There are planters and there are weeders in the garden of science. In Moral Minds, Marc Hauser bends his back in the former capacity and it remains to be seen just how green his thumb proves to be. Hauser is interested in how humans generate moral intuitions, why we have evolved the capacity to do so, and what predictions and precautions can be drawn from research on these topics. He argues that we are all born with a "moral faculty", a set of principles enabling us to automatically evaluate actions as permissible, obligatory or forbidden. Culture sets the parameters of these inborn principles in much the same way as, according to Noam Chomsky, language acquisition operates. The moral faculty thus enables individuals to learn their local moral system and constrains the diversity of possible moral systems. Hauser favors a cultural group selectionist explanation for the evolution of the moral faculty. By generating "universal and unconscious judgments concerning justice and harm", the moral faculty plays a crucial role in sustaining in-group cooperation and stabilizing reciprocity across a range of circumstances and currencies. Hauser believes his work in Moral Minds has practical implications and that "there is an urgency to putting this material together". Our moral faculty evolved, he notes, under significantly different circumstances to those in which we currently find ourselves. He warns us to expect some mismatch between the reasoned and explicit rules codified in culture and the intuitive deliverances of the moral faculty. Understanding our moral intuitions is important, then, because "we are more likely to construct long-lasting and effective policies if we take into account the intuitive biases that guide our initial responses to the imposition of social norms". Hauser optimistically suggests we may be verging on "a renaissance in our understanding of the moral domain". Whether a renaissance is imminent is questionable but, returning to the opening metaphor, we might at least look forward to a good growing season.

Much of Moral Minds is devoted to empirical work on the relationship between reason, emotion and morality. Hauser usefully distinguishes between how we behave morally and how we make moral judgments (between "moral performance" and "moral competence") and chooses to focus on the latter. In Chapter 1, we are introduced to three "creatures" — Kantian, Humean and Rawlsian — representing three ways in which reason, emotion and moral judgment may inter-relate. Hauser aims to discover which breed of creature we are, or, should we turn out to be hybrids, which mix and in what proportions.

Kantian creatures "deliver moral judgments based on conscious reasoning from relevant principles". Emotions, for this creature, are obstacles to be avoided in moral decision-making, a feature reminiscent of Kant's distinction between acting 'from' duty versus merely 'in accord with' duty. Hauser does not make particularly clear which principles are relevant for the Kantian creature. He quotes Kant's "universalizability" and "humanity" formulations of the categorical imperative, but quite deliberately does not go into the nuances of Kantian moral theory. This lack of clarity is no great impediment to Hauser's project, however, for which the connection between moral philosophy and moral psychology is of greater relevance. Developmental psychologists in the tradition of Piaget and Kohlberg have worked with a reason-based view of morality, which has lead, Hauser claims, to the current dominance in moral psychology of a "conscious moral reasoning" perspective. Hauser explicitly opposes that perspective.

Some philosophers may well wish to take issue with Hauser over his characterization of the Kantian moral creature. It is worth bearing in mind, though, that the crucial feature of the Kantian creature is that it reasons from principles to moral judgments and not that it reasons from specifically Kantian principles. Non-Kantian principles could be plugged into the 'Kantian' creature. Hauser intends only to provide a convenient label for one model of how we might be as moral judgers.

Hauser thinks it clear that we are not pure-bred Kantian creatures. Conscious moral reasoning from explicit principles is only sometimes the cause of our moral judgments. Hauser describes the phenomenon of moral "dumbfounding", in which subjects are unable to adequately justify their initial moral judgment of a given situation but staunchly maintain that judgment nonetheless, often appealing to "hunches". Dumbfounding effects have been uncovered by Jonathan Haidt and others, including Hauser himself with his online Moral Sense Test. For several years now, Hauser's research lab has maintained an online survey site that allows him to extensively sample folk moral judgments, examine the justifications subjects offer for their judgments, and determine the coherence of subjects' response patterns. The MST reveals that, when it comes to justifying moral judgments, the majority of subjects are "clueless". The phenomenon of moral dumbfounding shows, Hauser says, that our impression that we reason from principles to judgments can be "illusory". Conscious moral reasoning, he stresses, may merely provide post hoc justification for our moral judgments and have nothing to do with their genesis.

