Friday, 28 October 2011

Criminal evidence - a very quick introduction

The conviction today for murder of Vincent Tabak has raised the question of what sort of evidence should, and should not, be put to a jury.

My own view, and this is entirely subjective, is that if one is to have juries, one should trust them, or not have them at all.

If the evidence is relevant then it should be admitted to the jury.

Anything less is both somewhat artificial (and it then does not seem clear what the jury's function is, if it not to bring their worldliness to bear) and prone to disruption (especially in days of internet searches).

However, it may not be clear to non-lawyers what the basic rules of evidence are in criminal trials.

(Evidence is something which tends to show whether a fact is correct or not, and it can consist of witness evidence (either in writing or orally), exhibits, and expert evidence.)

In very general terms, there are three tests for criminal evidence:

1. Is the evidence relevant? In other words, does it go to either establishing or not establishing a fact which needs to be proved for the offence to be committed or a defence to be available. If the evidence is not relevant then it cannot and should not be used.

2. If the evidence is relevant to establishing or not establishing a relevant fact then the question is whether it is admissible. Here the starting point is whether the effect of the evidence is more prejudicial than probative. In jury trials (at the Crown Court), the criminal courts take a rather paternalistic view of what a jury would find prejudicial.

3. If the evidence is both relevant and admissible, then the question is what weight should be placed on that evidence. And with a jury one should never know, as the deliberations are secret. Judges in judge-only trials, on the other hand, will usually state openly why they prefer some evidence over other evidence.

The effect of this is that criminal trials are sometimes more a contest of what evidence is put before a jury rather than an exercise in assessing evidence.

Once a defendant knows what evidence will be before a jury, they may well plea guilty or not guilty accordingly.

A good defence lawyer will do their best with the evidence in court, but the masters are those who manage to get the evidence excluded in the first place, either on the basis of relevance or admissibility.

So in terms of evidence, as well as in terms of procedure, justice can be either carried or miscarried on essentially technical points, rather than on the overall facts of the matter or the black-and-white substantive law.


No purely anonymous comments will be published; always use a name for ease of reference by other commenters.


gyg3s said...

"If the evidence is relevant then it should be admitted to the jury."

Evidence has prejudicial and probative value: if the evidence has prejudicial but has no probative value, what's the point? Worse than 'what's the point?'; bear in mind that for a trial to be lawful it has to be fair; so, putting prejudicial evidence, with no probative value, in front of a jury is hardly fair.

(Case law ... R v Sang Lord Diplock's witterings).

A Brit Abroad said...

I think there is a statistical argument here as to whether something is relevant or not.

For example, If 1% of male computer owners have violent sexual imagery on their computer but only 1 in 10,000 (0.01%) male computer owners kill or rape women then the presence of violent sexual images on their computer adds very little to the evidence of guilt.
We can say this, even if 99% of male computer owners guilty of murder or rape have such material on their computer.

The question of weight placed on that evidence by the jury is important here. If most jurors don't have violent sexual imagery on their computer, the jury is likely to be sickened and look on the defendant in a bad light. Even if this "bad light" only makes the jury 10% more likely to convict, it is still massively-over stressing actual information gained from this evidence.

In an ideal world, we could trust a typical jury to correctly weigh this evidence a against other evidence, but I don't think we can. This is obvious from the number of people buying lottery tickets every week.

On the other hand, we might be able to infer something useful from the absence of such imagery on the defendant's computer... if experts could prove it hadn't been tampered with.

Conor said...

I'm not a lawyer, but understood that if a defendant claimed to be of good character, evidence of bad character was allowable. And there was evidence that he might have got sexual pleasure from strangling her. I don't have a problem with his use of prostitutes being inadmissible.
Even without this, I'm still surprised there wasn't a much earlier unanimous verdict. His account did not explain the number of injuries.

Fiona Hanley said...

Thanks for your subjective view David, that juries should be trusted. Thought I was going nuts trying to understand the logic involved in not admissing his taste for violent porn.

The argument against seemed to be should he be penalised for watching legal porn, and what if he was innocent. Well why is this stuff legal? I thought these days we took the view that human sexuality is a broad church with only one canyon-deep boundary: consenting adults. Child porn is not legal. Why is violent porn legal?

Elly said...

Fiona- 'extreme pornography' is not legal in the UK since the clause was added to the Criminal Justice Act in 2009 I believe.

Extreme pornography is counted as images which suggest potential serious injury to the breasts, genitals or anus, and/or potential to cause death.

But this is a very difficult law to enforce. Mainly because nearly all porn involves consenting adults and most makers of violent porn are very clear to show no harm is done to the participants. They include interviews with the actors and 'after shots' showing they are fine etc.

