoctoring’ can never be a legitimate alternative to doctoring, only a corruption of it.

But even if we grant the distinction between intrinsic and extrinsic motives, there remains the question of their interaction. Do extrinsic motives reinforce or ‘crowd out’ the intrinsic ones? Sandel sees mostly crowding out. He invokes Richard Titmuss' well-known hypothesis that paying people to give blood, as opposed to soliciting voluntary donations, undermines the charitable impulse and leads to less blood giving rather than more. But there is another, more hopeful possibility, suggested by the recent attempt of some struggling US schools to boost exam results by paying successful candidates. Sandel reports that some of these attempts worked, not for orthodox economic reasons, but because ‘the money had an expressive effect – making academic achievement “cool”.’

This example points to the importance of a third kind of motive, intermediate between the intrinsic and extrinsic – love of honour. Pupils who strive to do well in exams because the cash prize makes it ‘cool’ are clearly not into studying for its own sake. But neither are they just after the cash. What they want is recognition – recognition for having done something well. Wanting recognition for an achievement is internally connected to wanting the achievement itself, for recognition is gratifying only when there is something to be recognised. (A successful exam cheat could not take pride in a monetary reward, though he might appreciate the money.) Insofar as money functions as a mark of status, a signifier of social honour, it can complement, rather than undermine, intrinsic motivation.

Take another example. I recall a professor at my old Oxford college telling me smugly that declining academic salaries had purified his discipline, weeding out those whose hearts weren't in it. I thought, and still think, that this was rubbish. Falling pay has filled our universities not with Wittgensteins but with people of mean tastes and narrow outlook. It has made the profession unattractive to those many talented individuals who care not only about the truth but also about the world. The Wittgensteins will always follow their demon; what we have lost are the Freddie Ayers.

But these are complications. Sandel is right to concentrate on what is clearly the central case in our society – the corruption or crowding out of non-monetary by monetary motivation. What Money Can't Buy should be judged not on the completeness of its analysis but on its service in bringing to light one of the great, unspoken political questions of our time.