Prison population figures and rape

[major_clanger]

The prison population in the UK is just under 100,000. This seems to be commonly taken to be too high. But what should it be?

Let's take on crime for which prosecution and sentencing is topical and controversial: rape.

Statistics suggest that there are some 85,000 rapes in the UK every year. At present only a small fraction of those result in a conviction. But what if we could get that up to, say, a 50% conviction rate? If we err on the low side, that would be 40,000 rape convictions a year.

I am going to take it that nobody would think it a bad thing, given how frequent rape is, if we had 40,000 rape convictions a year.

The average length of a rape sentence is now eight years. As most prisoners on determinate (i.e. fixed-length) sentences serve half their sentence in prison and half released on licence, this means that the average time a rapist spends in prison is four years.

I am going to take it that nobody thinks that four years is an excessively long time for a rapist to be sent to prison.

If, every year, we were to imprison 40,000 rapists for four years, then the steady-state UK prison population of rapists alone is going to be 160,000.

So, is our prison population really too low? I suggest not: bearing in mind that there are other offences meriting prison time, it is perhaps half what it ought to be.

If we ever do improve rape convictions, we are going to have to build a lot of new prisons.

(I have neglected the points that some rapes are by serial rapists and so if 50% of rapes led to conviction there would be fewer rapes, even before any deterrent affect came into play. But I don't think this affects the figures too much. Also, we would need a lot more courts and jury trials as well as prisons.)

Comments

[hoiho.livejournal.com]

A clever argument, but, surely, a false one, as it's predicated on offences which there have not been convictions.

The absolute number is not the issue; it is the nature of who is locked up that is. The argument is that our prison population is too high for the offences for which there have been convictions.

Reductio ad absurdum: locking up, say, 100,000 petty shoplifters because there were 85,000 rapes is not a sensible policy.

[major-clanger.livejournal.com]

I'm not sure I see your logic. I don't dispute that there are many people in prison who shouldn't be; in particular, my experience of the criminal justice system has convinced me that the damage done by criminalising drugs far outweighs any benefit of doing so.

But my point was that even if you looked at just the number of people who realistically ought to be in prison for rape (I am sure we agree that there is no such thing as a minor or trivial rape) then that alone would yield a prison population considerably larger than the one we have at the moment for all offences taken together.

[hoiho.livejournal.com]

I don't disagree with your conclusion that if we convicted all rapists there would be many more in jail. But we haven't. So it's not relevant to the size current prison population.

Much of of which is made up of people who do not need to be imprisoned for public protection, and who will not benefit from being imprisoned. That is why the prison population is held to "too high", not because it's over a certain number.

[autopope.livejournal.com]

I broadly agree with you, but ...

ISTR studies show that the most effective approach to crime reduction is to maximize the probability of capture and conviction, not the severity of punishment. The current British approach to handling murder works well (90% ish clear-up rate) at keeping murder low; much better than the 18th century Bloody Code (no police, no systematic investigation, death penalty for nearly everything). It's the perceived probability of arrest that acts as a deterrent, not the severity of punishment (especially if arrest and punishment is rare, as it currently is with rape).

So: let's run a thought experiment in which some magic new breakthrough results in 90% of rapes leading to arrest and conviction. If the average sentence drops to just 2 years, we could expect the prison population to spike by an extra 80-90,000 prisoners for a year or so -- but then to fall back drastically, as the pool of potential or actual offenders get the message that arrest and conviction are almost inevitable. (I'm assuming here that the average rapist doesn't currently expect to get caught, and that most rapists would not consider one coerced fuck to be worth a subsequent year in prison.)

Does this reasoning make sense?

(If so, there may be interesting times ahead with the advent of ubiquitous monitoring networks and lifeloggers ...)

