Think about Belle from Beauty and the Beast. She is a smart, ambitious, independent young woman who trades her freedom for her father’s and over time comes to love the inconsiderate, dominating Beast who keeps her captive.
On one plausible reading, Belle’s case is a classic case of adaptive preference. By adaptive preference, I mean a preference that a person forms for an option in a limited set, that she would not have formed if other more expansive options had been available. And such preferences tend to raise problems for social and political philosophers and well-being theorists because they pull us simultaneously in two different directions: because they are the person’s own preferences, it seems that they are relevant to – perhaps even decisive in – determining what is good for her or how she should be treated; but because they involve settling for what she can get rather than a desire for what she would want if only it were available, they do not seem to capture what is genuinely good for her.
There are two obvious ways to respond to this dilemma: first, we could give an account of why adaptive preferences do indeed fail to capture a person’s well-being, and therefore hold that we lack important prudential reasons to satisfy them; or second, we could give an account of how they (perhaps surprisingly) succeed in capturing a person’s well-being, and therefore hold that do in fact call for satisfaction. In Must Adaptive Preferences be Prudentially Bad for Us? I argue that it is a mistake to see these options as mutually exclusive. Instead, I argue that both of these responses can be appropriate for the very same adaptive preference if we make our judgements, as we should, from two different temporal perspectives. While I show in the full paper how my argument succeeds across the range of standardly defended theories of well-being, I’ll here offer just the basic framework of the argument.
Here’s my main claim: That for at least some adaptive preferences, it can be the case both that 1) after the adaptive preference has been formed, the agent can have robust prudential reason to satisfy the adaptive preference rather than to act as the non-adaptive alternative preference would have recommended; and 2) that before the same adaptive preference has been formed, the agent would be made prudentially worse off by forming and acting in accordance with the adaptive preference than by forming and acting in accordance with the non-adaptive alternative. And to defend this claim, I’ll appeal to the mechanism of transformative experience.
By a transformative experience, I mean one from which a person emerges with a substantially altered set of preferences and values such that they can be meaningfully said to have a significantly different character or sense of self. While much has recently been said about transformative experiences in other areas of philosophy, the following features are most important for my purposes. First, transformative experiences are a function of the circumstances and events we actually encounter, and it can be difficult to know in advance which experiences will lead to fundamental personal transformations. While we sometimes choose experiences like parenthood that we can predict will change us significantly, we may also be deeply changed by events that we did not choose, anticipate, or even realize were occurring – and we can also find ourselves surprisingly unchanged by an experience that we expected to be transformative. Second, the changes involved can be more or less mutable. That is, transformative experiences will leave people differentially able to reverse the relevant transformation or undergo new transformations in that area. For instance, if someone becomes violently ill the first time they drink and consequently develops a strong distaste for gin, that distaste might decrease over time or might persist until death. Third, the changes involved might be more or less fundamental. That is, some transformations will anchor and organize our fundamental senses of self in ways that others will not. From these three features, it follows that not all transformations will be equally available to all persons. Not all circumstances will be available to all persons, which transformations arise from particular circumstances can be unpredictable, and – depending on the mutability and fundamentality of transformations already under-gone – certain further transformations in particular areas can be difficult or impossible.
In order to defend my main claim, I need to combine these features with the further claim that the development of an adaptive preference can be a transformative experience. And this further claim is highly plausible. Because the circumstances we actually encounter determine the transformations we undergo, and because adaptive preferences are a function of limited option sets, those limited option sets can lead to different personal transformations than their more expansive alternatives would have. So the development of an adaptive preference can be a transformative experience.
One might object that this cannot be the case because transformative experiences involve a change in response to values, while adaptive preferences simply involve a compromise with reality in cases in which value is made inaccessible by circumstance. But this is implausible. In even the most classic cases of adaptive preference, the reasons people have for endorsing their new circumstances often pick out things that we would count as genuine values in other circumstances. Consider the woman who becomes glad to have undergone ritual genital cutting because it gives her access to status, community belonging, or sexual purity. To be sure, it might be unjust or regrettable that this woman was only given easy access to and so transformed by these particular values, but transformative experiences are once again a function of the actual circumstances you encounter. So transformative experiences can yield adaptive preferences. And insofar as transformative experiences can be fundamental and immutable, adaptive preferences can be too.
