What happens to deathrow inmates found innocent?


3:59 pm - February 25th 2013

by Shantel Burns    


      Share on Tumblr

Since the death sentence was reinstated in the US in 1976, 140 people on death row – one in ten – are exonerated after their innocence has been proven.

What do the people who were wrongly imprisoned, sometimes for decades, on death row, say once they are found innocent?

One for Ten is a series of interactive, short documentary films which will tell the stories of ten individuals, from all across the US, that have been freed from death row after their innocence has been proven.

The films – which will be produced and broadcast over five weeks between April and May 2013 – aim to be an interactive documentary series by using social networking, user generated content and a strong media and charity coalition to allow others to be as involved as possible.

All films will be free to watch and share and will be made ‘live’, shot one day, edited the next and then uploaded online for people to comment, pass on and contribute with interview questions, music and artwork.

The team will be travelling across the US, from New York to California, in order to interview people found innocent and produce a new short film for release online every Tuesday and Friday.

Each of the films will look at a different failing in the American capital justice system that led to the wrongful conviction and incarceration of each of the individuals interviewed as part of ‘One for Ten’.

The pilot film shows ex-death row inmate Ray Krone talking about his time on death row and the injustice he was faced with during his trial:

‘One for Ten’ have just launched their crowd funding campaign to raise the much-needed funds to get the project on the road, for which you can donate here.

    Share on Tumblr   submit to reddit  


About the author
Shantel Burns is a News Editor at Liberal Conspiracy, and a publishing and journalism student and current affairs nerd. Blogs at: ramblepolitics.blogspot.co.uk too.
· Other posts by


Story Filed Under: News

Sorry, the comment form is closed at this time.


Reader comments


I simply cannot imagine the emotional turmoil this must cause, being wrongly convicted of an offence and then spending possibly years fearing for your life – just knowing that you are going to be publicly executed for a crime you did not commit. Then after all of the anguish caused – becoming a free man (or woman) and being left to fend for oneself outside of prison. Already institutionalised these men will have one hell of a job adjusting to the outside world. Innocent prisoners on the UK (of which I am sure there are many) at least do not have the threat of being killed hanging over their heads.

An worthy project. The hotly disputed morality of societal killing aside, the most fundamental problem with the death penalty for me has always been that mistakes cannot be corrected once it has been carried out. Having people on death row for years wondering when their day comes, depending at best on appeal processes and at worst on political expediency, is another weird side effect of this inherent contradiction of having the death penalty. It does nothing for closure for the victims’ relatives, which the death penalty claims to provide, but does not protect the innocent in all cases either.

I’m opposed to the death penalty on principle – whether people are guilty or not.

The irony is that those exonerated would have not have been found innocent if they hadn’t had nothing to lose.

It’s not unreasonable to assume that a similar proportion of innocent people are serving lesser sentences having plea-bargained their freedom away for a light sentence for a crime they did not commit, and that there are many, many more hoping that if they profess contrition for something they didn’t do they might get early parole.

140 is less than 1 in a million in the USA (still far too high, of course). The film *ought* to analyse the number of innocents *actually executed* which is zero or tiny. The campaign to clear Hanratty of the A6 murders ended with DNA testing proving his guilt.
And what about all the innocent victims of murder in the UK, like the teenager stabbed to death because the contract killers picked a house in the next street to their intended victim? Or the child killed while cycling home when two gangs decided to shoot it out across his path? Or the thousands of others?

Shatterface is right – plea bargaining is an instance of the injustice inherent in the US court system (not, just not, justice system) where rich and powerful DAs can bully the innocent into accepting punishment for fear of something worse to bolster the DA’s political career. Whatever you may say about the UK’s legal saystem it is far better than having elected judges and DAs

And what about all the innocent victims of murder in the UK, like the teenager stabbed to death because the contract killers picked a house in the next street to their intended victim? Or the child killed while cycling home when two gangs decided to shoot it out across his path? Or the thousands of others?

Yet despite executing vast numbers of (mainly black, poor and/or mentally disabled) criminsls the US has a vastly higher murder rate than the UK – which is pretty much conclusive evidence it is fuck all use as a deterent.

And since many of those who kill immediate members of their family then go on killing sprees to commit ‘suicide by cop’ or turn their guns on themselves giving people fuck all to loose increases the number of killings rather than reduces them.

And, by the way, your first example is fucking bizarre: they killed the wrong person – so if we had the death penalty they’d have been more careful and got the right guy?

Wtf?

7. So Much for Subtlety

2. Ben

An worthy project.

If they are honest about it.

The hotly disputed morality of societal killing aside, the most fundamental problem with the death penalty for me has always been that mistakes cannot be corrected once it has been carried out.

How can any mistakes be corrected? How can you give a man twenty years of his life back? You can’t. What you mean is that you do not have the stomach for thinking about it in one case because death is concrete, but you can’t imagine twenty years inside and so you’re fine with that.

Having people on death row for years wondering when their day comes, depending at best on appeal processes and at worst on political expediency, is another weird side effect of this inherent contradiction of having the death penalty.

No, it is a weird side effect of activists who deliberately clog up the appeals system in an effort to prevent any executions. They could execute them the day after the sentence if they liked – no waiting.

It does nothing for closure for the victims’ relatives, which the death penalty claims to provide, but does not protect the innocent in all cases either.

I would love to see the system that protects the innocent in all cases – not that these people are often innocent, just innocent of the specific crime they are accused of. The last man executed in Britain took part in a jail break and would have been automatically guilty of murder under the old felony murder rule even if he was not there. He was not innocent by any sane definition. As for the victims, who are you to speak for them? How do you know? I would expect that it gives the relatives a hell of a lot more peace of mind and closure than having to come back in eight years time to argue, usually fruitlessly, against parole.

6. Shatterface

Yet despite executing vast numbers of (mainly black, poor and/or mentally disabled) criminsls the US has a vastly higher murder rate than the UK – which is pretty much conclusive evidence it is fuck all use as a deterent.

Except Japan executes and Japan has an even lower murder rate than the UK. Which simply proves that you cannot compare dislike cultures. America has a culture of violence. Britain does not. Thus British people do not murder – although they murder a hell of a lot more since the death penalty was abolished. The famous study from Emory University showed that every execution deters between eight and twenty eight murders with an average of eighteen. Thus it works. If not in the general, then at least in the specific case.

And since many of those who kill immediate members of their family then go on killing sprees to commit ‘suicide by cop’ or turn their guns on themselves giving people fuck all to loose increases the number of killings rather than reduces them.

Actually I am not sure that is true but if it is, it saves us all a lot of money.

The death penalty works. The fact is liberals are simply so spineless they cannot bring themselves to do the right thing and so allow an endless parade of psychopaths out of prison to kill, rape and maim once more. There is no reason to do this. We don’t have to.

@ #6 Shatterface
Contract killers were not a feature of the British underworld when we had the death penalty. So it is not bizarre.
The data that Roy Jenkins presented to justify abolishing the death penalty failed to show that the suspension of the death penalty had any significant impact on the number of “non-capital” murders. However it did show a significant increase in the number of “capital murders” – statistically significant at the 0.0001% level. The increase in the number of innocent people murdered in the trial period was greater than the total number of people hanged since 1951 (Jenkins chose not to supply the number hanged under the Attlee government).
I am not sufficiently expert on the US situation to comment authoritatively thereon, but it is crystal clear than the death penalty acted as a deterrent in the UK.

“The film *ought* to analyse the number of innocents *actually executed* which is zero or tiny.”

A remarkable statistic. Do you have a source?

(Jenkins chose not to supply the number hanged under the Attlee government).

Are you suggesting this is a secret? Because that does strike me as somewhat unlikely.

Many people are for the death penalty, many against.
It’s just a private matter of morality or judgement IMO. Not really something for public campaigns – unless there are serious flaws in the justice process.

A lot of people in Bangladesh want those convicted of committing atrocities during the Bangladesh independence struggle to be excecuted. I don’t have any problem with that if it’s a popular choice.
I think it’s OK that India executed that guy who was convicted of taking part in the Mumbai massacre. I don’t see any particular worthiness of wanting to save his sorry excuse for a life. Now that he has been dispatched to history, life still goes on and a line can be drawn under that episode.
IF this is the popular will of the people.

Europe is different to the USA, and we should respect that difference. President Obama, who is considered by many to be a decent man, presides over the death penalty. This is a direct quote from him:

I believe that the death penalty is appropriate in certain circumstances. There are extraordinarily heinous crimes, terrorism, the harm of children, in which it may be appropriate.

