We should be cautious towards siren voices inviting the west to engage in global nation-building, says Samuel Brittan

There is a highly fashionable diagnosis of the causes and cures of international terrorism, which some of its exponents call "liberal imperialism". It runs thus: 1. The root cause of terrorism is poverty. Or as it is sometimes put, "Violence is the only weapon of those who have no other way of making themselves heard." 2. The best chance of alleviating world poverty is through a massive western aid effort - or "a new Marshall Plan". 3. Aid, however, is not enough. The world is full of "failed states". These will not use aid effectively without a concerted western attempt at what is called nation-building.

The three items of the creed are logically separate; but many exponents embrace them all. An example is a well-argued first article in the April issue of Foreign Affairs by Sebastian Mallaby on "Imperialism's virtues". Robert Cooper, foreign policy adviser to Tony Blair, also makes a plea for a "postmodern imperialism" (Re-Ordering the World, Foreign Policy Centre).

These views can be contrasted with a less ambitious liberalism, which was best expressed by by that great 19th century radical, Richard Cobden, in a letter to John Bright:

"In all my travels . . . three reflections constantly occur to me: how much unnecessary solicitude and alarm England devotes to the affairs of foreign countries; with how little knowledge we enter upon the task of regulating the concerns of other people and how much better we might employ our energies in improving matters at home." Neither view fits all cases. But the pendulum needs to swing back a good way in Cobden's direction.

The evidence against terrorism being a reaction to extreme poverty is overwhelming. Many of the terrorists come from well heeled backgrounds, are educated and proficient in information technology. The really poor are far too preoccupied with making ends meet. The appetite for violent change occurs after a takeoff towards a higher standard of living.

It is however fair to say that there is some link between revolutionary activities and perceived differences of income and wealth. A widening of income gaps is inevitable if some countries or some sectors within each country take off into modernity before others do. But as more and more people and countries become involved in the growth process, the gaps start narrowing again. (The mathematics of the process are simply explained by Nobel Laureate Robert Lucas in Economic Perspectives, winter 2000.)

It is untrue that countries cannot take off into self-sustaining growth without outside aid. Those who assert this might ask how the industrial revolution ever started. A more modest claim is that aid, by adding to the resources of developing countries, can make the transition to modernity less painful. The main reason for helping people in poorer countries is humanitarian. It only detracts from the cause if its exponents try to add that it will also save the economies of the west or keep terrorism at bay.

You can do more with $110 than you can do with $100 for your own resources. But we then have to look at the practical record: how much government-to government aid has been syphoned off either directly into the pockets of corrupt ruling elites or into prestige projects inimical to sound development. Or how much it has been used as a backdoor way for western countries to subsidise arms sales or engineering projects.

My own preference is for direct aid to people or small communities who are hungry or impoverished or who can, with a little help, be supplied with clean water and remedies against disease. The problem is that there are so many hundreds of non-governmental agencies, all appealing to our consciences, that we do not know which of them will send help where it is most needed and which will waste the highest proportion on their own administration or on various kinds of anti-capitalist propaganda. If only some publisher would be brave enough to provide a good charities guide.

It is the third aspect of the new imperialism - nation-building - which most alarms me. There are, on United Nations estimates, more than 100 countries with oppressive governments covering 3.6bn people or 60 per cent of the world's population. Human rights do not stop at frontiers; and if there is an opportunity to introduce them abroad we should do so. The trouble is that, for the reasons given by Cobden, these attempts are often self-defeating.

A study by Gary Dempsey for the Cato Institute (Policy Analysis, March 21) lists country after country, from the Lebanon to Somalia, where the US has entered in ignorance and retreated with ignominy. The worst example is Afghanistan, which was disrupted first by Soviet intervention, then by US support for the most extreme anti-Soviet and anti-modern elements in the country, and finally by Pakistan's internal security services which supported the Taliban faction.

Generalisations are difficult. But as Mr Dempsey writes, "The security of the US does not require a multi-ethnic, liberal democracy in Afghanistan. It requires only that the governments there be deterred from harbouring terrorists as the Taliban once did."

What makes me most suspicious of nation-building is that it is all too easily allied with the traditional desire of foreign ministries to preserve or extend every existing federation or union and establish new ones. The history goes back to British attempts to establish the now defunct Rhodesian and Caribbean federations, and to the warnings of George Bush, the former US president, to former Soviet republics against seceding from the Soviet Union, and western attempts to preserve Yugoslavia beyond the point of no return.

Unfashionably, I prefer Margaret Thatcher's view that it is wrong to prevent regions from breaking away from an unwanted union. I hope that she would apply this to Scotland.