Training Police Dogs

Police dogs trained to hunt for corpses react to the gases given off by decomposing bodies. These gases are principally hydrogen sulphide, hydrogen phosphide (phosphine), carbon dioxide and methane. Buried pig carcasses are used to train the dogs, since decomposing pig corpses produce the same mixture of gases as decomposing humans.

Central to this programme is the concept of evidence. When we seek to establish the truth about events in the past (or for that matter the present), we have to rely on evidence. Sometimes we may have access to the testimony of witnesses who observed the events taking place. Even this evidence does not constitute absolute proof as witnesses may have their own motives for giving the testimony they give. Furthermore, even if we were eye-witnesses we cannot be sure that we saw what we saw, or heard what we heard. Human senses are fallible, as too is the human memory. Often there may not be any witnesses, in which case we have to infer the facts of what happened from other information we ‘know’ to be true. For example we infer from the fact there are traces of the accused’s body matter (hair, blood, and so on) at the scene of the crime that it was the accused who perpetrated the crime. We might of course be wrong and there may be other explanations as to why the accused was there. Similarly, miscarriages of justice have occurred because wrong inferences were drawn.

The conclusion to be drawn is that we can never have sufficient reliable evidence about the past to be certain about any event. We can never have absolute proof. In effect we are dealing with degrees of probability. In the case presented in the programme, that of the missing Royal Marine, we are not even certain that a crime has been committed. All we can do is seek to establish a degree of probability as to whether or not a crime has been committed. What is striking in the programme is the variety of methods used to try to uncover facts from which inferences about what probably happened might be drawn. These range from geophysics and archaeology through to hypnosis and animal training.

A further point to remember is that the legal approach to evidence and the scientific approach are not always in harmony. For a person to be convicted of a crime in this country it has to be proved ‘beyond reasonable doubt’ that the person perpetrated the crime. Judges and lawyers are reluctant to give a percentage of probability for what is meant by ‘beyond reasonable doubt’. Scientists might well say, though, that they are 90% certain that this injury was caused by such and such a weapon. Whilst it is true that two scientists may argue, let's say, about whether British Beef is safe, they can at least debate the issue of safety in terms of specified numerical probabilities. This is not the case with the legal approach to evidence. When a defendant is acquitted by a jury, we do not know whether this is because the jury is 100% convinced of his or her innocence or whether it is because they are 99% certain that the defendant committed the crime but they are not sure ‘beyond reasonable doubt’. Finally, some evidence which a scientific approach would say is relevant to determining the probability of a defendant's guilt may be inadmissible in a court of law. For example if a defendant has a string of previous convictions for stealing from parked cars and is now charged with theft from a car, rationally viewed his previous convictions ought to be relevant to establishing whether he is guilty; yet such evidence would probably be excluded as it would be deemed to be likely to prejudice the jury against the defendant.