An Introduction to Evolutionary Ethics (17 page)

As in the case of the moral/conventional distinction, children seem to grasp this difference
despite
the fact that the difference is not at all apparent in their upbringing. How often do parents stop to spell out the difference between the “must” of necessity (as in “Bachelors
must
be unmarried”) and the “must” of duty (as in “You
must
return what belongs to others”)? And yet, kids get it. How could this be, ask supporters of innateness, if this knowledge were
not
innate?

Children also seem capable of appreciating the difference
intentions
make. I touched on this in the previous section. Consider some further evidence. A child is told that if Sam goes outside, Sam must wear a hat. The child is then shown four images depicting Sam. In the first image, Sam is inside without his hat on. In the second image, Sam is inside with his hat on. In the third image, Sam is outside without his hat on. In the fourth image, Sam is outside without his hat on, but his hat is being blown off by a gust of wind. The child is then asked: which image(s) show Sam being
naughty
? Now since images 3 and 4
both
depict Sam outside without his hat on, we would expect that some kids would say that Sam is being naughty in both images, that some kids would say that Sam is being naughty in image 3, and that some kids would say that Sam is being naughty in image 4. But the evidence does not show this. Instead, almost all children aged 4 and over say that Sam is being naughty
only
in image 3. Children evidently realize that being naughty requires
intentionally
being naughty. Since the wind causes Sam to lose his hat in image 4, he is not intentionally outside without his hat on. So Sam is not being naughty in that image. These results hold up even in places where it's rarely cold enough to need hats.

As in the case of language, these results reveal that children possess a rich understanding of morality – despite what appears to be a very limited exposure to moral discourse. But if, argue moral psychologists, what comes out of children is far more sophisticated that what goes in, then there must be a body of knowledge already inside children. Moral knowledge (or at least a core of it) is innate.

That said, it's worth repeating that even if morality is innate, this does
not
entail that morality is an evolutionary adaptation. Stepping back, we should underline that moral nativism, like linguistic nativism, is a story about what's present in humans from the beginning (so to speak). Adaptationism is a story about
how
what's present in humans got there. This means that one could accept moral nativism, but deny that morality is an adaptation. One could argue, for example, that morality is a by-product of other cognitive systems, that is, an
exaptation
(see §1.1. Or, for that matter, one could argue that God placed the knowledge in us at birth. At any rate, it's important to keep our options open as the evidence comes in. If the adaptation story sketched in the previous chapter does not in the final analysis receive the evidence it needs, we may decide to reject it. But this doesn't at the same time force us to reject moral nativism. We might find the evidence on that front compelling. Or we might not.

The hypothesis of moral innateness seems convincing when our attention is focused on what a small set of children know from a very young age. But what happens when we widen our focus? Recall from chapter 4 the variety of moral views we see in other cultures. Acts of killing that we find appalling are acceptable in other parts of the world – if not demanded; in most Western cultures, women are treated equally under the law, a practice that strikes some cultures as morally abhorrent. But doesn't moral variation across cultures mark an obstacle for the hypothesis of moral innateness? After all, if the instinct for morality is present in all humans, shouldn't we observe similar moral views wherever we find humans? This question has prompted some moral psychologists to pull back somewhat from the moral innateness hypothesis.

5.5 Switchboards, Biases, and Affective Resonances In making the case for linguistic nativism, I sidestepped a rather glaring fact: last time I checked, people speak different languages! But if language is innate, then shouldn't we expect to see everyone speaking the
same
language? No one doubts that the development of two arms and two legs is innate. And sure enough, humans around the globe have two arms and two legs. So how can linguistic nativists claim that language is innate if there are nearly 7,000 different languages spoken around the globe?

The answer requires getting clear on
what exactly
is innate. According to linguistic nativists (following Chomsky), native or natural languages – i.e., the language we actually speak and have learned from our parents – are not innate. What
is
innate, according to linguistic nativists, is the
grammar
underlying all these languages. For one of the striking discoveries of the last fifty years is that all natural languages share some common universals. Chomsky dubbed this deep similarity
Universal Grammar
. One way to appreciate the point is to imagine the space of logically possible languages (computer languages, for example); think of all the wild syntactic rules governing how to turn sounds into meanings that we could come up with. The amazing thing, say linguists, is that all natural languages spoken by humans fall into only a tiny, tiny sliver of that logical space. This is not to suggest, of course, that there aren't vast differences separating different languages. But these differences, it turns out, are only on the surface. Beneath the surface, the differences are negligible. And one way to think about these differences is in terms of switches or parameters. It will help to consider some examples.

Put crudely, what makes English English and not Italian, say, has to do in large part with how the switches are set. For example, Italian speakers (even young children) recognize that the sentences (1) Io vado al cinema.

(2) Vado al cinema.

are both acceptable. Both express the proposition that I am going to the cinema. But in English, while the sentence (3) I am going to the cinema.

is acceptable, the sentence (4) Am going to the cinema.

is not. The reason is that in English – but not in Italian – the subject of the sentence must be pronounced. Linguists refer to this constraint as the
Null Subject Parameter
. Every known language is either Subject Optional (as in Italian) or Subject Obligatory (as in English). For known languages, this exhausts the possibilities. Over the last forty years, linguists have identified many such parameters. In almost all cases, the parameter has but two settings.

Identifying these parameters and how they are set helps us sharpen the linguistic nativist's claim. What the child brings into the world, argues the nativist, is this Universal Grammar “switchboard.” What the child acquires from her environment is information about how those switches are to be set. Are the people in my environment regularly pronouncing the subject of declarative sentences or not? The language faculty is supposedly sensitive to this information (below the level of a child's awareness) and makes the necessary setting. What the child does
not
have to learn independently is that sentences have subjects and these subjects are either pronounced or not. That part of her education is over before she begins.

