22 Mar 2006

Rational self-interest

The sage is ruthless; he treats the people as dummies. LAO TZU

The main thesis of this article is so simple as to be almost trivial, yet if it is true, it will transform our understanding of the foundation of ethics. It is that self-interest is not rational.
I could almost stop there; is it not obvious immediately as soon as one thinks about it? But rational self-interest is the foundation of economics, contractarian theories and perhaps the whole of western liberal democratic capitalism. I must assume that it is not obvious after all and attempt to argue the case.
The Prisoner’s dilemma has a vast and growing literature, and I will assume the reader is familiar with the basic story, which comes from game theory, a branch of mathematics. A situation is defined such that the logic of self-interest leads inexorably to an ‘inefficient’ solution. As far as I am aware though, no one has ever suggested that this constitutes a proof, by reductio ad absurdum, that self-interest is not rational. The Stanford Encyclopaedia of Philosophy for example, argues thus:
However, inefficiency should not be associated with immorality. A utility
function for a player is supposed to represent everything that player cares
about, which may be anything at all. As we have described the situation of our
prisoners they do indeed care only about their own relative prison sentences,
but there is nothing essential in this. What makes a game an instance of the PD,
is strictly and only its payoff structure. Thus we could have two Mother Theresa
types here, both of whom care little for themselves and wish only to feed
starving children. But suppose the original Mother Theresa wishes to feed the
children of Calcutta while Mother Juanita wishes to feed the children of Bogotá.
And suppose that the international aid agency will maximize its donation if the
two saints nominate the same city, will give the second-highest amount if they
nominate each others' cities, and the lowest amount if they each nominate their
own city. Our saints are in a PD here, though hardly selfish or unconcerned with
the social good.
To return to our prisoners, suppose that, contrary to our
assumptions, they do value each other's well-being as well as their own. In that
case, this must be reflected in their utility functions, and hence in their
payoffs. If their payoff structures are changed, they will no longer be in a PD.
But all this shows is that not every possible situation is a PD; it does not
show that the threat of inefficient outcomes is a special artefact of
selfishness. It is the logic of the prisoners' situation, not their psychology,
that traps them in the inefficient outcome, and if that really is their
situation then they are stuck in it (barring further complications to be
discussed below). Agents who wish to avoid inefficient outcomes are best advised
to prevent certain games from arising; the defender of the possibility of
hyper-rationality is really proposing that they try to dig themselves out of
such games by turning themselves into different kinds of agents.
http://plato.stanford.edu/entries/game-theory/#Bas
(section 2.7)
I think this argument is false, and seriously underestimates the power of unselfishness. My understanding of people like Mother Theresa is that they are seeking to do God’s will and not to act in their own interest. Thus they do not identify ‘my poor’ and ‘your poor’ or ‘my good works’ and ‘your good works’ as separate things with separate values. They are indeed ‘different kinds of agents’ whose ordinary (but sadly rare) rationality leads them to the obvious ‘best’ overall solution for the poor in general. The logic of the situation does indeed demand that the agents’ interests are separate and different, and without this separation, the dilemma does not exist; this is the logic of self-interest. If the prisoner’s value each other’s well being equally with their own, there cannot be a separation of interests, and there cannot be a dilemma, ever.
It is worth noting that this cannot be achieved by identification with a group, as my team, my country, or my poor people; in such a case, the group remains vulnerable to PD situations in the same way as the selfish do-gooder does. Immunity to the PD could thus be seen as the criterion of rational/ethical living.
In order to be in a PD, or any game theory game for that matter, one has to know the score, one’s own and the other prisoner’s utility functions. One must in some sense then acknowledge the other’s agency on a par with one’s own. But then one must take one’s own utility as the overridingly significant one. There does not seem to be any possible rational justification for this. It is so prevalent that one might well be justified in assuming that the other is likely to take this position, but if I am irrational, that doesn’t make it rational for you to be like me.
If the argument I have just sketched out is correct, and self-interest is not rational, there are significant implications for ethics. The gulf between moral and factual statements – between is and ought - becomes far less problematic. ‘If you are hungry, I ought to feed you.’ Has the same structure and significance as ‘If I am hungry, I ought to feed myself.’ The latter is so obvious as to be almost a tautology; ought functions as a commonsense recommendation of the remedy for hunger – ‘Hurrah for food!’ is the meaning of hunger, as much as it is the meaning of ‘I ought to feed…’ and it is the same for anyone’s hunger, not just mine.
Of course I am not suggesting that I and most other people are not generally selfish, simply that our self-interest, however pervasive, is not rational and therefore does not require any ethical counter-argument. But because it is so universal, we take it to be the rational way to behave and then try to find explanations and justifications for everything that does not fit this model of rationality.
The whole enlightenment project, to which our western culture is heir, is founded on the idea of the rational self-interested man, the individual, participating in a social contract for mutual benefit. It has already been pointed out by feminist writers that this image becomes less clear when replaced by a woman, who is likely to be variously attached to and to contain other beings in the form of children. The individuality of the self, the privacy of the body and the mind, begins to break down. Our bodies come out of other bodies which they are part of, and our thoughts are likewise formed from the bones of the ancestors, Shakespeare, Plato, Enid Blyton, etc. Those who wish to maintain the privacy and copyright of their ideas should never read or listen to the thoughts of others; the mind is so terribly absorbent. Yet just this separate individual identity is a prerequisite of self-interest, and hence of any game of game theory.
The whole of ethics is counter-factual, because if ‘what ought to be’ is, there is no problem, and nothing further to be said. In this sense ethics arises from, and depends on, self-interest; if we were not selfish, there would be no need to make rules to forbid it. Because I am selfish, I see the benefits (to myself) of ‘cooperation’, and so the games begin. This is not the cooperation of Mother Theresa, a working together to the same ends, but merely a temporary and provisional mutual exploitation.
Perhaps I need to clarify that when I say that self-interest is irrational, I do not mean that it is irrational to eat when you are hungry, merely that it is irrational to eat all the food when I am hungry too. The awareness of other’s interests, which is the prerequisite of the PD game, is the fruit of the tree of knowledge in the Eden story, the game itself is the wilderness into which we are cast, and our shame is for the nakedness of our greed that is exposed by this knowledge. Ethics, then, is a bit of a fig leaf!
A Theory of Justice (Rawls, J. 1971) is a heroic attempt to embroider a fig leaf large enough to cover the entire western world. It is founded on the rationality of self-interest, and attempts to construct through a thought experiment, a reasoned contractual arrangement that self-interested individuals could come to in a certain idealised situation called ‘the original position’. One of the major requirements of the original position is that although the individuals are assumed to be rationally self interested, they are deprived of all knowledge of what these interests are and also of their position in the society whose rules they are to establish; he refers to this as the veil of ignorance. One might say that he obliges us to set up the rules of the game before we know which side we are on, and sure enough, it turns out that in these circumstances there is a pretty strong emphasis on the rules being fair. The veil of ignorance serves to prevent the operation of the self-interest assumed to exist.
The supposed rationality of self-interest has been exported wholesale to biology where it is tacitly assumed that survival of the selfish holds sway, at least genetically. When pushed, the geneticist will say that of course this is an analogy. Genes that survive tend to be ones that promote their own survival in some way, and this makes them appear as if they are selfish, although they have no self or any agency. On the basis of this ‘as if’, game theory is then applied to evolution, and so to the evolution of the human mind and from there to human groups and cultures.
In Natural Justice, Ken Binmore seeks to place Rawls’ theory of justice on a sound biological basis, and he acknowledges the analogical nature of the selfishness of genes at the outset.
Why, for example do songbirds sing in the early spring? The proximate cause is
long and difficult. This molecule knocked against that molecule. This chemical
reaction is catalysed by that enzyme. But the ultimate cause is that the birds
are signalling territorial claims to each other in order to avoid unnecessary
conflict. They neither know nor care that this behaviour is rational. They just
do what they do. But the net effect of an immensely complicated evolutionary
process is that songbirds behave as though they had rationally chosen to
maximise their fitness.
Binmore, K. (2005). Natural Justice, § 1.4 (his
italics)

