What sort of government would need to outlaw a postal workers' strike?

To restrict the right to withdraw labour would be as brutal a crime against democracy as abolishing the GLC In their new, sober, regulated form, the unions have begun to win back public respect

They really must be desperate! In a normal country, any politician who suggested banning strikes in the public sector would be led off by men in white coats. But in Britain, the proposal launched by Mrs Gillian Shephard, the Employment Secretary, and by a swarm of don't-quote-me Tory media feeders, is taken seriously in some quarters.

The idea seems to be that a ban on public-sector strikes might be added to the Tory election manifesto. The trial-balloons and kites have been sent up to see if this would be popular, in a summer when there are postal strikes and London Tube strikes, and the possibility of bigger industrial disputes on the railways. Fortunately there is no longer any time for such legislation to be put through in this Parliament before the election, and with luck the next government will be a Labour one anyway.

But it is bad enough that such an idea should even get off the ground. That tells a lot about the mental level to which this Government has sunk. It is not just a dictatorial and divisive idea, the recipe for a conflict which would make the miners' strike of 1984-85 and the "battle of Wapping" look like picnics. It is also a symptom of Britain's awful, growing insularity. No other European country would even dream of dreaming this up.

Well, not quite. I call President Lukashenko of Belarus, the man who cheerfully tore up his ballot paper in front of the television cameras to show what he thought of elections. Now here is a fine fellow who knows how to stand up to the unions on behalf of the travelling public. When the bus staff of Gomel and the metro workers in Minsk went on strike, he dissolved their union, sent police with rifles to arrest the strikers and had them incarcerated in an army camp.

The Tory leadership has been hankering for years to take this next step in their war against organised labour. Mrs Thatcher's legislation against trade union freedoms is now far in the past, long enough ago to reveal some unexpected effects. She certainly "broke trade union power", in the sense that the unions were evicted from national decision-making, although it was mass unemployment rather than the Thatcher laws that brought about the huge fall in union membership over the last 15 years. But the new laws also modernised the unions, transforming them from enormous, loose cannons into more democratic and law-bound instruments of self-defence. Labour was spared the task of trade-union reform (for which Tony Blair should thank Mrs Thatcher every night on his knees). And in their new, sober, regulated form, the unions have now begun to win back public trust and respect.

Into this situation comes the scheme to declare a list of "essential services", and to restrict the right of their employees to withdraw their labour. At its most ambitious, the list would comprise transport, gas, water, electricity, health-service workers from doctors through nurses to ambulance staff, the police and the fire service. This is nothing to do with the 1980s reforms, which included the requirement for strike ballots, a new disputes procedure and the abolition of the closed shop (which was not nearly as wicked or dreadful an arrangement as it's now painted). It would be a purely political act of force. Designed simply to criminalise centres of anti-government opposition, it would be as brutal a crime against democracy as the abolition of the Greater London Council and the other metropolitan authorities by Mrs Thatcher in 1986.

Such a step would violate elementary codes of labour relations. The ILO (International Labour Organisation), which represents a minimal consensus including the views of some very right-wing regimes, accepts that the right to strike can be restricted or prohibited in essential services "the interruption of which would endanger the life, personal safety or health of the whole or a part of the population". But its list of "essential services", while including hospitals, electricity, water, telephones and air traffic control, specifically excludes transport and the posts.

Banning, say, a power-workers' strike is plunging off down a path which leads to dark places. It implies that the power stations will be kept running, which in turn means either the mass recruitment of scab labour (an invitation to violence, and to disasters caused by untrained workers) or to the use of the armed forces. Beyond that lies the dinosaur practice of conscripting "essential" workers, so that they technically form part of the military and are subject to court-martial discipline. I well remember how even the Polish Communist regime drew back from that course, when a rail strike threatened to block the main Moscow-Berlin line in 1981. It might have led to civil war.

Germany deals with the problem in its own consensual way. The "social partners" (labour and employers) are bound by an elaborate code of collective bargaining and disputes procedure, which outlaws both unofficial strikes and lock-outs. There is no restriction on the strike right for essential services, but the Civil Service and senior state employees (the Beamte) are not allowed to withdraw labour.

Behind the German compromise lies an assumption which Britain has never fully accepted. It is the belief not only that the right to strike is a fundamental liberty, but that - if regulated - it is a useful right. It is one of the pillars of a balance between labour and capital, which limits the natural destructiveness of the uncontrolled market. Without strong trade unions and the right to withdraw labour, the web of free associations which we call civil society cannot survive.

Personally, I would go much further. The strike experience, as participant or close observer (and I have been both), is often an education in what's best about the human race. Yes, it is true that most workers are reluctant to strike; family and future are staked on a gamble which always means short-term loss of earnings and may - if the movement is defeated - mean years out of work. Yes, it is true that some strikes are misled by leaders who do not deserve the trust of their followers; it is hard to forgive Arthur Scargill for his errors in the conflict of the 1980s. But that is not the whole story.

A big strike constitutes a new, provisional sort of society. It is a self-managing society whose values are solidarity and fraternity, a new and independent space in which people open to one another and old hierarchies have to justify their leadership or be thrown aside. What I saw in the great shipyard strike at Gdansk in 1980, which changed the world, was not so different to the spirit of the Timex struggle at Dundee or even, in a small way, to the spirit of journalists picketing the doors of a newspaper office through nights of snow and rain.

It is a heavy business, only to be used in last resort. But those who have lived in a "strike community", even for a few days, will remember how these collective rebellions can bring back self-respect to individuals who had come to accept their own weakness in the face of bullying injustice. For those brief instants, there is a glimpse of what each of us could become, given the chance to unfold the powers which lie wasting inside us.

And the public are inconvenienced. Most people can remember the uncollected rubbish rotting along the pavements, the meals eaten by candlelight. Better that, though, than the inconvenience of a state which has abolished the right of workplace resistance, which forces its nurses and railwaymen and firefighters to work or face the riot police. Trade unions and the right to strike are the chain and anchor of democracy. They are still holding - just.