And, it is accepted though not formally acknowledged in law that many people have S and M interests in sex. Criminalising a whole group of people over their sexuality would be to go back to pre-1967 days when homosexuality was illegal.

Also there is no evidence that watching violent porn, or indeed participating in violent sex, leads to non-consensual violence.

I am a participant in violent sex and watching violent porn, and I am a very peaceful non-violent person.

Sorry for long comment.

Liam said...

Fiona, you're assuming violence cannot be consensual. I dint think he's being accused of having videos of scrape rapes.

Polly said...

@Fiona surely the answer is obvious: because the actors in said porn consent to the violence. Consensual BDSM is a very different beast to actual rape and sexual assault. As a submissive woman myself, I find the suggestion that all dominants/sadists/tops are murderers or rapists-in-waiting very offensive. In my experience, even in the most extreme forms of play, they take great care to make absolutely sure they don't do anything without the consent of the submissive partner and there is *always* an option to stop the scene. Putting otherwise perfectly normal, caring people who happen to enjoy consensual violent sex into the same bracket as people like Vincent Tabak is as silly as saying everyone who enjoys violent films or computer games is secretly longing to actually go out and shoot people in the face.

David Boothroyd said...

Never forget that if the Judge allows something as evidence over defence objections that it is prejudicial, and there is a conviction, then that will be item no. 1 on the grounds of appeal. Should the Court of Appeal take a different view from the Judge, that is almost certain to be an overturned conviction and a retrial order is not guaranteed.

James Medhurst said...

I was very surprised that evidence of using violent porn was excluded but I suppose it was because the defence case was that he killed Jo Yeates accidentally while trying to sexually assault her. Because both sides were in agreement that he was sexually violent, little probative value to the evidence remained. I dearly hope that, if he had run any other defence, the evidence would have been admitted.

Stephen said...

@A Brit Abroad: The statistical issue is very significant in this. It gets to a point where you might as well lead evidence about what the accused had for breakfast and it's not talking about the charges libelled at all anymore.

I think it's really hard to criticise juries for taking anything that they hear during a trial and considering it to be highly relevant for the sole reason that they heard it in evidence during a trial (and why else would the lawyers make you hear it if it wasn't important?)

Steve Evans said...

Allow me a short synopsis of my own history in this area. A few days after my daughter was born I was informed she was under an emergency protection order by a policeman and a social worker. What followed was three months of intrusion before I could get my daughter back (having done no wrong myself, it turned out my ex had a history of child abuse - the police & social workers weren't ALLOWED to tell me what this was). Then 2 years of SS involvement where they insist the ex be allowed access to K.

The ex had 3 diagnoses from her previous abuse cases, Munchausens by rpoxy, borderline personality disorder or borderline schizophrenia. I knew none of this (and still haven't forgiven myself for being sucked in by her lies).

Eventually C managed to steal my spare house keys and make a copy, her intention being to steal K and take her away.

She used these keys to gain access to my house, set fire to my duvet while I was asleep, and take K.

Fortunately a neighbour and friend who knew the history saw her entering the house and called the police, knowing I would never have given C a key.

At the trial for arson (the attempted kidnapping was dropped for reasons I will never understand) the judge did not allow the jury to hear ANYTHING about her trying to take K, ANYTHING about her mental illnesses or ANYTHING about her history of child abuse. I was told that should I mention any of those things there would be a mistrial and we'd all have to start again.

Given that, and a clever lawyer, it came down to my word against hers, and she was found not guilty. The same judge granted me a restraining order against C.

I am sure the rules of evidence have their place, but unless juries are allowed to hear ALL the facts relevant to the case (and I qould question how anyone could deny that the history was relevant) they can never reach a just decision.

The rules of evidence are broken, or being incredibly badly applied.

ivan said...

"Trust juries" - unfortunately we know that juries do have certain tendencies - they are liable to be sucked in by circumstantial and prejudicial evidence; they are liable to feel sympathy for the victim of a crime that occurred and convict someone, anyone. The prosecution does have to be very careful here, and it is the cases where they aren't that result in many of our egregious miscarriages of justice.

The whole point of justice is that it replaces the system of retributional vengeance with collectively deciding what is the best thing to do from the perspective of society.

Jimmy Mac said...

I am not familiar with the case and have just glanced at this so forgive me if I have missed the point. I would just like to make a quick observation on the stats (from a medical statistician) in response to @A Brit Abroad:

You say,
1% of m have v

where m is male computer users and v is violent imagery on computer

0.01% of m commit r

where r is a violent offence of relevance to the case

Is your argument that since very few people commit r then the proportion having v is of no value?