But two pages further on, ‘A player in the human game of life isn’t some abstract entity called “everybody”. We are all separate individuals, each with our own aims and purposes.’ The as though has vanished. Evolution works by trying everything and seeing what survives, and what survives best will be largely as game theory predicts to the extent that the theory can account for the complexities of the interactions between individuals and their species, and the environment. This latter seems to be somewhat neglected, perhaps because it is too complex for analysis, but on the face of it, the evolution of cooperation with the environment is widespread, and important to long-term survival. Dutch Elm Disease for example has been very successful in recent years, but is unlikely to outlive by long the elm trees that it is destroying. The individual gene does not exist and certainly cannot survive on its own, it is part of a genome that is part of a cell that exists in and is part of an environment. Cooperation is the norm and ‘selfishness’ is bound to be self-defeating.
Somehow the analogy that works to some extent in biology seems far less clear when reapplied to its origin, human minds. My own contention is that it is precisely the possibility of rational thought, as distinct from blind self-interest, that makes us at least potentially different. ‘Our species somehow learned to use culture as a form of collective unconscious or group mind within which to store the fruits of trial-and-error experimentation from the past, and to incorporate new discoveries made by individuals in the present.’ (ibid. § 1.5) This group mind, transcending the individual, is exactly what is missing from other species, preventing them from acting selflessly for the benefit of the species. The group selection fallacy is no fallacy if there is a group mind.

