'The first lopped-off limb makes you physically wince, but repetition reinforces artifice. Ten minutes later, the 35th simply makes you wonder how much time, effort and money Tarantino wasted on all these gore-squirting shoes.'

However, having just returned from seeing Kill Bill, Tarantino's latest offence, it is suddenly hard to remember what all the fuss was about.

Reservoir Dogs gave us a handful of unexceptionable shootings, Tim Roth spending most of the film dying from a shot to the stomach (which we didn't witness), and that infamous torture scene in which just about the only act of violence happens off camera.

Kill Bill on the other hand really is the 'killcrazy gorefest' that until now we may have thought Reservoir Dogs was. Uma Thurman's sword-wielding quest for revenge sends countless severed heads, arms and feet around the screen, all spurting blood like water pistols.

Personally, I disapprove of this, because it's boring and ludicrous. The first lopped-off limb makes you physically wince, but repetition reinforces artifice. Ten minutes later, the 35th simply makes you wonder how much time, effort and money Tarantino wasted on all these gore-squirting shoes.

Like Hitchcock, Tarantino has a great grasp of the fact that the most powerful movie violence is not in your face, but suggested, glimpsed out of the corner of your eye or simply unexpected. However he obviously he likes to throw the rulebook away sometimes and go psycho.

Whether such violence is cool, disgusting or tiresome is a matter of opinion, of course. But is it wrong? Are detractors judgmental killjoys wanting to impose their own squeamish taste on the rest of us? Or are Tarantino and his supporters irresponsible, corrupting us by feeding our lowest instincts, encouraging real violence and putting aesthetics before ethics?

There are two questions here which we need to deal with separately. 1) Is graphic movie violence wrong in itself? And 2) Is it responsible for real violence?

1) Is graphic movie violence wrong in itself?

Tarantino argues that "violence is totally aesthetic". "I love violence in movies," he told the 1992 Montreal World Film Festival, "and if you don't, it's like you don't like tap-dancing, or slapstick, but that doesn't mean it shouldn't be shown."

Obviously there is an aesthetic element, in that some people enjoy his films and others don't, but is such enjoyment healthy? Opponents argue that to present stomach-churningly explicit violence for entertainment is simply obscene, and corrupts public taste by pandering to its crudest impulses.

It is hard to be entirely consistent in this argument either way. On the one hand, many who find Tarantino's films sick love to watch war films and crime films with far higher body counts, if less blood.

Evidently then they also find cinematic killing harmlessly entertaining, conceding Tarantino's basic point that violence is aesthetic. The only question then is where you draw the line? What's more, some graphically violent films, such as Scorsese's Gangs of New York, are immeasurably more morally penetrating than the average Hollywood drama.

On the other hand, most of us who enjoy Tarantino's work could imagine seeing a film with an act of violence so sadistic, explicit, degrading and hateful that we would find it not simply unappealing ("That's worse than tap-dancing!") but offensive.

So both sides agree that there is a point where cool turns to sick. We disagree where that point is, and, though there may be various ethical factors, in the end it comes down to the effect it has on our stomach. It not entirely a question of personal taste or of morality, because in this case the two are inextricable.

Another complicating factor is that Tarantino's films are so stylistically brilliant, which sweetens the kill. The same graphic violence without the flair would be far harder to take. Those, unlike Senator Dole, who value great cinema are more forgiving of the nastiness, so again ethics and aesthetics are intertwined.

2) Is movie violence responsible for real violence?

However confident I am that I can happily watch actors knife each other without being corrupted, there are wider issues in film violence than my own personal conscience. If films are inspiring teenagers to copycat killings and rapes, and generally fuelling violence, then doesn't the question of whether they are sick or cool become rather irrelevant?

"My answer is that I can't worry about that," said Tarantino - honestly, if callously - in 1992. "As an artist, violence is part of my talent. If I start thinking about society, or what one person is doing to someone else, then I have on handcuffs." One might hope filmmakers would decide what effect their work might have and then act accordingly, rather than ignoring the question because it cramps their style.

