The target article by Baumard et al. uses their previous model of bargaining with outside options to explain fairness and other features of human sociality. This theory implies that fairness judgments are determined by supply and demand but humans often perceive prices (divisions of surplus) in competitive markets to be unfair.

When do humans become moral beings? This commentary draws on developmental psychology theory to expand the understanding of early moral behaviours. We argue that by looking at a broader range of other-oriented acts than what has been considered by Baumard et al., we can find support for the mutualistic approach to morality even in early instances of other-oriented behaviours.

Market models are indeed indispensable to understanding the evolution of cooperation and its emotional substrates. Unfortunately, Baumard et al. eschew market thinking in stressing the supposed invariance of moral/cooperative behavior across circumstances. To the contrary, humans display contingent morality/cooperation, and these shifts are best accounted for by market models of partner choice for mutually beneficial collaboration.

Studies of economic decision-making have revealed the existence of consistent contributors, who always make contributions to the collective good. It is difficult to understand such behavior in terms of mutualistic motives. Furthermore, consistent contributors can elicit apparently altruistic behavior from others. Therefore, although mutualistic motives are likely an important contributor to moral action, there is more to morality than mutualism.

Baumard et al. mischaracterize our model of individual and social choice behavior. We model individuals who maximize preferences given their beliefs, and subject to their informational and material constraints (Fehr & Gintis 2007). Individuals thus must make trade-offs among self-regarding, other-regarding, and character virtue goals. Two genetic predispositions are particularly crucial. The first is strong reciprocity. The second is the capacity to internalize norms through the socialization process. Our model includes the authors' model as a subset.

Mutualism provides a compelling account of the fairness intuitions on display in economic games. However, it is not yet clear how well the approach holds up as an explanation of all human morality. The theory needs to be tested outside the methodological neighborhood it was born in; such testing has the potential to greatly improve our understanding of morality in general.

Mutualistic theory explains convincingly the prevalence of fairness norms in small societies of foragers and in large contemporary democratic societies. However, it cannot explain the U-shaped curve of egalitarianism in human history. A theory based on bargaining power is able to provide a more general account and to explain mutualism as a special case. According to this approach, social norms may be more variable and malleable than Baumard et al. suggest.

We explain here how the natural selection theory of people's mutualistic sense of fairness and the biofunctional theory of human understanding are made for each other. We welcome the stage that the target article has already set for this convergence, and invite the authors to consider moving the two independently developed approaches a step closer to the natural selection level of biofunctional understanding.

Baumard et al. attribute to humans a sense of fairness. However, the properties of this sense are so underspecified that the evolutionary account offered is not well-motivated. We contrast this with the framework of Universal Moral Grammar, which has sought a descriptively adequate account of the structure of the moral domain as a precondition for understanding the evolution of morality.

Baumard and colleagues put forward a new hypothesis about the nature and evolution of fairness. In this commentary, we discuss the relation between morality and their views about fairness.

Altruism may be learned (behavioral evolution) in a way similar to that proposed in the target article for its biological evolution. Altruism (over social space) corresponds to self-control (over time). In both cases, one must learn to ignore the rewards to a particular (person or moment) and behave to maximize the rewards to a group (of people or moments).

Baumard et al. make a good case that a sense of fairness evolved and that showing this requires reciprocity games with choice of partner. However, they oversimplify both morality and the evolution of morality. Where fairness is involved in morality, other things are, too, and fairness is often not involved. In the evolution of morality, other things played a role. Plus, the motive for being fair originally was self-interest, not anything moral.

Baumard et al. argue that partner choice leads to fairness and mutualism, which then form the basis for morality. I comment that mutualism takes us only so far, and I apply the theory of competitive altruism in arguing how strategic investment in behaviours which make one a desirable partner may drive moral conduct.

We applaud Baumard et al.'s mutualistic account of morality but detect circularity in their articulation of how morality emerged. Contra the authors, we propose that mutualism might account for a sensitivity to convention (the ways things are done within a group) rather than for a sense of fairness. An ontogenetic perspective better captures the complexity of what it means to be moral.

Although mutually advantageous cooperative strategies might be an apt account of some societies, other moral systems might be needed among certain groups and contexts. In particular, in a duty-based moral system, people do not behave morally with an expectation for proportional reward, but rather, as a fulfillment of debt owed to others. In such systems, mutualistic motivations are not necessarily a key component of morality.

The target article convincingly argues that mutualistic cooperation is supported by partner choice. However, we will suggest that mutualistic cooperation is not the basis of fairness; instead, fairness is based on impartiality. In support of this view, we show that adults are willing to destroy others' resources to avoid inequality, a result predicted by impartiality but not by mutualistic cooperation.

Both evolutionary and developmental research indicate that humans are adapted to respecting property rights, independently (and possibly orthogonally) of considerations of fairness. We offer evidence from psychological experiments suggesting that enforcing one's rights and respecting others' possessions are basic cognitive mechanisms automatically activated and grounded in humans' sensory-motor system. This may entail an independent motivation that is more profound than considerations of fairness and impartiality.

Baumard et al. provide an intriguing model where morality emerges from the dynamics of partner choice in mutualistic interactions. I discuss evidence from human and nonhuman primates that supports the overall approach, but highlights a gap in explaining the human specificity of moral cognition. I suggest that an essential characteristic of human fairness is to override concerns about merit in favor of promoting the welfare in others who are needy.

Our discussion of the commentaries begins, at the evolutionary level, with issues raised by our account of the evolution of morality in terms of partner-choice mutualism. We then turn to the cognitive level and the characterization and workings of fairness. In a final section, we discuss the degree to which our fairness-based approach to morality extends to norms that are commonly considered moral even though they are distinct from fairness.