Kant claimed that a truly rational individual will necessarily observe one
particular categorical imperative: “Act only on the maxim that you would at the
same time will to be a universal law.” My mother had similar views. When I
was naughty, she would say, “Suppose everybody behaved like that?” Even to a
child, the flaws the flaws in this line of reasoning are obvious. It is true
that things would be unpleasant if everybody were naughty, but I’m not everybody
– I’m just myself. (ibid. § 3.2)


Listen to your mother, Ken; she is speaking with the authority of the group mind. You are not just yourself, except bodily, your mind is part of the group mind and therefore you are everybody. I cannot claim to speak for the group mind of course, but I am aware that my thoughts are an amalgam of cultural influences and personal experience, and that my ideas are entangled with those of Kant and Hume and Mr Binmore, and their thoughts are similarly entangled with others. All of this is the content of consciousness and is particular to each individual. But the nature of consciousness is the same for all of us and therefore not personal.
I tend to be in conflict with the group mind. Like Ken Binmore I want to maintain that I am just myself, and that principles like ‘love your neighbour as yourself.’ are no use to me unless all my neighbours live by the same creed – and even then I might take advantage of their ‘innocence’. But I can see in theory at least, that this separateness is not as real and absolute as it seems, and that from the point of view of consciousness as a whole, love your neighbour is both rational and natural, and my self-interest is an aberration.

It is this conflict between self and consciousness that gives rise to the phenomenon of ethics with its seemingly unnatural and unfounded demands, but since the group mind has evolved, it is futile to attempt to deny the fact and seek to return to a state of animal innocence. The apple has been tasted, and the conflict can only be resolved now in favour of the group mind, because the group mind is not in conflict and the individual mind is. It may be that the experiment will fail, and that we will follow the path of Dutch Elm disease and most species in history towards extinction; the blind watchmaker makes many mistakes. But the existence of the group mind could enable the watchmaker to see a little way down the road of evolution and perhaps make fewer.
The Cook’s dilemma is a situation I have encountered in real life. The Cook picks up a pan of pasta from the stove to take it to the sink and drain it. Only when he has picked it up and started to move, does he realise that the handles are too hot; his hands are being burnt. But if he drops the pot he will certainly scald his legs even more badly, not to mention losing his dinner. Does it make sense to talk about the self-interest of the hands, and the conflicting self-interest of the legs, and the stomach? My experience is that the hands do their best to get the pot safely put down somewhere, and the legs do not reproach the hands for any failing. The mind reproaches itself for not having the foresight to use a cloth, otherwise ‘everybodypart’ cooperates to help the afflicted and minimise the disaster. One does not need a theory of the evolution of cooperation to explain this mystery, in spite of the fact that the hands do not feel the suffering of the legs. There is no need of empathy on the part of the hands for the legs, nor for a sense of duty; these are purely mental affairs. Hands are not selfish because self hood too, is a purely mental affair. Indeed, selfish, independently minded hands would be a great nuisance and surely unlikely to survive long in the gene pool.

