1. Introduction

The aim of this article is to argue that bonobo societies exhibit social norms in the form of socially maintained normative regularities. To this end, I propose to apply the normative pluralism framework (^{Westra and Andrews 2022}; ^{Westra et al. 2024}) to the case of bonobos. This framework allows normativity to be analyzed without presupposing a single cognitive architecture or a strict definition of mental capacities, which makes it particularly suitable for studying non-human species.

The central argument I defend in this article is that if bonobos exhibit socially maintained patterns of behavioral conformity, then they display social norms. According to the normative pluralism framework, a behavior qualifies as a normative regularity when it constitutes a socially maintained pattern of behavioral conformity (^{Westra and Andrews 2022}; ^{Westra et al. 2024}). I argue that bonobos exhibit such normative regularities in certain social behaviors. I focus on three of these: first, inequity aversion. In cooperative contexts, this can be interpreted as a social response that reinforces a norm of equitable distribution. Second, reactions to interruptions of social activities. During activities such as grooming or play, bonobos tend to maintain social activities until a clear disengagement signal is given, and they use gestures to reestablish the interaction if it is interrupted. Third, female coalition formations. I argue that these coalitions operate as a form of third-party punishment aimed at protecting females and preserving group cohesion. Across these three behaviors, normative regularities can be observed, which supports the conclusion that bonobos manifest social norms within their societies.

Recognizing social norms in bonobos has implications for a philosophical debate within the field of moral psychology about what social norms are and whether they are unique to humans. In the literature on moral psychology, social norms are typically regarded as exclusively human (^{Westra et al. 2024}). Most theoretical proposals assume that social norms arise from forms of cognition and representation that only humans possess (^{Bicchieri 2006}; ²⁰¹⁷). Some have argued that social norms are possible thanks to “ought-thoughts” as cognitive representations that only humans can form (^{Nichols 2004}), to the human form of self-consciousness (^{Rochat 2015}), to unique human capacities for shared intentionality (^{Tomasello 2016}; ²⁰¹⁹), to gene-culture coevolution in human history (^{Richerson and Boyd 2005}; ^{Henrich 2015}; ^{Kumar and Campbell 2022}), or to specialized cognitive mechanisms for tool-making (^{Birch 2021}; ^{Sterelny 2021}). All of these positions assume that social norms constitute a relatively unified set of cognitive mechanisms unique to humans. However, there is no consensus in moral psychology about what this unified cognitive architecture underlying the human sense of normativity actually is (^{O’Neill and Machery 2018}).

For instance, one of the most influential approaches in moral psychology is that of ^{Cristina Bicchieri (2006}; ²⁰¹⁷⁾, who defines social norms as rules of behavior to which individuals have a conditional preference to conform. That is, individuals follow such rules only if (a) they believe that most other members of their group conform to the norm (empirical expectation), and (b) they believe that most members of their group believe they ought to conform to the norm as well (normative expectation) (^{Bicchieri 2017, p. 35}). The motivation to conform to a norm is framed in terms of desires, such as the desire to avoid sanctions, to gain approval, to please others, etc. (^{Bicchieri 2006, p. 24}). Consequently, Bicchieri’s approach conceives social norms as a social construct whose building blocks are mental states such as beliefs and desires, which are instantiated through expectations and rational preferences (2006, p. 11).

Bicchieri’s definition of social norms is problematic when applied to non-human primates, such as bonobos, insofar as it presupposes advanced forms of mentalization, including the attribution of propositional attitudes or mental states-specifically, the capacity to hold beliefs about what others believe. Such capacities are controversial in non-human animals, and there is no clear consensus regarding when and how they emerge in humans (^{Andrews et al. 2024}). As a result, Bicchieri’s approach becomes strongly mentalistic with respect to the psychology of norms (^{Westra and Andrews 2022}) and may over-intellectualize both human and non-human conformity to norms (^{Andrews et al. 2024}). In this vein, Bicchieri’s approach and other similar explanations of norms mentioned above presuppose psychological requirements-such as a homogeneous cognitive mechanism for attributing propositional attitudes like beliefs-that only humans could satisfy (^{Westra et al. 2024}).