Reason, for the "Humean creature", causes neither moral judgments nor moral behaviour. It serves only to compare various means to the creature's ends, recalling Hume's famous dictum that reason is and ought only be slave to the passions. For the Humean creature, "emotions ignite moral judgments [and] reason follows in the wake of this dynamic". Humean creatures develop, in the course of normal growth, a suite of emotions which, when triggered, cause moral judgments.

            How much of himself Hume would see in Hauser's Humean creature is questionable but not the important issue. As with the Kantian creature, Hauser intends only to provide a convenient label for one way in which reason, emotion and moral judgment may be related.

            Hauser points out a shortcoming in the Humean model, namely the lack of an account of "the evaluative process that triggers emotion", but he recognises that we do sometimes act like Humean creatures. Especially when confronted with harmful or disgusting actions, "we have the feeling that we [make moral judgments] quickly, unconsciously and without any apparent reflection upon explicit laws". This observation is not, however, strong support for the Humean creature, since Hauser here misses the possibility that making a moral judgment might feel like entering an emotional state without the moral judgment being caused by an emotional state. Automaticity does not imply emotional causation. In any case, Hauser thinks the predictions of the Humean model are not borne out by psychological research. If we were pure-bred Humean creatures, then damage to emotion-processing regions of the brain would knock out our capacity to make moral judgments; "a Humean creature needs his emotions to make moral decisions". Hauser interprets studies by Antonio Damasio and Shaun Nichols as strongly suggesting that emotionally-impaired subjects may have "normal moral competence but abnormal moral performance". If emotional impairments affect moral behaviour while leaving intact the capacity to make moral judgments then, Hauser argues, we cannot be twin to the Humean creature.

Hauser believes neither reason, nor emotion, nor both in concert, can fully explain the genesis of our moral judgments. Enter the Rawlsian creature, so named in homage to Rawls' suggestion in A Theory of Justice that "a correct account of moral capacities will certainly involve principles and theoretical constructions which go beyond the norms and standards cited in every day life". Rawls also suggested that "there may be deep similarities between language and morality, including especially our innate competences for these two domains".  The Rawlsian creature is equipped with what Hauser calls a "grammar of action", a set of principles for morally assessing actions in terms of their causes and consequences. The Rawlsian creature takes actions as inputs and gives moral judgments as outputs: permissible, obligatory, or forbidden. Actions are distinguished according to whether they are intentional or accidental, acts proper or omissions, and whether their consequences are harmful or beneficial, intended or merely foreseen. Emotions are downstream of this unconsciously-operative action-evaluation machinery. The Rawlsian creature's emotions are caused by its moral judgments and motivate its moral behaviour. Finally, this creature might engage in conscious reasoning to justify or somehow modify the judgments output by its moral faculty.

Although Kant or Hume scholars amongst Hauser's readership are best advised to look past the characterization of their respective creatures and let Hauser get on with his project, the same advice does not apply to those interested in Rawls, or at least not as strongly. Hauser believes both the Rawlsian creature and the real Rawls have been neglected; "many current discussions of the evolution of morality, and fairness in particular, either ignore Rawls or misinterpret him" and in moral psychology "there has been no serious engagement with the Rawlsian creature". Moral Minds is an attempt to redress this neglect by developing Rawls' analogy between language and morality and giving the Rawlsian creature some long-overdue attention.

Hauser has gathered evidence suggesting unconscious principles do in fact guide some of our moral intuitions. He devotes Part I of Moral Minds ("Universal Declarations") to presenting this evidence. Hauser's above-mentioned Moral Sense Test features prominently in Chapter 3 ("Grammars of Violence"), where he investigates the principles underlying intuitive judgments of permissible harm. The test introduces us to a cast of characters each caught up in a variant of the classic trolley problem. Hauser's cleverly constructed "moral dilemmas" prompt a pattern of intuitions consistent with the unconscious operation of a "double effect" principle guiding our judgments of permissible harm; "it is impermissible to cause an intended harm if that harm is used as a means to a greater good [but] it is permissible to cause harm if that harm is only a foreseen consequence of intending to cause a greater good". What matters to our intuitive judgments is not whether harms are "personal" or "impersonal" and thus differentially engage our emotions but instead whether harms are caused intentionally or merely foreseen. This double-effect principle operates unconsciously, Hauser says, because most subjects mention no such thing in their reasoning. Thus, Hauser garners significant support for the Rawlsian creature over its Kantian and Humean competitors.

