Navigation – Plan du site
Revue des livres
Comptes rendus

Cristina Bicchieri, Norms in the Wild. How to Diagnose, Measure, and Change Social Norms

Cédric Paternotte
p. 267-272
Référence(s) :

Cristina Bicchieri, Norms in the Wild. How to Diagnose, Measure, and Change Social Norms. Oxford: Oxford University Press, 2016, 264 pages, ISBN 978-0190622053.

Texte intégral

Afficher l’image
Crédits : Oxford University Press

1Since the publication of The Grammar of Society twelve years ago, it is fair to say that Cristina Bicchieri has become possibly the world’s main philosophical specialist on social norms. Her approach has then become increasingly empirical, both in her search for experimental evidence supporting her theory and her interest in real-world phenomena. Norms in the Wild is the result of such developments, as it stems from her work for the World Bank regarding how undesirable social norms such as female genital cutting or open defecation, should be concretely tackled (that is, modified or suppressed). Indeed, the book's aim is to “avoid the sort of failures that interventions have met in the past.” (105)

2The book offers the rare, interesting example of a philosophical theory that actually bears practical fruits—that becomes applied and even useful. In the first chapter, Bicchieri recalls her account of norms—namely that they exist when a number of individuals prefer to act in a certain way because they expect others to do the same (empirical expectations) or/and because they think that others expect them to act in this way (normative expectations). In order to change a behaviour you have to know which factors influence it; social norms amount to behaviours that depend on both empirical and normative expectations. So, in practice, preferences and expectations of various kinds have to be measured first—the topic of the second chapter. Factual beliefs, empirical and normative expectations are relatively easy to measure through questionnaires. But the key to detecting social norms is to find out whether people's preferences depend on these expectations. Laboratory experiments often cannot be performed in the field; hence the resort to hypothetical scenarios or vignettes—involving the subject or fictional characters—in order to check what may happen when expectations are modified. Although not ideal, these works as a “proxy for manipulating expectations in the lab” (105).

3The third chapter deals with the appearance, modification and disappearance of norms. Such changes typically involve three necessary ingredients: shared reasons for change; the solution to a collective action problem—as there are typically incentives not to deviate from an existing social norm due to punishment; and the collective coordination on a specific change. Beyond that, norm creation and norm abandonment require different processes. For a norm to arise, normative expectations must be manufactured first; for a norm to disappear, empirical expectations first have to become weaker—all have to start believing that less and less people follow the norm. Social norms may arise from descriptive ones (practices fuelled by conformity) that gradually acquire a normative dimension, or from the need to solve a problem of cooperation. But the suppression of social norms proves to be the most problematic part. Bad norms are often perceived as better than their alternatives, or their alternatives are simply unknown. Different cases call for different methods, including: mentioning empirical evidence, appealing to the moral worth of alternatives, making the incoherence between a norm and other values salient through discussion. The resilience of undesirable behaviours may be surprising—new ways of interpreting a social category may simply be added to previous ones without replacing them. Bicchieri mentions (140-141) the telling example of this Tanzanian village in which what is expected from a good wife came to include working and providing for her family not instead of, but in addition to, cleaning, meal preparation, social activities and marital duties.

4The fourth chapter lists a number of concrete tools with which one may try to change a social norm, the most efficient of which are not necessarily the most intuitively appealing. As changes need to be simultaneous in a population, all such tools rely on publicity. Legal means, and top-down interventions more generally, work only when institutions are trusted and when changes remain moderate (close to the existing norms). Economic incentives typically lack efficiency because they degrade or corrupt the behaviour they promote. But media-based solutions such as progressive soap operas or nudging information campaigns have been seen to lead to measurable changes; and so has deliberation, which allows individuals to openly discuss taboos, at least when it is not distorted by power dynamics or perceived as manipulative.

5Despite all these possibilities, norm change is hard because the first steps are. Accordingly, the fifth and last chapter discusses “trendsetters”—the first movers—and the ways in which they influence norm change. Populations always are heterogeneous: their members differ with respect to their sensitivity to norms, their autonomy, their perception of and sensitivity to risk... First-movers may be more autonomous than others; once they deviate, the risk of deviance is perceived as diminished by others, which may lead to more deviance, and so on. So, individuals of the trendsetter type (autonomous, risk and norm-insensitive—also measurable, in principle, through surveys and vignettes) should be targeted first by interventions that aim to change norms. Once they reach a certain threshold, population-level change may be triggered, often quite suddenly. One favourable scenario for norm change involves the formation of small trendsetter groups, the behaviour of which then propagates by way of public commitments (in small communities) or media interventions (in larger ones)—which Bicchieri highlights as particularly efficient.

6The reader ends up with a nuanced, context-dependent view of norm change. There is no magic bullet for getting rid of undesirable social norms. Rather, there are several possible changes, various factors on which to intervene and a number of tools by which such interventions may be performed. Theory only leads you so far; at some point, the local peculiarities of the targeted behaviour, of the community that has adopted it, of the reasons why they so behave, including the links with their underlying beliefs and values, all become relevant and call for a tailor-made process.

7Accordingly, in what follows I would like to use Bicchieri's book to discuss the links between theory and practice more generally; although Bicchieri's initial account of social norms never was armchair philosophy, it mostly focuses on definitions and model-based insights.

