Who can you trust?

Politicians frequently agonise over whether voters find them trustworthy. But the more important que

Trust, David Trimble observed recently, is "overrated and frequently misplaced. The issue in politics is, rather, can you do business with the other side?" The Northern Irish politician was discussing efforts to put the Good Friday Agreement into practice, a process which, in Northern Ireland, has also illustrated that the greater issue is not whether trust is built among politicians, or whether people trust politicians, but whether people trust each other.

That question is overshadowed in public debate by politicians who are more concerned about the degree of trust they themselves enjoy. Thus we hear, expressed frequently and hollowly, concern about loss of public trust and avowals of intention to rebuild it. The emotional undertone to these protestations is that people should feel warmth towards, as well as confidence in, those who govern them. In other words, people should feel the same way about their relationships with those who seek or hold political power as they do about their relationships with each other.

This personalising of civic relations, overestimating their emotional content, then spills over into ideas about relationships within politics. It may underpin views such as one expressed by Jonathan Powell, which sees trust as a means to achieve co-operation rather than an outcome of it (which Trimble was criticising in his Guardian review of Powell's memoir when he made his remark above). But where trust is concerned, relationships among politicians and relationships among people are fundamentally alike. Trust is an expectation that others will incorporate one's interests into theirs, as the American political scientist Russell Hardin insists. Sometimes that will be with love or goodwill; sometimes it will just be business.

By starting off from self-interest, the real value of trust becomes clearer, and so do ways to build it. Throughout history and before it, across geographical and cultural distance, trade has proved its worth as a starting point. One party wants to eat; the second has food to sell. If the buyer pays the agreed price and is happy with the goods, the exchange may be repeated. As the exchanges proceed, mutual trust will develop, opening up possibilities for more extensive co-operation: the buyer may be prepared to make larger orders, for instance, and the seller may offer credit.

Beyond that, the relations between the two of them will be relaxed rather than tense, tolerant rather than suspicious, and cordial rather than impersonal. Much of what is meant by quality of life is expressed in these contrasts. If everyday exchanges typically have these qualities, society is in good shape. The individuals who enjoy these exchanges are also likely to be in better shape. Lives led at odds with those around one, continually stressed and on the defensive, are likely to be not only unhappier but shorter. Stress tells on the arteries as well as the mind.

The chances that any particular exchange succeeds are heavily influenced by the success of everyday exchanges in general. Where suspicion prevails, individuals will find it hard to co-operate no matter how rationally they calculate it would be in their interests.

The sociologist Diego Gambetta describes the failure of taxi drivers in the Sicilian city of Palermo to organise an efficient radio-despatch system, in contrast to their counterparts throughout mainland Italy. In these systems, the despatcher announces the customer's location and nearby drivers report their positions; the job is allocated to the driver nearest the fare. For the arrangement to work, drivers have to trust each other to report their locations truthfully. A driver who frequently claimed to find himself close to customers would arouse suspicion, because, "in Sicily, there is nothing as suspicious as luck".

Eventually the drivers agreed a compromise under which they gathered in car parks to be allocated customers, who paid the price of endemic mistrust in higher fares and longer waits.

Palermo drivers were not necessarily more mistrustful than drivers in other cities, but they were, in fact, less likely to act on their suspicions. Elsewhere, drivers could police cheating by going to announced locations: if they got there before the driver who had claimed the job, they could report him, and he could be punished by disconnection from the radio network. Gambetta suggests Palermo drivers may have been unwilling to check out fellow drivers in case the cheat turned out to be under Mafia protection.

Mafias, however, were not the cause of social mistrust. According to Gambetta, they originally arose as a solution to it. He quotes a rueful 19th-century Naples coachman lamenting that he had bought a useless horse because of the absence of his camorrista (Neapolitan Mafia) protector, who would normally have overseen and guaranteed the transaction. Protection was originally a genuine service, rather than a racket, offered in conditions where levels of dishonesty were so high that buyers and sellers felt unable to make arrangements between themselves.

