A good news story for once, on the face of it: significant falls in violent crime, according to figures from hospitals across England and Wales.

This could be linked to the fact that we appear to be drinking less alcohol. Or at least, that is the slant being put on the story by most media outlets today.

But in the same week, Britain’s biggest police force reported a slight increase in violent injuries.

Has there really been a fall in violence and if so, does it have anything to do with alcohol?

What’s new today?

Researchers at Cardiff University have revealed a 12 per cent drop in serious violence in 2013, based on data from 117 emergency rooms at hospitals and walk-in centres across England and Wales.

A 12 per cent reduction means there were an estimated 32,000 fewer hospital visits due to violence last year. Incidents of violence were down across all age groups, with the biggest falls among the groups at most risk: youths and young adults.

The sample is big, covering about a third of the emergency units in England and Wales, and concentrating on medical treatment given to victims of violence gets us around many of the perennial problems that dog the official crime statistics.

These figures are harder for the police to fiddle and many victims who might not want to report crimes are still included.

What about crime figures?

The Met Police released figures this week showing a slight rise in violent offences. But that’s one year in the life of one police force. Overall, the nationwide stats show violent crime is at a historic low.

And as we’ve found before, there are serious concerns about the accuracy of crimes recorded by the police themselves, so much so that they are no longer considered official national statistics.

There is an alternative: the Crime Survey for England and Wales (CSEW), based on data supplied by victims, although that has its own shortcomings.

We spoke to a number of experts about whether crime figures are completely unreliable now and, unsurprisingly, we got a range of opinions, ranging from extreme scepticism to the view that, if both the big sets of data point in the same direction, the trend is likely to be real.

The Cardiff researchers make that point today too, looking at police-recorded crime, the CSEW and hospital admission data alongside its own findings.

All four measures come from different sources, but they all say that there has been a significant long-term fall in violence.

Is it because we’re drinking less?

That’s one hypothesis everyone has seized on today, although the Cardiff academics mention it only as a possibility.

They stress that the reasons for the fall “are not clear, but are likely to be multi-factorial and complex”.

Recent decreases in the affordability of alcohol and its consumption are mentioned as a possible cause along with “structural factors such as unemployment, poverty and inequality in addition to public health and criminal justice interventions to prevent violence locally”.

The team also notes that violence dropped in regions where “information-sharing was most developed”, suggesting that the way the authorities work with each other can make a real difference.

There is good evidence globally of a correlation between alcohol intake and violence, as the World Health Organization summarises here.

Many victims of violent crime in Britain consistently say that the person who assaulted them had been drinking – 47 per cent in the latest survey.

And a 2012 study from Glasgow found that the number of outlets selling alcohol in a neighbourhood was a significant factor in predicting local crime rates, even allowing for deprivation.

But the problem with singling out changes in alcohol pricing as the driver of recent change in the UK is that violence appears to have been falling over many years in many different countries.

Academics have come up with a number of intriguing theories to explain this trend, including high abortion rates, changing police tactics, rises in the prison population and the removal of lead from petrol, but all these remain theories.

Angels instead?

Some thinkers, including the Canadian cognitive scientist Steven Pinker and the British historian Ian Morris, have found evidence of a massive, consistent drop in violence in cultures across the world throughout the whole of human history.

http://embed.ted.com/talks/steven_pinker_on_the_myth_of_violence.html

Both say rates of violent death have plummeted since the Stone Age. In his 2011 book, The Better Angels of Our Nature, Pinker says that you can track the decline over millennia, centuries and decades, concluding that we are probably living in the most peaceful era ever.

He draws on the work of criminologists who trawled through centuries of English court records and found that the average resident was about 20 times more likely to be murdered in the 13th century than the 20th.

If this global and historical pattern of falling violence is correct, it would be very difficult to single out short-term changes in the price of drink as a significant cause.

Pinker turns to much broader sweeping historical narratives to try to explain this apparent shift in behaviour. He variously suggests that the rise of the nation state, increased trade and improving literacy rates could be the answer.

Or it could be that people are simply becoming better at empathising with others and thinking rationally.

The verdict

Today’s figures are probably more trustworthy than the crime stats, so a big fall in violent crime should be cause for celebration.

Pinning down the reason behind the drop is less easy. Note that the researchers who did this analysis don’t actually say they are sure that less drinking is the main factor, although the correlation is difficult to ignore.

Academics from many different disciplines have found convincing evidence that violence has been falling generally for many years around the world, but few would claim to have come up with the definitive reason why.