8It is interesting to note that the practical application of a theory may only require some aspects of that theory. Accounts are not unified, tightly integrated wholes, especially in the social sciences; part of them may be empirically idle. Indeed, Bicchieri's account of social norms is only partly applied ‘in the wild’. Bicchieri (2006)’s account contained a definition of social norms as well as a game theoretic model of the change in preferences that constitutes them, which showed how and how much the motivational structure of interactive situations may change when social norms come into play. This formal, quantitative aspect is absent from her new book, for at least two reasons. First, the game-theoretic model had little predictive import—it depended crucially on an individual's norm sensitivity, which is difficult to assess quantitatively in general. Second and more importantly, when dealing with norm change in practice, one is interested in determining which interventions are causally efficacious, that is, whether an intervention would make a difference. The intensity of the causal effect is only secondary—of course, the stronger the better, but the main issue is to find out what works to begin with. Note that Bicchieri’s definition of social norms and her game-theoretic analysis were already independent from one another in her original account, which explains how the former can be retained even when the latter is dropped.

9Moreover, the specific labels that one may provide to different kinds of socially influenced behaviours (conventions, descriptive norms, social norms...) also become irrelevant in practice. What matters is that we have a list of the possible causally relevant factors and of their relations. Bicchieri mentions, as fundamental building blocks, factual beliefs, empirical expectations, personal normative beliefs, normative expectations, and preferences. We may, then, be interested in their comparison; for instance, a gap between personal normative beliefs and normative expectations may reveal a situation of ‘pluralistic ignorance’ (more on this below), which is much easier to suppress than full-fledged social norms. Or we may be interested in their dependence: for instance, that of preferences on normative expectations. All this has implications on the possibility and nature of a successful intervention—but the precise labelling of sets of ingredients does not. In other words, practical application may trump semantics: what matters is that our distinct relevant sets of factors somehow carve the causal reality at its joints. Semantic disagreements need not impinge on applications in the wild.

10Another interesting link between theory and practice is that the latter can highlight new theoretical properties of the target phenomena. This is not the basic claim that observation provides evidence for theories, but that once interpreted and probed in the light of our theory, phenomena may reveal new links between theoretical properties or new aspects on which the theory is still silent. For instance, once the empirical/normative expectation distinction is in place, it becomes possible for Bicchieri to notice that for norms to appear, empirical expectations must change before normative ones, but that for norms to be abandoned, the converse holds—theoretical features that had not been identified beforehand. Another useful lesson gained from practice is that our normative nature abhors a vacuum. It is harder to suppress a norm altogether than to replace it by another; and a good way to fight a bad norm is to reframe its alternatives positively by piggybacking on other shared values of the population (for instance by finding ways to describe deviant behaviours positively, e.g. as honourable, as signs of purity, etc.—see 139-140).

11A third important lesson from practice is that the stabilisation of a norm crucially depends on what happens at the micro-level and becomes more likely as top-down interventions fade or disappear. For instance, for deliberation to favour a new norm, it is crucial that the debating individuals come to abandon an undesirable behaviour by themselves (159); and deliberation facilitators are more efficient if they share the individuals’ core values rather than question or oppose them. In another telling example, Bicchieri highlights the fact that the abandonment of open defecation by the adoption of public latrines was only successful when communities decided themselves to monitor and sanction deviant behaviours, and collectively chose the kind of sanction to be adopted (116). In other words, micro-level processes are integral to the success of any change of social norms. This betrays, once again, the need for sensitivity to local conditions and the limits of external paternalistic attitudes.

12Let me mention one minor criticism though. For such a strongly empirically-minded book, it is surprising that Bicchieri in places still uses purely theoretical notions or model-based considerations. Consider the chapter on trendsetters, for instance, which is actually more focused on norm dynamics—on the ways in which new behaviours can cascade out in a population of norm-followers. This is made easier by the heterogeneity of the population: norm-insensitive people deviate first, then influencing others. But there is no prior trendsetter type that could be identified and targeted; what counts as a trendsetter depends on the size of the minimal community needed for a global change, which is in turn context-dependent. This reminds me of Bicchieri’s previous work on “pluralistic ignorance” (see the fifth chapter of The Grammar of Society, itself based on Bicchieri and Fukui, 1999) which stemmed from a model in which norm adoption and disappearance heavily depended on the co-presence in a population of two idealised, sharply distinct types—trendsetters and pure conformists. Even if there is no such clear-cut distinction here anymore, the use of the trendsetter term and of formally-inspired considerations about the distributions of various kinds of sensitivities in the population strike me as pedagogically problematic, and as the last remnant of theoretical considerations with limited empirical consequences.

13None of this is meant to take anything from Bicchieri’s work, which will be interesting to many, whether they be interested in concrete social change, in seeing a philosophical account bear practical fruits, or in the philosophy of social science in general. The book is packed with examples, some of them well-known and others quite surprising. Its modesty and dismiss of turnkey solutions is also laudable—changes of social norms demand processes which must be context-tailored, but which can only be built from a somehow limited toolbox. All this is work in progress—but in the best sense of the word.

Haut de page


Bicchieri, Cristina. 2006. The Grammar of Society: The Nature and Dynamics of Social Norms. Cambridge: Cambridge University Press.

Bicchieri, Cristina and Yoshitaka Fukui. 1999. The Great Illusion: Ignorance, Informational Cascades and the Persistence of Unpopular Norms. Business Ethics Quarterly, 9: 127-155.

Haut de page

Pour citer cet article

Référence papier

Cédric Paternotte, « Cristina Bicchieri, Norms in the Wild. How to Diagnose, Measure, and Change Social Norms », Œconomia, 8-2 | 2018, 267-272.

Référence électronique

Cédric Paternotte, « Cristina Bicchieri, Norms in the Wild. How to Diagnose, Measure, and Change Social Norms », Œconomia [En ligne], 8-2 | 2018, mis en ligne le 01 juin 2018, consulté le 20 février 2019. URL :

Haut de page


Cédric Paternotte

Sorbonne Université, équipe SND.

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