To: Trevor P.
From: Geoffrey Klempner
Subject: Values and the idea of a 'unique valuational perspective'
Date: 15 April 2002 13:10

Dear Trevor,

Thank you for your letter which I received on 3 April, with your fourth essay for the Moral Philosophy program, in response to the question, 'What are values? Where do they come from? How are values integrated to form a "unique valuational perspective"?'

Asking, 'What is a good arrow?' is a useful way to focus discussion on the questions of the nature of values and where they come from. As you succeed in demonstrating, the arrow can be considered 'good' for this or that purpose: for use in an archery competition, or for hunting wild boar, or as a prop in a film about Robin Hood. However, our values, as opposed to the things we value for this or that purpose, are those objects or states of affairs which we choose as *ends* rather than merely as means. So, for example, one might value a sport, such as archery for its own sake and not merely as a means of improving one's skills as a warrior, should one ever be called up to fight.

In the program, I take particular care to separate the additional question of the basis on which we value other persons as ends rather than mere means (in the Kantian sense). To value another person is to value their pursuit of their values: but the prior question, as I see it, concerns the nature of an individual's values. Moral value presupposes that each individual has ends or goals which they themselves value as worthy of pursuit.

Ayn Rand is interesting in this context as an example of a philosopher who vehemently rejects the Kantian notion of moral value. According to Ayn Rand, I have no reason to value the values of others as such. My only reason for being moral is that I have come to appreciate that the moral way of life is life enhancing *for me*.

A value is *anything I like*. I don't distinguish here between a laudable value, like doing philosophy, and mere hedonism or thrill seeking. A theory of values, in this sense, should explain why something is 'thrilling' or a pleasure for us, as well as explaining what there is about philosophy, or some other laudable pursuit, that gives us a reason for wanting to do it.

I agree with Ayn Rand's point that it is impossible to see how a robot would value anything, but the reason is not the robot's indestructibility. The phrase, 'artificially constructed intelligent agent' doesn't tell us anything about the creature's desires or needs. There is no reason why such a creature should necessarily desire its own survival for the sake of survival, as we do (notwithstanding the examples of 'Blade Runner' and 'Hal' in '2001 a Space Odyssey').

Evolution by natural selection has given us a *nature*, on top of which, as you remark, culture has added its desirable 'ends' which have 'evolved' in a similar struggle for survival. If we built an intelligent robot, we would also have to give it a 'nature' of its own. But what would that entail? Making the robot like us? But a robot could only be 'like us' if it enjoyed something analogous to a human existence, including the experience of being a helpless infant cared for by a mother/ father, then a child and so on. What would it mean to 'give' the robot desires or values? Our values are deeply structured by our natural and cultural history. On what basis (other than imitation - the 'easy' solution) could values be programmed, as it were, from scratch?

(Just to give an example, Asimov in his 'Robot' sci fi novels, describes a set of laws which all robots are programmed to follow which put human beings first in every conceivable circumstance. Are we to understand, then, that the only reasons for action which such a creature is able to give itself are those which relate to the benefit of human beings? This idea goes beyond the most extreme 'slave mentality' that is ever found in human beings: for even the most downtrodden slave has something that they want and need, other than merely to please their master.)

I certainly do not think that Dawkins' idea of 'memes' is the last word on the nature of values. He is right about one essential point: the fact that a 'meme' has survived is no proof that it is *good for us*. We are the disposable survival machines for memes, just as we are for genes. So, for example, Dawkins' example of the 'meme' which 'makes' parents teach their children that they must believe in X and will go to hell if they do not teach the same precept to their children.

However, I do not accept the reductionist view that values are never more than memes. Our ability to recognize the 'value' of a valid as opposed to an invalid argument presupposes capacities for reason and reflection, which cannot be reduced, as Dawkins wishes to do, to blind imitation.

All the best,

Geoffrey