Wednesday, 17 February 2016

Do People Choose What is Good for Them?

From Blick
Jerry Coyne has a post up titled How Iranian women would dress if the theocracy disappeared. It shows a number of (western) fashionably clad Iranians from the 1970s. It seems obvious that (some) women would dress differently if there were no theocracy in Iran, although there is perhaps an implicit assumption behind the title that the women depicted before the theocracy are free from societal coercion, which seems unlikely; we are all subject to conformation pressures, after all. An apologist for Islam might point out that the extra flesh displayed by the 70s women indicates sexism in the 'western' East just as the covering of flesh indicates sexism in the modern-day Islamic East. As one of the commenters, Heather Hastie, says, "Things are better in the West, but there are still big problems.".

I think a *diversity* of dress is indicative of a free society, but we observe that humans are clubbable animals so they are not making these choices in a vacuum (if they are making choices at all, pace free will debates). Some people might choose to show more flesh, some might choose to show less flesh; the reasons for their choices can range from fashionable to political.

Serious scholars have observed that some women *chose* to don the veil in the 1970s ("In the 1970s, often to the consternation of parents and siblings, certain progressive young Arab women voluntarily donned the veil." - Should we defer to the autonomy of these women in the choices they make? Or should we complain that their choices are somehow malformed? What is to stop the Muslim apologist suggesting that it is women in the West who are slaves to culture?

These questions give me a chance to discuss some philosophy of well-being. Some think that final values (those things we *should* pursue) reside in objective things, like life, knowledge, community, health, wealth and so on; Martha Nussbaum argues for a flavour of this.  Others argue that we should simply satisfy our preferences; Harriet Baber, for example. Simplistically, the problem with the first idea is determining what that list is, if it is objective. One problem with the second idea is: what to do if people 'prefer' to embrace normally 'illiberal' things, like the veil - we should not object because we should let people follow their preferences? An objection to this sort of preferentism is that preferences can be malformed in some way; these so-called 'adaptive' preferences might, therefore, pose a problem for preferentists. They may, for example, cause difficulties if we want to improve the conditions of people in apparently straitened circumstances. I shall examine those difficulties and illuminate them through Nussbaum’s arguments and Baber’s replies.


Preferentism, or ‘preference-satisfaction welfarism’ (Barber, 2014, p.47), is the theory that final value resides in a person’s well-being, and well-being is delivered by satisfying a person’s preferences. By final value I mean what people should rationally direct their efforts to attaining; as Baber says, ‘what’s basic or fundamental, good in and of itself’ (The Open University, 2014a, 0:30). It contrasts with hedonism, which also proposes well-being as the final value, but delivers it with happiness; and with objectivism, which instead suggests there is an objective list of things in which final value resides, such as life, knowledge, community, health and so on. Preferentism benefits over objectivism in avoiding the dangers of an authoritarian imposition of values that are not universally agreed upon, but opens itself up to charges of relativism – perhaps anything could have final value under this theory.

The meaning of adaptive preferences is not altogether clear. Alex Barber says a ‘person’s set of preferences is adaptive if he or she has taken it on in response to limited options’ (Barber, 2014, p.63). But some of these preferences appear unproblematic, such as someone wanting to be a football referee rather than playing, say, because they lack footballing ability. Nussbaum’s conception of the term reflects her theory: she suggests an objective list of capabilities that all humans should have and want. These ‘Central Human Capabilities’ (Nussbaum, 2001, p.154) include things like being able to have good health, being able to laugh and play, and being able to enjoy a life of normal length.

Nussbaum writes:
People’s liberty can indeed be measured, not by the sheer number of unrealizable wants they have, but by the extent to which they want what human beings have a right to have. (p.160)
Presumably the perfectly free, then, will want Nussbaum’s capabilities. So if someone does not want the capabilities that, she suggests, constitute basic human rights that would reflect their lack of liberty. For example, any preference that threatens the ability to have good health reflects a restriction on that person’s freedom to choose, so would be adaptive. And while considering Amartya Sen’s work, Nussbaum mentions ‘life-long habituation’ and says ‘most of the interesting cases do involve life-long socialization and absence of information’ (p.161).