I think I agree with him.
http://2012.presidential-candidates.org/Obama/Capital-Punishment.php

12. the a&e charge nurse

On the flip side the chingford skinhead says “I have kept track year by year since the death penalty was suspended then abolished of the number of people who have been killed by persons previously convicted of homicide. It has averaged three people a year. About 150 people killed because their killers have been freed to kill again”.
http://blogs.telegraph.co.uk/news/normantebbit/100181673/a-return-to-capital-punishment-it-is-time-we-thought-once-more-about-the-deterrent-effect/

13. Ultimo Tiger

@7

You do know Japan executes very few people don’t you?

In fact, they sometimes let people like this man out of jail:

http://en.wikipedia.org/wiki/Issei_Sagawa

You might also want to factor in:

1-Japense culture
2-The allegation that Japanese police are less than honest in how they get people to confess.

“The death penalty works.”

So why does Texas need to execute so many people?

And is it working in places like India and Pakistan? Or China?

“it saves us all a lot of money.”

Well, using that logic, NHS Mid Staff’s is doing a great job.

By the way, before you accuse me of being a “spineless liberal”, I’m slightly to the right of Ian Smith and would quite happily s*** on the breakfast of most people on this website.

14. Robin Levett

@john77 #8:

Contract killers were not a feature of the British underworld when we had the death penalty

Interestingly, the last person to be sentenced to death in the UK or dependencies was Anthony Teare – who was, you guessed it, a contract killer.

As for whether the death penalty deters; comparing the UK before and after the Chatterley trial is hardly going to produce any sensible figures.

You can though compare different populations within the same country with and without the death penalty; in the States. Those without the death penalty have lower homicide rates than those that do; so the stats don’t bear you out.

@ Robin Levett
I have looked at the stats that Jenkins produced. Have you?
May I suggest that you should do so before saying the stats do not bear me out?

By the way, before you accuse me of being a “spineless liberal”, I’m slightly to the right of Ian Smith and would quite happily s*** on the breakfast of most people on this website.

Still to the left of So Much For Sanity, though.

@ #10 Jimmy
*You* try to find the numbers!
Not a secret but it didn’t suit Jenkins to draw attention because they were higher in six years of Labour government than in thirteen years under the Conservatives [this can be deduced from articles about Albert Pierrepoint even if the actual number cannot].

@john77

I did. It wasn’t very hard. The mumbers were higher because prior to the Homicide Act 1957 the death penalty was mandatory for murder. After 1957 most murders were punishable with imprisonment.

Those without the death penalty have lower homicide rates than those that do; so the stats don’t bear you out.

I think you can still argue a cultural difference between north and south – as in, the death penalty’s not the only reason that Texas has more murders than Vermont.

Wasn’t there a state that abolished and then re-instated the death penalty? That presumably would have something useful to say about its efficacy.

Hmm. Turns out that all US States temporarily had to abolish the death penalty after a Supreme Court ruling (the Furman ruling). There was then a four year moratorium before the Supreme Court ruled on new death penalty legislation, after which 38 states resumed executions.

So we have an ideal way of checking whether capital punishment works in theory (and whether actual executions work in practice) to deter crime. There’s been a paper on it as well.
http://www.sfu.ca/~allen/deter.pdf

I’m not a statistician (and for what it’s worth I’m opposed to the death penalty on good solid idealistic and philosophical grounds) but it would seem clear from that paper that the death penalty has a significant deterrent effect, and that executions have an even stronger one. In the US, the death penalty deters.

“And what about all the innocent victims of murder in the UK, like the teenager stabbed to death because the contract killers picked a house in the next street to their intended victim?”

The way you write that it sounds as if you think had the killers been competent and got the right person it wouldn’t have been an innocent victim. A bit like the phrase ‘innocent bystanders’, its somewhat unfortunate.

I seriously doubt whether the death penalty would have had any effect on this case. The killers were drug addicts, persistently in trouble with the law, unemployable and by all accounts would struggle to reach the IQ level required even by Texas.

What would have made a difference would have been longer sentences for their earlier violent offences (which in one case was actually domestic violence which is notoriously under punished) combined with a rigorous drug rehabilitation and education programme aiming at turning them into useful members of society. I think the right would have supported the first part of that, but I doubt whether they’d support taxpayers having to spend money to achieve the latter.

@ #18 Jimmy
Go on then – tell us!
” The mumbers were higher because prior to the Homicide Act 1957 the death penalty was mandatory for murder. After 1957 most murders were punishable with imprisonment.”
NOT TRUE.
Firstly: prior to 1957 the death *sentence* was mandatory. The death penalty was only imposed if the conviction was upheld on appeal and appeal for to the Home Secretary was rejected. Execution was a semi-judicial semi-political decision and the Labour Home Secretaries approved more executions than the Conservative ones.
Secondly, the change in legislation in 1957 after six years of Conservative government could only have been the cause for more executions under the six years 1945-51 of Labour government than under 13 years 1951-64 of conservative government if the number of executions after that was negative. That sort of transparent disingenuity (aka bullshit) does nothing for your credibility.

@ 21 Planeshift
You are misinterpreting – the intended victim was also innocent. I gave those examples because they had recently hit the headlines and not even Sally would dispute their innocence.
As to your last comment – I am under the impression that right-wingers are more keen to wean addicts of “dangerous drugs” and left-wingers more inclined to legalise drugs and feel less need to wean addicts off them; also right-wingers are more enthusiastic about spending money on “law and order”. Admittedly there are a number of mean selfish individuals masquerading as right-wingers as they think that gives them an excuse for failing to contribute to society but you will find a greater number (albeit usually – but not always eg Jimmy Carr, Kinnock junior, Rothschild junior, Margaret Hodge – with less money) on the left who also make excuses for avoiding contributing.
Your habitual attempt to depict “the right” as evil and “the left” as virtuous is irrelevant to this debate.

@22

I think you’re trying to make bricks without straw here. The rate of executions under Churchill is not significantly different to that under Attlee, the rate only drops later. The Commons under Attlee in fact passed abolition, only the Lords saved the rope. Rather than fight the Lords, Chuter Ede set up a Royal Commission which reported in 1953, the year Bentley was executed and Evans was posthumously cleared. Rather than give in to the abolitionists, the government opted to reduce death sentences rather than scrap them.

“There’s been a paper on it as well.”

More than one I think you’ll find.

@ Jimmy
IF you know the numbers, what are they?
What are you hiding?

@ #14 Robin Levett
Your example is, if not deliberately misleading, misleading through gross negligence. Anthony Teare was not even born when the death sentence for murder was abolished in the UK.
He could be usedas an example by those in favour of the restoration of the death penalty.

@ 24 Jimmy
“Rather than give in to the abolitionists”
What?!? There was not then, and probably never has been, a majority in the country for abolition. Even adding together the handful of Tory and Liberal abolitionists to the whole of the Parliamentary Labour (whipped) woulds not have got sufficiently near a majority to embarrass the government.

@john

Sorry I missed that. I believe it was 108, including 3 for treason.

“There was not then, and probably never has been, a majority in the country for abolition.”

True, which was why Attlee did not invoke the Parliament Acts to ram it through the Lords.

@ #29 Jimmy
Thank you. One step forward. Executions determined by a pro-abolitionist Labour Home Secretary were far higher than under Conservative Home Secretaries. One of the arguments for having a Conservative government is that Labour backbenchers prevented Michael Howard from being more than half as illiberal as David Blunkett.
@ #24
I’m not making bricks without straw*: you’re pretending to make bricks without clay. You are trying to ignore facts that do not suit your viewpoint

*although that is the normal procedure in England

@john77.

I’ve taken the trouble to collate the stats for all 20th C Home Secs up to the abolition of the mandatory penalty. These are for civilians executed for murder in England and Wales. You’ll see they don’t bear out the party political point you seek to make:

Prior to the law changing in March 1957 the numbers are as follows

Butler: 0 in 2 months
Lloyd George: 13 in 27
Maxwell Fyfe: 51 in 36
Chuter Ede: 98 in 74
Somervell: 0 in 2
Morrison: 58 in 55
Anderson: 12 in 13
Hoare: 19 in 27
Simon (2): 14 in 24
Gilmour: 26 in 32
Samuel (2): 10 in 13
Clynes: 16 in 21
Joynson-Hicks: 70 in 55
Henderson: 6 in 9
Bridgman: 18 in 15
Shortt: 52 in 33
Cave: 22 in 25
Samuel (1): 4 in 11
Simon (1):10 in 8
McKenna: 52 in 43
Churchill: 22 in 20
Gladstone: 54 in 50
Akers-Douglas: 71 in 29
Ritchie: 24 in 20

“Contract killers were not a feature of the British underworld when we had the death penalty. So it is not bizarre”
The Krays
“But it is crystal clear than the death penalty acted as a deterrent in the UK.”
Did it act as a deterrant in Georgian England

@ #31 Jimmy
Thanks
I was *not* trying to make a party political point – I was merely moaning about the dishonesty of Roy Jenkins in the abolition debate. Unless you mean my claim that Labour backbenchers in opposition are better at holding governments to account because most Tory backbenchers feel it is, or seems to be, lacking in patriotism to attack the government (the few exceptions are those like Admiral Keyes in 1940 and Douglas Hogg in the Iraq War whose patriotism, and that of their families, is above suspicion).
The numbers *do* show that executions were at a higher rate under Labour which, as I said, didn’t suit Roy Jenkins. If I was picky I should point out that executions were lower when Churchill was Home Secretary but I realise that you were referring to the period when he was PM not when he was making the decisions on clemency.

