A story which has received wide coverage recently is the drop in violent crime in the UK. The Guardian has a good summary and comment here www.theguardian.com/society/2014/apr/23/alcohol-prices-violence-study-binge-drinking
Like other articles, it makes the link between the increasing cost of alcohol and decreasing violence. It’s difficult to get to the bottom of the relationship between alcohol and violence. There is no doubt that there is a relationship, but it’s almost impossible to prove causality. The only way really to do that would be to take some fairly large groups of randomly selected people, give one group no alcohol, the next group a little bit of alcohol, the group after quite a lot of alcohol, and the final group tons of alcohol, and then see what happens, and if one group would behave more violently than another. For good measure, you’d have to set the experiment in a realistic situation, but you’d also have to exclude anything else which have been shown to have links with levels of violence, and these could be internal such as tendencies in a person’s personality, and/or external such as hot weather….. to do this would be very expensive, difficult, and it probably would not pass the strict ethical rules which surround scientific testing. However, testing things in this way, ie by randomised controlled trial, is considered the ultimate in terms of scientific evidence, because it demonstrates cause. Without it we cannot say that alcohol causes violence.
Yet there clearly is a link. Alcohol is implicated in the vast majority of domestic violence cases, and is around half of all violent crime, including assault and rape. Alcohol makes people less inhibited – maybe it turns of the checks and balances which prevent violence in some people. However we all know that different people behave in different ways when drunk, and often these are polar opposites: sad, happy, argumentative, withdrawn, gregarious. Difficult as it might be to explore the relationship between alcohol and violence, we need to keep trying, and if a randomised controlled experiment is not possible, then we need to use other ways to investigate.