by Dan Sperber
So, to answer the question posed in the subtitle of Cosmides’s article and in the title of this section, we agree that natural selection has shaped how humans draw all kinds of inferences and has produced a wide variety of specialized inferential modules. One of these, we add, is a reason module. The justifications and the arguments that the reason module produces may have, embedded in them, conclusions relevant to all domains of knowledge and action. This virtual domain-generality does not, however, make reason, and the organization of human mind generally, any less modular.
What functions does the reason module fulfill? We have rejected the intellectualist view that reason evolved to help individuals draw better inferences, acquire greater knowledge, and make better decisions. We favor an interactionist approach to reason. Reason, we will argue, evolved as a response to problems encountered in social interaction rather than in solitary thinking. Reason fulfills two main functions. One function helps solve a major problem of coordination by producing justifications. The other function helps solve a major problem of communication by producing arguments. (Our earlier work has been focused on reasoning and on developing, within the interactionist approach, an “argumentative theory of reasoning.”)9
The Challenge of Coordination and the Justificatory Function of Reason
Human cooperation is exceptional not only by its scale but also by the open-ended variety of the forms it takes. Other animals may have a few types of cooperative interactions in their behavioral repertoire with little or no place for creative improvisation. What each cooperator may expect of the others is largely predetermined. When these expectations are not met because of the incompetence or the defection of one of the cooperators, cooperation is likely to fail. Humans, on the other hand, rely less on predefined expectations. Existing forms of cooperation are deployed with great flexibility and often readjusted on the fly. New forms are often tested. This flexibility and creativity of human cooperation can be highly advantageous, but only if massive cognitive resources are invested to secure effective coordination.
To achieve the degree of fine-grained coordination that their multiple forms of cooperation require, humans need mutual expectations that have to be constantly updated to remain reliable. Members of a party of warriors, or of a sports team, for instance, readjust or even redefine their tactics and their mutual roles on each occasion. They must, moreover, be able to rely on each other not only when things go as anticipated but also when they don’t. So must, in different ways, friends, spouses, coworkers, and business partners.
Even among competent cooperators, there may well be differences in the understanding of the common goals and of the part each individual must play. Differences of interests, instead of leading to defection or cheating, are often recognized and handled on the basis of mutual commitment to what is seen as fair; here too, however, there may be differences of interpretation leading to failures of mutual expectations.
For humans, knowing what to expect of each other is a crucial cognitive challenge. How is this challenge met? How do humans succeed in forming, if not perfect, at least adequate mutual expectations? The most common answer consists in invoking two mechanisms: norms at the sociological level, and understanding of the mental states of others at the psychological level.
At the level of the social group, there are shared norms of various kinds: moral, legal, religious, prudential, technical, and so on. They may be explicitly codified or not and enforced with sanctions or not. These norms regulate a great variety of social interactions.10
Some norms aim directly at securing coordination in a way that is beneficial to all the people involved. Without traffic rules, for example, driving a car would be absurdly risky. Very precise rules of coordination, however, tend to be highly specialized. The kind of interaction traffic rules regulate, for instance, is unlike any other; at every moment, drivers have a relatively narrow, well-defined range of options. If they don’t coordinate their decisions, their lives are at risk. Effective coordination is in the interest of all of them.
Legal, moral, or religious norms, on the other hand, contribute to all kinds of social coordination in a great many ways, but securing effective coordination is not their sole purpose and needn’t be their main one. These norms leave much room for interpretation and many relevant issues untouched. In some milieus, for instance, not arriving late at a dinner party may be a faux pas. People, moreover, don’t at all times have the same interest in these various norms’ being obeyed. Elected officials coming up for reelection, for instance, overtly encourage all their constituents to follow the norm and vote while making it harder to do so for those less likely to reelect them.
Do the many and diverse norms found in one’s society constrain people’s options to the point of indicating what people can expect from each other in the fine-grained manner needed for effective coordination? Of course not! There are, for instance, many norms—legal, moral, religious—telling spouses what they can expect of each other. Still, to achieve the level of mutual expectations that daily life coordination requires, spouses need to understand much more about each other than the fact that they are supposed to abide by a socially sanctioned set of norms.
In many forms of social interaction, some degree of creativity and improvisation is socially condoned or even required. You throw a party, for example. Whom should you invite? Conventional or moral norms might suggest that you invite people who have invited you in the past, so perhaps you should invite Olga, who invited you to her party last month. For other possible guests, norms of reciprocity won’t help at all, but there are still issues of coordination to be addressed. If you invite Diego, for instance, you had better not invite Ruth. If you invite Ruth, she will want you to invite Chao.