The Problem

Nussbaum thinks that adaptive preferences pose a problem for preferentism because they prevent us from challenging institutions that threaten the capabilities she considers central, because ‘some existing preferences are actually bad bases for social policy’ (p.154). In particular, she is addressing the limited options women have in many societies. Her approach recognises the deeply-ingrained nature of certain power structures so that even the victims of those power structures consider them desirable.

This hints at a wider concern about preferentism: that without some objective standard it’s difficult, if not impossible, to push for change where people express satisfaction with the status quo.


Harriet Baber, following J.C Harsanyi, draws a distinction between a person’s manifest preferences (those they express and reveal by their behaviour) and their true preferences. For a person’s preferences to be true, she says:

1) they must be fully informed, so that those that result from misinformation are deformed;
2) they ‘must be free in the broadest sense’ (Baber, 2007, p.107), so that those that manifest as a result of high passion, for example, are deformed;
3) they cannot derive from moral obligations.

Baber thinks Nussbaum ignores the distinction between manifest and true preference above and also the ‘dispositional nature of preference’ (p.108); that our behaviour sometimes does not reveal our true preferences - for example, many will shop at Claire’s Accessories when they might rather shop at Tiffany’s. Baber says that:
"Adaptation" is irrelevant: if I want something, getting it is good for me regardless of how I came by that desire; if getting what I choose does not benefit me, it is because what I chose is not something that I want. (p.110)
Baber denies that the genesis of a preference matters, whilst also allowing (in the 3 points above) that a preference’s genesis can be flawed – an apparent contradiction.

The quote does reflect preferentism’s ‘apparent ass-backwardness’ (The Open University, 2014a, 10:20), which suggests that what is good is what we prefer, not that we prefer what is good. What follows is that an individual’s true preference does not represent an unjust state of affairs; which further suggests that people’s true preferences define the justice or injustice of institutions.

Adaptive True Preferences?

Baber argues plausibly that we all have a preference ranking, and that to choose, and to express, a sub-optimal preference does not show that the preference is deformed or adaptive; someone may just consider it the choice that achieves the best outcome in the circumstances. People ‘have a certain fundamental character represented by their preference rankings’ (9:55), revealed if someone jumps at an opportunity if given it. She cites evidence from Nussbaum’s work to show that Nussbaum’s subjects (Nussbaum makes a study of women in non-western cultures) do betray a series of preference rankings, since they jump at the chance to exercise political power when given the chance. The extent to which people would jump at such opportunities, Baber says, exposes how unjust a state of affairs is.

So here she presumably agrees with Nussbaum that some preferences are bad bases for social policy. Ultimately, however, Baber is still committed to a final set of preferences that is fundamental, and Nussbaum can target those preferences as adaptive whilst granting that expressed preferences can change according to circumstance. Consider the following diagram:

Pa to Pz is a person’s ordered set of preferences, Pz being the final preference;
Pm to Pz measures the injustice a person suffers, per Baber;
Pz to P? is the putative adaptive preference suggested by Nussbaum, suggesting an additional injustice.

Preference Inception

Baber speculates:
...if [Jayamma] were offered a promotion or a raise she would jump at it, since there is no reason to think that she is any different from most people who prefer more money to less money and would rather not spend their days hauling bricks if other options were available. (Baber, 2007, p.111)
Well, perhaps, but the notion that some people are not driven by money and prefer simpler, more basic, work is not so outlandish that Baber can assume this is not the case for Jayamma. Baber thinks it’s a fundamental preference of her own that she would never like to go shopping for clothes (The Open University, 2014a, 9:40). Presumably she goes shopping for clothes occasionally, but that is just a manifest preference; her true preference is never to go shopping for clothes. It seems likely that the subjects of Nussbaum’s research, like Jayamma, also only occasionally go shopping for clothes. Jayamma may have a manifest preference to occasionally go shopping, because of her limited options; but her true preference may be to go shopping or to not go shopping (like Baber); we don’t know.

But should we believe Baber when she says she doesn’t want to go shopping, and not Jayamma, if that’s what she says? Nussbaum has an account that can answer this, whilst Baber’s seems inadequate.

