Navigation – Plan du site
Revues des livres
Comptes rendus

Anna Alexandrova, A Philosophy for the Science of Well-Being

Marc Fleurbaey
p. 409-412
Référence(s) :

Anna Alexandrova, A Philosophy for the Science of Well-Being, Oxford: Oxford University Press, 2017, 248 pages, ISBN 978-019930051-8

Texte intégral

Afficher l’image
Crédits : Oxford University Press

1Subjective well-being studies have thrived in the last decades, with the development of happiness questionnaires and the recent enthusiasm of economists for subjective data. This wave has left many skeptics on the side, for whom the underlying theory of well-being is unappealing and the empirics is not cautious enough. In these troubled times, this book makes a very welcome effort to connect the dots between theory and empirics, and to put some order in concepts and methods.

2By and large, this book is primarily a plea for more attention being paid to philosophy by scholars of well-being, who these days seem to be mostly in psychology and in economics. But in the first part, the book also turns toward philosophers and argues in favor of an empirical turn to the philosophy of well-being.

3The first part of the book, indeed, criticizes the philosophical orthodoxy that views well-being as “that general, all-things-considered evaluation.” Instead, the author proposes to replace it by a contextualist approach for which well-being may have a specific definition depending on who and where and when the concept is deployed. To flesh out the thesis, this part ends with a chapter on child well-being. In the case of children, it is indeed clear that there are specific considerations about personal development that do not appear similarly for adults.

4The author even denies that contextualism should relate to a single overarching theory of well-being. It is true that philosophy offers competing theories of well-being, and there is little hope that convergence toward a single theory will ever occur. Somehow, the author believes that various contexts may call upon diverse theories. She opposes the “vending machine” view, according to which a single theory could offer the concepts and measures needed for various contexts, to the “toolbox” approach, according to which different theories provide different notions that can be relevant in various contexts. Unfortunately, she does not venture to suggest which theories may be needed, and how theories and contexts should be associated. Fleshing out the architecture of theories, concepts and contexts would be the most compelling defense of the thesis.

5The author’s preferred approach, as illustrated in the child well-being chapter, is what she calls “mid-level theories.” A mid-level theory is adapted to the object of study (child or adult, country or individual, sick or healthy person…), to the subject doing the study (government, friend, caregiver…), and the possibilities for changing the situation of the object of study (the less one can do for the person, the lower the standard of well-being). The author acknowledges that not many philosophers of well-being may be willing to go that road, since it brings a lot of empirical and pragmatic considerations into the picture.

6Let me try to relate this approach to my own questioning about how well-being has been approached in Rawlsian and post-Rawlsian theories of justice. In these theories, there is a central need to compare people’s advantage and to define priority for the worse off in the morally relevant way. One reading of such approaches to interpersonal comparisons is that they are not about well-being but about something else. A fair share of resources, an adequate set of opportunities, seems to refer more to the means for flourishing than to well-being itself. However, given that resources and opportunities are multidimensional, one would like to measure these objects in a way that serves the interests of the recipients, and therefore one may want to mobilize theirs values and desires in the measures. Can one understand such a particular context (a theory of justice focusing on means to flourishing) as calling for a particular value- or preference-sensitive measure of well-being in terms of resources and opportunities? Such a measure is unlikely to be useful for other contexts. This could be an illustration of the author’s thesis that a single theory of well-being is unlikely to serve all purposes in all contexts.

7The author’s example about child well-being is compelling, but perhaps too disconnected from philosophical theories to really illustrate the author’s thesis. One would like to see where and how philosophical theories are useful, and in the case of children, sociological and psychological considerations about child development and affect seem to be the main source of inspiration for rejecting simple theories that draw on hedonism, subjectivism and objective lists. The case of child well-being, moreover, is problematic because it is not clear that any precise theory of child well-being can be constructed without first deciding how adult well-being should be defined. If childhood is only a temporal subset of a person’s life, child well-being should be about promoting overall lifetime well-being. If overall lifetime well-being is mostly about emotions, then equipping the child with the means to access good emotions should shape the developmental part of the measure, and current emotions would feed the measure for its present-time component. Other approaches would see it quite differently on either count. The author does discuss a lifetime inductive approach for a special theory, namely, subjectivism (i.e., relying on people’s values and desires). But it is not obvious why some kind of lifetime induction is not logically necessary if the measure of child well-being is meant to be part of a consistent set of measures.