Your future interactions with others will depend on many small, interconnected decisions you make. So what should guide you in solving coordination problems that norms leave unanswered? Here the standard answer is: mindreading. You have to understand the states of mind of Olga, Diego, Ruth, and other possible guests in order to anticipate their reactions and secure the conditions for a successful party that will enhance rather than compromise your future relationships with others.
Just like norms, mindreading plays an essential role in coordination. Still, the picture is far from complete. What is blatantly missing is the fact that individuals don’t just infer what they can expect from each other on the basis of what they know of other people’s minds and of the norms they share. People’s mutual expectations are reviewed, discussed, negotiated in detail. Many decisions on how to interact are themselves taken interactively.
Gossip provides rich evidence of what can be expected of third parties. Individuals, however, need not be passive objects of gossip. They can participate in the ongoing conversation about themselves, explain and justify their views and their decisions, and, in so doing, to some extent, safeguard their own reputation. Officials trying to suppress the vote of constituents unlikely to reelect them justify their actions by saying they are just trying to prevent voting fraud. Guests arriving too late at a party justify themselves by explaining how they were delayed. You can tell Diego that, alas, you had to invite Ruth because she had invited you before; you can tell Ruth that, of course, she can come with Chao because her friends are your friends; and so on.
By giving reasons to explain and justify yourself, you do several things. You influence the way people read your mind, judge your behavior, and speak of you. You commit yourself by implicitly acknowledging the normative force of the reasons you invoke: you encourage others to expect your future behavior to be guided by similar reasons (and to hold you accountable if it is not). You also indicate that you are likely to evaluate the behavior of others by reasons similar to those you invoke to justify yourself. Finally, you engage in a conversation where others may accept your justifications, question them, and invoke reasons of their own, a conversation that should help you coordinate with them and from which shared norms actually may progressively emerge.<
br />
Reducing the mechanisms of social coordination to norm abiding, mindreading, or a combination of these two mechanisms misses how much of human interaction aims at justifying oneself, evaluating the reasons of others (either those they give or those we attribute to them), criticizing past or current interactions, and anticipating future ones. In these interactions about interactions, reasons are central.11
Justificatory reasons, in fact, bridge the gap between norms and mindreading. When we justify ourselves, we present our motivations as normatively apt, and we present norms as having motivating force. In other terms, we psychologize norms and “normalize” mental states. In doing so, our goal is not to give an objective sociological or psychological account of our actions and interactions; it is to achieve beneficial coordination by protecting and enhancing our reputation and influencing the reputation of others.
The role of reasons in social coordination has often been highlighted in philosophy, psychology, and the social sciences. The dominant view, however, is that attributing reasons is the most elaborate form of mindreading. We would argue that as far as mindreading goes, the attribution of reasons is typically misleading. The causal role it gives to reasons is largely fictitious; the reasons people attribute to themselves or to others are chosen less for their psychological accuracy than for their high or low value as justifications. The explanatory use of reasons, we suggest, is in the service of its justificatory use: it links reasons to persons so that good reasons are seen as justifying not just a thought or an action but also the thinker of that thought, the agent of that action.
The ability to produce and evaluate reasons has not evolved in order to improve psychological insight but as a tool for defending or criticizing thoughts and actions, for expressing commitments, and for creating mutual expectations. The main function of attributing reasons is to justify oneself and to evaluate the justifications of others.
The Challenge of Communication
Even more than human cooperation, human communication stands apart by its scale, its diversity, and the complexity of its mechanisms. Other animals use a small repertoire of signals to convey a few basic messages about the here-and-now: warnings about the presence of a predator, threats to competitors, and mating calls, for example. Humans have languages with huge vocabularies and powerful syntax with which they can produce an unbounded variety of linguistic utterances. With the help of language, they are able to communicate simple or complex ideas about any conceivable topic, about events distant in space and time, about general facts, or about abstractions, topics that are absent from animal communication and that are at the center of human cultural knowledge.
Humans, who have by far the richest codes, use them to communicate even more than what they manage to encode. As we have already noted, when speaking, people do not fully encode what they mean. What they do is give partly encoded evidence of their meaning. Hearers infer the speaker’s intended meaning from this linguistic evidence taken together with contextual information. When, for instance, Azra tells Marco, “This case is too hard,” he understands her to mean that the legal case on which he had chosen to write an essay for the tort law class is too difficult and that he should select an easier case to write about. In so understanding Azra, Marco goes well beyond decoding the sentence she uttered—a sentence that in other contexts could be used to convey utterly different meanings—but he does not go beyond the ordinary pragmatic procedures of comprehension we all use all the time.