In contrast, the normative pluralism framework has been proposed as a methodological and conceptual approach for understanding social norms as the product of a diverse set of underlying cognitive, affective, and ecological processes of varying degrees of complexity, which makes it possible to identify social norms beyond the human dimension (^{Westra and Andrews 2022}; ^{Westra et al. 2024}). In this sense, this perspective seeks to advance previous debates on animal normativity (such as those of ^{Lorini 2018}; ^{Danón 2019}; ^{Andrews 2020}). This approach is pluralistic insofar as it accepts that the cognitive processes involved in normative psychology can be highly diverse. Instead of treating social norms as a homogeneous category of mental representations, the normative pluralism framework suggests that the mechanisms of normative cognition may be quite heterogeneous, and therefore there is no single set of cognitive mechanisms responsible for normative cognition. In this sense, since social norms are likely produced by a broad range of cognitive mechanisms, many of these mechanisms must be shared with non-human animals. Identifying a normative dimension in the social practices of bonobos supports this framework and extends its empirical scope, demonstrating its applicability beyond human beings.

Although similar approaches have been applied to chimpanzees (^{Fitzpatrick 2020}; ^{Danón 2024}), this framework has not yet been directly applied to bonobos. While the empirical evidence on the social behaviors of bonobos-both in the wild and in captivity-remains more limited than that available for chimpanzees (^{Hare and Yamamoto 2017}), I argue that the existing data support the claim that some of their social behaviors can be analyzed from a normative perspective.

The article is organized as follows: in section 1, I present the normative pluralism framework and the definition of normative regularity, which is based on two key elements: (1) the existence of a pattern of behavioral conformity and (2) the presence of social maintenance mechanisms. In section 2, I analyze how these two components are expressed in the inequity aversion observed in bonobos. Section 3 focuses on social activities such as grooming and play, which are likewise interpreted through the notion of normative regularity. Finally, in section 4, I examine female coalition formations, highlighting how these alliances can act as forms of social sanction.

2. The Normative Pluralism Framework

The normative pluralism framework proposes understanding social norms as normative regularities. According to this approach, rather than grounding social norms in discrete cognitive mechanisms, we should focus on identifiable behavioral patterns. This makes it possible to recognize that the psychology of social norms can emerge from multiple distinct processes. What these processes share is not a common cognitive architecture but a causal role: they act as forces that stabilize and maintain such normative regularities. In this sense, a social norm is not something separate from a normative regularity; they are, in fact, the same thing.

The normative pluralism framework defines normative regularities as “a socially maintained pattern of behavioral conformity within a community” (^{Westra and Andrews 2022, p. 9}). This definition is structured around three main components. The first is the concept of patterns of behavioral conformity. These describe observable regularities in the behavior of members of a community. One of the central motivations underlying this definition of social norms is that many standard accounts in moral psychology overlook the fact that social norms are identifiable behavioral phenomena. A key feature of this approach is that such patterns are aspects of the social world that can be recognized without needing to know the underlying psychological mechanisms that produce them. This idea aligns with ^{Dennett’s (1991)} concept of real patterns, according to which certain phenomena can be considered real if they are stable, identifiable, and have predictive power, even if their internal mechanisms are unknown or diverse. This motivation makes it possible to define social norms in primarily behavioral terms before advancing claims about the kinds of psychological processes that might give rise to such norms. This point is precisely what enables a pluralistic account of the underlying psychological processes. Once social norms are identified as social and behavioral phenomena, then it becomes clear why one must allow for the possibility that a plurality of distinct psychological processes may underlie and sustain social norms across different species.

