Scientists, God love them. What kind of a life would it be without boffins picking up gene samples in Papua and telling us California's going to fall into the Pacific and Fulham will be under water? The answer is we'd all be bored silly and our taxes would go down.

But this is not an attack on science; or if it is, only on the kind of science that makes Big Statements on Little Evidence. For what diverts me most about this sort of science is its massive certainty. You will not find historians saying that the Second World War was caused by the hysteria of laboratory rats trapped in Ilfendingen, but you will find "scientists" to tell you that Asians are passive because they drink tea, and the West is gung-ho and hyperkinetic because it drinks coffee.

N-n-n-no doubts about it at all, folks! You want to know why the Japanese are (or were until recently) so good at flogging their products and making the stuff we wanted? Look no further. They underwent a dietary change in the aftermath of defeat, pouring all their tea into their insalubrious sewage system and, as conquered nations always do, flattered the victors by adopting their national beverage, coffee. Or perhaps you'd care to consider why Americans are so aggressive about getting going in the morning and starting up new businesses (to patent those Papuan genes, for instance) only to lose interest in what they're doing a few hours later?

Well, now I can tell you: it's that their coffee, ingested in one great whacking early dose, turns them on with a huge dose of caffeine, the wearing off of which (an hour or two) simply leaves them as mentally deflated as Bill Clinton. Whereas those slick, charming Italians, taking the stuff in small, concentrated doses, have time for the arts, stay smart all day and night, and have invented the game of the future, the politics of absurdity.

There is not a sentence in here so far that is not "verifiable" in some way by a scientific study, and the reason this engages my attention today is that I recently read a study which said that coffee was not just one of the most widely consumed products in the world, but also one of the most studied. It seems that if you go into whatever data bank it is that contains such matters, you'll find - in the past 20 years alone - some 60,000 studies on coffee. And - here is the lovely bit - they don't agree. Whether it's good or bad for you, whether it stimulates or depresses, or why it should do either or both, whether it induces cancer or repels it.

Why those 60,000 studies? The easy answer is that coffee is a mild stimulant that helps make one more attentive, better able to concentrate, and simultaneously relaxed. Like all such natural products that people have been ingesting for centuries, if you come into the office dithering about the 19 things you should be doing and not getting any of them done, coffee will give you, temporarily, a sense of priority - part of what we call "alertness". By and large, in modest quantities, this is what coffee does, tobacco does, drink does.

But, of course, these substances work differently on different people: maybe that's why it takes 60,000 studies to tell us what we know already. For instance, that people with low blood pressure, or a slow metabolic rate, can have an espresso before going to bed with no ill result; whereas their contraries are going to have trouble sleeping and produce the sorts of dreams that keep some analytic practises going.

There, I've dropped a scientific "fact" on you. But is it one? Not really. What science can do, however, is tell you what something like coffee is made up of (beans are eight to 12 per cent water, ten per cent sugars, between 1.1 and 4.5 per cent caffeine, 12 per cent fats, 6.8 per cent chlorogenic acid, 4.1 per cent ash, and the rest nitrogenised (12 per cent) and un-nitrogenised (18 per cent) substances). It can also tell you what happens to it when it is processed - in the case of coffee, roasted: water goes down to one per cent, sugar to two, the caffeine stays equal, and we get 25 per cent grey cellulose, 30 per cent glucide derivatives, 14 per cent lipids... You want to hear more?

What common sense will tell you is that if every day 2.5 billion cups of coffee are drunk, there must be something in its favour. And there is something wonderfully pragmatic and experienced about the Italian espresso, just about a hundred years old now. It should contain 6.5g of freshly-ground coffee (roughly 50 beans), made at 94 degrees centigrade, at a pressure of nine atmospheres, and take 30 seconds to filter through. Now there's information we can use. But even more important, to my mind, is that appearance, as with all foods, is as good a guide to flavour as any chemical test yet devised. A sick bean you can see; a sick bean you should discard. Sick science you should discard - before you come to think that the tea-drinking Chinese are unaggressive, or we allow Dunkin' Donuts to colonise the People's Republic