Hauser also searches, in Chapter 2 ("Justice for All"), for principles underlying intuitive judgments of fairness. Appealing to experimental economics, he argues that "fairness is a universal principle with the potential for parametric variation". Cultures set fairness parameters differently due to differences in social organization and local ecology. For example, take the ultimatum game. In this game, one player (the proposer) begins with a sum of money and offers the other player (the respondent) either some proportion of that sum or nothing at all. If the proposer offers something, then the respondent either accepts the proposed proportion and the proposer keeps the remainder, or the respondent rejects the proposal and both players end empty-handed. When members of slash-and-burn horticulturalist cultures play this game, proposers offer low and respondents reject rarely, reflecting, Hauser suggests, the relatively minimal role cooperation plays in their subsistence style. Individuals from cultures in which competitive gift-giving is prevalent, by contrast, propose fairly high offers and as repondents routinely reject. The principle of fairness, Hauser thinks, "has parameters concerning the responsible agent, the original source of the resources, the dependency on others for acquiring the resources, and the option of rejecting an offer". It would be interesting to see if future work could uncover a "dumbfounding" effect in ultimatum game players like that shown by Moral Sense Test respondents. The principles guiding moral intuitions are, after all, supposed to operate unconsciously on the Rawlsian model of moral judgment.

The empirical studies of moral intuitions about violence and justice that Hauser surveys are certainly intriguing and provide ample motivation to further pursue the hypothesis that humans are most closely akin to the Rawlsian creature.

Hauser is especially keen in Moral Minds to specify the design of the moral faculty and the course of its development. He attempts this in Part II ("Native Sense"), and most directly in a chapter entitled "The Moral Organ". That title is somewhat misleading, the chapter being largely taken up with enumerating members of that faculty's "support team". Many capacities that are not exclusive to the moral faculty are nevertheless required for its operation. In particular, action perception and mindreading are vital if the Rawlsian creature is to assess actions in terms of causes and consequences. Hauser surveys work in infant and child psychology on "the ABCs of Action", describing how we unconsciously analyse events into discrete actions and interpret those actions by attributing beliefs and intentions to agents and objects. The support team roll call continues: self-awareness, the ability to share attention and engage in pretence and play, to delay gratification and to inhibit impulses more generally, to experience empathy and sympathy. Even innate numeracy skills get a mention. Keeping in mind Hauser's own distinction between moral competence and performance, one could wish for greater clarity here regarding which team members are supposedly necessary for moral judgment and which 'merely' support moral behavior. More importantly, one might wonder where, amidst all this, the moral organ itself lies.

Hauser posits "a circuit specialized for recognizing certain problems as morally relevant and others as irrelevant" but admits that no current studies "pinpoint a uniquely dedicated moral organ". He cites advances in brain imaging technology as just one of the things necessary before such pinpointing is even a possibility. A case of premature publication? Not necessarily. Moral malfunctions may indirectly illuminate the design of the moral organ. For this reason, Hauser takes a walk over the well-trodden testing ground that psychopaths provide for theories of morality.

Hauser holds that psychopaths are "perfectly lucid about their actions". Hence, he thinks, the Kantian model cannot account for the moral malfunction evident in psychopathy. If humans were Kantian creatures, then psychopaths, their reason being intact, would not display moral malfunctioning. One might well think Hauser moves too quickly here. He assumes that engaging in justificatory behaviour is sufficient to qualify psychopaths as rational in the important Kantian-creature sense of "reasoning from relevant principles". This assumption is questionable. Indeed, whether psychopaths suffer rational defects is an ongoing debate amongst pyschologists and philosophers. Hauser himself even reports that psychopaths are often glib and superficial and their reasoned justifications betray "an unparalleled egocentrism". At the very least, it seems, some relevant principles would need to be specified before psychopaths' lucidity could rule out the Kantian model of moral judgment as Hauser himself described that model.

Hauser takes the prevailing view of psychopathy to be that an emotional impairment prevents psychopaths properly distinguishing moral from conventional transgressions thus making them more likely to morally transgress. Given the prominence of emotions in the prevailing view, the Humean model looks in good shape. Hauser criticises the prevailing view, however. He thinks (along with Nichols) that the idea that what distinguishes moral from conventional transgressions is heavier affective loading cannot account for actions being heavily negative-affect-laden but not forbidden or even morally evaluable at all; explosive diarrhea, say, or projectile vomiting. Emotions, Hauser tells us, "cannot do all the heavy lifting when it comes to deciding between conventional and moral events". The Humean model thus also has trouble accounting for the psychopaths' moral malfunction; it cannot be just an emotional impairment that leads psychopaths to misjudge and misbehave. In the course of this discussion, though, Hauser allows that "at least some of our moral judgments — perhaps only those handling norms against harm and disgust — may emerge from our emotions". He thus assigns the emotional Humean creature a quite significant role in generating moral judgments, which he set out to argue was done by an unconscious Rawlsian grammar of action.