34. So Much for Subtlety

13. Ultimo Tiger

You do know Japan executes very few people don’t you?

How is that even remotely relevant? They don’t need to. They punish criminals. Thus they have few criminals. They punish by, among other things, executing them. Thus Japan proves that you cannot make simplistic arguments against the death penalty.

You might also want to factor in:

1-Japense culture

As that was my point, I am not sure I would bother to do it again. If you can’t be bothered reading what I wrote, I see no need to repeat myself. But I will – murder rates are determined by culture first and foremost.

2-The allegation that Japanese police are less than honest in how they get people to confess.

I am sure. Which suggests that perhaps we ought to copy them. When British police verballed criminals we had a lot less crime. There is a case here for a blind eye.

So why does Texas need to execute so many people?

Because it is full of people who think murder is an appropriate solution to their problems. Presumably they would murder more if they did not have the death penalty.

And is it working in places like India and Pakistan? Or China?

Probably. We would have to check by asking them to suspend the death penalty for a few years to see if crime goes up.

Yet despite executing vast numbers of (mainly black, poor and/or mentally disabled) criminsls the US has a vastly higher murder rate than the UK – which is pretty much conclusive evidence it is fuck all use as a deterent.

Of course it is no such thing.

36. Robin Levett

@john 77 ’15:

I have looked at the stats that Jenkins produced. Have you?
May I suggest that you should do so before saying the stats do not bear me out?

Way to miss my point.

One of the few things I agree with SMFS on is that murder rates are a function of the society you are looking at. Comparing murder rates before and after abolition is not comparing like with like; early 1960s UK society was very different from late 60s, early 70s, late 70s etc. That period was one of enormous social change. Trying to tease out any deterrent effect from the numbers you have is not as easy as simply comparing murder rates before and after abolition.

@ #36 Robin Levett
How many times do I have to repeat that Doctor Who is fiction?
The data on capital and non-capital murder presented before Parliament debated the abolition of the death penalty did NOT include any data from the 70s, 80s and 90s. So your claim that 70s were different is simply a red herring. If one wanted to bring in societal change one would expect the mid-60s to show a continuation of the decline in the murder rate due to increasing affluence and continuing distance from the two world wars and the Korean War. The data was as like-for-like as practicable and the choice of the time zones should have helped the abolitionist case instead of thoroughly refuting it.
Your so-called point that there are different murder rates in different areas of the USA is another red herring since I specifically stated in #8 that I was referring to the UK where the evidence is crystal clear.

38. Robin Levett

@john77 #27:

Your example is, if not deliberately misleading, misleading through gross negligence. Anthony Teare was not even born when the death sentence for murder was abolished in the UK.

Certainly on the mainland; but Teare was tried, convicted and sentenced on the Isle of Man, where the death penalty was mandatory until 1993. He killed a 22-year old woman with a mental age of 15. While, from abolition in the UK until then, successive Home Secretaries had invariably commuted death sentences to life imprisonment, it wasn’t automatic. Teare was a contract killer who was risking capital punishment.

@ #38 Robin Levett
Firstly, you are repeating your error after I have pointed out that you are wrong. The Death *Sentence*, not the death penalty, was mandatory and the press, if no-one else, expected Teare to be sent to jail instead – which he was.
Secondly the murder rate in the Isle of Man prior to formal abolition of the death penalty was about one in a dozen years. The UK homicide rate peaked at nearly one thousand every year and has since declined to 550. The population of the Isle of Man is more than one thousandth of that of the UK. Not all homicides are murder and not all murders faced the death sentence after 1957 but we have a frequency per head difference of around sixteen times! [The murder rate in IoM has risen since 1994, albeit not to UK levels].

The argument here appears to be a classic post hoc ergo propter hoc fallacy. Increasing homicide rates are far mnore likely to be caused by other factors, drugs being perhaps the most obvious. The argument might be stronger if the numbers for non capital murders remained constant after 1965, but they didn’t. They rose too.

@ #40 Jimmy
But not significantly – which was Jenkins’s excuse for abolishing the death penalty for capital murders.

42. Robert A. Fouts

How many innocent people have been executed? How many people were not exonerated and set free because they had already been executed?
In the ancient code of Hamarabi the Judge in the case would have been made subject to the same sentence.

43. Chaise Guevara

@ 42

“In the ancient code of Hamarabi the Judge in the case would have been made subject to the same sentence.”

Good thing we don’t live under that code, then, given that the US uses jury trials, and that generally failing to do your job perfectly is not seen as a capital offence.

44. So Much for Subtlety

36. Robin Levett

One of the few things I agree with SMFS on is that murder rates are a function of the society you are looking at. Comparing murder rates before and after abolition is not comparing like with like

Sorry but it is, or rather it depends on which abolition you are looking at. It is the closest we can get to looking at a pure change with no other factors at any rate.

early 1960s UK society was very different from late 60s, early 70s, late 70s etc. That period was one of enormous social change.

I am not sure. It may be true that mass immigration has made a huge difference – as we can see with the Chris Hulne trial where the jurors clearly did not understand either basic English or the basis on which a jury works – a jury with just two White people. But the changes in the 1960s were probably cosmetic at first and took a while to work their way through society. Most people just carried on as they had. By the late 70s? Perhaps. But crime and especially murder kept rising and rose from Day One of abolition – even before that. So that is a big cultural shift. We said we did not care that much about murder and so more people did it.

Trying to tease out any deterrent effect from the numbers you have is not as easy as simply comparing murder rates before and after abolition.

True up to a point. Although I get the feeling if the numbers worked your way you would be saying something else. However the nice people at Emory did that – every execution deters an average of 18 murders in the US. We should be executing too.

45. Robin Levett

@SMFS #44:

early 1960s UK society was very different from late 60s, early 70s, late 70s etc. That period was one of enormous social change.

I am not sure.

Having lived through the period, I am.

46. Robin Levett

@john77 #39:

Firstly, you are repeating your error after I have pointed out that you are wrong. The Death *Sentence*, not the death penalty, was mandatory and the press, if no-one else, expected Teare to be sent to jail instead – which he was.

I was right the first time, and you wrong. In Manx law, the death penalty was mandatory; it was only the exercise of the Royal Prerogative by successive Home Secretaries that prevented executions in general on Man, and Teare’s execution in particular.

The press expected that, after the event; but deterrence doesn’t work on the press, after the event. Would you carry out a particularly unpleasant murder relying entirely upon a politician being prepared to reject the red-top demand for your blood?

Secondly the murder rate in the Isle of Man prior to formal abolition of the death penalty was about one in a dozen years. The UK homicide rate peaked at nearly one thousand every year and has since declined to 550. The population of the Isle of Man is more than one thousandth of that of the UK. Not all homicides are murder and not all murders faced the death sentence after 1957 but we have a frequency per head difference of around sixteen times! [The murder rate in IoM has risen since 1994, albeit not to UK levels].

FWIW the UK homicide rate peaked at a little under 1,000 per year; but without Dr Shipman, it would’t have gone over 800 or so, and would have plateaued between 2001 and 2005 at a little short of 800.

You seem to be arguing against yourself somewhat here. You have argued that the deterrent effect of capital punishment prevented the existence of contract killers before abolition, and that Teare isn’t a counterexample because while the death penalty hadn’t formally been abolished on IoM when he committed his murder, it had effectively been abolished many years earlier by the exercise of the Royal prerogative by UK Home Secretaries.

Now you seem to be suggesting that the IoM murder rate has drifted upwards since formal abolition because of formal abolition.

Did the formal existence of the death penalty on the isle of Man up to 1994 deter or not? If so, how do you account for Teare, a contract killer. If not, how do you account for the rise in Manx murder rates since 1994.

Come to that, the UK murder rate is now in the mid-500s, having been in the mid-300s in the last years before final abolition in the rest of the UK; and is now far closer to that rate than to the peak plateau rates. What is the “deterrence” narrative here?