[major-clanger.livejournal.com]

To an extent I agree with you, but I suggest that there's an important distinction between murder and rape. Very few people who commit murder don't believe that murder is a crime. By contrast, very many men who rape women clearly don't think that what they are doing is what has (quite shockingly) sometimes been called 'rape rape'. Indeed, one of the most persistent and disturbing myths about the Assange affair is that the conduct he is accused of is not 'really' rape (it is, under both Swedish and English law.)

What we need, and what might well happen alongside a radical improvement in the rape conviction rate, is clear and firm education of what rape is.

[biascut.livejournal.com]

I agree, and I generally tend to the view that the standard of proof required for a successful prosecution means that the law on its own is too blunt and inadequate an instrument for preventing rape.

But at the same time, I think you are taking "many people don't think that forced sex is a serious crime" and "there are too few prosecutions for forced sex" as two separate propositions, when it seems very likely to me that they are strongly linked.

[andrewducker]

Locking people up for those actions might educate people that what they were doing _is_ rape.

(Of course, public adverts would help too.)

[brixtonbrood.livejournal.com]

Lambeth have done a very pleasing campaign in tube stations along the lines of "Getting it On: Get Off Me - Real men know the difference"

[philmophlegm.livejournal.com]

On the subject of "rape rape", see also Evans, Ched:

http://philmophlegm.livejournal.com/235723.html

[brixtonbrood.livejournal.com]

My instinct is that multiple offenders are a much more significant factor than you're allowing for, and that a small proportion of men are systematically using alcohol and rape myths to assault many women, in the knowledge that what they're doing is morally wrong (whether or not they label it as rape) but the chance of successful prosecution is small.

I agree that improving the probability of being prosecuted and/or convicted would make a big difference. In the riots the thing that turned the corner wasn't the harsh sentences, it was the fact that people were demonstrably being pulled into court and tried almost immediately - suddenly people stopped looking on looting as a risk-free crime.

[purplecthulhu.livejournal.com]

I agree with you on both points - see my own analysis of numbers below.

Moreover, I think the factor that has brought about a reduction in the rate of drink driving is not the conviction rate (probably still too low) but the continued education campaign that has made drink driving much less acceptable than it used to be.

What Lambeth is doing is great, but this needs to be nationwide and just as hard hitting as drunk driving adverts. This will doubtless make many people uncomfortable since it concerns sex, and that might be why such a campaign hasn't happened yet. Let's hope it does.

[major-clanger.livejournal.com]

Absolutely, yes. We have made drink-driving reprehensible among most people; much the same can be said for smacking children in public. We need to do the same for sexual misconduct. I've seen a few campaigns, but there needs to be more.

[biascut.livejournal.com]

I've read research to that effect too - it argued that the idea that a significant number of rapes are perpetrated by men who don't realise they are raping is wrong. It was based on focus groups with men in the US army in which they discussed forced sex. They found something that something like 95% clearly recognised it as wrong and were absolutely firm that they would never do that. The remaining 5% said that of course they would, and in many cases had. Furthermore, they were adamant that they were perfectly normal, and that all men would do the same. Can't find it on a brief google, though.

Oh, I don't think this is the study I was thinking of - I'm sure I read one done with army personnel - but it finds something very similar.

(Sorry, editing again! Mystery solved - that link leads to this, which looks at two surveys - one is college students, and one is US Navy personnel. They both have similar findings.)

[brixtonbrood.livejournal.com]

I think even the men who do it know it's wrong - just like people always knew drink driving was wrong - but there's "evil wrong" and there's "don't get caught wrong" like cheating in an exam - and some repeat rapists, just like some drunk drivers (even today), are categorising their actions as type 2.

[biascut.livejournal.com]

Yes I agree - the links I've posted are making a distinction between the idea that a significant number of rapes are one-off events committed by men who just "went too far" or "didn't know she didn't want to" and who are horrified at the idea that they might have forced someone, versus the research findings that certain men are repeat predators who are deliberately contriving situations where they have the opportunity to rape. So yes, they know it's wrong, but they think it's wrong in the "teehee, look what I got away with" rather than "serious crime" way.