Psychopathy, Hauser thinks, is problematic for the Kantian and Humean models but provides a good test case for the Rawlsian creature. He says psychopaths "appear to deliver normal moral judgments" and suggests that their action-evaluation machinery (their "genetically endowed moral competence") may be intact even though their moral performance is impaired by an emotional deficit that leaves them unmotivated to act. Hauser's position here is somewhat confusing. Throughout Moral Minds, he takes moral judgments as the measure of moral competence. He accepts the prevailing view of psychopathy insofar as he agrees that psychopaths fail to properly distinguish moral from conventional transgressions. He thus seems committed to denying that psychopaths make normal moral judgments; they do not employ a distinction even small children understand. It is unclear in what sense Hauser supposes the psychopath's moral competence to be intact and unclear, as a result, that the Rawlsian creature fares better than its competitors.

Hauser supports one of the key claims in Moral Minds, that we have a dedicated moral organ, with an argument that contains some contentious premises, some dubious moves, and doesn't convincingly eliminate the Rawlsian's competitors. Hauser's case is likely to be especially unsatisfying to any who doubt his three-way moral bestiary was complete to begin with. The charge of premature publication may, again, seem appealing. Hauser can be excused, though, on grounds of understandable excitement. The findings in support of the Rawlsian creature reported in Part I are certainly enough to motivate his, and future, work on pinpointing dedicated action-evaluation machinery.

Hauser's extensive roll call of moral faculty support team members might disappoint those hoping to see the moral faculty spotlighted. It does, however, allow him to give some specificity and substance to the idea that the building blocks of human morality are evolutionarily ancient. In Moral Minds Part III ("Evolving Code"), he asks which parts of the moral faculty (broadly construed to include its support team) evolved prior to the emergence of our own species. He surveys an impressive range of research, reporting studies on birds, primates, elephants, dolphins, rats and bats and more. Hauser contends that many members of our moral faculty's support team are shared with non-human animals. He presents evidence that some animals also know our "ABCs of action", along with the case for mindreading and self-awareness in nonhumans. Key building blocks of morality are, in this sense, evolutionarily ancient.

Hauser makes a few eyebrow-raising uses of moral language when discussing nonhuman "proto-morality". His casual reference to the "evolved rights" of mink, for one thing, should jar readers. The issue is not the superficial one of incautious expression but rather a deep and genuine uncertainty, in which Hauser is not alone, about how moral language ought to be applied to animals. Hauser adventurously muses on the possibility of full-blown nonhuman morality but, recognizing the formidable difficulty of testing for it as opposed to "merely socially coordinated behaviors", is content to conclude that "such questions exist and are worth addressing". Indeed.

Hauser's cautious "interim report" is that nonhuman animals have only some of the precursors of human morality. They also, Hauser concludes after surveying yet another interesting array of studies, display only a fraction of the cooperativeness observed in humans. Reciprocal interactions in nonhumans are, at best, restricted to "a single commodity, within a single context, and the timespan for exchange is remarkably short". Hauser's review of the animal cooperation literature is at times very entertaining; any under-appreciated IT support workers amongst his readership can look forward to some solace from his story of social hierarchy, dominance, aggression and a popcorn machine amongst macaques. Nonhuman animals do of course cooperate in many ways. However, Hauser says, humans seemingly uniquely engage in large-scale cooperation among unrelated individuals. We are also seemingly unique in maintaining stable reciprocal relationships across a variety of circumstances and currencies. The mismatch between human and nonhuman "building block" sets, along with the mismatch in cooperative capacity, suggests to Hauser that the evolutionary function of our moral faculty is to allow large-scale cooperation and generalized reciprocity.

