In the Guardian – I think only online – is an adaptation of that blog post about falling murder rates in LA.

As a self-confessed amateur of this topic, I have taken the opportunity of buying Chris Williams’ book on History and Crime.  I have seen an earlier chapter on the difficulty of extrapolating from crime stats, and for me it confirms Chris’s comment here:

As far as we can make out, there’s a massive variation in reporting rates for all types of crime in the UK, with the long-term trend being for far more crimes – violent and non-violent – to be reported. We may well be getting more violent – we may well be getting less violent – we may well be getting more screwed up by equal or ebbing amounts of violence – something else may be happening: but in any event, the numbers are a very bad way of attempting to measure this.

But I think murder rates are reasonably unambiguous and do not tell a picture of Broken Britain.

Now I know this topic winds people up.  I’m ducking out till I’ve received Chris’s book.


Favourite comments so far.


have lived in the same part of inner city Manchester for about 18 years now. An area of multiple deprivation, high crime, gangs, guns, drugs, all the rest. I drink in rough pubs, walk around on my own late at night, know a few dodgy characters.I feel vastly safer now than I did 15 – 18 years ago. It’s ages since anyone I know was mugged, whereas I used to hear about it all the time. I see fewer fights and puddles of blood in the road than I used to.

Edict of Nantes

You are called Giles, I suspect you don’t live in a place that is particularly violent

And once they stop fulminating about my (admittedly posh) name, reasonable Paul , who thinks I may have made a better fist of it, says


Can i suggest you re-read the final para of my post.I said the fear of
crime was actually greater than the risk,And when i asked what was
fuelling the fear-maybe the media?-i meant with regard to the media,s
possible role in fuelling unnecessary there are not murderers,
rapists, child abusers etc lurking in dark alleys in every community

Finally, I think ‘testy’ puts it far better than I have

I hope you understand how meaningless this sought of study is. The category ‘violent crime’ is simply too broad in scope and definition to mean anything congrete. Tolerance of aggression and violence is constantly shifting dragging our conception of ‘violent crime’ along with it. So where you say that ‘violent crime’ has increased by 44%, I say what is the character of today’s ‘violent crime’ versus the ‘violent crime’ of ten or even twenty years ago.

Oh, finally finally.  This comment is so thorough on the BCS that I am putting it here just to bookmark it.


5 thoughts on “Me, elsewhere, ducking below the parapet

  1. Oooooh. no.

    “But I think murder rates are reasonably unambiguous and do not tell a picture of Broken Britain.”

    Murder rates are terribly ambiguous. They depend upon hte efficacy of traumatic health care. Something which has been increasing very rapidly in the past couple of decades (in the UK largely as a result of what we learnt in NI but that’s another story).

    Just look at the differences in dead and wounded ratios in war now…..

    1. Well, yes and no. Whether someone actually dies matters; you don’t want infant mortality measured by how many kids would have died given constant technology. If your intention with murder rates is to identify the trend in attempted murderous acts then you have more of a point. But then the ability to save the victim improving matters for the cases where you find a saveable victim, not, say, what happens with the prostitute killer of Ipswich. So the improvement in that ratio can only affect a proportion

      Finally, while battlefield injured:dead ratios will be sometimes relevant, over the longer term a lot of this will be down to changes in technology that don’t have a direct bearing on civilian homicide, surely. The wounded becoming gangrenous, for example: murder through stabbing that turns into gangrene is not a big part of the stats now, but in faraway and hard to reach battlefields it may have been.

      So while I think this factor is worth noting, I don’t think it moves it beyond my hurried phrase of ‘reasonably’ unambiguous

  2. Well…..OK, I know this is a newspaper piece but I’d say that a 70% drop leads to some ambiguity:

    SPRINGFIELD, Mass., Aug. 11 (AP) — Improvements in emergency care over the last 40 years have helped to lower the death rate among assault victims by nearly 70 percent, a new study says.

    “People who would have ended up in morgues 20 years ago are now simply treated and released by a hospital, often in a matter of a few days,” said Anthony R. Harris, a sociology professor at the University of Massachusetts at Amherst who led the study, published in the journal Homicide Studies. Researchers from Harvard Medical School also took part in the report, which looked at crime data from 1960 to 1999.

    Deaths from criminal assaults dropped almost 70 percent over the 40 years, with annual declines of 2.5 percent for firearm and knife assaults and 3.5 percent to 4 percent for other assaults, including poisoning and arson, the researchers found.

    In 1960, the police nationwide recorded 9,110 homicides and 154,320 aggravated assaults. That translated into 5.1 homicides per 100,000 people and 86.1 assaults per 100,000. The researchers found 5.6 percent of those assaults ended in death.

    In contrast, in 1999 the police recorded 15,522 homicides, a rate of 5.7 per 100,000, and 911,740 aggravated assaults, a rate of 334 per 100,000. But only 1.67 percent of the 1999 assaults ended in death.

    The researchers noted that there was a slight increase in deaths from assaults in 2000 and 2001 and said they were studying it.

    Their study credited a variety of medical advancements for the improved death rate, including the development of 911 services, rapid stabilization and transportation of trauma victims, better training for emergency medical technicians, and more hospitals and trauma centers, particularly in rural areas.

    Without the medical advances, the researchers calculated, 45,000 to 70,000 homicides would have been recorded annually nationwide.

    The results are consistent with common sense, said Dr. Stephen R. Thomas of the Harvard Medical School, a specialist in emergency medicine who worked on the study. Advances in trauma care since the Vietnam War have increased survival rates for a wide array of injuries, Dr. Thomas said.

    He said that the study could raise questions about how crime statistics are analyzed, and that researchers should consider whether medical care has improved when assessing local changes in crime rates.

    1. Agreed, acknowledged. I need to take this into account in my thinking. In America in particular this must be significant. Though, note, not the reason for 1the 1995-2009 LA figures, I would say.

  3. Given that the paper addresses data up to 1999 you’re certainly correct there. But if we’ve seen a decades long trend we wouldn’t be terribly surprised to see it continuing for a further 15 years….

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s