Even without knowing him personally, it should be obvious from the title of his book on school choice, ‘How Not To Be A Hypocrite’, that Adam Swift is an interesting guy. This is the sort of moral philosophy that tries to reach out from academia and give us a practical guide for the difficult choices we face in practice, including the question: should we ban private schools?
Banning private schools is an infringement of freedom, but allowing their existence is an offence to meritocracy. While Swift clarifies five types of objection to private schools, what really interests me in this post is the response of another excellent philosopher, Elizabeth Anderson. Anderson argues in response to Swift that fair chances are not the same as equal chances – an argument that seems to have a striking similarity to some of the public attitudes I referred to in a previous post.
(Quick note: I here use ‘private’ and ‘private’ in the US not UK sense. The UK meaning of ‘public schools’ schools seems designed to confuse as many people as possible…).
What’s wrong with private schools?
Swift’s problems with private schools are fivefold (according to Anderson):
- Efficiency – private schools allow less able students to rise to senior positions in society, despite their lack of (relative) ability.
- Queue-jumping – private schools students unfairly jump public-school competitors in the queue for valuable universities and jobs.
- Peer group effects – more motivated students are split from less motivated ones, and the less-motivated students suffer from being surrounded by other less-motivated students.
- Parental voice – many parents who care most about education are split from the public school system, who therefore have fewer pressures to improve.
- Solidarity – students are segregated along the lines of wealth, thereby undermining social solidarity.
If you’re like me, then this initially seems persuasive. However, Anderson argues that only the last argument here is a valid objection to private schools. For the ‘peer group’ and ‘parental voice’ objections, her main counter is that these are not likely to be that large (p108) – and are therefore too weak to restrict the liberty of parents. (Readers may find this as unconvincing as I do…). But it is her two other arguments I found provoking and thought-provoking.
The construction of ‘merit’
As gets repeated endlessly, the term ‘meritocracy’ actually comes from a satire of the idea by Michael Young. Young defined merit as ‘IQ plus effort’, which I take as meaning ‘innate’ and ‘constructed’ ability respectively; the complexities in what ‘innate ability’ means I leave for another occasion.
Often when talking about meritocracy, we refer to only to natural ability (as in Leon Feinstein’s ‘killer chart’). Yet when we consider constructed ability, the problem of ‘efficiency’ isn’t actually an objection to private schools at all. In fact, according to the ideal of the meritocracy, it is desirable for private school students to jump over public school students with the same innate ability – as long as their effort and education lead them to have greater levels of achieved merit (the ‘efficiency’ argument above). As Anderson puts it (p102):
“Meritocracy does not care whether people are meritorious because they are ‘natural born’ talents or were born with a given temperament, or because their parents invested huge effort in developing their talents, and relentlessly pushed them until they internalized their parents’ ambitions. It just wants the most productive workers. How they came to be that way is no concern of the meritocrat.”
What I find particularly powerful here is Anderson’s scepticism that ‘innate ability’ has any greater worth than achieved ability. The accidents of genetics have no deeper moral value – or in Anderson’s nice phrase (p103), “we have no reason to defer to any aristocracy, whether ‘natural’ or artificial.”
The meaning of ‘fair chances’
What really struck me, though, was Anderson’s rejection of the idea that ‘equal chances’ were a central aim for meritocrats (the ‘queue-jumping’ argument above). She argues from her wider theoretical perspective that what really matters is chances that are fair enough to ensure democratic equality (p106):
“Democratic equality is egalitarian in its conception of just relationships among citizens, but sufficientarian in its conception of justice in the distribution of resources and opportunities. What is important is not that everyone has equal opportunities to acquire resources and fulfilling jobs, but that everyone has ‘enough’. The ideal of democratic equality specifies how much this is: enough to secure the conditions of citizens’ freedom and civic status as equal to other citizens.”
Anderson clarifies that everyone must be entitled to a ‘fair chance’ to get a fulfilling, well-paying job, but that this does not mean an ‘equal chance’ – partly for the reason above (that it contradicts the efficiency basis of meritocracy) and partly because it denies more-motivated parents the opportunity to help develop their children’s talents. She therefore denies that the queue-jumping argument has any force, instead saying that from the perspective of democratic equality, the main problem is one of solidarity. (Which she argues is insufficiently strong to lead them to be banned, problematic though it is).
The point here is not the philosophically validity of this claim (with which I disagree), but rather how far this seems to fit the public attitudes in the JRF report I described a couple of weeks ago. People know that there are barriers that make it easier for some people to succeed – but they nevertheless feel that everyone has a chance to succeed that is fair enough. Arguably this ties in with the extensive sociological evidence that education is both the main means of social mobility and simultaneously the main means of the reproduction of inequality.
If people are to be convinced that there are unfairness that demand action, then we not only need to show inequities in life chances, but that these violate the threshold of what is acceptable. And it is this that most conventionally-presented statistical coefficients fail to do.