@ #46 Robin Levett
If the death *penalty* was mandatory why were none of the three murderers hanged? Judges are mandated to *sentence* people.
As to Shipman, firstly you need to learn to do arithmetic a bit better: 215 deaths in 24 years is approximately 9 per annum, so does not reduce an annual rate of 1,000 to 800. Secondly even if your wildly incorrect figure had been correct that would only reduce the mainland ratio from 16 times to 13 times the Manx ratio.
There are two parts to deterrence, as the erudite paper on US experience explains, the stated policy and actual executions. It shows that both have a statistically significant effect on the number of murders. Rather than my writing several paragraphs why don’t you just read it?
The current UK murder rate is still more than 50% higher than it was prior to abolition, but the rate for “capital murders” is a multiple of the pre-abolition rate. You are using Jenkins’ ploy of pretending that the existence of the death penalty should be judged on the rate of non-capital murders which is transparently suggestio falsi.
You say that you lived through the sixties and seventies: so did I and the greatest part of social change came in the 70s *after* the abolition, but gun crime rose sharply during the suspension period which can be directly attributed to the suspension of capital punishment for murder using firearms. If you are ascribing the rise in murder rates to the “sexual revolution” you are stretching credibility beyond breaking point. The secular trend of capital murders prior to 1965 was downwards. It is verging on the ridiculous to suggest that social change between the first half of the 1960s and 1965-8 was the sole cause for such a rise in murders (in some sub-categories an infinite percentage rise).

48. Robin Levett

@john77 #47:

Just a very quick one for now; the ONS say that Shipman’s murders were recorded in 202-3, and were responsible for the spike that year:

Figure 7 and Table 2.01 show the numbers of homicides, as taken from the Homicide Index, for the last 50 years. Homicides increased steadily from 1961 up to the early 2000s (the peak in 2002/03 includes 172 homicides committed by Dr Harold Shipman). The number of homicides has shown a general downward trend since 2002/03.

The above is taken from page 24 of the ONS publication “Focus on: Violent Crime and Sexual Offences, 2011/12″

http://www.ons.gov.uk/ons/dcp171778_298904.pdf

Just to note, if you cite Ehrlich (1975) or Dezhbakhsh and Shepherd (2003), who use his methodology, in a death penalty context, you’ve been reading too many dishonest far-right website. And if you call them “erudite”, you’re clearly a gibbering fuckwit who’s never been within 100km of a university.

The vast majority of criminologists accept the conclusion, put well by Donahue and Wolfers, that there is no evidence justifying *any* concrete conclusions on whether the death penalty deters or encourages crime, and that anyone who claims otherwise is laughably biased.

The current UK murder rate is still more than 50% higher than it was prior to abolition, but the rate for “capital murders” is a multiple of the pre-abolition rate.

Eh? “non-capital murders” only started to exist as a category in 1957. Before that, all murders were capital. And yet there was no rise in the murder rate from 1957-1961.

Anyone who claimed the maximum sentence for robbery affected the frequency of adultery would be laughed to scorn.
John B and Robin Levett deserve the same

If robbery and adultery had been legally treated as the same offence from time immemorial until 1957, and also from 1965 onwards, with only an eight-year period 50 years ago when they were regarded as different offences, then yes, it would be completely reasonable to hypothesise that sentencing for one would affect perceived sentencing for the other.

@ #52 John B
A sudden attempt to pretend you are talking about Victorian sentencing policy when the debate is about deterrence. Trying to muddy the waters to hide the fact that the statistical data on the rise in in “capital” murders during the “suspension” (in practice early abolition) of the death penalty is clear and unambiguous. While the change in “non-capital” murders was not significant the rise in “capital” was significant at the 0.00001% level.
If you don’t believe that punishments deter, do you propose to abolish all fines and prison sentences, restore Fred Goodwin’s knighthood, abolish the ICC and War Crimes legislation etc?

54. Robin Levett

@john77 #47:

If the death *penalty* was mandatory why were none of the three murderers hanged? Judges are mandated to *sentence* people.

And prison authorities and executioners are mandated to follow the orders of the Court; absent intervention, the 3 murderers would have been hanged. None of them was hanged because (for the third time) successive Home Secretaries intervened and exercised the Royal Prerogative to commute the sentence to life imprisonment. But that exercise is not as of right, it is wholly discretionary. If the law had taken its course, 3 murderers would have been executed.

As to Shipman, firstly you need to learn to do arithmetic a bit better: 215 deaths in 24 years is approximately 9 per annum, so does not reduce an annual rate of 1,000 to 800.

Already dealt with in my earlier comment today; take this up with the ONS…

Secondly even if your wildly incorrect figure had been correct that would only reduce the mainland ratio from 16 times to 13 times the Manx ratio.

Demonstrating that the issue is one of societal differences.

There are two parts to deterrence, as the erudite paper on US experience explains, the stated policy and actual executions. It shows that both have a statistically significant effect on the number of murders.

The last execution on the Isle of Man for murder was in 1872; how much deterrent effect did that have in, say 1992?

It is fair to say that at that time, Manx law required the Crown positively to order execution for the death sentence that had been passed to be effected – but it was changed because Victoria didn’t like that.

As for Dezhbakhsh and Shepherd (the “erudite paper”), their analysis is pretty poor, as shown by the Donahue and Wolfers paper that John B has referred you to. I’d started reading D&S, but could see no obvious logic to their methodology, so put it aside until I could actually devote a little more thought. Tbh, D&W seems to trash D&S by applying D&S’s methodology to states that had had no death penalty either before or after the moratorium and coming up with the same evidence of deterrence…

Rather than my writing several paragraphs why don’t you just read it?

I had started; but I am sure that you’d draw my attention to the relevant parts. While you’re at it, a reference for the Roy Jenkins document that you’re castigating would be welcome.

The current UK murder rate is still more than 50% higher than it was prior to abolition, but the rate for “capital murders” is a multiple of the pre-abolition rate.

Is it? So far as I am aware, capital murder only existed for eight years. That isn’t long enough to develop reliable statistics. Again, there is no certainty that what we might now classify as equivalent to capital murders would have been so classified at the time such a concept had legal existence, for two reasons: firstly, classification itself is/was not an exact science, and secondly (for obvious reasons) any doubts as to classification would be resolved in favour of non-capital murder.

You are using Jenkins’ ploy of pretending that the existence of the death penalty should be judged on the rate of non-capital murders which is transparently suggestio falsi.

Nope; I am denying the existence of “capital murder” as a category upon which to base statistics.

You say that you lived through the sixties and seventies: so did I and the greatest part of social change came in the 70s *after* the abolition, but gun crime rose sharply during the suspension period which can be directly attributed to the suspension of capital punishment for murder using firearms. If you are ascribing the rise in murder rates to the “sexual revolution” you are stretching credibility beyond breaking point. The secular trend of capital murders prior to 1965 was downwards. It is verging on the ridiculous to suggest that social change between the first half of the 1960s and 1965-8 was the sole cause for such a rise in murders (in some sub-categories an infinite percentage rise).

I disagree; there was further significant social change into the 1970s, but to suggest that it eclipsed social change during the 1960s is frankly ludicrous. To take but one example, household car ownership (of one or more cars). From 1961 to 1971 it went from 31 to 52%; 1971 to 1981 it went on to 60%. (https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/71279/tsgb-2012.pdf)

Capital punishment stats over such a short period have little validity because of inherent variability, as was universally recognised in the 1960s debates on capital punishment; I don’t say that societal changes solely led to the increase in capital murders. I say that there was a constellation of factors (including societal changes), among them being some question of whether there was a real rise in the underlying rate that needs explaining, given the small numbers and variability involved.

55. Robin Levett

@john77 #53:

If you don’t believe that punishments deter, do you propose to abolish all fines and prison sentences, restore Fred Goodwin’s knighthood, abolish the ICC and War Crimes legislation etc?

I think you’ll find that both John B and I accept the general proposition that punishments deter; but if you were paying attention, you’d know that that isn’t the issue. The issue is whether capital punishment deters murder more than does imprisonment or any other option.

56. Robin Levett

@john77 #53 (contd):

…and could you produce the analysis in support of this:

While the change in “non-capital” murders was not significant the rise in “capital” was significant at the 0.00001% level

57. Robin Levett

@Tim J #20:

You ought to read the Donahue & Wolfers paper that John B references at #49; it does rather trash Dezhbakhsh and Shepherd – for example, by showing that the evidence for deterrence that they identify around the moratorium appears also to exist in states with no death penalty at any time during the period. Which is curious, if D&S are right.

@ #56 Robin Levett
Yes, given time to search – it was over forty years ago and the stuff, if I still have it, is in the loft. I can remember the most startling bits which include my not being able to calculate an exact answer because my statistical tables only ran to six figures after the decimal point and the number of three categories, including murders of prison officers – which is where the difference between execution and life imprisonment is most simply differentiated as a deterrent – had infinite percentage increases.
I have no idea if there is an interlink to a document that I read more than a dozen years before the commercial internet started up.

59. Shatterface

And since many of those who kill immediate members of their family then go on killing sprees to commit ‘suicide by cop’ or turn their guns on themselves giving people fuck all to loose increases the number of killings rather than reduces them.

Actually I am not sure that is true but if it is, it saves us all a lot of money.

