In a recent essay in Time magazine ("Dictatorships and Double Standards", September 23, 2002), Charles Krauthammer argued that the Bush Administration is correct to ally itself with authoritarian regimes in the struggle against international terrorism, especially if the alternative in those countries " to autocracy was not democracy but often totalitarianism." This conclusion is especially appropriate in considering what needs to be done with respect to Indonesia.

Indonesia is the largest and most populous Islamic country in the world. It is also a political morass. The central government is weak and corrupt. Effective leadership is nonexistent, and while much emphasis is placed on the shortcomings of President Megawati Sukarnoputri, which seem real enough, it should be remembered that her two immediate predecessors were not appreciably better.

What there is of a civil society is in a feeble state. The democratic process is a cynical game, one recently characterized by Adrian Vickers as one "where nothing is as it seems, the truth is easily manufactured and no one takes responsibility." With the partial exception of the military, all state institutions are weak.

Violence is endemic throughout the vast sprawling country: in Aceh, in Borneo, in Sulawesi, in the Moluccas, in Papua, and now in Bali. The existence of extremist Islamic organizations - perhaps the only significant organizations in Indonesia that represent a combination of strong convictions, organizational discipline and firm goals - is well documented. If this combination of circumstances continues, Indonesia is set to become both a major breeding ground for anti-Western terrorism, and an agent that will, either deliberately or inadvertently, destabilize the whole of southeast Asia.

It is a matter of vital concern to Australia, the United States and the world that neither of these outcomes becomes a reality. This means that the first priority is achieving stability and order in Indonesia as soon as possible. This, in turn, requires supporting those elements in the Indonesian system capable of furthering that end. But the unfortunate truth is that the main, if not the only, institution capable of doing this is the Indonesian military, a brutal and deeply tarnished institution. It was under the military rule of Suharto that Indonesia experienced the only decades of stability that it has so far enjoyed. They were decades of corruption and suppression, but also of increasing prosperity and stability. There is the depressing possibility that this is as good as it will get for a country like Indonesia, that the Suharto period - or at least the first twenty years of it - may seem in retrospect to be the country's golden era. In any case, in the interest of combating terror and maintaining the stability of the region, Australia, the United States and the other liberal democracies of the Western world need to be prepared to swallow hard, subordinate the causes of democracy and human rights and move to strengthen the role of the military and work with it to suppress terrorism.

This unpalatable conclusion will be strenuously resisted by many who would argue that a properly functioning democratic system would be more conducive to stability and order than any military government. But the trouble is that, while in the long run circumstances favoring democracy may emerge, there is no sign that such a democratic system is going to be available in the foreseeable future. One can also be sure that it will be argued that all this grossly exaggerates the threat represented by Islamism in Indonesia; that Indonesian Muslims, unlike Arab ones, are moderate and averse to violence; and that in so far as there are exceptions to this they constitute only a tiny and unrepresentative minority.

As for the first part of this argument, the facts hardly seem to support it. Throughout the country's half-century history, substantial numbers of Indonesian Muslims have held beliefs that have led them to resort readily to violence. The Darul Islam guerrilla movement of the 1950s in West Java did so for a decade, in its effort to convert the country from a secular to a Muslim state. Again, in the 1965-6 massacre following the failed Gestapu coup, orthodox Muslims played a very prominent part in the killing of hundreds of thousands of their fellow countrymen - a more prominent part than anyone else apart from the military. Today, Muslims are fighting and killing Christians in Sulawesi and the Moluccas.

As to the second part of the argument - that only a small minority of Muslims is truly fanatical and committed to terrorism - while this may be true, it is not necessarily reassuring. In the early days of most extremist mass movements, the dedicated activists are small minorities. At the beginning of 1917, for example, membership of the Bolshevik Party was only 23,600, in a Russian empire of over 130 million. In circumstances where those in power lack conviction and will, while a small but organized minority is full of passionate intensity, there is every likelihood that the latter will prevail and will convert the majority, especially if they are allowed some initial successes.

The dilemma facing policymakers in Canberra and Washington is a real and distasteful one. If a stable and orderly Indonesia can only be achieved by working with and through the Indonesian armed forces - unsavory though they may be, and bitterly opposed as such a policy would be by influential sections of public opinion - will Western governments be prepared to adopt such a policy? This is a current form of an old and crucial political question: Will he who wills the end be prepared to will the means?

Owen Harries, the founding editor and editor emeritus of The National Interest, is a senior fellow at the Centre for Independent Studies in Sydney. This article is adapted from a piece that originally appeared in The Australian.