‘The pundits and gurus who claim to know the uniquely Good or Right way to do things are just windbags and blowhards; the reality is that we have only our own likes and dislikes to guide us.’ (ibid. § 12.8) This is very true and applies to philosophers, biologists, and game theorists as well. Fortunately, it appears that our likes include things like giving to disaster funds, preserving the environment, and an interest in higher states of consciousness than self-interest. The prisoner’s dilemma is endlessly fascinating because it encapsulates the human condition. Paradise can be seen but selfish selves cannot reach it. Self-interest cannot get there and that is what makes it paradise. The saints are already there; the gurus claim to know the way but they are just windbags – the self cannot get there, not I, nor you, nor them.
If self-interest is not merely irrational, but is the main source of irrationality, justification ceases to be a problem. Rather than a theory of morals, what stands in need of explanation is immorality – why are we immoral and irrational? Why do we spend most of our energy fighting and destroying each other and the world? Why are we so concerned with this nasty little thing called self?
In other contexts, self-interest is known to be a distorting factor. Politicians are required to declare their interests, so that they can be allowed for. Scientists take steps to remove the potential bias of self-interest from their experiments. In fact we all know that self interest is not rational, but it is in our selfish interests to convince ourselves otherwise, because rational is good and true and right, and that’s what I want to be, or to think of myself as being. But my self is precisely that which I think of myself as being; If I think of myself as having rational self-interest, that is what I am. If I come face to face with the fact that I am fundamentally irrational, thought and self can no longer be sustained. There is an ending of self as thought, and therefore of self-interest. This is a psychological death and the beginning of rationality. Thought as self is tricksy though, and will usually convince itself that this has already happened and that now, ‘I’ am rational. This of course, is another delusion, another irrationality.
In the everyday sense, games are founded on pretence. Kittens pretend they are trying to catch a ball of wool; footballers pretend they want to kick the ball between the posts without using their hands. But thought is unable to distinguish between reality and pretence. When I lose a game of chess, it feels as though I have lost something real. The game is over, but I carry on the pretence and take it as real. Indeed so prevalent is this delusion that it enables people to make a real living from playing games. Indeed the economy is itself another game. Let’s pretend, but all the time forever, that we all want to collect these tokens called money, and whether the ball is ‘in’ or ‘out’ in this game of tennis shall determine who gets the most tokens. And this game is called ‘living in the real world’. And indeed one cannot survive in the culture without taking account of the fact that everyone is playing this game and taking it for real. I would like to suggest that there are other things we could do, other games, or no game, but I won’t be throwing my money away just yet, any more than I would start sun bathing in the penalty area during the Cup Final.
In fact a lot of people spend a good deal of their lives doing other kinds of things without any great thought about their self-interest and this is the secret of a contented life. It is also a problem for game theorists and economists because these people are not ‘playing the game’. The only solutions in game theory are that either they have self-interests that have not been accounted for, or they have made some miscalculation. That they are doing something else entirely, without ulterior motive seems to make no sense, not to a rationally self-interested individual.
Sometimes, philosophy is a game of who can beat whom in an argument, a game played for ‘real’ stakes of money and status. Is there, beyond that a desire for truth, a desire to think straight and see straight for its own sake? Desire seems all encompassing at times, but if I eliminate all the game desires that I have pretended to have and then forgotten the pretence, there seems to be very little left, apart from physical comfort, to be not too warm or too cold, to move when I have been still for a while to rest after activity to eat when hungry, some sexual release – the body is quite simple and easily satisfied. Does the mind have its own desire, to be active and busy, is that why we spend our lives playing games?
When a lion chases a deer, it does not think about the feelings of a deer; there is hunger, excitement, speed and strength. But for myself, once I have come to an awareness that the deer has feelings like mine, I can only continue to hunt at the cost of a mental separation – I am not the deer - and the denial of the feelings that I do in fact have for the deer. There is no way back to the innocence of the lion; either I enter the world of a separate self, full of fear and loneliness, or else I am a vegetarian. For the lion, there is no separation. ‘Lion’ is ‘chasing a deer’. My hunting is different: ‘I’ am ‘chasing’ (a frightened deer). I have to separate myself from the fear, and so also from the deer. In the biblical story of the fall, it is sex where this separation is first manifested – they saw that they were naked and were ashamed. It’s the same thing. Once one is aware that the other has feelings, sex is either consensual, or it is rape – there is knowledge of good and evil. And the wages of sin is this psychological separation as a self, and therefore eventually the death of that self.
The secret of happiness is to be unselfish, but when I, a selfish person, hear about this, of course I want this happiness for myself, so I start to practice unselfishness through self denial, or doing good to others, through self flagellation, fasting, meditation, prayer or whatever. It is all to no avail though, because I am still pursuing my separate, selfish ends. For the selfish self, the only end is death, but I would face even that if I could reach this thing – and even my death would be one more selfish act, this is the stupidity of suicide. The road to heaven is always barred to me – and that’s what makes it heaven.
What is ethical is that which does not separate self from other, because from separation comes suffering. Is this not rational? The original position that I wish to propose, is not that I do not know whether I am to be the doctor, the foetus, the mother, or the legislator, but that these are not separate positions at all, but one single identity. I might choose to abort a part of myself, but I would need a really pressing reason to do so, as I would to amputate my own leg. As in fact I am separated, and am not the foetus, the mother or the doctor, I take it to be most ethical not to seek to legislate either.





Binmore, K. (2005a). Natural Justice. Oxford: Oxford University Press.
Rawls, J. (1971). A Theory of Justice. Cambridge, MA: Harvard University Press.
http://plato.stanford.edu/entries/game-theory/#Bas (section 2.7)

2 comments:

keylawk said...

Very grand to be downwind from some of your winding.

But how helpful was Kant's Critique if we still evaluate theory on its "rationality" and reason?

unenlightened said...

I'm not sure I understand Kant, but I don't find his categorical imperative very useful in making rational moral decisions. The categories are where all the difficulties arise.
The rules for treating humans, animals, trees, and rocks, are very different but the moral issues come at the borderlines of identification:
Is a foetus human?
Are infidels human?
Are black people human?
Are animals equal to humans?
Does Kant say you can't rationally step on an ant unless you can will that a mile high monster will step on you?
It's all too easy to separate and dehumanise - the immigrant, the disabled, the paedophile, the heretic, and I can universalise to my heart's content because it doesn't in fact apply to me or mine, but only to those 'others'.Once the first separation of myself from the world is made, there does not seem to me to be any escape - hense my interest in identity.