Thursday, November 22, 2007

Infant Morality

We are now being told that infants have a sense of morality because they tend to favor dolls representing a helpful character over dolls representing adverse characters.

The results are from a study in which a character was shown struggling but failing to get up a hill, the character being helped up the hill by another, and the character being hindered in its attempt to get up the hill by a third. Those infants - between the ages of 6 and 10 months - showed a preference for the helpful character.

From this, we are told that the children have an innate sense of right and wrong.

That implication is nonsense.

The infants have shown a capacity to recognize helpful and unhelpful with respect to an immediate goal. However, the 'rightness' or 'wrongness' of the action is another story.

Let's say that there is some threat at the top of the hill - something that would harm the individual if the individual reached the top. The person helping the individual up the hill knew of this threat and was doing so out of a desire to see the being harmed. While, on the other hand, the person pushing the individual down the hill was trying to save him from this danger. These types of variables have a significant effect on the rightness or wrongness of the various actions. However, there is nothing in this for the child to have any capacity to 'sense'.

This suggests that, whatever the child is sensing, it is not 'right' or 'wrong' -not if we can make change the rightness and wrongness at will without changing what the infant senses. It tells us that the child senses something else.

A great many evils are committed in the world by people who measure their reaction to something and immediately jump to some sort of unwarranted conclusion about 'right' or 'wrong' as if this is something that people have the ability to perceive directly. A person imagines a woman having an abortion, has a negative reaction to it, and from this alone condemns that action. Another imagines eliminating all of the world's Jews, discovers that he likes the idea, and concludes from this that those who would condemn that action must be mistaken.

Both of these are built on the philosophy that we have a capacity to sense right and wrong, when right and wrong is not an entity to be sensed.

This leads directly to another problem with this type of implication. If we are going to say that somebody is 'sensing right and wrong', then don't we need a theory of right and wrong that explains how it is something that can be sensed, and explaining how this particular sense organ works? Without such a theory, how do we know that the individual is, in fact, sensing right and wrong, and not something else that happens, in certain (mostly manufactured) circumstances to coincide with right and wrong?

In this case, the people conducting the experiment have no theory of right and wrong. They are using vague concepts, which then allows them to alter the shape of the concept to fit the theory. In order to 'prove' capacity to sense right and wrong, they look at what the individual senses, and conclude that what the infant favors must be (interpreted in such a way that it can be described as) right, and what the infant disfavors must be wrong. Consequently, any agreement between what the infant perceives and what is right or wrong is entirely manufactured.

Let us look at a couple of additional examples from the animal kingdom.

At some stage, lion cubs acquire an ability to determine which antelope in a herd to go after. They are 'attracted to' chasing certain antelope and are prone to avoid chasing others. We may assume that years of evolution have helped to shape these dispositions. However, there is absolutely no justification for leaping from the observation that lions are disposed to favor chasing one type of antelope that they are sensing a property of "deserves to be chased" or, in the case of those the lion avoids, they are perceiving a property of "deserves not to be chased".

A kitten, soon after being born, can be placed on the edge of a table and know not to step over the edge. What the kitten has is a sense of danger - a built-in aversion to certain visual stimuli that probably has some type of genetic component - since those who lacked this disposition were more likely to die. However, there is nothing in this that justifies a researcher making the claim that the kitten senses that it is immoral to step over the edge of the table. The claim that these children are sensing some sort of moral quality in the actions of the two individuals is similarly unjustified.

Where are these researchers getting the idea that these creatures are sensing some sort of moral quality?

Of course, if somebody wants to believe something strongly enough, they can easily ignore the logical problems associated with that belief. We see it all the time in religion, where the desire to belief brings individuals to simply ignore even the most blatant contradictions and factual errors.

It appears to me that the desire to find some type of biological component to morality is so strong that the biologists who are involved in this research, and the lay population that follows and repeats these findings, are similarly adept at simply sweeping aside the fact that the assumptions that underlie these claims makes no sense. Yes, infants have the capacity to acquire perceived preferences for different states of affairs. Yes, the probably evolved a disposition to favor states that tended to favor the survival and genetic replication of their ancestors. However, there is still an unbridgeable logical leap from these observations to a conclusion that these researchers have discovered (1) some sort of moral property in nature, and (2) an organ capable of accurately and directly perceiving that property.

Those types of entities are no more real than gods and intrinsic values.


Martin Freedman said...

Good points but you really need to answer the most important work in this area (sort of) recently namely Marc Hauser's work and his Moral Grammer. Unless I am mistaken, you have not dealt with this properly. Your post only glances on Hauser and his argument for a genetic basis of a moral grammar which is not the same as having the moral sense per se nor having one with fixed parameters (we certainly agree on the invalidity of arguments for these). Your argument that there are no genes to reason over trolleys and such is correct but besides the point. and a book Moral Minds: How Nature Designed Our Universal Sense of Right and Wrong by Marc Hauser.

(BTW I am dubious of the notion even of a Moral Grammar, as I am dubious of Chomsky's - not his great mathematical analysis of language syntax but his - non-evolutionary innate Universal Grammar claims. Anyway I am interested on what you have to say on the matter of Moral Grammars not Chomsky!).

Alonzo Fyfe said...


Okay, I'll make a concentrated effort to give a more dedicated response to Hauser.

However the main point remains quite simple. Hauser looks at whether certain 'moral judgments' are common among populations.

However, the one thing he has not done - and the one thing he has to do in order to claim that he is actually studying morality (rather than moral claims), is to also show that these common answers are the right answers. To the best of my knowledge, he does not even have a theory of right answers, let alone applied such a theory to his study of morality to show that the subjects of his experiments are coming up with right answers.

Hume's Ghost said...

Hauser goes out of his way to explain that he is not saying that having a moral sense means that our intutions yield correct prescriptions for ethical behavior.

But evidence has amply demonstrated that our brains are wired to process ethical dillemas in a particular way ... Hauser points out that understanding our brain's biases might help us understand and overcomes situations where we have ethical shortcomings.

Hume's Ghost said...

From Moral Minds:

"The only way to develop stable prescriptive principles, through either formal law or religion, is to understand how they will break down in the face of biases that Mother Nature equipped us with."

Hume's Ghost said...

Doggonit .. last comment ... I forgot about a this from an interview with Discover magazine

You mention honor killings in cases of infidelity, but sometimes the victim may simply have been caught in public talking to a man who is not her husband. As a Western woman raised in the liberal tradition, I think that is immoral. Yet in societies where honor killings are acceptable, the decision to kill the woman is deemed morally correct. Why?

Let’s go back to language. You’re a speaker of English. In French, the world "table" is feminine. Why? Isn’t that weird? Isn’t that incomprehensible? For an English speaker, that’s the most bizarre thing in the world! It’s incredibly hard to learn. Yet are the French weird? They’re not weird. They speak another language.

The analogy to language is to me very profound and important. When you say, “Look, it’s weird that a culture would actually kill someone for infidelity,” it’s no
different than us making a language that’s got these really weird quirks. Now, here’s where the difference is crucial. As English speakers, we can’t tell the French: “You idiots. Saying that a word has gender is stupid, and you guys just change the system.” But as we have seen historically, one culture telling another culture, “Hey, this is not OK. We do not think it is morally permissible to do clitoridectomies, and you should just stop, and we’re going to find international ways to put the constraints on you”—now, that’s whoppingly different. But it also captures something crucial. The descriptive level and the prescriptive level are crucially different. How biology basically guides what people are doing is one thing. What we think should happen is really different. That just doesn’t arise as a distinction within language.