Friday, January 18, 2008

E2.0: Jonathan Haidt: Moral Intuitionism

This is the tenth in a new series of weekend posts taken from the presentations at the Salk Institute’s “Beyond Belief: Enlightenment 2.0.”. I have placed an index of essays in this series in an introductory post, Enlightenment 2.0: Introduction.

This posting concerns a presentation made by Jonathan Haidt, Jonathan Haidt, an Associate Professor of Psychology at the University of Virginia. When I first listened to Haidt’s presentation I found a lot there that I wanted to comment on – too much to fit into the format for this series. Fortunately, many ideas Haidt presented also showed up in a New York Times article this week, “The Moral Instinct,” giving me an opportunity outside of the conference to report on those elements.

Particularly, Haidt gave three propositions about morality that he more-or-less asserted were beyond dispute:

(1) Morality is a natural phenomenon that can and should be studied by the methods of science.

(2) Much of morality is innate (“structured in advance of experience”)

(3) Much of that structured by kin selection (the ethic of care) and reciprocal justice (the ethic of justice/fairness).

And, as it turns out, while I agree with the first one, I entirely disagree with the second, and substantially disagree with the third. I hold that the idea of an innate morality is a contradiction – like round squares. Our innate dispositions can either be justified by some outside standard (in which case morality rests entirely with the outside justification, not with the innate disposition), or it cannot be justified by appeal to an outside standard (meaning that our innate position is nothing more than a desire to do things that harm others for no good reason and to feel good about it – in other words, it is not morality).

In presenting his case for intuitionism, Haidt quotes David Hume

We speak not strictly and philosophically when we talk of the combat of passion and of reason. Reason is, and ought only to be the slave of the passions, and can never pretend to any other office than to serve and obey them.” (David Hume, 1739)

I happen to agree with this statement. I have said many times that we cannot reason about ends, we can only reason about means. However, every passion (or desire) not only identifies an end, but it is a means to the fulfillment (or to thwarting) other desires. So, we are not prohibited from applying reason to ‘the passions’ to determine which conflict with other passions and which are in harmony with them. Those other passions give us reason to promote some passions and inhibit others.

This leads to a conclusion that Hume himself endorsed, that the quality of a virtue is determined by the degree to which it is pleasing or useful to self or others. Or, in other words, a desire is good to the degree that it tends to fulfill the desires – either directly (pleasing) or indirectly (useful) – of self or others.

This does not at all lead to Haidt’s intuitionism.

Heidt wants to replace ‘passions’ with ‘intuitions’. Whereas I replace ‘passions’ with ‘desires’. The difference between a ‘desire’ and an ‘intuition’ is that an ‘intuition’ imbeds a proposition that has a truth value. A desire imbeds a proposition that the agent wishes to make true. A moral intuition that killing the people in the next village and taking their property implies support for the proposition that it is morally permissible to kill the people in the next village and take their property. A desire to kill the people in the next village and take their property does not support any moral conclusion.

How does Haidt defend his intuition? He does so by noting all sorts of situations in which he can demonstrate that what people are in fact doing is ‘justifying their intuitions’ – cases in which the moral judgment comes first, and the reasons for adopting them come afterwards as ‘rationalizations’ for the moral position.

I do not see why we cannot come up with a system of religious intuitionism the same way. You take a religious statement that somebody accepts, you demonstrate that there is absolutely no justification for that belief, you force the person into a position where they say, “I cannot disagree with you rationally; yet, I know that God exists and that is all there is to it.” Now, all we need is for Haidt to come along and state that these fundamental religious propositions that cannot be defended by reason are our ‘religious intuitions’ – a knowledge of God that is written directly into the mind (presumably by God himself).

Consider your response to be if some theist were to defend those fundamental propositions of theism that he cannot demonstrate to be true – propositions that command him to do harm to others – on the basis of some ‘religious intuition’ by which he can simply know that those religious propositions are true without proof . . . without evidence . . . without justification, claiming that ‘justifications’ when they occur are merely ad hoc.

This gives us a reduction ad absurdum of the form of reasoning that Haidt is trying to use in defense of moral intuitionism. The type of evidence that Haidt is providing does not justify believing in the types of entities that his theory postulates – moral intuitions (in the first case), or religious intuitions (in the second).

In addition, a researcher may be able to find example after example of cases where subjects are inclined to (for example) use the logical fallacy of ‘affirming the consequent’. He may be able to take brain scans of people committing the fallacy of affirming the consequent and see just what parts of the brain are involved. It may be the case that a theory that includes affirming the consequent is the best theory for explaining and predicting the behavior of test subjects. Yet, with all of this, ‘affirming the consequent’ is still a fallacy. It remains a fallacy no matter how often it is used or how well researchers do in predicting its use.

Haidt’s research, which he claims shows that people reliably engage in a pattern where moral intuition leads to judgment which leads to ‘reasoning’ (or coming up with fallacious claims based on false premises in support of the judgment), can never support the conclusion that passion or sentiment alone can actually justify a moral judgment (a conclusion that others may be legitimately harmed).

In fact, this is a problem with a great deal of moral reasoning. Agents tend to jump far too quickly from a desire to inflict certain types of harm to the conclusion that they are morally justified in doing so. They ‘justify’ this leap by stating that God wrote those moral rules directly into their brain. But God, in this sense, is just an invention that allows one to act on one’s desires without guilt. Moral intuitionism works the same way. By calling an impulse to act in ways harmful to others a ‘moral intuition’ rather than a ‘desire’, one can pretend that the actions that the desire motivates the agent to perform are justified.