Therefore, we do not need to assume a single psychological architecture underpinning social norms. It is enough to recognize that certain behavioral recurrences within a community work as social norms because they help predict actions, explain interactions, and coordinate expectations. Thus, the existence of a social norm does not depend on all agents sharing the same cognitive representation of the norm, but rather on the presence of a stable and effective behavioral pattern. Once this pattern is detected, the pattern is the norm: no deeper psychological grounding is required to justify its status.

The second key component of the definition is social maintenance. Not every behavioral pattern constitutes a normative regularity; for it to operate as a social norm, it must be sustained by some form of social pressure that promotes conformity. A pattern is considered socially maintained when individual adherence to it can be explained by the social reactions that instances of conformity and non-conformity elicit. In other words, social maintenance refers to social responses or behaviors through which a community incentivizes conformity to the norm and discourages non-conformity (^{Westra and Andrews 2022}). More specifically, the normative pluralism framework distinguishes between two broad types of social maintenance mechanisms: negative and positive. Negative social maintenance comprises two forms of punishment: second-party and third-party punishment. Second-party punishment consists in negative responses by individuals who are directly affected or harmed by a violation of a social norm, whereas third-party punishment refers to negative responses by observers who are not directly affected by, or involved in, the transgression of the norm (^{Westra and Andrews 2022}; ^{Andrews et al. 2024}). Third-party punishment is generally regarded as a strong empirical marker of the presence of a social norm (^{Westra et al. 2024}). Examples of negative social maintenance include physical retaliation, signals of disapproval, protest, negative affective responses, withdrawal from the violator, and related behaviors (^{Andrews et al. 2024}). Positive social maintenance mechanisms, by contrast, refer to social incentives aimed at promoting conformity to the norm. Examples include social inclusion in group activities, gains in social status, explicit verbal praise (in human cases), differential treatment and pedagogical practices such as teaching and facilitating social learning (^{Westra and Andrews 2022}; ^{Andrews et al. 2024}). This means that community members not only observe and reproduce the behavior but also actively reinforce it, whether through approval, disapproval, reward, or punishment. In this way, social maintenance distinguishes normative regularities from mere customs or collective habits by highlighting their normative role within the social fabric.

The third and final component of the definition is the concept of community. This term refers to the set of relevant individuals who exhibit the pattern of behavioral conformity and participate in its social maintenance. In this sense, a social norm does not exist in the abstract but always within a specific group whose members share such regular behavior. This group can be all members of a population, a dyad, a subgroup, all members of the same species, or even members of different species. The specification of the community depends on the behavioral pattern under study. It is important to emphasize that the same individual can simultaneously belong to multiple normative communities and, therefore, conform to different norms depending on the context. This reinforces the idea that social norms do not depend exclusively on stable internal dispositions but are sensitive to the specific dynamics of the communities in which they operate. Recognizing the role of the community thus underscores the situated and plural nature of normative regularities.

It is worth noting that normative regularities may have a cultural component, a phenomenon that ^{Westra et al. (2024)} describe as communal normative regularities. Certain behaviors understood as cultural variations-namely, community-specific traits transmitted through social learning-might be regulated by social norms. These communal social norms are defined as “suites of normative regularities that serve to distinguish different animal communities behaviourally within a given species” (^{Westra et al. 2024, p. 1069}). As such, communal normative regularities enable the characterization of communities in terms of culturally distinctive normative patterns. This cultural dimension of social norms is closely connected to group identity as a normative phenomenon, often taking the form of implicit principles such as “What we do around here” (^{Andrews et al. 2024}). Evidence from bonobos, as with other great apes, supports the existence of community-specific cultural variation across both material and immaterial domains (^{van Leeuwen et al. 2020}), both in captive and wild populations. Approximately fourteen such behaviors have been identified, some shared with chimpanzees and nine unique to bonobos (^{Hohmann and Fruth 2003}). Cultural variation in captive bonobos has been documented in grooming, play, and group-specific sociability (^{Tokuyama et al. 2019}; ^{van Leeuwen et al. 2024}), as well as in emotional expressions such as consolation, which vary across communities (^{Brooker et al. 2025}). Moreover, wild bonobo groups display distinct hunting preferences despite sharing the same geographical space, a pattern plausibly explained by conformity to group-specific social norms (^{Samuni et al. 2020}; ^{Whiten 2020}).

