Virtue Signaling

by Geoffrey Miller

  Young-adult peak in trait expression. For sexually selected behavioral traits, conspicuous virtue-displays should peak in young adulthood, at the peak of mating effort. They should be low before puberty, should increase rapidly thereafter, and should decline gradually as individuals shift their time and energy from courtship to parenting.

  Alternative mating strategies. Individuals lacking the sexually-attractive virtues should more often pursue ‘alternative mating strategies’ that try to circumvent mate choice by the opposite sex and mate-guarding by same-sex rivals, just as animals with lower-quality ornaments tend to do. This may include increased use of short-term opportunistic mating, deceptive affairs, sexual harassment, sexual stalking, and/or sexual coercion. If so, we might predict vicious cycle effects in which initially virtue-deficit individuals fail in the mating market, and adopt increasingly desperate and exploitative alternative mating strategies, which further undermine their virtues. This is not a surprising prediction for behavioral ecologists, but it offers a view of sexual aggression quite different from the dominant feminist model in the social sciences (‘rape as a crime of patriarchal aggression, not sex’), and it may lead to more effective anti-rape interventions.

  Mate Choice for Moral Virtues:

  Mate preferences. All else being equal, virtues should be favored in mate choice. They should be highly valued aspects of potential mates that individuals are motivated not just to judge passively by observation, but to probe actively, by arranging various socio-sexual situations that test the specific virtues of potential mates. (Such virtue-tests are a central plot device in sit-coms and romantic comedies.) For example, a potential mate’s honesty and fidelity may be tested by arranging for a same-sex friend to flirt with them and report the outcome. Further, virtues as good genes indicators should be favored more often in short-term relationships and extra-pair copulations, and by women at peak fertility near ovulation. Virtues as good parent/partner indicators should be favored more often by individuals seeking long-term relationships, by individuals who already have children from previous relationships, and by women at lower fertility phases of the ovulatory cycle.

  Positive assortative mating. In species with social monogamy such as ours, individuals should assortatively mate with respect to virtues, because the competitive mating market should ensure that high-virtue individuals prefer each other, leaving lower-virtue individuals no choice but to settle for each other.

  Sexual derogation and gossip about trait values. If virtues are valued in courtship, same-sex rivals should selectively derogate each other with respect to virtue deficits such as lying and cheating. Also, in social species such as ours with collective mate-choice that takes into account the views of family and friends, gossip about potential mates should focus considerable attention on virtues as fitness indicators. Virtue should be praised and vices condemned, especially when people discuss potential mates for their kin and friends. This may sound tautological – what are virtues but socially-praised traits? – but the mate choice model can lead to more specific predictions about sex differences and relationship context effects on the sexual derogation of virtues. For example, men seeking a short-term copulation value female promiscuity as a virtue (probably under a morally palatable euphemism such as ‘fun,’ ‘liberal,’ or ‘adventurous’), whereas men seeking a long-term relationship value female chastity as a virtue. Thus, a clever female will derogate or gossip about a short-term sexual rival as a ‘frigid conservative,’ or a long-term marriage rival as ‘an insatiable nympho.’ Thus, the sexual attractiveness and moral valence of specific rival traits should reverse under predictable conditions.

  Example: Sexual Fidelity as a Moral Virtue

  For example, suppose a researcher hypothesizes that sexual fidelity evolved by sexual selection through mutual mate choice (rather than through kin selection, reciprocal altruism, or group selection). Fidelity might minimize the spread of sexually-transmissible pathogens, the risk of cuckoldry (a male investing in offspring that were sired by another male), and the costs of polygyny (a female losing investment in her own children if a male sires children by another female). How could this hypothesis be tested? It might be easiest to go in reverse order of the 12 criteria listed above: start with mate choice for the virtue, then phenotypic features of the virtue, then genetic bases of the virtue.

  A first step would be to investigate mate choice for fidelity. Do surveys, interviews, and experiments show that people prefer sexually faithful mates, all else being equal? Yes: although males are attracted to promiscuous females as potential short-term mates, neither sex respects high levels of promiscuity in potential long-term mates. Also, jealousy research shows that men and women across cultures react very negatively to sexual infidelity, yet are highly motivated to discover it. Do people verbally derogate their sexual rivals for being unfaithful? Yes: moral derogation of sexual rivals (using morally charged terms such as ‘skank,’ ‘slut,’ or ‘sleazeball’) is a common mating tactic. Do courting people often display their likely future fidelity to potential mates? Yes: lovers typically make impassioned, adaptively self-deceptive declarations of infinite, eternal, exclusive love.

  If the moral trait shows most of these adaptive mating-related features, then the researcher might progress to phenotypic studies of sexual fidelity as an individual-differences dimension. Are there stable individual differences in the likelihood of fidelity versus infidelity, or is infidelity driven entirely by chance and opportunity? Research on the opposite of fidelity, the personality construct of ‘sociosexuality’ (interest in promiscuous, short-term, or extra-pair mating), confirms there are stable individual differences in this trait dimension. Is fidelity positively correlated with other desirable moral virtues and fitness-related traits, such as kindness, conscientiousness, agreeableness, mental health, longevity, and intelligence? (This question becomes complicated, since individual of higher mate value will be sought more often for short-term, extra-pair copulations, so will be tempted by more opportunities for infidelity. Mate value and infidelity opportunities would have to be carefully statistically controlled in studies of fidelity’s correlations with other moral virtues.)