And Thomas Macaulay Miller talks about the way that other men and women tell the serial rapists that their activities are low-risk, by supporting the situations that allow them to continue and disbelieving their victims.

[major-clanger.livejournal.com]

What I have found in my work in both criminal defence and child neglect work is that there are some people with quite shocking views of what 'normal' behaviour is. You hear comments like:

- 'Everyone nicks stuff from the shop when they can.'

- 'Everyone drives without insurance.'

- 'Everyone smacks their kids'. (For values of 'smacking' = serious bruising)

In my experience this is often down to a combination of family and friends with few or no good role models, limited education and long-term reinforcement of the 'normalcy' of such behaviour from peers.

[biascut.livejournal.com]

Oh yes, absolutely - similarly people who believe that everyone cheats on their partner, or that all teachers would sleep with the vulnerable but beautiful 15-year-old if they got the opportunity, and so on. Which is exactly why all the stupid jokes about the idea that a drunk girl is an "opportunity" matter so much.

[ms-cataclysm.livejournal.com]

I noticed that the article that you linked to says that the average sentence for rape is longer than the average sentence for murder which surprised me. Does murder really attract an average effective sentence of under 4 years ( 8 years with 50 % discount)or is this a massaged statistic ? Am boggled.

[major-clanger.livejournal.com]

Manslaughter, not murder - and there is an important difference.

You have to compare statistics with care because rape (unless for a second offence or where there are exceptional aggravating circumstances) attracts a determinate sentence, whereas murder results in a life sentence.

What 'life sentence' means is that the convicted person is sentenced to a totality of a life sentence, comprising a minimum term to be served before release followed by life on licence. This contrasts with a determinate sentence of N years, which usually takes the form of N/2 years to be served in prison followed by N/2 years on licence. For murder, a typical minimum term is 8 to 12 years.

Manslaughter has a sentencing range from nothing to life, because it covers such a wide range of circumstances. Guidance here suggests determinate sentences of a year for death resulting from a punch (where no intent to seriously injure or kill) to five years for recklessly running someone down and dragging him under a car. In essence, manslaughter is the crime of inadvertently killing someone, albeit in a blameworthy way, whereas nobody inadvertently rapes someone.

(Yes, I am simplifying the law on manslaughter, but I'd have to write an article to cover all the legal aspects.)

[purplecthulhu.livejournal.com]

I think your estimate of the effect of repeat rapists on these figures is an underestimate.

If there are 85000 rapes a year in the UK, then a 50% clearup rate would, as you say, results in convictions of about 40000 individuals (assuming that the criminal justice system is 100% effective in convicting the guilty and exonerating the innocent - but that's a different story).

However, if, as seems likely, rapists don't do it once but do it, on average, twice, then your 40000 drops to 20000 individuals. As with most things, the number of rapes per rapist is likely to be a power law, so an average of 2 is probably an underestimate.

[major-clanger.livejournal.com]

I would like to see research on this, but I suspect there is indeed a very skewed distribution: a lot of people who commit rape once (often in a way they probably consider not really to be rape) and a small number of repeat offenders.

Something we have to be very careful of is drawing misleading assumptions about the number of perpetrators from the frequency of incidents. It's quite possible for a large fraction of women to be victims of sex offences without anything like the same percentage of men being offenders. But equally that means that a small number of repeat offenders can result in a lot of incidents, which is why it's important to tackle the problem in, for instance, environments such as fandom.

[unwholesome-fen.livejournal.com]

Rebecca Watson (skepchick.org) quoted some statistics when she spoke to Cambridge Skeptics in the Pub a couple of years ago. Unfortunately I didn't take notes, and a quick trawl back through the blog hasn't turned up an equivalent article. My (admittedly hazy) recollection is that the stats suggested that a surprisingly (to me) large proportion of rapes for which there was a conviction were perpetrated by a relatively small number of serial rapists. I'll let you know if I track down an actual citation.