3.9: The capability approach and adaptive preferences

As we saw in the previous section, a widely-voiced reason offered for rejecting the happiness approach as an account of wellbeing is the phenomenon of adaptive preferences, which has been widely discussed in the literature (e.g. Elster 1983; Sen 1985c, 3, 1992a; Nussbaum 2000; Teschl and Comim 2005; Burchardt 2009; Khader 2009, 2011, 2012, 2013; Conradie and Robeyns 2013). Phenomena of mental adaption are a problem if we take happiness or desire-satisfaction to be our account of wellbeing. Yet we also concluded in section 3.7.3 that the capability approach sometimes boils down to a desire-fulfilment account of wellbeing. Hence we need to ask: how do processes of adaptation affect the desire-fulfilment view of wellbeing, and what are the implications for the capability approach?

In the most general terms, preferences formation or adaptation is the phenomenon whereby the subjective assessment of one’s wellbeing is out of line with the objective situation. Two persons who find themselves in the same objective situation will have a very different subjective assessment, because one is happy with small amounts of ‘objective goods’, whereas the other is much more demanding. In the capability literature, the general concern is with deprived persons who, over time, adapt to their objectively poor circumstances, and report a level of subjective wellbeing which is higher than the objective circumstances warrant.

The idea of adaptation can take different forms. Jon Elster (1983) referred to one particular type of adaptation, in which being unable to fulfil a preference or realise an aspiration leads one to reject that preference or aspiration. This phenomenon is known as ‘sour grapes’: the fox who cannot pick the grapes, because they are hanging too high for him, starts telling himself that they are sour anyway, and no longer desires to eat them. On Elster’s account, adaptation occurs at a non-conscious level, as a reaction to the painful process of cognitive dissonance that a person who can’t fulfil her unreachable desires or aspirations feels. Elster’s notion of adaptive preferences only refers to a process, and makes no reference to an objective notion of wellbeing. These psychological aspects of adaptation are echoed in Sen’s reference to this phenomenon, when he writes that “considerations of ‘feasibility’ and of ‘practical possibility’ enter into what we dare to desire and what we are pained not to get” (Sen 1985a, 15). Adaptive preferences are a reason for Sen to reject a focus on mental metrics, such as utility or happiness, as the metric of wellbeing. After all, someone who is in an objectively dire situation may have adapted to that situation and learnt to be pleased with little. As Sen (1985c, 21) puts it, “A person who is ill-fed, undernourished, unsheltered and ill can still be high up in the scale of happiness or desire-fulfilment if he or she has learned to have ‘realistic’ desires and to take pleasure in small mercies”.

Serene Khader (2011) has argued that not all cases that we tend to consider as cases of adaptive preferences fit Elster’s conceptualisation. Khader believes that an account of adaptive preferences must make reference to an objective notion of flourishing, even if that notion remains vague and only focusses on basic flourishing (since there is more intercultural agreement on what basic flourishing entails). She develops the following definition:

An adaptive preference is a preference that (1) is inconsistent with a person’s basic flourishing, (2) was formed under conditions nonconducive to her basic flourishing, and (3) that we do not think a person would have formed under conditions conductive to basic flourishing. (Khader 2011, 51)

A similarly perfectionist, but much less systematically developed account of adaptive preferences can be found in Martha Nussbaum’s work. She understands adaptive preferences as the preferences of people who do not want to have items of her list of capabilities, whereby these preferences are deformed due to injustices, oppression, ignorance and unreflective habit (Nussbaum 2000, 114).

What questions do adaptive preferences raise for the capability approach? At the very minimum, they raise the following questions: first, why would adaptive preferences pose a problem for capability theories? Second, do we have any evidence about the prevalence of adaptive preferences? And third, how can capability scholars deal with adaptive preferences in their capability theories and applications?

Let us start with the first of these questions: why would adaptive preferences pose a problem for capability theories? There are at least two reasons. The first lies in module B2, the selection of dimensions. If that selection is done in a participatory or democratic way, then it may be vulnerable to adaptive preferences. A group that is systematically socialised to have low aspirations and ambitions will perhaps not put certain capabilities on its list, thereby telling themselves that they are unachievable, whereas objectively speaking they are achievable, albeit perhaps only after some social changes have taken place. The second reason is that a person with adaptive preferences may objectively have access to a certain capability, but may believe that either this capability is not available to her, or else that she should not choose it, and hence she may pick from her capabilities set a suboptimal combination of functionings. If we then assume that this person (or group) has non-adaptive preferences, then we will wrongly interpret the choice not to exercise certain capabilities as a matter of personal agency, which a capability theory that focusses on capabilities rather than functionings, should respect. The capability approach by default regards adults as agents rather than patients, but this may be problematic in the case of adaptive preferences.