Therefore, an investigation into normative regularities must clearly determine these three components: it should begin with the identification of (1) patterns of behavioral conformity and specify whether such patterns are subject to (2) forms of social maintenance within a (3) specific community (^{Westra et al. 2024}). In other words, it is not enough to observe a regularity; it is necessary to establish that this regularity is sustained through social mechanisms that incentivize conformity and discourage non-conformity. Regarding the third component of the definition, this article focuses on the case of bonobo communities, whether in captivity or in the wild. Therefore, the emphasis lies on identifying the two remaining components of the definition. I proceed to argue that these two fundamental aspects manifest in three domains of bonobo social behavior: inequity aversion, reactions to interruptions of social activities, and female coalition formation, which constitute examples of normative regularities.

3. Inequity Aversion in Bonobo Societies

In this section, I argue that responses to inequity aversion in bonobos constitute forms of social maintenance. Such responses should not be understood merely as affective or instrumental reactions to unfavorable distributions, but rather as indicators of an underlying behavioral pattern. Specifically, responses to inequity can be conceived as forms of social maintenance that reveal the presence of an implicit norm regarding the equitable distribution of resources. This norm should not be understood as an explicitly formulated or reflexively internalized rule, but rather as a behavioral pattern that guides and structures cooperative interaction between partners, especially in contexts where benefits must be distributed proportionally to effort.

Inequity aversion can be defined as the manifestation of negative reactions to the unequal distribution of rewards among individuals who have performed similar tasks (^{Ritov et al. 2024}; ^{Radovanović et al. 2025}). These responses can take various forms, one of the most common being the refusal to continue participating in the activity. In such cases, the subject chooses to withdraw upon observing that their partner has received a more valuable reward for the same effort (^{Brosnan and de Waal 2003}). Two modalities of inequity aversion are distinguished: advantageous and disadvantageous. Advantageous inequity aversion refers to situations in which the agent receives more than their partners. Conversely, disadvantageous inequity aversion refers to situations where the agent receives less than their partners for the same effort (^{Brosnan 2006}).

Cooperative dyadic interactions are commonly used to experimentally evaluate responses to inequity. These involve two individuals collaborating on the same task. In this type of experimental setup, a token exchange task is employed. This task requires participants to hand over a token to an experimenter in exchange for a reward. This allows for a comparative examination of individuals’ responses to different distributions of benefits.

Although studies on inequity aversion in bonobos remain limited, two recent investigations have addressed this issue (^{Verspeek and Stevens 2022}; ^{Radovanović et al. 2025}). Both studies focused on analyzing bonobos’ responses to disadvantageous inequity situations, using the token exchange task paradigm mentioned earlier. To date, no empirical evidence has been found supporting the presence of advantageous inequity aversion in this species (^{Verspeek and Stevens 2022}). In these experiments, cooperative interactions among bonobos were structured based on significant social bonds, predominantly mother-offspring relationships as well as relationships among adult females.

4. Reactions to Interruptions of Social Activities in Bonobo Societies

In this section, I argue that bonobos’ reactions to interruptions of social activities, such as grooming and social play,² manifest normative regularities. In particular, I propose that these primates follow social norms that regulate both the initiation and termination of social activities. Moreover, the use of communicative gestures to encourage the continuation of an interrupted interaction constitutes a form of social maintenance. Thus, the fact that bonobos actively prompt their partners to resume social activities through specific signals (^{Pika and Zuberbühler 2008}; ^{Heesen et al. 2020}, ^2021a, ^2021b; ^{Bangerter et al. 2022}) indicates the existence of an implicit norm that reinforces the obligation to complete the initiated activity.