This view of language, say moral nativists, provides a model for how we should understand moral development. Like Mikhail, the psychologist Marc Hauser has been vigorously pushing this idea. According to Hauser, just as a child's grammaticality judgments emerge from a universal linguistic grammar whose parameters are set by local conditions, the child's moral judgments emerge from a “universal moral grammar, replete with shared principles and culturally switchable parameters” (2006: 43). What the child inherits, in short, is a moral switchboard. What the child picks us from her environment is how to set each of the many parameters that make up a moral system. If Hauser is correct, “every newborn child could build a finite but large number of moral systems. When a child builds a particular moral system, it is because the local culture has set the parameters in a particular system” (2006: 298). Hauser sees numerous examples of cross-cultural universals tweaked by distinct cultural norms.

Take killing. According to Hauser, a child is already equipped with a principle that prohibits harming others. A child does not have to learn that harming others is forbidden. What a child must learn are the exceptions, if any. Are there many exceptions to the general prohibition? Or only a few?

In some environments, killing other people is regarded as almost universally prohibited. With the exception of self-defense or maybe capital punishment, killing others is presented to the child as
never
OK. In this environment the switch would be set at ALWAYS PROHIBITED. In other environments the moral status of killing is much more complicated. It appears to be a function of several sub-parameters. Is the person to be killed a member of my tribe? Is the person female? Has this person offended my family's honor? Has this person humiliated me? In environments like this, the switch is much lower, at SOMETIMES PROHIBITED. But it's part of the nativist's claim that the moral switchboard puts constraints on possible judgments – for example, there is no setting for NEVER PROHIBITED. Among the thousands of human cultures studied to date, not a single one exhibits an indifference to harming or killing others. The variation we see appears to be constrained in predictable ways. Analogously, there are many settings on the language “switchboard” that we will not observe since they are disallowed by the rules of Universal Grammar.
⁶
According to Hauser, “our biology imposes constraints on the pattern of violence, allowing for some options but not others” (2006: 132).

Hauser believes that similar stories can be told about justice, fairness, incest, and infanticide. In all these cases, it appears that humans everywhere (from a very young age) possess an innate sense of what's required and what's forbidden. The variation we observe is explained by parametric variation tweaked by distinct cultures.

But Hauser's principles-and-parameters view of moral development is a bit too rich for some tastes. The philosopher Chandra Sripada is prepared to offer a
kind
of nativist story, but it's far less imposing than the one suggested by Hauser. Sripada's dispute with Hauser concerns what the data force us to accept.
Linguistic
nativists believe we're forced to accept the view that language is innate by reflecting on how difficult the language-learning task is for kids. (Recall our discussion in the previous section.) Sripada is willing to accept this view of language. But unlike Hauser, Sripada does not think that the data force us to accept the principles-and-parameters model of moral development. Why? Because the
moral
-learning task is not nearly so difficult. Moral norms, notes Sripada, are not “far removed from experience in the same manner as the hierarchical tree structures and recursive rules of human grammars” (Sripada 2008: 328). After all, kids are being told by their caregivers what they can and cannot do quite regularly. So while there may be some distinctions requiring explanation (e.g., the moral rule/conventional rule distinction), the moral-learning task confronting children is not nearly as daunting as the language-learning task. Also, in the former case, kids have the advantage of having a language with which caregivers can
instruct
kids on what is right and wrong. Not so in the case of language. Finally, Sripada is concerned that the cross-cultural variation in moral norms is simply too great to be accounted for by the Hauser story. It must be the case then that children come into the world with more flexibility than Hauser supposes.

In place of the principles-and-parameters model, Sripada proposes what he calls the
Innate Bias Model
of moral development. This model allows for more flexibility in moral development because it attributes less to the moral mind. What children possess innately, according to Sripada, are certain
biases
or
dispositions to favor
some norms over others. In other words, human minds are hard-wired to find some social rules more attractive than others. This smoothly explains the moral similarities we observe across different cultures. Sripada cites incest as an example.

There's good evidence that humans possess an innate aversion to engaging in sex with anyone with whom they have spent prolonged periods of their childhood. We refer to this as the
Westermarck aversion
, after the Finnish sociologist Edward Westermarck who first proposed the mechanism. But note that an innate aversion to incest is not the same thing as an innate
moral prohibition
on incest. (Hauser is prepared to argue that children possess an innate moral prohibition on incest.) According to Sripada, the innate aversion will lead to a sort of group disgust in reaction to an act of incest. And this group disgust will lead over time “to the emergence of new moral norms that forbid the offending action” (2008: 336). The Westermarck aversion is just one example of an innate aversion to a given practice that “would enhance the ‘attractiveness’ of a moral norm” forbidding that practice. If this story is correct, moral prohibitions on killing, stealing, lying, and so on, grew out of innate aversions shared by every member of the group.

The philosopher Shaun Nichols has argued for something quite similar: “Normative claims [some of which will be moral claims] that are ‘affect-backed,’ that prohibit an action that is emotionally upsetting, will be better remembered than non-affect-backed normative claims” (2004: 128). Memory plays a special role here for Nichols. For norms that are remembered enjoy
cultural survival
, and this explains the near-universality of moral thought. According to Nichols, what we see in the moral domain is not so much the work of biological evolution as of
cultural
evolution. Cultural evolution involves the selection of
ideas
– not genes. (Richard Dawkins dubbed these units of selection
memes
.) Which ideas get selected? Whichever ideas happen to survive and spread to the minds of others.
⁷