8In a nutshell, the author’s argument for mid-level theories may put the finger on what is perhaps an important flaw in the philosophical orthodoxy of well-being (namely, that there is one true concept of well-being), while running the risk of overselling contextualism and pragmatism, leaving too little room for real philosophical debate in important contexts such as the construction of a theory of justice.

9In the second part of the book, the author casts a critical eye on well-being studies as they are being made. She first rejects the idea that such studies can be purely objective. The reason is that the selection of a notion of well-being is necessarily a normative step, making any statement about it a “mixed claim”, i.e., a statement of fact about notions that are value-laden. The methodology she advocates involves deciphering the value presuppositions in the concepts and measures that are used, check if they involve controversy, and if so, consult the relevant parties. An example she provides is the following. Does growth promote well-being? According to empirical studies, this is more the case for satisfaction with life than for emotional happiness. Since there is controversy about hedonism, one should consult the relevant parties about which measure of well-being is to be chosen.

10Here, I would like to express frustration when a philosopher defers to “the relevant parties” in the case of disagreement. Indeed, when there is disagreement, one would also like philosophers (or other relevant experts) to make suggestions for possible solutions or compromise. I wonder why one could not simply say that in each context, one should rely on the relevant normative approach. For instance, for defining priorities between different people in social policy, one may want to rely on a theory of justice. The fact that there is disagreement between theories of justice does not exempt philosophers from continuing to refine these theories.

11Another chapter discusses the measurability of well-being, scrutinizing Hausman’s claim (in Valuing Health, Oxford University Press, 2015) that well-being is not measurable because it involves too complex an aggregation of the multiple goods that make up a good life. The author concedes that an encompassing notion of well-being is unlikely to be measurable, but argues that contextual notions as designed in mid-level theories are more likely to be measurable, because they are less ambitious.

12The most interesting part of the book, in my view, is the discussion of construct validation, a topic introduced in this measurement chapter and analyzed extensively in the last chapter. The author analyzes the usual approach as validating a measure when 1) it is inspired by a plausible theory of the concept; 2) subjects reveal that the measure tracks the concept in survey answers; 3) other data corroborate the correlation between the measure and the concept. The author notes that this “coherentist” approach is likely motivated by the absence of an axiomatic basis relating the measure to the concept. But in the last chapter she shows how construct validation is often performed by largely skipping step 1 and being sloppy about steps 2 and 3. She lambasts psychometrics for theory avoidance, making a plea for going back to theory first in order to check how the measures are related to the relevant target concepts. An interesting example provided is the list of affects retained in some measures of hedonic happiness, and which appears to miss obviously important affects such as joy, sadness, anxiety, stress, after a factor analysis has selected a list in a sort of blind (statistical) way. The author could also mention weak correlations with other data used to validate particular measures, or arguments that measures that are related in similar ways to objective data (such as income or marital status) must be equivalent. It makes a lot of sense to encourage greater reliance on theory and better checking the direct empirical link between concept and measure.

13This book appears like a first step in the construction of a more ambitious project. One would like to see the two parts of the book reconnected to one another in the development of a set of theories and concepts that can be useful in relevant contexts, and for which measures which are rigorously tied to the concepts can be obtained from accessible data. Another possible project, starting from this book, would revisit the existing measures and check how well or badly they reflect particular concepts of specific theories, and how useful they can be in particular contexts. This book lays the foundations, defines the basic notions, and it remains to be seen if this can inspire an innovative approach that would change the conversation, either in philosophical theories of well-being or in empirical studies of well-being.

Haut de page

Pour citer cet article

Référence papier

Marc Fleurbaey, « Anna Alexandrova, A Philosophy for the Science of Well-Being », Œconomia, 8-3 | 2018, 409-412.

Référence électronique

Marc Fleurbaey, « Anna Alexandrova, A Philosophy for the Science of Well-Being », Œconomia [En ligne], 8-3 | 2018, mis en ligne le 01 septembre 2018, consulté le 21 janvier 2019. URL : http://journals.openedition.org/oeconomia/3383

Haut de page

Auteur

Marc Fleurbaey

Princeton University

Articles du même auteur

Haut de page

Droits d’auteur

Licence Creative Commons
Les contenus d’Œconomia sont mis à disposition selon les termes de la Licence Creative Commons Attribution - Pas d'Utilisation Commerciale - Pas de Modification 4.0 International.

Haut de page
  • Logo Association Œconomia
  • Logo CNRS
  • OpenEdition Journals