  The genetic studies of infidelity would be the hardest to perform, but often the most informative. Would twin and adoption studies show that the propensity to infidelity vs. relationship stability is heritable? After controlling for overall mate value, would higher genetic inbreeding and paternal age reduce fidelity, suggesting a role for partially recessive harmful mutations in driving infidelity? Would one find positive genetic correlations between the tendency to fidelity, and the mate preference for fidelity, as might be expected if there has been sexual selection for the trait?

  Clearly, the sexual selection hypothesis for moral virtues is eminently testable. However, it requires new ways of thinking about costly signaling and sexually-selected adaptations that are well understood by evolutionary biologists, but that have been slow to permeate evolutionary psychology. In particular, evolutionary psychology still emphasizes criteria such as low cost, high efficiency, high modularity, low phenotypic variance, low heritability, and reliable development to identify psychological adaptations. These criteria are often appropriate for survival adaptations, but not for sexually attractive signals. For naturalistic moral philosophy to benefit most from recent evolutionary insights, it must not only increase its appreciation of sexual selection’s power, but also expand its understanding of how to analyze costly signaling adaptations, by using, refining, and expanding the twelve expected features of moral virtues listed above.

  Implications for Normative Ethics

  Normative ethics is supposed to help us distinguish right from wrong and good from evil. It tries to achieve a ‘reflective equilibrium’ between three things: (1) possible universal moral principles, (2) derived moral implications that would apply in particular situations, and (3) human moral intuitions that react to those principles, implications, and situations. The hope is that normative ethic
ists can articulate a set of universal, coherent, consistent moral principles that yield intuitively acceptable moral implications across all possible situations, and that thereby embody a rational distillation of human moral sensibility. Almost all moral philosophers accept that this is the legitimate goal of normative ethics, though debates still rage between consequentialists and deontologists, act ethicists and virtue ethicists, etc. However, if moral virtues rose through sexual selection, this reflective-equilibrium approach to normative ethics will probably continue to fail – as it has for 2,500 years – for at least three reasons.

  First, suppose human moral intuitions evolved as part of our person-perception system for inferring stable, morally valenced, mating-relevant personality traits from observable behaviors. If so, moral philosophers are trying to do ethical alchemy: trying to refine unconscious, domain-specific, person-perception adaptations (the base metal) into verbally articulated, domain-general, universal moral principles (the gold). This is likely to be an uphill battle. One problem is that we seem to have a dual-process system of moral judgment, as we do in so much of person perception and social attribution. Our ‘hot’ moral intuitions usually precede ‘cool’ moral reasoning, as Jonathan Haidt has argued. These hot moral judgments are often driven by morally judgmental emotions that figure prominently in sexual relationships, such as anger, disgust, jealousy, embarrassment, shame, and gratitude. Our moral judgments may also arise from implicit social attitudes that may be very hard to consciously articulate into normative-ethical principles.

  Second, if our person-perception system relies on social-inference heuristics that are fast, frugal, and pragmatic, then our moral judgments will often violate procedural norms of rationality derived from logic, statistics, and rational choice theory – norms such as consistency, transitivity, and completeness. There are deep decision-theoretic reasons why it may be impossible to derive a set of consistent, coherent moral preferences from the operation of such social-inference heuristics. To know whether this is a fatal objection to the reflective equilibrium approach, we need to learn a lot more about moral judgment heuristics in the context of person perception research.

  Third, human moral intuitions evolved to assess people’s stable moral virtues in ancestrally typical, fitness-relevant situations, and to guide ancestrally feasible forms of social response such as forming friendships or mateships, gossiping about liars, punishing cheaters, or ostracizing psychopaths. There is no reason to expect our moral intuitions to show consistent, logically defensible reactions to evolutionarily novel moral dilemmas that involve isolated, hypothetical, behavioral acts by unknown strangers who cannot be rewarded or punished through any normal social primate channels.

  For example, we often seem cognitively paralyzed by many current debates in reproductive bioethics. How should we feel about abortion, sperm donation, egg donation, surrogate pregnancy, human cloning, genetic testing, or genetic enhancement? Different framings of these issues will activate different, domain-specific moral intuitions. This is precisely why rhetorical metaphors are effective in such moral debates. For example, ‘genetic enhancement’ may seem pernicious fascism if we view it as a limited resource that will be appropriated by the powerful for their nefarious ends, or it may seem democratically liberating if we view it as a natural extension of good genes mate choice, for those whose own sub-optimal mate value precludes getting good genes from a willing partner.

  Is there any neutral, rational position from which we can judge such issues, without assimilating them to one or another of our domain-specific moral intuitions? Probably not: rational decision-making depends upon subjective utility functions that must be supplied either by the genetic imitation of ancestral utilities (‘gut instinct’), or the social imitation of peer utilities (‘learning,’ ‘social norms’). Gut moral instincts will be mute or misleading guides to moral dilemmas raised by new technology, and moral conformity to peer opinion will be biased by vested political, corporate, and media interests that define the current ‘ethical issues’ in their own interests.