So we can conclude that adaptive preferences can pose a problem for capability theories in which the choice of dimensions is made democratic, or in which we focus on capabilities rather than functionings. But a critic may raise the question: do we have any evidence about the prevalence of adaptive preferences? Is this not a theoretical problem invented by philosophers who like complex puzzles, or by western scholars who pity the lives of poor people in the Global South?

There are at least two answers to be given to this question. The first is that there are indeed good reasons to be very careful with the conclusions we draw when studying adaptive preferences, especially in a context with which one is not familiar. Serene Khader (2011, 55–60) provides a nuanced and convincing discussion of the various mistakes that can be made when we try to identify whether a person or group of persons living under unjust conditions expresses adaptive preferences. There are at least three ‘occupational hazards’ that those trying to identify adaptive preferences may make: we run the risk of psychologizing structural constraints, of misidentifying possible trade-offs between various dimensions of wellbeing that a person makes, or we may be unable to recognise forms of flourishing in very different culture or class settings. All this shows that thinking about adaptive preferences needs to be done with great attention to contextual details and in a very careful manner; it is not an analysis that can easily be done by applying a rigid formula. Scholars should therefore be very cautious before concluding that someone or a group shows adaptive preferences, and carefully investigate alternative interpretations of what they observe, since otherwise they run the risk of seeing adaptive preferences where there are none.

Having said this, it is clear from the literature that adaptive preferences are a genuine phenomenon. For example, Serene Khader (2011) discusses real cases of groups of women who had adaptive preferences. Tania Burchardt analysed the 1970 British Cohort Study and found that “among those able to formulate agency goals, the aspirations expressed are conditioned by their socio-economic background and experience” (Burchardt 2009, 13). She also found evidence that adaptation may play a role in the selection of functionings from one’s capability set, since among the sixteen-year-olds who have the capability to continue full-time education, the choice whether or not to do so is highly influenced by past deprivation and experiences of inequality. Burchardt rightly concludes that if the influence on people’s choices is so systematically related to previous experiences of disadvantage, that this is a case of injustice. Hence the need, for capability theorists and not just for those endorsing the happiness approach or the desire-fulfilment theory of wellbeing, to take processes of adapted preferences and adapted aspirations seriously. On the other hand, as David Clark (2009, 32) argued in the context of development studies, adaptive preferences may not be as widespread as some capability theorists make it out to be: “the available evidence only provides limited support for the adaptation argument and is not always easy to interpret”. Given the “occupational hazards” that those trying to identify adaptive preferences face (Khader 2011, 55–60), it is important not to ‘see’ adaptive preferences where there are none. In conclusion, the capability scholar will have to balance the tricky tasks of neither ignoring processes of adaption, nor making the adaptation problem bigger than it really is.

This brings us to the last question: can the capability approach deal with these issues? Given that capability theories and applications can be very diverse, we will need different methods to handle the issue of adaptive preferences for different capability theories and applications.

In the context of action-research, small-scale projects and grassroots strategies, what is required above all is deliberation and interaction with people of whom one may be worried that their preferences may show signs of adaption, as exemplified by Ina Conradie in her project with women in a South African township (Conradie 2013; Conradie and Robeyns 2013). Khader (2011) has developed ‘a deliberative perfectionist approach to adaptive preference intervention’, in which a practitioner who suspects that a group of people has adaptive preferences will first attempt to understand how the suspected preferences affect their basic flourishing. This must be done via deliberative processes — a strategy that we also see in Conradie’s research. If the practitioner has good reasons to suspect that some of the preferences are adapted, she can involve those with the alleged adaptive preference in a discussion and together search for a strategy for change. Note that there is an interesting parallel here with the grassroots-based development model that has been proposed by Solava Ibrahim (2017), in which ‘a conscientization process’ is an integral part of the development process. In this process, a person reflects critically on her life, develops aspirations for better living conditions, and makes a plan of action to bring about the desired change (Ibrahim 2017, 206). While, as Ibrahim rightly notes, adaptive preferences and aspirations may provide a challenge for this conscientization process, they are also very likely to be challenged and hence changed via such a process.

What about capability applications that involve the empirical analysis of large-scale datasets? How can adaptation be dealt with in those applications? Here, the capability approach needs to use insights from the disciplines that have built most expertise in large-scale adaptation processes, such as sociology and social policy studies. Based on the insights from those disciplines, we know the likely candidates to be dimensions of adaptation — such as social class, caste or gender. We can then use indicators of those dimensions to study whether preferences and aspirations systematically differ, as in the earlier mentioned study by Tania Burchardt (2009). But it is clear that this can only help us to identify adapted preferences or adapted aspirations; it will not always tell us whether for each application it is possible to ‘launder’ the data so as to clean them from processes of unjust adaptations.