This, then, explains how the first part of my main claim can be true. When adaptive preferences are the result of immutable or fundamental transformative experiences, satisfying them can bring us more prudential value than acting in alternative ways that do not jive with the character or preference set we have actually developed. While different theories of well-being will offer different explanations for why this will be so, and while I cannot include those explanations here, I show in the full paper that this claim holds true across the range of theories of well-being most commonly defended in the literature.
In order for the second part of my claim to be true, we need to appeal to just two further ideas. The first is that in order to count as a full theory of well-being, a theory must tell us both what has prudential value and how to compare the prudential value of various lives. The second is that both transformative experiences and the development of adaptive preferences are events that occur over time in response to contingent experiences that could have been otherwise. So even if a particular adaptive preference becomes fundamental or immutable as the result of a particular transformative experience, the person could have initially been differently and non-adaptively transformed, so that she developed a character and preference set that allowed her to access a greater amount of prudential value from developing and satisfying non-adaptive preferences. Again, I cannot include here the various explanations that different theories of well-being offer for this fact, but they are included in the longer paper.
Let’s return, then, to Belle. Is it prudentially good for her to love and make a life with her Beast? If I am right, it might not be appropriate for us give a simple yes or no answer. Insofar as the Beast is inconsiderate and controlling, then we should almost certainly think that she would be prudentially better off if she never met him, and instead developed relationships with respectful and egalitarian partners and friends. But if the experience of actually coming to love the Beast is sufficiently immutably or fundamentally transformative, then after the fact continuing to love him might genuinely be the best that she can prudentially do.
_____________________________________________________________________________________________________________
Rosa Terlazzo is an Assistant Professor in Philosophy at Kansas State University. She works in moral, social, and political philosophy, with a particular emphasis on:
- The concept of adaptive preferences
- Well-being and transformative experiences
- Autonomy
- Non-ideal theory
Thank you for your post, Rosa.
I understand why you have qualified your thesis to cover only immutable adaptive preferences: otherwise it might well be in a person’s best interest to simply undergo a new transformative experience and acquire “better” preferences.
I don’t understand, however, why or what work the “fundamental” qualifier is doing. Can you clarify?
Here’s an example to illustrate why I’m skeptical that the “fundamental qualifier” is relevant. Suppose Stan is uncoordinated and weak sometime around 4th grade through 8th grade. As a result, it becomes part of his fundamental self-identity that he avoids exercise whenever he can; he even develops rationales (viz. motivated reasoning) for why exercising is an inferior or un-choiceworthy activity. Consider Stan’s younger sister Stacy: Stacy has the same body type as Stan growing up but develops a non-fundamental preference against exercise. It seems to me that whether exercise is good for Stan or Stacy covaries with whether each of their adaptive preferences is mutable. So far as my intuitions go, it has nothing to do with whether their adaptive preferences are fundamental.
Avi
Thanks for the comment, Avi. I think you’re right that it isn’t fully clear from the blog post what work fundamentality is doing – the dangers of writing in the 1,000-1,500 word range. The broad idea is that fundamental preferences anchor much of who you are and what you care about. Given their connections to so much else in your life, you might have good obviously prudential reasons not to change them, because doing so would require related changes in other preferences and parts of your life that would be prudentially harmful in sum. And that might be true even if that preference were mutable and you could be made prudentially better off if you changed it in isolation. The claim is of course contingent on how fundamental the preference is, what other parts of one’s life it anchors, what changes to other parts of one’s life would be required by changing the fundamental preference and acting on its replacement, etc – but even if it doesn’t apply in all cases I think it plausibly often will.
Hi Rosa! Interesting post.
I am interested in the familiar question of whether cases of adaptive preference give us good reason to reject preference-based account of well-being, and subjectivist accounts more generally. So I wanted to ask you your views on that.
Here is one way to put the three relevant Belle outcomes:
THE LIFE: Neither Belle nor her father ever gets tangled up with Beast. Belle develops relationships with respectful and egalitarian partners and friends.
LEARN TO LOVE: Belle’s father gets tangled up with Beast. To free her father, Belle agrees to a life of captivity with Beast. But over time she adapts to these circumstances and comes to love her life, and to love the inconsiderate, dominating Beast who keeps her captive.