Hauser apparently wants to dovetail his view with the cultural group-selectionism of Robert Boyd and Peter Richerson. The focus of Moral Minds is predominantly on the development of the moral faculty, though, and Hauser's comments on its evolution go by fairly quickly. It is simply too early to assess whether Hauser plus Boyd and Richerson equals "a solution to the paradox of human cooperation". It is, however, worth pointing out that Hauser's account of the evolution of the moral faculty need not be tied to the theory of cultural group selection. Selection for reciprocity, in particular, can occur at many biological levels. Hauser's clearly thinks, though, that there are interesting links between his work and that of Boyd and Richerson. Developing the ideas in this section of Moral Minds would be a worthwhile future project.

Moral Minds ends with a discussion of issues Hauser thinks likely to arise out of his work so far. He quite sensibly rejects three "classically held beliefs" about the relationship of biology and morality. A biological explanation for moral judgment need not imply, and in Hauser's own case certainly doesn't imply, "predetermined outcomes, thereby eliminating free will". Nor need it claim that moral principles are somehow "encoded in the DNA". Those two classically held beliefs rest on (frustratingly persistent) misunderstandings of the relevant biology. Finally, biological perspectives on morality should not be opposed on the grounds that "only religious faith and legal guidelines can prevent moral decay". Hauser's brisk dismissal of this idea is akin to Richard Dawkins' in The God Delusion (in fact, these two books could be illuminatingly read together). There are, however, more vexing issues that remain live despite Hauser's confident assurances.

Hauser insists throughout Moral Minds that his project is descriptive; he is not making prescriptions for human behavior. He also insists, though, that his work on the moral faculty will help us "navigate between descriptive and prescriptive principles". This would be simple sense, and unsensational, if all he meant was that some reasoned prescriptions will be more at odds with the intuitive deliverances of our moral faculty than others, and that we are better off knowing of such obstacles than not. Hauser has something stronger in mind, though. Most books in this genre make early mention of the Naturalistic Fallacy (in some guise) and Moral Minds is no exception. "We are not entitled to move from the natural to the good", Hauser tells us; "the equation of good with natural [is] fallacious". He nevertheless thinks an understanding of our moral grammar "bears on our approach to the prescriptive principles of what ought to be". Once we uncover the moral faculty's principles, he says, "we may use these principles to guide how we consciously reason about morally permissible actions". It is certainly tempting to read Hauser here as saying that laying out our grammar of action could give us insight into what is 'really' right and wrong, our intuitive biases aside.

It is confusing just what kind of connection Hauser is trying to establish  between descriptive and prescriptive moral projects. He is perhaps best understood as recommending that we reason pragmatically when constructing our normative theories, eyes open to the difficulties of implementation our intuitive biases will create. How reason-responsive our moral judgments are, though, is left distressingly unclear at the end of Moral Minds. The Rawlsian, Kantian and Humean creatures all receive some support from Hauser's work. In his view, we are a mix of all three. Our initial moral judgments are generated by unconscious principles, we can reason in favor or against these judgments, and emotions can promote or hinder our acting in line with our moral judgments. Hauser's work could be useful as moral 'navigational aid', however, only if the Kantian creature retains a significant role even after the Rawlsian creature has taken up its share of the psychological stage. How many lines the Kantian is left with (or if it even has a speaking part at all) will likely be a lively debate taken up in the wake of Moral Minds.

Further, there is a difference between reason changing our moral intuitions and reason influencing our moral behavior. It seems that a consequence of Hauser being right is that certain intuitive moral judgments could persist even while we rationally reject them and behave accordingly. In that case, Moral Minds implies that some rationally desirable normative policies may come only at the cost of 'moral dissonance', so to speak. Of course, if Hauser's right, whining about it won't help. Still, we should recognise that the connections between descriptive and prescriptive moral projects are not as clear or constructive as is optimistically suggested.

Finally, Hauser considers the Rawlsian model of moral judgment attractive insofar as it seems amenable to "a pluralistic position…that recognizes different moral systems and sees adherence to a single system as oppressive". It would be unfair to make too much of Hauser's closing remarks, but one may struggle to see how being "as perplexed by another community's moral system as we are by their language" could be a good thing. As repeatedly stressed, however, drawing out the implications of Hauser's work will be an interesting and worthwhile Endeavour.

With Moral Minds, Marc Hauser has sown many seeds. It remains to be seen which are watered, which wither, and which grow wild.

 

© 2007 Ben Fraser

 

 

Ben Fraser, Ph.D candidate, Research School of Social Sciences, The Australian National University, Canberra, ACT, Australia

Categories: Ethics, Psychology