Either way, manifest or true, Nussbaum can suggest the preference has been habituated, and there is another preference (P?) which the subject is not free to prefer. And she could suggest that Baber’s true preference is habituated too. Maybe her upbringing has prevented her from appreciating the joys of shopping; perhaps she has had an ascetic, academic upbringing that has prevented her from appreciating some of the finer things of life, like beautiful clothes?

Baber says that Nussbaum ‘doesn’t seem to realise how little room to manoeuvre most of us have’ (The Open University, 2014b, 3:30). According to Baber we all have revealed preferences that change according to the room we have to manoeuvre, but we also have true preferences that are, if not cemented in, still our final preferences.

But those final preferences are dispositional and ultimately down to the individual, as Baber concedes when she recounts the story of the Harvard academic who has chosen to spend her time counting blades of grass rather than something more apparently worthwhile, like teaching her students or writing papers. In the end, being a preferentist, she says ‘...De gustibus. Keep counting.’ (The Open University, 2014a, 12:00).

Unjust Institutions

But some true preferences seem to result from institutions that are obviously unjust. Baber struggles with the example of human trafficking and the story of Srey Mom, for instance, ‘rescued’ by Nicholas Christof from a brothel, but who then returned to it. ‘It is not so clear’, she writes, ‘that it would have been better for Srey Mom to go back to her village or get an honest job sewing sneakers’ (Baber, 2007, p.120). This is the same problem I suggested might arise with Jayamma. We just don’t know what someone’s true preference is, and it’s possible Srey Mom’s state of affairs is her truly preferred state of affairs or just a manifest preference.

This presents two problems for the preferentist:

1. While Baber maintains that the difference between manifest preference and true preference is the measure of how much a person would change their circumstances, given the option, (Pm-Pz), we don’t know what that is ahead of time, so it’s difficult to know what actions to take to relieve individual situations.
2. Even if we grant the preferentist account, it is resistant to any objective measurements, such as health indicators and life expectancy. While preferentists must commit to satisfied preferences being the good, there will be no impetus to improve those measures (except insofar as people prefer them).

As a straight matter of fact, then, if one’s goal is to improve those indicators, preferentism has a problem. Baber would reply that our goal should not be to improve those indicators, per se. Even if the preferentist grants some adjustment to people’s revealed preferences for problems in their provenance, ultimately the preferentist is committed to prioritising autonomy, from wherever it springs, over any objective measures of well-being. In the end, a preferentist like Baber must bite the bullet and accept that a human-trafficked prostitute’s situation can be just.


Preferentism would deliver a world where people’s autonomy is observed and a wide range of diverse lifestyles and cultures would be accommodated and respected. However, it presents an epistemic problem when we are confronted with an apparent case of injustice – is it really unjust? And, further, it is plausible that individual autonomy is good, but the notion that what we choose just is good relies on an idealised self rather than the messier self of real life; we are none of us causa sui, and we have all been habituated to a degree.

As a liberal Millian type of character, I want to encourage self-expression, even if it doesn't reflect my values - in fact, *because* it doesn't reflect my values. But simply deferring to agents' autonomous wills does not seem workable without some anchoring in our state of being; some recognition of our human condition must be included in any account of rational action to avoid a rational relativism which can be destructive to human lives. So I'm a little sceptical when folk claim how things would be, given the removal of some obstacle to freedom, even if we agree that the obstacle (like theocracy) should be removed; unadulterated choice is still not available, and something like Nussbaum's 'Central Human Capabilities' is needed if we are to navigate our way toward a healthy society.

Baber, H. E. (2007) ‘Adaptive preference’, Social Theory and Practice, vol. 33, no. 1, pp. 105–26.
Barber, A. (2014) Reason in Action (A333 Book 2), Milton Keynes, The Open University.
Nussbaum, M.C.. (2001) ‘Nussbaum on adaptive preferences and women’s options’ in Barber, A. (ed) (2014) Reason in Action (A333 Book 2), Milton Keynes, The Open University.
The Open University (2014a) ‘Baber on welfarism and adaptive preferences (Part 1)’ [Audio clip], A333: Key questions in philosophy. Available at (Accessed 11 Jan 2015).
The Open University (2014b) ‘Baber on welfarism and adaptive preferences (Part 2)’ [Audio clip], A333: Key questions in philosophy. Available at (Accessed 11 Jan 2015).

Read more »