Anonymous said...

Another excellent critique. I don't know if you've read or commented on Haidt's piece he wrote for the Edge website, but it caused quite a bit of controversy. It can be found here:

Anonymous said...


For reasons not clear to me, you seem incapable of distinguishing between psychologists and philosophers. When a psychologist says something is innate, he or she says nothing about its moral correctness, only about its origin.
If you accept (and perhaps you do not) that a situation in which one may fulfill or thwart the desire of another is at least a potentially moral situation. And if having a desire to fulfill or thwart the desires of another is a characteristic of one’s morality, then one could ask where this desire came from. If we can prove empirically that these desires were not learned, then they are innate.
This says nothing about what is a morally correct choice, only the origin of a desire.
All I believe psychologists are saying is research shows many of the feelings people have about right and wrong, good and bad, and what we ought and ought not to do, are innate feelings that were not specifically taught.
What that probably means is that having these feelings in some situations had some evolutionary benefit in the past. In no way would they say they prove any sort of moral justification for anything, or even that they have any evolutionary value today.
You are the one who keeps building up and knocking down the straw man that an intuition, or instinct, or innate characteristics justify harming others. I am quite confident all these you criticize hold exactly the same position as you that their research findings in no way justify the mistreatment of others.

Alonzo Fyfe said...

atheist observer

I, too, am quite confident that the people that I criticize hold that their research does not justify the mistreatment of others.

But . . . what kind of treatment is 'mistreatment'? How do we determine whether an act counts as 'mistreatment' or not?

In fact, the proposition that mistreatment is justified is a contradiction in terms. Mistreatment is, by definition, that which cannot be justified.

The question is not whether mistreatment can be justified, but whether a certain act counts as mistreatment.

And how do we know?

Intuitionists say that we can know that something counts as mistreatment (or not) by appeal to our 'moral intuitions'. Psychologists may say that they are merely describing some phenomena in nature. However, when the phenomenon they claim to be describing is, "the method of reaching correct moral conclusions", then they are necessarily saying something about the correctness of conclusions reached using the methods they describe.

Anonymous said...

I have not seen any of the instances you address as meeting your “they claim to be describing is, "the method of reaching correct moral conclusions" criterion. They say they have found evidence that humans often rely on moral “intuitions” to make moral decisions. They would be the first to tell you that decisions based on intuitions can easily be wrong. In fact that is one of the major themes addressed in introductory psychology.

Alonzo Fyfe said...

athiest observer

Intuitionism means "an ability to know without reason."

It is applied to faculties that are alleged to be able to give one knowledge (correct conclusions) by directly 'sensing' that they are true rather than through the application of reason.

Some allowance is given for the possibility that intuitions are fallible. However, if intuitions yield a conclusion of A but reason provides no defense of A (or no compelling reason to believe not-A), then the agent is still considered justified in believing A based on the intuition alone.

If an agent has a 'moral intuition' that X is wrong, but cannot come up with a reason or an argument to justify the belief that X is wrong, the 'intuition' justifies his belief, unless evidence can be found against the claim that X is wrong.

When we are talking about an 'intuition' that somebody else may be harmed, it is particularly dangerous to be suggestion that a mere feeling that harming somebody is the right thing to do is evidence (even if fallible) that, in fact, harming him is the right thing to do unless somebody else can come up with a reason NOT to harm him.

Anonymous said...


Aren't you playing exactly on reader's moral intuitions when you caution against, for ex. a religious intuitionism that would say "God tells me I can harm others" etc.?

You might believe you have a rational argument that harming others is wrong or should not be done. But you didn't offer one in this particular post, so your writing as it stands, just plays on our intuitions that harming others "becuase God tells me so" without any rational justification, for ex., is a counter-intuitive, even Mad thing to do (which I suspect is for most religious people or atheists).

Anonymous said...


the problem I have with your argumentation is that you suppose it is possible to derive an ethical system from reason alone by considering the facts of nature.
But Hume famously pointed out, it is not possible to ground moral fundamental propositions (like harm of others is bad, pleasure is good, I ought to give up my own interest for others) from amoral facts about the world. These moral postulates are in fact nothing more than a projection of our deep feelings upon a morally indifferent reality.
You give an analogy of theistic intuitions to show the ridiculousness of relying upon ones intuitions to get knowledge, but I believe that each philosophers commit the very same mistake when they are positing the validity of FUNDAMENTAL beliefs.

What is the evidence that I ought to give up my own pleasure for the sakes of other ? Because it is selfish to do otherwise . What is the evidence that I ought to not be selfish ? Because if everybody was selfish, human societies could not flourish. What is the evidence that human flourishing ought to be pursued ?
And I could continue further or further, but at the end of the day, nobody has ever been able to justify an "ought" from an "is", from a certain point, it is necessary to recognize that some moral principles is fundamental and can not be deduced from natural facts.
Otherwise, all what one gets is an infinite regression of moral justification which are completely circular and question begging.

Eneasz said...

Hi Anon. Desirism rejects the existence of magical "ought" statements that are separate from the reality of "is". To quote:

Desire utilitarianism does not have unsupported foundational ‘oughts’. It has desires, ... states of affairs, and relationships whereby the propositions that are the objects of those desires are true or false in any given state of affairs.

That’s it

Please see: here and here.

When you claim:
[Without a fundamental ought], all what one gets is an infinite regression of moral justification which are completely circular and question begging.

You simply show that you are not yet fully knowledgeable about ethical theories, as there are several which don't fit this criteria.