  Basically, there is no compelling reason to think that our moral intuitions have any true normative credibility as guides to genuinely moral behavior. Of course, there may be evolutionary reasons to expect that species-typical human moral intuitions would tend to maximize inclusive fitness under ancestral conditions. However, that is quite different from claiming that they are normatively justifiable in any broader sense. For example, moral philosopher Peter Singer has made some rationally compelling but emotionally counter-intuitive arguments about animal rights, euthanasia, and infanticide. In such cases, it seems impossible to reach a reflective equilibrium between our gut moral instincts and our scientifically informed normative judgments.

  This is not to say that rationally adjudicated principles of normative ethics are impossible to achieve – only that most humans are likely to find such principles emotionally uncompelling and cognitively incomprehensible. The analogy to higher mathematics may be instructive. Humans did not evolve cognitive capacities for manipulating abstract symbol systems to prove difficult mathematical theorems. Only a tiny minority of humans with extraordinarily high intelligence can do so, given years of rigorous training. When Terence Tao was awarded the 2006 Fields Medal in mathematics for his contributions to partial differential equations, combinatorics, harmonic analysis, and additive number theory, the popular science press could not hope to convey the substance of his contributions in these areas. They could only hint at the magnitude of his genius, by mentioning that Tao was promoted to full professor at UCLA at age 24. Of course, there are a few thousand mathematicians in the world who can understand Tao’s work and reach a consensus about its creativity, but to ordinary folks, good harmonic analysis work is indistinguishable from bad harmonic analysis work.

  Normative ethics seems likely, at best, to become a discipline like higher mathematics – a small world of like-minded geniuses pursuing consensual moral truths that remain forever beyond the moral imaginations of most humans.

  In the light of these problems, consider two different forms of a typical normative-ethics question. Abstract form: Is it morally right to assassinate a genocidal war criminal? (Perhaps – many have praised the attempted assassination of Adolf Hitler by Colonel Claus von Stauffenberg on July 20, 1944). Personal form: Suppose there is a 21st century head of state who ordered his country into a fraudulent and illegal war that resulted in thousands of needless civilian casualties, but who is almost certain to avoid accountability to the International Criminal Court in The Hague. Is it morally right for a woman to feel sexually attracted to a man who succeeded in killing the wicked head of state, with a single head-shot from a Barrett M82A1 .50-caliber sniper rifle at 800 meters on a windy day? The personal form is much more specific about the identities of the moral judgment-maker, the morally-judged individual, the civilian victims, the nature of the assassination, and the fitness-relevant, socio-sexual implications of the moral judgment. These details should and do matter in making adaptive mate-choice judgments about the moral virtues of snipers. A woman who knows her ordnance might admire the sniper’s good genes indicators, such as his resourcefulness (the M82A1 costs $7,775 retail), his physical condition (the 13-kilogram, 145-cm-long rifle is hard to carry), and his marksmanship (the 800-meter head shot was near the rifle’s maximum effective anti-personnel range of 1000 meters). Yet she may equally worry about his good dad indicators: his vigilante action may reveal psychopathy, paranoid schizophrenia, bipolar disorder, impulsiveness, fame-seeking narcissism, or high-risk sensation-seeking. She can only tell by gathering further information about his virtues, both moral and non-moral – which is a central function of prolonged human courtship.

  5. The Google Memo

  It was the summer of 2017, and I was pissed off.

  America was drifting away from its free speech principles. In 2014 through 2016, I’d been fighting for free speech in my university’s Faculty Senate, without success. I’d been in touch with
the good folks at the Foundation for Individual Rights in Education (FIRE), and with the Heterodox Academy (founded in 2015) – both strong advocates of free speech. Many of my Leftist colleagues argued that the whole concept of ‘free speech’ was an obsolete relic of a bygone patriarchy, and that ‘freedom’ was nothing more than a bourgeois illusion hiding sexism, racism, fascism, colonialism, transphobia, etc. I saw the administration’s behind-the-scenes attempts to prohibit Milo Yiannopoulos from speaking on campus in January 2017. I heard University Counsel claim that Public Law 92-318 (‘Title IX’) takes precedence over the First Amendment to the US Constitution.

  I’d also become more active on Twitter since early 2016, and saw the ‘social justice warrior’ culture taking over public discourse. But I’d naively hoped that censorious virtue signaling was limited to mainstream media, social media, and academia. I never imagined that it could take over American corporate life – much less the most powerful company on Earth, run by some of the smartest humans on Earth.

  Then the Google Memo went viral, in July 2017. You remember the story. You remember the insanity.

  Quillette was an up-and-coming online magazine that was developing a reputation for defending free speech rights, and for its skepticism about ‘woke’ virtue signaling. Its editor, Claire Lehmann, saw that the Google Memo debate was almost entirely based on signaling, not substance. She asked four scientists who knew about sex differences research to go a bit more substantive, to comment on the Google Memo’s scientific claims. I was one of those four, and here’s what I wrote.