So it doesn’t matter if people go on killing sprees before being shot or killing themselves because the money we save is more important than the extra deaths?

@ #55
” The issue is whether capital punishment deters murder more than does imprisonment or any other option.”
It has been demonstrated, beyond reasonable doubt, that it does. The issue is whether you are willing to accept that or whether you will continue to argue about the impact of the death penalty on those to whom it does not apply. Both of you persistently ignore inconvenient facts and waffle on in an attempt to conceal reality.
Deterrence is not uniform – e.g. Ian Brady wanting to die while Myra Hindley wanting to be released – and I, personally, think a genuine life sentence is crueller than execution, but I think the law should make protection of the innocent a priority.

61. Robin Levett

@john77 #58:

Yes, given time to search – it was over forty years ago and the stuff, if I still have it, is in the loft. I can remember the most startling bits which include my not being able to calculate an exact answer because my statistical tables only ran to six figures after the decimal point and the number of three categories, including murders of prison officers – which is where the difference between execution and life imprisonment is most simply differentiated as a deterrent – had infinite percentage increases.

Over 40 years ago takes you back to say 1970, only 5 years after temporary abolition began – our moratorium, except ours was followed by abolition, not reinstatement, in 1970. So you are calculating over only a 5 year period; or 3½, as Lord Gardiner pointed out in the abolition debate in 1969.

And prison officers? Between 1965 and 1969, I understand that only one prison officer was killed on duty – in a Borstal. There were far too many – one is too many – killed susbequently, but that wasn’t about capital punishment but about the PIRA.

There were 4 police officers killed during that period, all in 1966, in only two incidents.

That really isn’t going to produce anything approaching statistical significance; so I would be very interested in seeing your calculations showing virtual certainty that there was a marginal deterrent effect to capital punishment.

#60:

”The issue is whether capital punishment deters murder more than does imprisonment or any other option.”

It has been demonstrated, beyond reasonable doubt, that it does

I am reasonable, and I doubt it. Your trump card – Dezbhakhsh & Shepherd – seems to find some numbers to support a change in murder rates with changes in executions, but those rate changes also appear in states where there was no change in executions numbers because there were no executions.

The Canadian murder rate figures mirrored the USA ones, albeit at a lower level, until the late-1970s, but Canada hasn’t had the death penalty (with limited exceptions) since 1961, and haven’t executed anyone since 1962. When they did execute, however, they executed at 10 times the current Texas rate (per murderer). The change in the late-70s showed USA murders remaining high (with capital punishment and executiosn re-imposed) while Canadian murders (with no policy change) dropped.

All of that fails to take into account any effect on a jury of the existence of the death penalty; in the Lords abolition debate in December 1969 (http://hansard.millbanksystems.com/lords/1969/dec/17/murder-abolition-of-death-penalty-act), the Lord Chancellor pointed to figures from Sotland showing convictions for murder doubling (after abolition) as a proportion of murder charges, while culpable homicide convictions on murder charges dropped by 50%. That is an indication, although no more than that, that juries will convict of murder more readily when they are not deciding on the life or death of the accused.

The suggestion that a proposition that those who study this for a living consider is unclear at best is demonstrated “beyond reasonable doubt” is a difficult one to accept. As Donahue and Wolfers put it:

In light of our reanalysis of the data, we would strongly urge a reassessment of what is known or knowable about the impact of the death penalty. And we do not mean simply to raise “legitimate questions,” but rather to urge fundamental reconsideration of whether existing data can be sufficiently informative as to form the basis of capital punishment policy at all. The estimated effects of capital punishment on homicide rates change dramatically even with small changes in econometric specifications. Aggregating over all of our estimates, it is entirely unclear even whether the preponderance of evidence suggests that the death penalty causes more or less murder.

As for:

Both of you persistently ignore inconvenient facts…

Rightbackatcher.

@ #61 Robin Levett
Well, since I told you (#8) that this was the data assembled by Roy Jenkins to support his decision to abolish the death penalty it would take us to 1970 wouldn’t it?
And I did point out (#37) that it was comparing mid-60s to previous period excluding the ’70s
A

Continued
As to your claim that “Your trump card – Dezbhakhsh & Shepherd” – you are more stupid than I can believe from your previous posts or a deliberate liar or completely careless about reading serious comments from your opponents. I have repeatedly stated that I am basing my argument on UK data provided by the abolitionist Roy Jenkins, not US data which I have not studied to the requisite degree of rigour. It is palpable that none of the abolitionists of this thread has even tried to study the relevant data.
Your data on the sentencing in Scotland by juries is TOTALLY IRRELEVANT to the question of whether the death penalty acts as a deterrent to cri8minals.

64. Robin Levett

@john77 #63:

Ignoring the unnecessary abuse.

I have repeatedly stated that I am basing my argument on UK data provided by the abolitionist Roy Jenkins

And I have repeatedly asked for a reference to that data. We can’t debate based on some calculations you did 40 years ago and put in your attic along with the source data.

I would in any event say that if the Jenkins data was for pre-and post the 1965 Act, then there is just far to little there to ascribe any statistical significance at all; read the references in the abolition debate to the 1949-53 Royal Commission’s position on the variability of the data in general. The further idea that you could draw any reliable conclusions from the figures relating to the much smaller subset of capital murder is absurd. We are talking about data in single figures, for FSM’s sake.

Let us suppose for an instant that someone, anyone, on this thread wants to read the truth.
Then note that for the most favourable, to abolitionists, assumption about probability distributions for murders, to wit the Poisson distribution, there are three possibilities for the likelihood of murders of prison officers and the other two categories, one of which I think was murder with firearms, that Lambda is zero, that Lambda is one or that Lambda is more than one.
If Lambda is zero for any of those threew then the probability of a single murder is zero so that proves capital punishment is a deterrent. If Lambda is one then the probability of no murder in any year is 1/e, so the probability of no murder in seven years 1958-64 (actually it’s more like eight years 1957-65 but I’m giving Jenkins the benefit of every doubt) is one in e^7 (1/1096.3) and the probability of that happening three times is 1 in 1.3 billion. If Lambda is greater than one the probability is even lower – for Lambda=2 for all three it would be 1 in 1.7 quadrillion. That is apart from the rise in other capital murders, the details of which were less memorable.
Murders of police officers increased from 6 in nearly eight years 1957-early 1965 to eight in three-and-a-bit-years late 1965 to 1968. The probability calculation are more complex but it should be obvious that a rise from 0.75 pa to over 2 p.a. is rather less than 50:50 – for, instance using the simplest assumption that Lambda=0.75 gives a 13% probability moving the overall probability down from 1 in 1.3billion to 1 in 10 billion.
Obviously I was understating the case when I said the probability was less than 0.00001% – that was because at the time I did the analysis I was using six-figure tables.

For historic interest, try this HoC debate in April 1973 on a private member’s motion, proposed by Edward Taylor (Conservative), to restore capital punishment for murder involving the use of firearms or explosives and for the murder of a police or prison officer.
http://hansard.millbanksystems.com/commons/1973/apr/11/restoration-of-capital-punishment-bill

Roy Jenkins responded.

The motion was defeated by a substantial majority: For 178, Against 320.

@ #64
I do not indulge in *unnecessary* abuse. You are not as stupid as your posts on this particular thread would imply. I want you go back to thinking.
My polite comments have been met with stupidity and falsehoods. What should I do next other than point out that your posts are totally incompatible with honest debate?
There is such a thing as information without an internet reference. If you genuinely did live through the 60s and 70s you should know that.
Your comment on police officers killed is somewhat at variance from the data on your beloved internet http://en.wikipedia.org/wiki/List_of_British_police_officers_killed_in_the_line_of_duty
There were, excluding accidents EIGHT, not four, policeman killed in the period mid 1965-8 or seven excluding the one killed in February 1965 (after an expectation that Labour might abolish or suspend the deat5hg penalty but before it legislated to do so), compared to the five in 1957 to 1964. QGM Stanford was stabbed in late 1965, PC Bradley and PC Drake were run over, not accidentally, in addition to the four you admit to in 1966 {For clarity the website specifies that PC Ackerman and PC Rippingdale as well as several officers in other periods were accidentally run over)
You claim to have sourced data for murders of prison officers which, on its own, provide evidence at the 0.1% significance level for the effectiveness of deterrence. So your attempted denial implies stupidity or a complete lack of understanding of statistics or intellectual dishonesty.
FYI It is quite possible to draw sound statistical conclusions from data in single figures; victorian actuaries deduced mortality rates for assured lives aged over 90 despite the numbers of assured lives at any age over 90 being countable on their fingers when the first mortality tables were produced. Maybe you know better and can replace the Instituter of Actuaries and The Royal Statistical Society with the Robin Levett school of bigots or maybe not.
If your lack of understanding of statistics is that bad, how dare you insult me by questioning my statistical analysis? If not, why?