This is basic epidemiology. To infer a causal relationship one must first demonstrate an association, to do this one must collect data on both those that commit event r and those that do not in an un biassed way and then compare the proportion that have trait v in both populations using tried and tested statistical methodology.

The issue will always be the leap from an apparent association and determining causality but there is plenty experience of this, good and bad, in the literature. I don't think the "stats" you quote add any value to the debate.

A Brit Abroad said...

@Jimmy Mac:
No, it is obvious to everyone that the fact Vincent Tabak had violent sexual imagery on his computer does add some evidence of his guilt - that is why so many people have kicked up a fuss about the evidence being ruled inadmissible.

They don't appreciate the finer point here, namely that for any evidence to be presented, it must tend to prove something that is not a given (probative value), and that we should consider the effect of that evidence on the jury's tendency to convict (prejudicial value).

Let's take a guess and say that only 1 in a 100,000 male computer owners without violent sexual imagery on their computers kill someone in the course of a sexual, whereas 1 in a 1000 male computer owners with such imagery murder someone. On the face of it, it looks like the presence of sexual imagery makes someone 99% more likely to have committed the crime, and thus the evidence should be admitted.
But this ignores:
1) the false positives problem - namely that only 0.1% of those with such imagery will be guilty of murder.
2) The vast difference between the prejudicial value and the probative value.

Let's come back to what the evidence was being used to decide. Tabak admitted strangling the victim, but was arguing that causing death was an accident. (i.e. he was arguing that the offense was manslaughter, not murder). So the only evidence that will be admitted is evidence that will add probative value to determining whether the death was intended, or not. Probability-wise, evidence will add probative value if it increases information about the likelihood of Tabak's story being true.

Consider some of the posts on this blog from people who like S&M. These posts indicate that they are likely to have sexual imagery on the computers similar to that that was found on Tabak's computer. The presence of such imagery would therefore support a claim that the death was accidental, if Tabak had been arguing that the sex was consensual S&M sex.

However, I understand he was not arguing that, he was arguing that he had mis-read the signals, became alarmed by the victim's screams and tried to silence them through strangling but without the intent of killing her.

Pretty much the only thing to prove is whether intent to kill existed. So, to determine whether the presence of the strangulation porn on Tabak's computer has probative value, you have to ask yourself this:

What proportion of people who killed someone deliberately by strangulation in a sexual act have such porn?
What proportion of people that killed someone accidentally by strangulation in a sexual act have such porn?
The relative difference between the two will give the probative value of the evidence.

I suspect that the sort of person likely to misread the victim's signals is quite likely to have such porn on their computer, as is the sort of person that went to the flat with the intention of murder. Possibly, the more cold-blooded murderers are actually more likely to remove all trace of such porn than people guilty of manslaughter in the heat of the moment.

Because of this, the evidence has little or no probative value, and therefore was correctly excluded, especially since it would have significant prejudicial value.

Jimmy Mac said...

Thank you @brit_abroad, I now do understand your point.  I still haven't read much about the case and only had a passing interest as you used the word statistics. Although you used an artificial statistic (you called it a guess) to demonstrate your point.  I can now see you weren't making any judgement based on statistical methodology either correctly or incorrectly applied. 

Dropping the "statistics", I be lieve your point,  is: even if  "feature A"  is very much more common in persons that commit this crime than in persons who don't commit this crime, this is irrelevant as the case is hinging on whether there was *intent* to commit the crime. 

Thank you for your detailed response especially the definition of evidence, I think I agree with your conclusion. I have just been selected for jury service so I hope to put this logical thinking to good practice.

Just to be clear, if "feature A" was scientifically (using statistically sound methods) shown to be different in people who had intent compared to people that did not have intent then it would be admissible, presumably, along with the scientific study. 

James said...

Correct me if I'm wrong, but my understanding was that the images found on Tabak's computer were viewed after the murder.

If that is true then the probative value of the images is much reduced. He may have been viewing these images for a number of different reasons, none connected to the motive of the murder. For example, if his defence was true (but as we know, rejected by the Jury) he may have been effected psychologically by the victim's death and triggering a not before known interest in violent porn.

There may have been medical evidence, but that's not the sort of information the press might get to hear about.

If the probative value of the images was therefore weak then the prejudicial effect would be far greater and have too much of an influence on the Jury.

If the images were found to have been viewed before the murder, then the argument against admitting the evidence would be considerable weakened.

However, the pathway for admitting the images would be via one of the gateways in s.101 of the Criminal Evidence Act 2003 and other considerations come into play, in particular propensity to commit acts of a similar nature. Having a propensity to watch violent porn is all together different predilection.