A social activity can be defined as a cooperative action between at least two individuals who coordinate their individual efforts with the aim of achieving a common goal (^{Heesen et al. 2021a}, ^2021b). Reactions to interruption refer to the behavioral and communicative responses that arise when an ongoing social activity is disrupted. Such responses do not occur automatically but are mediated by communicative signals exchanged between participants, which encourage the continuation or restoration of the activity (^{Bangerter et al. 2022}).

Social activities are structurally marked by entry and exit phases. These phases are characterized by the production of communicative signals that indicate the initiation or termination of the cooperative interaction. Such signals may include mutual gaze, physical contact, and vocalizations. In particular, mutual gaze recurrently marks the beginning and end of the social activity. Similarly, the termination of the action tends not to be abrupt but is also preceded by closing signals (^{Heesen et al. 2021a}).

In bonobos, reactions to interruptions have been studied in the context of two types of social activities: social play and grooming; in the case of social play, bonobos engaged in various playful activities with both conspecifics and humans. These interactions involved an object and required both participants to perform alternating turn-taking sequences, as well as to coordinate their attention toward the task, the object, and the other agent (^{Pika and Zuberbühler 2008}). Grooming, on the other hand, is a social behavior characteristic of primates that involves cleaning a partner’s fur by removing parasites, dirt, or other impurities, and likewise requires active coordination between two individuals. Therefore, it is considered a relevant type of social activity for analyzing reactions to interruptions.

Experimental studies designed to evaluate reactions to interruptions in bonobos consisted of artificially interrupting these social activities and observing subsequent behavior. Interruptions occurred in two ways: in some cases, one of the participants voluntarily withdrew from the interaction; in others, the interruption was caused by external factors. The aim of these designs was to examine whether bonobos would resume the same activity with their original partner, whether they attempted to encourage the participation of the partner who withdrew, or whether, on the contrary, they initiated a new interaction with a third party. The results indicate that, following an interruption, bonobos tend to resume social activity with their original partner, rather than with just any individual. Moreover, they do not resume the actions arbitrarily but accurately reproduce prior roles, spatial positions, and specific behaviors (^{Heesen et al. 2021a}). For example, during grooming, they continue attending the same areas of the body as before the interruption. In general, it was observed that bonobos are more likely to resume social activities than interrupted individual activities. Additionally, bonobos who interrupt a social activity tend to produce communicative signals to restore the interaction. Rather than unilaterally abandoning the interaction, bonobos clearly mark the entry and exit phases of social activities through mutual gaze and physical contact (^{Heesen et al. 2021a}, ^2021b). Significantly, individuals encouraged their reluctant partners to resume their role in social interaction through communicative gestures (^{Pika and Zuberbühler 2008}).

5. Female Coalitions in Bonobo Societies

In this section, I argue that female coalition formation in bonobos can be interpreted as a form of social maintenance that reveals the existence of an underlying norm. In particular, these coalitions operate as a third-party punishment mechanism: their formation follows a consistent pattern in which, whenever a male exhibits aggressive behavior toward females or offspring, a collective response arises to sanction him. This suggests that these alliances are guided by a normative regularity aimed at protecting group cohesion and ensuring the well-being of its most vulnerable members.

Female coalitions consist of strategic alliances between two or more females aimed at targeting a conspecific, typically a male within the group (^{Tokuyama and Furuichi 2016}). It has been documented in both captive and wild bonobos that females frequently form coalitions, which may be dyadic or involve more than two individuals, including unrelated females (^{Surbeck et al. 2025}). These coalitions are neither random nor reactive in a merely individual sense; rather, they tend to be activated in response to prior aggressive behaviors by males toward females and offspring (^{Tokuyama and Furuichi 2016}). In a recent observation, researchers documented an adult male bonobo being repeatedly attacked by a coalition of several adult females. The authors hypothesized that the incident was likely triggered by the male’s aggression toward an offspring two days prior (^{Pashchevskaya et al. 2025}).