@ #66 Bob B
Jenkins, in his speech, assumed Hanratty was innocent: as a result of a campaign backed by abolitionists DNA evidence has proved Hanratty’s guilt; all he could say for Evans was that a jury might not have deemed him guilty beyond reasonable doubt if they had known Christie was also a murderer; the third possible innocent is a mystery as no possible innocent who has been convicted and hanged has, to my knowledge, been identified. I trust(ed) Jenkins less far than I can(could) throw him (he weighed a lot more than I did).

@ Robin Levett
If you want the raw data you can ask the Home Office
Only a cheat or a moron would keep demanding an internet reference to hard copy that was last printed more than a dozen years before the internet was created.

70. Charlieman

@36. Robin Levett: “One of the few things I agree with SMFS on is that murder rates are a function of the society you are looking at. Comparing murder rates before and after abolition is not comparing like with like…”

Sorry Robin, but the argument above suffices and there is no benefit from conducting a protracted debate about murder rates under different sentencing practices for murder.

To understand whether capital punishment would act as a deterrent in the UK today, we need an experiment conducted in current society. That sounds like an impossible task but it is unwise to apply limits to human ingenuity.

Armed with data about deterrence (yes or no answer), we could move the debate forward to the morality of capital punishment, or to the inconclusive capital sentences raised in the OP.

71. Charlieman

@67. john77: “FYI It is quite possible to draw sound statistical conclusions from data in single figures; victorian actuaries deduced mortality rates for assured lives aged over 90 despite the numbers of assured lives at any age over 90 being countable on their fingers when the first mortality tables were produced.”

The actuaries were able to deduce mortality rates based on the deaths of younger people. The insurance companies made their money from policies sold to middle class people who were expected to die in their late 50s or early 60s. The number of people who lived to 80 or 90 years was tiny as you note, so the actuaries made a modest guesstimate knowing that the result would not bust the company. If the policy holder lived to 100 years, the actuaries reduced the benefit or increased the cost.

We can’t/shouldn’t apply the same guesstimate techniques to murder rates, motoring accidents at a specific location or infections acquired in hospital. We really need solid data and when such data does not exist we should not guess.

If you want a bad example of small data sample abuse, I suggest looking at the qualification for erecting a speed camera. A length of road may average one accident per year over fifteen years but if there is one freak year with three accidents, a camera may be erected. (My sympathy is for the local residents who paid for the camera, not for anyone caught speeding.)

@ Charlieman
Actuarial science was first developed to calculate the value of annuities on which the British government was making horrendous losses by selling them too cheap. So mortality rates at higher ages needed to be calculated: they did not use “modest guesstimates”. By Victorian times, the science had advanced and the Friendly Societies that provided sick pay and pensions to the old and sick had actuaries who calculated mortality rates at ages up to the mid-nineties; one of the biggest, Manchester Unity even had separate mortality rates calculated according to the member’s job. Next time check your facts first.
On speed cameras the misuse of data lies in the claim that they reduce accidents: although in a small percentage of cases this may be true, the reduction in 90-odd% of the examples is pure random variation in the number of accidents and has nothing to do with speed cameras.

What matters in the deterrent case for capital punishment is not whether the person executed actually committed the crime but only that most believe that the person executed was guilty.

The problem with that is, as this shows, there have been a frighteningly large number of miscarriages of justice:
http://www.innocent.org.uk/

For those who maintain an interest in executions and executioners, Granada made a feature-length movie about Albert Pierrepoint, who was the principal executioner in Britain for decades. It has been posted on YouTube (1/10):
http://www.youtube.com/watch?v=J1lqYoBS5kQ

Although billed as Britain’s last hangman, that is not strictly true as Pierrepoint had resigned from the job a few executions before hanging finally stopped. In his memoirs, Pierrepoint said that he no longer believed that capital punishment was a deterrent. One experience that had led him to that conclusion was that of hanging a man whom he had known socially for years who knew of Pierrepoint’s part-time job as the official hangman.

75. Robin Levett

@john77 #67:

I’d typed a very long reply, but lost it.

Headlines: 2 of the eight “capital murders” in your analysis were pre-abolition, so should be counted as pre-abolition murders; 2 were apparently not murders (the roadblock incidents), unless Lord Gardiner the Lord Chancellor, was misinformed about the law, which I doubt; and one was a killing by a 14 year old, which could not be a capital murder.

The “Shepherd’s Bush murders” were one incident involving 2 killers killing 3 policemen. I’d personally count that as two, not three, data points, sicne deterrence acts on murderers, not murders.

You can add the Linwood bank robbery murders, to get to anything between two (incidents) and five (dead policemen); although since it was a single killer (a former police officer, ironically, or maybe not), I’d say three.

I count 6 in the 8 years of capital murder (one of the killings was a double murder by a single killer – so you’d make it 7); and 6 in the 8 previous years.

I’m not seeing an increase after abolition.

As for Victorian actuaries, they were working out the age distribution of a population of 10s of millions, with the assistance of regular census returns. The real question is how close the table predictions were to the actual numbers over short periods of, say, 4 years. Do you have the figures?

And you might want to reflect on the tones of our respective contributions to this thread; for example, calling me either a cheat or a moron because I ask for a reference which you don’t supply isn’t considered polite in my circles…

@ #75 Robin Levett
There are none so blind as those who will not see. You see no increase. Even if you insist on taking the date of Royal Assent to the Act rather than the date when the Commons passed the bill, you get six in less than three years (from December 1965 to 1968) against six in nearly nine (from 1957 to December 1965). More than two per annum is an increase over less than one per annum – like 100+%! You want to count the Shepherd’s Bush incident as two although the first site I looked at says “three career criminals, Harry Roberts, John Witney and John Duddy brutally murdered three police officers who were trying to question them at the roadside in Braybrook Street, London. They were all convicted and given life sentences” Even on your own basis that should count as three. I stated that I had excluded those accidentally run over. Those that were deliberately run over count as capital murder, which included “Murder in the course of or for the purpose of resisting, avoiding or preventing lawful arrest or effecting or assisting an escape from lawful custody.
Murder of a police officer in the execution of his duty or of a person assisting him.”
I was not including the Linwood murders because they came after the 1969 abolition bill and I had based my earlier post on the data provided by Jenkins ahead of the debate on that bill.
You have chosen to ignore the data on murders of prison officers, which alone demonstrate that deterrence works – statistically significant at the 0.1% level.
Most Victorian Actuaries were NOT “working out the age distribution of a population of 10s of millions, with the assistance of regular census returns” They were calculating the premiums (for life insurance), capital sums (for annuities) or contributions (for Friendly Society benefits) needed to maintain solvency and the reserves that needed to be held to pay for the benefits contracted to be paid in the future. No, of course I don’t have data stored in my attic on how mortality in every single period of four year for annuitants, assured lives and members of Friendly Societies compared with the mortality tables. One can assume that the tables were more than adequate since both the Life Assurance Companies, which were net beneficiaries of the reduction in mortality
resulting from improved sanitation and the Friendly Societies whose liabilities for pensions were more significantly increased (the poor sanitation had affected mortality rates in the working class more than in the middle classes) both survived, with failures concentrated among those small Friendly Societies that had failed to take advice from an Actuary.
Implying that I am a liar because I am old enough to read hard copy is hardly polite. When you quote sources that quote chunks of the data that I use without an internet link that implies that you *should* know that there isn’t an internet link to the paper to which I was referring. Yet you persisted in demanding a link that you knew, or should have realised, did not exist. I don’t expect “Sally” to follow Queensberry rules but *you* set yourself up as an individual with high standards. To say that 1965 to 1968 is eight years looks like cheating but I assume, since you are not Sally, that this was just a twitch of the eye since you were comparing it with eight years 1957-65.

77. Robin Levett

@john77 #76:

You see no increase. Even if you insist on taking the date of Royal Assent to the Act rather than the date when the Commons passed the bill

It doesn’t matter which date we pick; both Third Reading and Royal Assent postdated both the 1965 murders of policemen. Royal Assent as 8 November 1965; the Commons didn’t accept the last Lords Amendments until 28 October 1965.

More will follow.

@ #77 Robin Levett
Do you really mean the Third Reading? That was before it went to the House of Lords. You quote a date after the Bill came back from the Lords. Not the same thing. I said when the Commons passed the Bill.
You are not just quibbling – you are actually changing the date.
Not that it changes the reality that there was a significant increase in the frequency. If I played your game and treated Inspector Pawsey and Sargeant Hitchens as one incident that would make it only three from 1957 to 1964 inclusive, compared to four in 1966 alone.

79. Robin Levett

@john77 #78:

Do you really mean the Third Reading? That was before it went to the House of Lords. You quote a date after the Bill came back from the Lords. Not the same thing. I said when the Commons passed the Bill.
You are not just quibbling – you are actually changing the date.

