Despite the non-involvement of major manufacturers such as the United States, Israel, China, India and Pakistan, the treaty banning cluster bombs signed last week in Dublin stigmatises their deployment and is thus a good thing.
The British armed forces clung to their cluster bombs like a baby to its rattle and some suspected they were trying to sabotage the treaty on behalf of their American friends (who were not there, of course).
But Prime Minister Gordon Brown overruled them, in the end, and Britain was among the hundred countries that agreed in Dublin last Friday to a treaty banning cluster bombs.
Well, it doesn't actually ban all cluster bombs; just the current designs that leave large areas littered with unexploded bomblets that go on killing civilians for years after they are dropped.
Israel dropped some four million bomblets on Lebanon during the last three days of the 2006 war, for example, and more than 30 people have been killed by them since the war ended.
If someone designed a cluster bomb whose bomblets all exploded reliably on impact, or at least within 48 hours of landing, then it would presumably be legal since it mostly killed soldiers.
The major producers of cluster bombs - the United States, Russia, Israel, China, India and Pakistan - were not even at the Dublin conference and have no intention of signing the treaty. But it's a start.
This sort of treaty does not really work by legal compulsion. The countries that sign the treaty are legally bound by it but even for them there is no enforcement mechanism.
For those that don't sign the treaty, there are no formal constraints of any sort.
But by "banning" a particular weapon, the smaller and less militarised countries can exert a real moral pressure on those nations that insist on retaining it. It wouldn't work if one of those countries felt that its very survival was threatened, but that hardly ever happens.
In lesser emergencies, when a country is choosing which weapons to use from a broad range of options in its arsenal, the fact that cluster bombs are now seen as illegal by a majority of the world's states could have a major influence on which weapons get chosen.
They won't admit it, of course.
The Pentagon issued a statement saying that "while the United States shares the humanitarian concerns of those in Dublin, cluster munitions have demonstrated military utility and their elimination from US stockpiles would put the lives of our soldiers . . . at risk".
But this statement would be even truer of nuclear weapons, which have excellent military utility against troops but also kill everybody else in the vicinity.
Cluster bombs would have been quite useful in the environment they were originally designed for, which was industrial-scale warfare in central Europe or on the Korean peninsula.
If they exploded high enough to let the bomblets scatter properly, a few well-placed cluster bombs or shells could destroy dozens of soft-skinned military vehicles and blunt the attack of an entire mechanised infantry battalion.
A few hundred could stop an army corps.
But that kind of war never happened and where cluster bombs have actually been used is in little wars against low-tech opponents: by the US in Cambodia, by Russia in Afghanistan, by the US again in Kosovo and Iraq, and by Israel in Lebanon.
They are not particularly effective against the sort of targets that are on offer in that kind of war, but what the hell, we have them, let's use them.
Unfortunately, whether by accident or by design, the bomblets have this curious propensity not to go off right away.
Between 10% and 40% of the hundreds of bomblets released by the average cluster bomb or shell fail to detonate on hitting the ground, and lie there until - weeks or months or years later - a farmer drives over it in his tractor, or a kid comes along and picks it up.
It is estimated that 40% of the casualties of cluster bombs are not soldiers but children.
So why do some countries cling to these things, while others are willing to let them go? If you look at the list of the hold-outs, it is mainly the countries that just might, in some remote but dreadful contingency, have to face a mass assault by motorised forces: US forces in Korea, Indian or Pakistani forces in the Punjab, the Israelis against Syria (although the Syrians would have to rebuild their forces first), and Russia and China mainly against each other. None of these contingencies is at all likely to occur, but the rule in military affairs (as in much else) is better safe than sorry.
None of these countries signed the 1997 treaty banning anti-personnel land mines either and they are not going to give up their cluster bombs.
So of what use is the treaty? More than you might think.
Cluster bombs are now stigmatised as immoral and (for most countries) illegal weapons, and governments that do use them will have to pay a high public relations price.
That certainly would not deter those countries if they would make a real difference militarily, but that has not been the case in most instances where they have been used in the past. What the treaty really does is to shift assumptions so that international public opinion will see a country that uses cluster bombs as being in the wrong.
As a result, there will be instances where a country that possesses them decides not to use them.
The treaty is not a waste of time.
Gwynne Dyer is an independent London journalist. His latest book, After Iraq, was published recently in London by Yale University Press.