More recently and more conventionally he told Empire magazine, "Violent films don't turn children into violent people. They may turn them into violent filmmakers, but that's another matter altogether."

But whether movie violence does inspire real violence is very hard to say. Sifting out the social and cultural influences behind a high school shooting is virtually impossible. Even if you found that the majority of violent criminals watched Tarantino films, how would you know whether the films made the viewers violent or, being naturally violent for some other reason, they therefore enjoyed violent films? If studies revealed that men with shaved heads are statistically more likely to commit assault, would that mean that you could reduce crime by outlawing headshaving?

In the absence of hard evidence either way you could even argue, quite plausibly, that media violence provides a healthy channel for our violent instincts, making us less likely to turn really violent. After all, wasn't it those newspapers that most deplored Tarantino that got most excited about war in Iraq, and those that most appreciate him that opposed it? If US Republican leaders were persuaded to stop worrying and love gruesome movies, might they be less driven to drop bombs on people?

The difficulty of coming up with any solid proof allows liberals to argue for the complete artistic freedom of dazzling auteurs such as Tarantino, unhampered by censorship, self-censorship or censoriousness. This attitude suits me as I love these films, but are we not letting ourselves off the hook too easily?

It has sometimes been pointed out that such liberal arguments are uncannily like those of the tobacco lobby in defence of cigarette adverts, which of course liberals deride. It is hard to prove a concrete connection between advertising and the number of cigarettes smoked, so advertisers ask to be given the benefit of the doubt. If we find it instinctively obvious that advertising encourages smoking, then why do we resist that instinct when it comes to movies encouraging violence?

Tobacco traders argue that advertising increases brand recognition but not smoking. Daniel Batt has pointed out on Shoot the Messenger that Tarantino uses the same argument: he says in Jackie Brown that when Chow Yun-Fat used a 9 mm Baretta in John Woo's The Killer, that was all the underworld wanted to buy - but denies that films get more people shot.

Ben Elton argues powerfully against Tarantino's stance in his novel Popcorn. "Are you one hundred per cent absolutely sure" a real killer asks the Tarantino figure "that no matter how many times you show a sexy murder to a rock and roll soundtrack you have no effect on the people who watch?"

"I am an artist. I cannot ask myself that question."

"Why can't you? If you won't take responsibility for your actions, why should we take responsibility for ours?"

It seems that Stanley Kubrick came to agree, withdrawing A Clockwork Orange before its original cinema run was completed, apparently because of a spate of obvious copycat assaults. However even this anecdotal evidence for the influence of films is questioned by The Observer's Philip French, who argues that Kubrick never gave a reason for the film's disappearance, and that the alleged assaults are mythical, first talked of only five years afterwards.

Research

What we need then is some serious research into the effects of screen violence, and in fact there is plenty. The US Surgeon General released a vast report on youth violence in 2001, surveying a large number of different kinds of study. The report found that there was indeed substantial evidence that media violence can increase aggressive and violent behaviour (though violence is much harder to test than aggressiveness) of those children and young people who are that way disposed for other reasons, although "many questions remain".

How we should react to such findings is hard to say. Do they make it any easier to say whether a given film or scene is likely to inspire specific acts of violence? Do they give weight to the idea of a "climate of desensitisation" where every act of screen violence is part of the problem and therefore a rolling back of 30 years of media liberalisation is called for?

That was certainly not among the report's recommendations: it called rather for better education of parents in monitoring viewing and for proper tests of other policies to reduce the effects of media violence. The report, after all, looked at its effects on children, not on society as a whole.

On the other hand, it is not good enough for filmmakers - or fans - to shrug such findings off as unwelcome and therefore irrelevant, as Jack Valenti, president of the Motion Picture Association of America seemed to do, when he told the LA Times, "The scientific evidence is murky. The conclusions of some of these people don't measure up."

We need to think seriously about our responsibilities where film and TV are concerned, as well as our freedoms. And it's nice to have something to think about as you sit through all that tedious dismembering.