Trust is a Hormone?

Terrence Kealey in today’s Times (21 November) writes a speculative piece on the biochemistry of trust and introduces me, at least, to the hormone oxytocin. I am not completely convinced by the argument. Consider this extract:

“But the basis of wealth is trust. Only if people can trust each other will they trade, and without trade there is no wealth. Yet, if they can, rational people will cheat and not honour their contracts…”

The first part is fine, but once Terrence Kealey turns to describing “rational” behaviour I am left uneasy. Keeping the discussion solely to behaviour to do with economic relationships his point carries conviction but if we were to switch to behaviour to do with more serious crimes (for fraud is a crime) I wonder how many would share his clinical description of what is rational.

Would, for instance, engaging in rape because of self or induced eroticism be encompassed in rational behaviour? A taxi driver sees an opportunity to overcharge a foreigner and takes it; he sees an opportunity to rape a passenger and takes it? Just because a behaviour has a reason, be it selfishness or lust, is it rational?

The second part of the extract above continues: “… which is why people try to avoid doing business with those determinedly rational people, the taxi drivers, businessmen and politicians of large parts of Africa. It is because he supposed that all people were as rational as African taxi drivers that Thomas Hobbes believed in tyranny, because only tyrants could enforce contracts: “Covenants without the sword are but words.”

I agree that contracts need enforcement. They need enforcement, nowadays through legal process, not with a tyrant’s sword - Friedrich Nietzsche spoke of the earth being drenched red with the blood of humans who failed to keep their promises. Keeley continues:

“Adam Smith feared that Hobbes was right, but he also understood that human beings had concepts such as fairness and trust which could override rationality and so enforce contracts: “Nobody ever saw a dog make a fair and deliberate exchange of one bone for another with another dog.”

I find this argument a muddle. Keeley describes criminal acts as ‘rational’ and implies that self-restraint is irrational, or at least non-rational. Breaking contracts is a peculiar form of rationality if it discounts to zero the consequences of the ‘rational’ action. Burning a house down to keep warm is not rational even if the likelihood of judicial punishment was near zero, if the consequence of the fire is that the person is exposed to minus 10 degrees temperature in the depths of winter.

Prisoner’s dilemma shows that defection makes the players worse off compared to co-operation, but that the inability to trust leads individuals to the defection that makes them both worse off. Introduce justice into the context – and Adam Smith always understood the importance of justice for the working of markets and for society (‘without justice it would crumble to atoms’, “Moral Sentiments”). I suggest Keeley confuses rationality with the passions.

The reporting on the hormone oxytocin is interesting:

“Today we increasingly understand that neither fairness nor trust are abstract ideas — rather, they are rooted in biochemistry. And in a paper published recently in Nature by a group of neuroeconomists led by Michael Kosfeld in Zürich, oxytocin was recognised as a hormone of trust.”

Selfishness is one of the passions and humans access many passions; we have mixed motives behind our behaviours. Smith criticised those (e.g., Francis Hutcheson) who saw only the worthy passions as appropriate for behaviour. He also criticised Bernard Mandeville (“Fable of the Bees”) for focussing solely on the passion of selfishness. “Self-love”, said Smith, “is not enough” (Lectures in Jurisprudence). Modern economists based the entire edifice of mathematical models of behaviour on the sole premise of selfishness (misknown as self interest).

That neuroeconomists follow suit does not validate their conclusions.
Keeley concludes:

“The problem of explaining human co-operation has long troubled philosophers. Economists and psychologists have tried to explain it by game theory, and they have demonstrated how rational but selfish people will learn to co-operate in their common interest. In other words, they believe: “We must hang together or we’ll hang separately”. Terms such as zero sum or non-zero sum games are now part of every educated person’s vocabulary.”

Those who have problems explaining co-operation are in difficulties because of their assumptions that it is rational to behave selfishly and with criminality. The assumption causes the trouble Keeley says they feel. Game theory does not alter the assumption. In Prisoner’s Dilemma, the absence of trust, leaves the players with no choice but to defect: ‘I defect not because I want to but because I must”. This is quite different from the motive of the rational selfish player: ‘I defect not because I must but because I want to’. The difference highlights my concerns with the rationality assumption driving selfish criminality.

Trust arises from a learning process. It has nothing to do with rare acts of altruism. Trust to become a regular behaviour requires reciprocity – it has to work for it to pay off into repetitive behaviours. A single trusting player in a community of selfish egoists would not survive; but two trusting players would not just survive, but would breed (reproduce) successfully. They would out survive the egoists because of their small advantage. Reproductive advantage is the key to natural selection.

“Biochemistry is therefore showing that we are not solely selfish, strictly rational people who co-operate only because of the enlightened self-interest of game theory. Rather, we possess molecules such as oxytocin that produce genuine altruism and even self-sacrifice. Oxytocin, in short, is a molecule of virtue.”

I agree: we are “not solely selfish, strictly rational people who co-operate only because of the enlightened self-interest of game theory.” I am not sure that ‘enlightened self interest’ properly explains the learning process, or that a hormone released, apparently, during orgasm explains why two individuals in the stone age – may be beforehand, during the million years or more of the Brutes preceding the appearance of the first Homo sapiens – exchanged some things instead of violently trying to seize them from each other.

With property, said Smith, government enforced property rights, at first lightly as befitted the primitive sense of property in the gatherer-hunter economies of the time. By the time property reached today’s complexity, justice was a deep part of the social fabric. As children socialise they learn about both trust and justice. If they experience dosages of oxytocin while learning that is interesting and worthy of research and reporting, but it is going too far to assume that the entire edifice of human activity is down to a hormone.

For one thing it would not explain why for millennia oxytocin was a minority activity in the age-old contest between plunder and trade, nor why it was so little in evidence during seventy years of State socialism. For another, why does trust operate so well, as it does, in market economies? Disparities in oxytocin production in the various populations? Or was Smith more perceptive?