MISERABLE: Belle’s father gets tangled up with Beast. To free her father, Belle agrees to a life of captivity with Beast. But she doesn’t have to like it. And she never does.
You would say, of these three possibilities for Belle, that THE LIFE is the best and that LEARN TO LOVE is better than MISERABLE. That seems like a reasonable view, and indeed an illustration of how it can be both wise to adapt your preferences if you’re stuck in a bad situation but better to be out of the situation entirely.
But do you think that a subjectivist/preferentist about welfare can agree? It is easy to see how a subjectivist can deliver the result that LEARN TO LOVE is better than MISERABLE. But do you think a subjectivist account can plausibly deliver the result that THE LIFE is better than LEARN TO LOVE?
Hi Chris,
Yes, I wonder about this, too. What I end up saying in the paper is that yes, subjectivist accounts *can* tell us that THE LIFE > LEARN TO LOVE, but contingently – it’s going to rely on THE LIFE involving (say) more opportunities to form and satisfy stronger preferences, or on the particular judgements that your idealized desire account lets you make about the particular case in question. But like I said, it’s contingent – while that will often plausibly be the case for worrisome adaptive preferences, I definitely haven’t shown that it will always be. And in the longer version of the paper I make it clearer that I’m showing that there is a *set* of worrisome adaptive preferences that needs to be evaluated in these two ways, not that we’ll get this verdict for all of them.
I know that answer sounds a bit weasely, but I think that here the kind of question I’m interested in is very much influenced by the fact that I ultimately mostly care about the political and social philosophy elements of the problem – and showing that there is this set gives us enough justification for the kind of treatment I think we ultimately owe people who we rightly take to have worrisome adaptive preferences.
But I’d like a more complete and satisfying answer for the pure prudential question, too – let’s talk more about it sometime soon!
Hi Rosa, Luc Bovens is working on some new things on adaptive preferences that might support your view, so you might want to check out what he has to say. In some way, I think the miserable life might be more praiseworthy even if it is not as good (prudentially) as that in which Belle adjusts and is happy (supposing she could go either way). I think she has more integrity in the miserable life. Perhaps this is something idiosyncratic about me, though :). Thanks for the post!
Hi Nicole,
I think you might absolutely be right about that – I think we should allow for the possibility that adaptive preferences can be bad in lots of different ways (from the perspective of integrity, autonomy, well-being, injustice, etc), and that a lot of those different ways can come apart in given cases. I only wanted to make a claim about prudential goodness here, but I do think it’s an interesting further question how integrity fits into prudential goodness (I suspect it will have at least a small role to play on most accounts, but that it won’t be an overriding role on most if any?).
Thanks for the tip of Luc Boven’s stuff – I’ll shoot him an email. And thanks for putting together such a great blog series here!
Hi Rosa. Thanks. Yes, let’s do talk about it more sometime soon. In the meantime, I agree that the subjectivist can say that THE LIFE > LEARN TO LOVE only contingently, and so only in some versions of these cases. I’m inclined to think that this is good enough, and that we should therefore not reject subjectivism on these grounds.
As a follow-up to your response to Heathwood, is there a social initiative or government agency that you think is mistaken in their actions due to implicitly not accepting one of the two parts of your thesis? Or is it more a mistake that’s confined to the academic literature?
Chris, I think you’re right that this is good enough – especially because (as I said above to Nicole) there are so many other, and often more relevant, ways to judge the goodness of preferences and their objects. I’m happy to say that prudential goodness pretty much always matters, but that often other kinds of goodness matter more.
And Avi, I afraid I’m probably not qualified to comment on particular work done by particular agencies and organizations! But I don’t think this is purely an issue for the philosophy literature. Insofar as claims about prudential goodness are used to justify certain kinds of legislation, I think that we need to take into account the point in preference development that people affected by that legislation are at. So grant for a moment that women are socialized to value care-taking, and to be more likely to want to stay home with children than men. If we’re proposing policies to change that, at least in part on the grounds of the prudential good of those women, then I think we need to take seriously the possibility that lives like that might both a) be genuinely prudentially good for a lot of the women currently living them, and b) not be the prudentially best lives that girls and younger women could be socialized to value and pursue.