I accept that Commons Third Reading predated the second murder. But the Commons did not “pass the bill” until October 1965; the chronology was:

PC Russell was shot 10 February 1965

Commons Third Reading was 13 July 1965.

Lords Committee Stage commenced 5 August 1965.

DS Stanford (QGM) was stabbed on 20 August 1985.

Lords Third Reading (as amended) was 26 October 1965.

Commons sitting on 28 October 1965 approved the Lords amendments.

Royal Assent was 8 November 1965

Both murders, then, predated the Lords Third Reading and, necessarily, therefore, predated the Commons approving he Bill in its final form.

Are you saying that a Borstal absconder (for that is who stabbed Stanford) was sufficiently clued up as to parliamentary procedure to believe it likely that although he was subject to the death penalty as the law stood when he stabbed Stanford, by the time he was tried the Bill would have completed its Committee stage in the Lords, and then…etc etc.

He killed before either the Commons or the Lords passed the bill into law.

Not that it changes the reality that there was a significant increase in the frequency. If I played your game and treated Inspector Pawsey and Sargeant Hitchens as one incident that would make it only three from 1957 to 1964 inclusive, compared to four in 1966 alone.

No; it would make it 6 from 1957 to 1964, as compared with 1 in 1966 – the “Shepherds Bush murders”. You accuse me of ignoring the facts, but you refuse to accept the fact that PC Armstrong was killed by a 14-year old, which was not a capital murder.

80. Robin Levett

@john77 going back to #76:

There are none so blind as those who will not see. You see no increase. Even if you insist on taking the date of Royal Assent to the Act rather than the date when the Commons passed the bill, you get six in less than three years (from December 1965 to 1968) against six in nearly nine (from 1957 to December 1965).

No. 9 November 1965 (Royal Assent) to 16 December 1969 (Lord Gardiner’s speech) is slightly over 4 years; and there were not 6; there were 3, taking the Shepherd’s Bush murders as 3. Insp Bradley and PC Drake were not capital murders. Killing != murder.

And it was 7 (taking Pawsey and Hutchins as 2 incidents -although no-one was convicted of their murder ;-)) between 1957 and 1964:

1958:
PC Summers
DI O’Donnell

1959:
DS Purdy

1961:
Insp Pawsey
Sgt Hutchins

1965:
PC Russell
DS Stanford

You want to count the Shepherd’s Bush incident as two although the first site I looked at says “three career criminals, Harry Roberts, John Witney and John Duddy brutally murdered three police officers who were trying to question them at the roadside in Braybrook Street, London. They were all convicted and given life sentences”

Joint enterprise. Only two actually pulled triggers; Witney didn’t have a firearm.

I stated that I had excluded those accidentally run over. Those that were deliberately run over count as capital murder, which included “Murder in the course of or for the purpose of resisting, avoiding or preventing lawful arrest or effecting or assisting an escape from lawful custody.

Murder of a police officer in the execution of his duty or of a person assisting him.”

What you persistently ignore is that neither of those marked in the Wikipedia page we’re looking at were apparently deliberately run over. PC Drake was stood alongside a car the other end of which was struck by an escaping lorry; what exactly happened to Insp Bradley (save that he was “run down” at a roadblock by an escaping car) I’ve been unable to track down, but the Lord Chancellor did not include either him or Drake in his list. I am guessing that both were convicted of manslaughter, which would not have carried the death penalty at any relevant stage.

You have chosen to ignore the data on murders of prison officers, which alone demonstrate that deterrence works – statistically significant at the 0.1% level.

I’ve now established that the Borstal trainee was 19 when he killed the prison officer. That doesn’t change the fact that a conclusion relying upon a single incident cannot be statistically robust; independently of the statistical techniques involved.

Implying that I am a liar because I am old enough to read hard copy is hardly polite.

And if you can point to me doing so I will withdraw and apologise. I have certainly not intentionally done so. I have asked for “a reference” to your data on a number of occasions. I have never asked for “a link”.

7. So Much for Subtlety wrote -

“not that these people are often innocent, just innocent of the specific crime they are accused of.”

Yeah totally agree, I’ve seen the mugshots of these ‘innocent’ people, most have eyes that are just too close together or their eyebrows meet in the middle so they must be guilty of sommit!

“it is a weird side effect of activists who deliberately clog up the appeals system in an effort to prevent any executions. They could execute them the day after the sentence if they liked – no waiting.”

Yep that’ll sort it all out, if only we could get rid of the activists there would be no-one too ask awkward questions about silly stuff such as innocence or guilt.

82. Robin Levett

@john77 #76:

Three quick comments:

First, killing != murder. The two killed at roadblocks were not included in the LC’s figures in the Lords debate, presumably because the defendants were convicted of lesser offences, such as manslaughter, rather than murder; I have as yet been unable to establish what the verdicts were, but the news report I have located (link will follow if required, but you can google for it) suggests that John Mellon’s defence was he was tryign to get through a gap in the block, not to kill the policeman in the car he hit; which would, if accepted, negative the encessary mens rea.

Secondly, 14 year olds could not commit capital murder; so the stabbing by a 14 year old cannot be included either. Which leaves 3 in a single incident; where only two actually did the killing.

Thirdly:

To say that 1965 to 1968 is eight years looks like cheating but I assume, since you are not Sally, that this was just a twitch of the eye since you were comparing it with eight years 1957-65.

Please re-read my post and identify where I referred to 1965-68 as eight years. I referred to the 8 years of capital murder, and the 8 years preceding them.

And fourthly (I lied); the very fact that we are arguing about single incidents which make such huge differences to he figures does rather imply that your statistical conclusions, even if based on correct data (and for the reasons I have given, they are not) are not robust.

@ Robin Levett
It is utterly plausible that they were omitted from Lord Gardiner’s figures in the HoL debate because it didn’t help his case to include them. 1964 marked a watershed in politics in many ways, one of which was the weighting given to politics over that to given to jurisprudence in the choice of a Lord Chancellor. As I said I carefully omitted the two cases of “accidentally run over” and, on the data shown, these look like capital murders. I have neither accepted nor denied that one incident was a 14-year-old because I haven’t data on that and I haven’t had time to search for it (unfortunately I occasionally have to work). I have consistently looked at 1965-8 data because, as far as I can recall, that was the data Jenkins used and if it was 1966-8 (which I doubt) that does not make a sufficient difference to the conclusion. You seem to have decided three murders only count as one after I pointed out that your claim that they counted as two on the grounds that there were two murderers failed because three men were convicted of the crimes.
One of the websites ( http://www.capitalpunishmentuk.org/abolish.html ) refers to the Home Office Report (Murder 1957-68) and shows a graph of capital murders more than doubling post-suspension while non-capital murders initially declined. In every single year 1965-8 the number of capital murders was higher than in any year 1957-64 and the average was more than double. Incidentally, said website appears, from the text, to be that of an abolitionist group.
*YOU* are arguing about individual cases in one subset of data *after* I have pointed out that the other subsets are, even without that one, statistically significant at any level that could be measured with six-figure tables. [If you want to academic rigour, that has to be measured with a Chi-squared test, not just multiplying out the results from the Poisson distributions, but I cannot replicate that calculation without digging out the data. It *was* significant at the 0.0001% level]. I think that implies my statistical conclusions are robust at the Rocky Marciano/Daley Thompson level.
All those convicted of capital murder in 1964 who had not been executed prior to the election had their sentences commuted to “life imprisonment” by the Labour government, so the Death Penalty was effectively suspended from October 1964.

84. Robin Levett

@john77 #83:

Mellon was convicted of culpable homicide on November 23 1967 (ie not murder, so not capital murder):

http://news.google.com/newspapers?nid=GGgVawPscysC&dat=19671123&printsec=frontpage&hl=en

Michael Wyber was 14 when he stabbed PC Brian Armstrong

http://icnewcastle.icnetwork.co.uk/1000expats/expatsfeatures/2002/11/20/in-the-line-of-duty-50080-12389880/

On Inspector Bradley; you can assume that a Lord Chancellor would lie on an issue on which he would easily be caught out – I won’t.

I am dealing with the subset of data that you are prepared to share.

And this:

All those convicted of capital murder in 1964 who had not been executed prior to the election had their sentences commuted to “life imprisonment” by the Labour government, so the Death Penalty was effectively suspended from October 1964.

Are we talking about deterrence?

85. Robin Levett

@john 77 #83:

You seem to have decided three murders only count as one after I pointed out that your claim that they counted as two on the grounds that there were two murderers failed because three men were convicted of the crimes.

No; the three murders I refer to between 1965 and 1969 are the “Shepherd’s Bush” murders.