There are those - particularly, in Italy, to the north - who are inclined to blame this kind of dysfunction on collective character flaws. Others find the roots of dispositions towards trust in culture or patterns of social structure.

Francis Fukuyama notes how Catholicism's fervent invocation of the family can encourage families to trust family members but, in socially adverse conditions, hardly anybody else. The result may be a tense social ecology of mutually suspicious clans. Fukuyama attributes the industrial success of Japan, from among east Asian countries, to the readiness of its companies to form strategic alliances: this, he argues, has its roots in a traditional pragmatism that allowed Japanese families to co-opt competent outsiders into enterprises. By contrast, he suggests, Chinese enterprise is limited by a more rigid adherence to family limits.

Trust is an outcome not just of tradition, but also of policy. Gambetta points to the destructive effects of the Spanish Habsburgs and Bourbons, who controlled southern Italy in the 18th century through "the promotion and selective exploitation of distrust". If their efforts to divide and rule can be blamed for the failure of Palermo taxi drivers to organise themselves efficiently more than a century after Italian unification, it bodes ill for societies whose more recent communist rulers also did their best to prevent citizens from making arrangements among themselves. In Cuba, for example, it may be easier to let a market economy arise than to undo the effects of neighbourhood Committees for the Defence of the Revolution, which enforce party discipline in place of civil society.

In liberal societies, policy tends to be preoccupied with how best to promote healthy social relations between people of diverse outlooks and backgrounds. The political scientist Robert Putnam notes, with regret, that in the United States, the more diverse a locality, the less likely its inhabitants are to say that by and large they trust other people. In highly diverse San Francisco, the figure for people who trust their neighbours a great deal is about 30 per cent; but it is around 80 per cent in a rural South Dakota county where "celebrating 'diversity' means in viting a few Norwegians to the annual Swedish picnic".

There is nothing particularly surprising about such findings. As trust is the expectation that others will take your interests into account, much depends on having confidence in your ability to predict how others will behave. This confidence will be supported by shared culture and norms: if you know that others have grown up with similar ideas about how to behave, you can answer the question of how they are likely to act by asking yourself: "What would I do if I were in their position?"

Without this confidence, people may be discouraged from interacting with those different from themselves.

Diversity and poverty

Nothing in these findings implies that trust cannot be achieved within diversity. Thinking about trust in terms of respective interests suggests the obvious strategy: promote it by minimising conflicts of interest. Since diversity often correlates with poverty, it may require material investment. The larger the cake, the less people's interests in pieces of it will conflict. And it may be just as important to reduce inequality as to reduce deprivation, particularly when inequalities are perceived between groups.

Measures to reduce inequality have to be seen to be opportunities for all, not handouts for some, particularly, for example, where everyday trading relationships are blighted, such as mistrust between customers and shopkeepers from a different ethnic minority and the belief that one is prospering at the expense of the other. The promotion of trust does not sit comfortably with encouraging identity politics.

Even if true, the reports of warmth between Sinn Fein's Martin McGuinness and the Democratic Unionist Party's Ian Paisley affirm that rule. Northern Ireland's agreement for politicians to share power permits their constituencies to share the province while having as little to do with each other as ever. If the politicians trust each other to uphold the deal, it is so their constituents don't have to trust each other in everyday life.

In the province and in cities around the world, trust will develop when people see that it is in their interests to build relationships with others of different cultures. They will have to build on the elements in their cultures that they have in common - which may mean building a new, shared, culture. This does not mean facing in surmountable barriers in human nature, despite what those who dislike diversity frequently imply; just the normal difficulties that arise in any relationships that are worth building. Trust is the reward of self-interest properly understood.

Marek Kohn's "Trust: Self-Interest and the Common Good" is published by Oxford University Press (£10.99). He will take part in "Trust in Me?", an event at the Royal Institution of Great Britain, London W1, on Monday 30 June