Opinionista Jacques Rousseau 22 March 2011

Natural selection and principled prejudice

“Discrimination” is one of the most ambiguous words in English. We respect the “discriminating wine connoisseur”, but revile those who discriminate on the basis of gender or race. The distinction goes beyond mere semantics, and seems to lie in whether or not discrimination is on the basis of principled judgement. Is it just our nature or a moral choice?

Only a very brave or a very foolish person would claim to have no prejudices. I’m not talking about the conscious decisions we make to discriminate, for these are often justified, but the relatively thoughtless, perhaps instinctive, preference for one sort of thing over another, whether that thing be a type of animal, a football team or a variety of insect.

In conversations about prejudice – particularly racism and sexism – we sometimes make the mistake of treating discrimination as universally bad, in a moral sense. This view would be simplistic and false, in that making value judgements and having preferences are definitive of much human activity, and allow us to surround ourselves with people and things that maximise our happiness. Where possible, I choose, for example, to not surround myself with bigots, and that is a form of discrimination.

What makes my act of discrimination morally acceptable is that it is principled. More importantly, perhaps, that it is principled in a consistent way. If we (loosely) define a bigot as someone who holds unthinking prejudices that are unresponsive to evidence, then it seems clear enough that the prospects of fruitful dialogue are lessened with such people, and that it is, therefore, a relative waste of time to engage with them.

But what if we are evolutionarily predisposed to prejudice? Would that mean that some amount of tolerance for racist and sexist attitudes is justified, much as people argue we should be tolerant of the religious, because the notion of the supernatural is also a “natural” disposition?

Authors such as Steadman and Palmer have argued that beliefs in the supernatural tend to promote cooperative social relationships because they communicate an acceptance of social influence, much as family cooperation is enhanced by a child’s willingness to accept the guidance of a parent.  If religious behaviour does increase cooperation, it becomes plausible that this strategy is rewarded  by natural selection.

Analogously, other forms of “groupism” could be equally strong contenders for natural selection. Distinguishing between us and them – however defined – allows for a preferential allocation of resources, and for a rapid identification of threats. And given the short time that Homo Sapiens Sapiens has been around, even our relatively advanced cognitive abilities are no guarantee that we should have shaken off once functionally useful (even if arbitrary), but now only arbitrary, heuristics by which to distinguish “ingroups” from “outgroups”.

Vision is a primary sense by which humans navigate the world, and we can thus explain that differences such as skin colour and sex were extensively used to make these group distinctions. These visual differences are no doubt part of the explanation as to how we continue to make these distinctions. The fact that we might do so, however, says nothing about what we should do, or what we are capable of doing.

It is this fundamental error – of taking something that is “natural” or historically advantageous in an evolutionary sense and thinking that it provides a guide as to how we should act, or what we should believe – that underpins so-called scientific racism. It is a fundamental error for two simple reasons: First that race or sex was always just shorthand for “looks like me”, and second because natural doesn’t always mean good, and unnatural doesn’t always mean bad.

On the first point: For groupism to work, we simply need some way of differentiating between us and them. The fact that we now know that colour, race and ethnicity have no intrinsic biological significance can’t obscure the fact that at some point in our history, it was all we had to go by, and also that it actually had a role to play in our survival. We can do better now, for example by distinguishing on the grounds of language, culture and belief. And we can do better still, by recognising that there are common features to all sentient beings – such as the ability to feel pain and pleasure – that make the universe of “us” also include animals other than humans.

Furthermore, the threats that confront us now are rarely, if ever, motivated by these crude markers. Our threats largely emerge from ideologies, dictatorships, confusions between selfishness and rational self-interest and so forth, rather than from the mere belief that people who look different are inferior or threatening.

But this should not stop us from recognising that these sorts of discrimination need to be fought, and unlearnt, because they might well be natural dispositions. Fascinating research from Laurie Santos at Yale University suggests that the distinctions we make between us and them “may date back at least 25 million years, when humans and rhesus macaques shared a common ancestor”.

The monkeys in her study showed a clear preference for ingroup versus outgroup individuals, and also showed a clear association between ingroup members and “good” versus outgroup members and “bad”. For example, “the monkeys stared longer at sequences in which outgroup individuals were paired with positive objects like fruit suggesting that this association was unnatural to the monkeys”. The study is well-worth reading in full, but one summary would be that the tendency to have a negative disposition to outgroup members might be more difficult to eliminate than we’d prefer.

This brings me to the second reason why this version of the Naturalistic Fallacy – that natural means good, and unnatural means bad – was and is always a poor argument for continued discrimination. It’s easy to think of examples of things that are natural, yet bad. Cancer is one, and I’m sure we can find an island full of people who would cite tsunamis as another. Conversely, chemotherapy is (mostly) good yet unnatural, as are the spectacles or contact lenses that some of us wear to avoid walking in front of those (unnatural) motor-vehicles.

In an effort to prop up various prejudices – against homosexuality, for example, or euthanasia or stem-cell research – people tend to pick and choose when to apply the principle that what’s natural or not can inform morality. And principles driven by this sort of expediency are never principled – they simply expose underlying bigotry. Likewise, those who might be tempted to think that Santos’ work justifies racial discrimination would be equally unprincipled, because many humans (I’m reluctant to say “most”) have recognised that whatever pragmatic reasons once justified this discrimination no longer apply.

Even the monkeys in Santos’ research seem capable of realising this. While they divide the world into us and them, they do so in a flexible, real-time sense. Monkeys that move from in- to outgroup only weeks previously are viewed prejudicially, and recent immigrants from outgroups are treated favourably. Seeing as we’re all humans, isn’t it about time we caught up with the other primates? DM

Note: Santos’ research used variants of Harvard University’s implicit bias tests. You can check the extent to which you hold implicit biases relating to criteria like race, gender, sexuality and religious views on their website.


EMI records refused to allow the Beatles' Here comes the Sun to be placed on the Voyager spacecraft's record.