Hugh McLachlan: Beyond reasonable doubt

Sally Clark leaves court with husband Stephen after winning an appeal against her conviction for the murder of her two sons. Picture: Reuters

We should be wary of the current proposal from the Scottish Government to abolish the requirement of corroboration in criminal trials in response to the low rate of convictions in rape cases.

There are two distinct, fundamental questions at issue. On the basis of uncorroborated evidence, could it ever in principle be known – beyond all reasonable doubt – that someone committed a particular crime? In practice, could we reasonably expect juries to fairly gauge the guilt of accused people on the basis of such uncorroborated evidence?

Unlike historians or scientists, members of a jury are not merely required to assess whether particular statements are true or false on the basis of evidence that is relevant to such truth or falsity. They should make such an assessment only on the basis of evidence of which the presentation and evaluation are consistent with a fair trial.

The rational, objective testing of a theory is not the same thing as the fair trial of a person. Consider how, in general, we are able to establish any matter of fact.

Knowledge is true justified belief. Why do we believe that, for instance, the earth is more or less spherical? What justifies our belief to the extent that we claim to know that it is true?

There is no “killer fact” or single, isolated piece of compulsive evidence. Sphericalness, even in small objects, is not directly observable. We cannot simultaneously see the near and the far side of a golf ball, far less the far and near side of the Earth.

We know that the Earth is more or less spherical because that belief coheres systematically with a whole set of other beliefs that seem to have good evidence in their support.

For instance, photographs taken in space from various angles show the profile of the Earth to be circular more or less.

No-one has ever reached an edge of the planet Earth.

It is possible to fly from the North Pole in what seems to be a straight line and return to the North Pole.

Theories of the movement of what we call “planets” based on the assumption that the Earth is roughly a sphere produce predictions about eclipses and such like which turn out to be true.

Suppose that we are at sea and looking toward the horizon. If another ship approaches us, we see initially only the tip of the funnel. Gradually, the rest of the funnel, then the upper deck and the remainder of the visible parts above the water line appear to us.

The view that our knowledge is formed, like a building, upon a sure foundation seems dubious. It is more plausible to think of it – as Wittgenstein suggested – as being like a nest where particular individual beliefs are akin to twigs that lend support to and are supported by other such twigs.

In a sense, all items of evidence can be considered to be circumstantial. Sometimes the combined weight of them can be so overwhelming that we are justified in saying that, beyond all reasonable doubt, particular people are guilty of the particular crimes for which they were tried.

To claim to know something or other on the basis of only one item of evidence would be dubious. To claim to know that someone was guilty, beyond all reasonable doubt, of having committed a crime on the basis of only one item of uncorroborated evidence would be alarming.

In principle, uncorroborated evidence might be used appropriately in criminal trials – but only if it features as part of a wider case against an accused person. There are grave dangers in an over-reliance upon it, particularly if it is the uncorroborated testimony of an expert witness or an alleged victim.

We know from experience that expert witnesses can make mistakes. We know that “evidence” can sometimes be dishonestly planted or manipulated.

Even when a particular sort of evidence could be used correctly, there is a danger that it will be misinterpreted. We know from experience that not only members of juries but also lawyers and judges sometimes misunderstand the nature and significance of statistical claims.

In the notorious prosecution of Sally Clark for the murder of her two babies, the case against her was based on the suggested improbability that two of her babies could have had natural cot deaths. Among the fallacies that were involved was the isolated focus on one particular unlikelihood. A comparison with other relevant, unlikely occurrences and events should have been undertaken.

For instance, although it is improbable that two babies of the same mother would die spontaneously of cot death, it could also be said to be highly improbable that a mother would murder her two babies.

Furthermore, it is fallacious to use statistical improbability as evidence that a particular event has not occurred. Highly improbable events do occur.

Someone will win the lottery no matter how unlikely it is to win. After the draw is made, we cannot use the improbability of winning as evidence that he or she did not win it.

Truth is not the same as truthfulness. Those witnesses who are truthful do not always say that which is true. One can be, for various reasons, honestly mistaken in one’s testimony.

If an alleged victim and an alleged rapist give inconsistent accounts of the event at issue, they might both be speaking truthfully. It might be the case that neither is lying – even if what one of them says is false.

Furthermore, accused people who tell lies when they give evidence are not necessarily guilty of that of which they are accused.

It seems inconceivable that it could be rationally concluded that, beyond all reasonable doubt, someone was guilty of a particular crime on the basis solely of the uncorroborated contested evidence of an alleged victim, even if the truthfulness of the testimony is not in question.

Juries must consider only relevant evidence. Furthermore, they must give the relevant evidence an appropriate weight. Otherwise, they cannot give someone a fair trial.

To give someone a fair trial for rape is not merely to come to the correct answer to the question: did he rape the alleged victim? It is to come to this answer for appropriate reasons when this is the correct answer.

It is the role of the courts to try those who are accused fairly and justly. It is not their function to secure convictions. If the laws and procedures are fair and just, we should accept the verdicts of the courts, whatever they turn out to be.

We should not tamper with the procedures in order to produce particular outcomes.

• Hugh McLachlan is professor of applied philosophy in the Glasgow School for Business and Society at Glasgow Caledonian University