Securing comprehension is, however, only half of the goal of communication. A speaker typically wants not only to be understood but also to be believed (or obeyed), to have, in other terms, some influence on her audience. A hearer typically wants not just to understand what the speaker means but, in so doing, to learn something about the world. This occurs when the hearer not only understands what he is told but also accepts it. Azra’s intention in telling Marco “This case is too hard” was to have her implicit advice accepted. His goal in paying attention to what she was saying was to gain some guidance from it. Still, Marco may have understood what Azra was telling him and disagreed. In human communication, comprehension does not automatically secure acceptance. (Do you, for instance automatically believe everything you understand when reading a book like this one? Of course not!)
Without acceptance of the information communicated, communication wouldn’t be beneficial to either communicators or addressees. If it were not beneficial to both, communication could not evolve. Would, then, a disposition in addressees to automatically accept communicated information ensure optimal communication? Far from it. Imagine what communication would be like if people automatically believed or did what they were told. Communicators might be quite satisfied; their ability to influence gullible and docile listeners would be without limit. Addressees, on the other hand, accepting everything they were told, would be prey to all kinds of misinformation and manipulation, not a satisfactory condition at all.
Automatic trust would make more sense only if there were a corresponding disposition in communicators to be automatically trustworthy. The eighteenth-century Scottish philosopher Thomas Reid affirmed that God had equipped humans with such a pair of dispositions:
The wise and beneficent Author of nature, who intended that we should be social creatures, and that we should receive the greatest and most important part of our knowledge by the information of others, hath, for these purposes implanted in our natures two principles that tally with each other. The first of these principles is a propensity to speak truth …. [The second principle] is a disposition to confide in the veracity of others, and to believe what they tell us.12
An omnipotent creator might have decided to make humans both trusting and truthful. Natural selection is much less likely to produce and maintain such perfect harmony. It will, if possible, favor communicators who do not hesitate to deceive trusting receivers when it is in their interest to do so. Likewise, it will favor receivers who are not easily deceived.
Communication can still evolve under natural selection when the biological interests of emitters and receivers are, if not identical, at least sufficiently aligned, such as in the case of honeybees (which in the case of sterile workers contribute to the propagation of their genes by all helping the queen reproduce). Communication can also evolve when it is limited to topics of common interest where the communicator wouldn’t gain by deceiving its audience or the audience by distrusting the signal, such as when a female baboon signals to potential mates, by means of a large swelling of her posterior, that she is ready to mate.
Human communication, however, is definitely not limited to topics of common interest where truthfulness and trust are mutually advantageous to the interlocutors. Linguistic signals can be produced at will to inform or to mislead. Unlike the honeybee’s waggle dance or the female baboon’s sexual swelling, linguistic signals are not intrinsically reliable. Human communication takes place not only among close kin or cooperators but also with competitors and strangers. Lying and deception are in everybody’s repertoire. Even children start expecting and practicing some degree of deception around the age of four.13
Humans stand to gain immensely from the communication of others. Without communicating a lot, they couldn’t even become competent adults or lead a normal human life. As a result of their dependence on communication, humans incur a commensurate risk of being deceived and manipulated. How, notwithstanding this risk, could communication have evolved into such a central, indispensable aspect of human life?
Communication is a special form of cooperation. The evolution of cooperation in general poses, as we saw in Chapter 9, well-known problems. It might seem reasonable to expect that theories that explain the evolution of human cooperation might also explain the evolution of communication.14 Do they really?
Actually, communication is a very special case. In most standard forms of cooperation, cheating may be advantageous provided one can get away with it. For any given individual, doing fewer house chores, loafing
at work, or cheating on taxes, for example, may, if undetected, lower the costs of cooperation without compromising its benefits. This being so, cooperation can evolve and endure only in certain conditions—in particular, when organized surveillance and sanctions make cheating, on average, costly rather than profitable, or when the flow of information in society is such that cheaters put at risk their reputation and future opportunities of cooperation. For communication to evolve, however, the conditions that must be fulfilled differ considerably from this.
Although this is hard to measure and to test, there are reasons to think that communicators are spontaneously honest much of the time, even when they could easily get away with some degree of dishonesty. Why? Because when humans communicate, doing so honestly is, quite commonly, useful or even necessary to achieve their goal. People communicate to coordinate with others, to organize joint action, to ask help from friends who are happy to give it. In such situations, deceiving one’s interlocutors, even if undetected, wouldn’t be beneficial; it would just be dumb.
Still, in other situations that are also rather common, a degree of deception may be advantageous. People generally try to give a good image of themselves. For this, even people who consider themselves quite honest routinely exaggerate their virtues and achievements, use disingenuous excuses for their failings, make promises they are not sure to keep, and so on. Couples who most of the time may be cooperative and honest with one another lie about their infidelities. Then there are cases where candid communication is generally seen as just incompetent—these include politics, advertisement, propaganda, and some businesses where honesty results in too many missed opportunities. Being lied to may be less frequent than being told the truth, but when it happens, it can be quite costly.