It has been observed that these coalitions are not limited to the intragroup context. In encounters between different groups, females may form alliances with females from other groups to collectively respond to male threats or aggression (^{Tokuyama et al. 2019}). Furthermore, coalitions exhibit a hierarchical structure, with older females tending to support younger ones. Recently, it has been proposed that female coalitions also constitute a behavioral strategy aimed at increasing female power relative to males, thereby reconfiguring the group’s hierarchical dynamics (^{Surbeck et al. 2025}).

6. Conclusion

I have argued that bonobos exhibit behaviors that can be interpreted as social norms. To do so, I have followed the definition of social norms proposed by the normative pluralism framework, according to which social norms are normative regularities (^{Westra and Andrews 2022}; ^{Westra et al. 2024}). I have pointed out that these social norms can be identified in at least three typical social behaviors found in bonobo societies: responses of inequity aversion, reactions to interruptions of social activities, and female coalition formation. In each of these, I have identified two fundamental components of normative regularities: (1) the pattern of behavioral conformity and (2) the forms of social maintenance that sustain it.

It can be concluded that the available evidence shows that bonobo societies possess implicit social norms in the form of normative regularities governing fundamental aspects of their society, such as equitable resource distribution, the completion of social activities, the protection of important members of their societies such as females and offspring, and the preservation of social structure.

Moreover, bonobo societies actively engage in the social maintenance of these normative regularities, whether through negative responses-such as protests against situations of inequity-or through positive responses, like the use of communicative gestures to encourage conspecifics to uphold their activities. All of this indicates that social norms are important to members of bonobo communities and that these norms play a significant role in their social interactions.

A potential objection to the present account is that the available evidence does not clearly establish the existence of social maintenance mechanisms in the three cases discussed, and that the behaviors in question are therefore over-interpreted as normative regularities. For example, it might be argued that evidence for inequity aversion is confined to token-exchange experiments, that bonobos do not exert social pressure to sustain joint activities in wild contexts, or that female coalitions do not function as forms of third-party punishment against male aggression. This concern can be addressed by noting that the objection rests on the assumption that a normative reading introduces explanatory commitments that go beyond the available evidence. However, the present account does not posit additional mechanisms or entities beyond those already required to describe the observed behavioral patterns and social responses. Rather, it offers a minimal normative interpretation of the same empirical regularities, one that captures their stability, their sensitivity to specific forms of deviation, and their socially mediated maintenance over time. Accordingly, it is sufficient, for the purposes of attributing a minimal and pluralistic form of normativity, to identify relatively stable behavioral patterns whose violation elicits social responses aimed at correction or restoration. On this view, experimental evidence of inequity aversion supports sensitivity to violated fairness expectations under conditions of equal effort; reactions to interruptions are evidenced by socially maintained regularities in social activities, where premature withdrawal prompts communicative repair; and female coalitions need only be systematically activated in response to specific forms of aggression, rather than functioning as punishment in all cases. The identification of implicit social norms in bonobo societies strengthens the normative pluralism framework and provides additional empirical evidence on animal normativity, complementing previous analyses in chimpanzees. This reinforces the thesis that certain behavioral regularities, sustained by social maintenance mechanisms, can be interpreted as social norms even in non-human species.

The possibility remains open that future research will identify new normative regularities in other areas of bonobo behavior. Advancing along this line will not only allow for a deeper understanding of the social dynamics of this species but also enable a more precise grasp of the structural complexity of social norms in animal communities, thereby facilitating meaningful comparisons with the normative dimension in humans.

Likewise, it would be pertinent to extend approaches inspired by normative pluralism to other species of great apes. This framework could enrich philosophical elaborations in the field of the psychology of norms by providing comparative evidence that supports discussions on the nature and origins of normativity within the primate lineage. Together, these advances would contribute to the development of more comprehensive, rigorous, and coherent theoretical frameworks aligned with an evolutionary perspective on social norms in both human and non-human communities.³