It's a good argument: If our kids are old enough at 18 to vote and be sent to fight wars for us, why shouldn't they be allowed to have a drink at the pub if they want to?

Put that way, it's hard not to see the raising of the drinking age as inconsistent and hypocritical.

But if great harm to our children and those around them is the consequence, are we asking the right question?

Voting, with most of the other things we sanction at 18, isn't hazardous to health.

War zones are, of course, but whether or not we consider the sacrifice of young lives to be justified depends on how we feel about the justness of the war, or war itself.

But what of alcohol, ranked in a study published in The Lancet last year as the most harmful of all recreational drugs, on the basis of harm to the user and others - way ahead of heroin and crack cocaine?

It's well known that teenage drinking is linked to increased risk of road accidents, injuries and death, crime, violence, sexual risk taking, mental health problems, suicidal behaviour and a higher risk of being a victim of drink-driving or alcohol-fuelled crime.

Most worrying is the effect of even moderate levels of alcohol consumption on the still developing adolescent brain. "Underage drinking," as the US Surgeon General has warned, "can cause alterations in the structure and function of the developing brain, which ... may have consequences reaching far beyond adolescence".

So what level of harm are we prepared to tolerate? If heroin and crack were legal, would we serve them up to our children in the interests of teaching them how to use drugs responsibly?

Even for adults, the news is bad.

A large European study published in the British Medical Journal in April concludes that a "considerable proportion of the most common and most lethal cancers is attributable to former and current alcohol consumption".

According to the researchers, "there is no sensible limit below which the risk of cancer is decreased".

As for the health benefits of red wine - we've been kidding ourselves. Even though "light to moderate alcohol consumption might decrease the risk of cardiovascular disease, and mortality, the net effect is harmful".

Red wine was thought to be the reason for the so-called "French paradox" - low rates of heart disease despite high rates of smoking and a diet swimming in saturated fats.

But as a French doctor lamented, "man is not only a heart, and to improve the coronary arteries of a group of low-alcohol consumers is not a goal for public health if it has to pay with a high rate of liver cirrhosis, cardiomyopathy, road accidents, violence, hypertension and nervous diseases produced by alcohol".

It turns out the death rate from cirrhosis of the liver in France is about four times higher than New Zealand's.

Analyses by the Law Commission last year and the office of the PM's chief science adviser Sir Peter Gluckman this year make a strong case for change. The latest report adds weight to calls for increasing the price of alcohol, raising the legal drinking and purchase age, and tighter controls around liquor outlets, licensing hours and advertising. But as for educating our way to a better alcohol culture, the Gluckman report says the weight of the evidence "suggests that reasonable and rational appeals" aren't effective.

There is, however, "good evidence from US-based research that raising the minimum age to 21 and adequately enforcing the law leads to reductions in alcohol-related harm".

The US example is instructive. In 1971, when the voting age was dropped from 21 to 18, it made sense to also lower the drinking age. If young men were to be drafted into the Vietnam war, the least a grateful country could do was give them a vote and a drink.

But the dramatic increase in highway crashes, injuries and deaths caused by drunk teenaged drivers, and the stark contrast between states that maintained a drinking age of 21 and those with lower ages forced a rethink. In 1984, all states were told to raise their drinking age to 21 or face funding cuts.

Alcohol consumption among teens fell significantly, even despite weak enforcement, and while teenage drinking remains a problem, the gains have held.

We lowered the drinking age in 1999 - and despite the evidence of significant damage since, advocates of reform are swimming against a powerful cultural tide.

The recent focus on King's College and what we imagine to be a small, binge-drinking segment of our society misses the point. The cultural change we need goes much wider.

If young people today can't conceive of a good time without alcohol, it's because their elders can't either. It takes a strong will to resist our booze-loving culture. Young people can't do it on their own.