One of the websites ( http://www.capitalpunishmentuk.org/abolish.html ) refers to the Home Office Report (Murder 1957-68) and shows a graph of capital murders more than doubling post-suspension while non-capital murders initially declined. In every single year 1965-8 the number of capital murders was higher than in any year 1957-64 and the average was more than double. Incidentally, said website appears, from the text, to be that of an abolitionist group.

Capital murder under the Homicide Act had no single qualifying principle; it was a grab-bag. Why killing with a firearm or explosive was considered to be more heinous than killing with poison, for example, is beyond me.

If you want to establish whether capital punishment was a marginal deterrent, therefore, you need to look at the three categories separately.

The three provisions relating to prisons, the police etc (s5(1)(c)-(e)), form one reasonably cohesive category. They, despite your protestations, show no significant increase after suspension, so either marginal deterrence wasn’t a factor in their previous level, or there are other factors operating.

So are there any factors that explain why killings in the course of theft, and killings with firearms or explosives, should increase to the extent implied by these figures? Was there a greater availability of firearms/explosives, for example? Indeed, is the increase, given that these are subsets of much larger sets, even significant. If thefts doubled over the period, then one might expect killing in the course of theft to mirror that even without loss of marginal deterrence following abolition.

86. Robin Levett

@john77 #85:

Following on my final point; Jim Callaghan, who was the Home Secretary who guided final abolition through the House (not Roy Jenkins), made the following point in the Commons debate on 16 December 1969:

My statisticians have done the best they can and have made an estimate, but they explain very fairly in paragraphs 20 and 21 of their report the problems of estimation and their conclusions from the figures since 1965 are, as they say, “inevitably an over-estimate”. Subject to that important caveat, there has been an increase in so-called “normal” murder and it began in the ‘sixties, but it began before abolition. The year 1964 is a key date if one looks through the figures. Within the “normal” category, the proportion of estimated “capital” murder is higher than it was in the early ‘sixties, though only slightly higher than it was in the late ‘fifties.

But what does that prove? Before anyone jumps to the conclusion that the absence of capital punishment causes an increase in capital-type murder, let him recall that capital-type murder is frequently a higher proportion of abnormal murder than of normal murder. In the last two years, the largest single block of capital-type murder has been murder by shooting followed by suicide. The estimated figures of capital murder may tell us something about choice of weapon, but to me they convey no clear message about deterrence or the penalty for murder.

http://hansard.millbanksystems.com/commons/1969/dec/16/murder-abolition-of-death-penalty

Note the emphasised sentence. The lack of a marginal deterrent effect of capital punishment would hardly have been an issue in these crimes.

He also confirms that there had been only three capital murders of policemen in the period 1965-68 (and we know there were none in 1969 until after his speech):

Since abolition there have been four murders, three of them in one shocking incident at Shepherds Bush. The fourth was stabbed by a boy aged 14; and the Act would have given no protection against that. Mercifully, no police officer was murdered on duty in 1967 or 1968.

@ Robin Levett
This is one of the days I need to work so comprehensive reply to follow BUT one criterion for capital murders was those where there was generally a high chance that the killer might repeat the offence so (i) deterrence was more needed and (ii) execution was the more humane version of the limiting forms of preventive detention. It only looks like a grab-bag to you because you have omitted to think about the reason for their selection. e.g. if you are serving a life sentence for killing a prison officer what is to stop you killing another unless you are put in a straitjacket or chained up until you die?
Secondly I was around in 1967 and noticed the devaluation and job-swap, but I am sure that I remember that the report was commissioned by Jenkins. Callaghan, as you note, was honest enough to admit that the figures showed an increase in capital murders although he had not personally analysed them and realised how clear the message was. Callaghan was actually quite a good Home Secretary
Thirdly I DID look at the FIVE categories separately and applied a rigorous statistical analysis to the data provided by the Home Office.
You have a touchingly naive faith in the speeches of Labour politicians as a source of truth, whole truth and nothing but the truth. Your emphasised sentence carefully excludes murders in 1965 1966 and much of 1967. Incidentally, the first sentence of Callaghan’s previous paragraph includes the 1965 data, as I had based on the Report, in the post-suspension category so your quote implies that my memory of comparing 1965-8 with 1957-64 was done on the basis of the report commissioned by Jenkins

88. Robin Levett

@john77 #87:

his is one of the days I need to work so comprehensive reply to follow BUT one criterion for capital murders was those where there was generally a high chance that the killer might repeat the offence so (i) deterrence was more needed and (ii) execution was the more humane version of the limiting forms of preventive detention. It only looks like a grab-bag to you because you have omitted to think about the reason for their selection. e.g. if you are serving a life sentence for killing a prison officer what is to stop you killing another unless you are put in a straitjacket or chained up until you die?

It would have looked like a grab-bag to the Royal Commission, just as it did to those commenting on it in the debates on abolition.

Why would use of explosives or a firearm be more likely to be repeated than use of a knife?

Your emphasised sentence carefully excludes murders in 1965 1966 and much of 1967.

So what were the relevant figures for 1965 and 1966 (the figures given excluded 1969 – which you would know if you’d used the link I gave)?

Incidentally, the first sentence of Callaghan’s previous paragraph includes the 1965 data

I’m not quite sure what you mean by “includes the 1965 data”; that Callagahn did not include the 1965 figures in “post abolition data” is clear from his earlier comment that:

“Two were murdered in 1965, before abolition became law but while it was under discussion. In the case of one of those, the person concerned was found unfit to plead”

taken with his comment that:

Since abolition there have been four murders, three of them in one shocking incident at Shepherds Bush. The fourth was stabbed by a boy aged 14; and the Act would have given no protection against that. Mercifully, no police officer was murdered on duty in 1967 or 1968.

@ #88 Robin Levett
“Why would use of explosives or a firearm be more likely to be repeated than use of a knife?”
You really weren’t attending in 1957. I was still at school but I did notice (I was periodically called upon to prevent bullying so I had an academic interest in whether deterrence worked). Murder by firearm involves a degree of premeditation (except, just possibly, for a soldier in a combat situation in wartime). Only a highly skilled marksman can shoot to wound, without a serious risk of killing. Stabbing with a knife is far more likely to wound than kill (even when trying to kill it is more likely to wound – look at battle data). Burglars carried knives to open windows and slit open containers: they only carried guns to shoot people. Incidentally there was a feeling that hanging was not a deterrent to “domestic” murders, as they were called in 1957 (Callaghan rightly disliked the phrase “normal murders”) with at least one tabloid referring to the (often, but not always, apocryphal) phrase “I’ll swing for her”.
I did read the link that you gave. It seems that you skimmed it since you say “I’m not quite sure what you mean by “includes the 1965 data”; Callaghan said “A good deal has been made of the juxtaposition of the figure of 67 murders that were, or might have been, capital in 1961-64, and a wholly hypothetical figure of 154 that might have been capital in 1965–68″
A 130% increase. On a like-for-like basis as he is including all the cases that *might* have been capital in 1961-4 including those by unconvicted minors. An increase of more than 21 innocent victims per annum i.e.
over 10 times the reduction in executions of murders found guilty “Executions fell to two a year for murders recorded in 1962, 1963 and 1964.” Abolitionists seem to be basing their argument on a value judgement that the life of a guilty murderer is worth more than ten times the life of an innocent victim.
Secondly the link includes “If we are to do this, we must compare the trend of crime before and since the suspension of the death penalty. All executions were stopped after the Second Reading of the Bill in December, 1964. We therefore have the experience of four calendar years-1965 to 1968—upon which to base our judgment.” Which is why the Home Office Report includes capital murders in 1965 when the death penalty was no longer a deterrent, and why Mr Callaghan, unlike yourself, include 1965 in post-abolition data (and tries to duck out of including 1965 in either category for murders of policemen because, as I said, the data is statistically significant at the 0.1% level).
“Table 10 shows that, comparing the last four years with the previous four years, the number of abnormal capital murders increased by about 48 per cent. whereas the number of normal capital murders increased by over 200 per cent.”
“The number of capital murders has risen by 125 per cent., whereas the number of non-capital murders has increased by only 3 per cent.”
@ #85
“The three provisions relating to prisons, the police etc (s5(1)(c)-(e)), form one reasonably cohesive category. They, despite your protestations, show no significant increase after suspension, so either marginal deterrence wasn’t a factor in their previous level, or there are other factors operating.”
Complete and utter bullshit. Firstly they don’t form a single cohesive category and secondly the increase is highly significant even if you group them. As to police you can have 11 in 19 post-war years or 5 in 8 years 1957-64 prior to suspension of the death penalty and 6 in 4 years subsequently. If you don’t want to listen to me go to *any* honest statistician and ask whether a more-than-doubling is significant. If you include prison officers it gets to 7 in 4 years against 5 in 8 years, nearly trebling. Prison Officers separately is 1 in 4 years versus none in 8 which is significant at any level covered by six-figure tables.


Reactions: Twitter, blogs




Sorry, the comment form is closed at this time.