Sally Clark
August 9th 2002 New Law Journal vol 152 No 7044
Editorial by Dr. B. Mahendra.
Convicted by flawed science
"How can you be sure she is not the killer of those children?"
"We cannot ever be sure she did or did not do the deed. But what we have
are rules by which, to secure a guilty verdict , the Crown has to prove
its case beyond all reasonable doubt."
One develops a practised ease in remarking thus when a case has
attained the status of a cause celebre as has now the case of Mrs Sally
Clark. She, a solicitor, had been convicted in November 1999 of the murder
of her two infant children. It soon became clear that it was going to be
one of those cases where lingering doubt is raised and left in the minds of
observers. In Mrs Clark's case the doubts concerning the scientific
evidence given at trial-- both statistical and pathological-- were cast
soon after conviction (see New Law Journal, 4 February 2000 p 125) and ever since, in
an almost relentless march of news, these doubts have gained in such
strength that now more than a whiff of injustice hangs in the air.
An expert for the Crown, Professor Sir Roy Meadow, set down a figure
of one in 73 million for a double death to have occurred in the
circumstances of Mrs Clark's family through natural causes; a happening
translated to lay minds as occurring once every 100 years. This figure has
been challenged virtually from the time Professor Meadow, a paediatrician
and not a statistician, uttered it in court. It has now been conceded, in
the scientific community at least, as likely to have misled the jury. In a
recent statement the Royal Statistical Society (RSS) drew pointed
attention to the misuse of statistics in the courts in general and in the
Sally Clark case in particular. It stated that Professor Meadow's view was '
statistically invalid (and) has no statistical basis.' This is because the
figure assumes that 'normal' cot deaths arise independently in families.
Not only is there no empirical basis for this assumption but, as the RSS
points out, there are very strong a priori reasons for assuming that there
are unknown genetic or environmental factors that predispose families to '
cot deaths'. In fact, the empirical evidence points in the opposite
direction. A recent study at Manchester University had reportedly discovered
a genetic basis to 'cot deaths', associated with a three- fold increase in
the risk of 'cot death' in those possessing it compared with the normal
population. This may mean that 'cot deaths', like many disease
conditions, have a propensity to run in families. As the RSS states, the
true frequency of families with two cases of 'cot deaths' may be 'very
much less incriminating than the figure presented to the jury at the
Clark) trial.' It went onto say this case 'is one example of medical
expert witness making a serious statistical error, one which may have had a
profound effect on the outcome of the case.' Aggrieved by the statistically
semi-numerate usurping of the role of experts, the RSS goes onto demand
that only 'appropriately qualified' statistical experts be allowed to give
this kind of evidence.
The other strand of tainted evidence involved the testimony of the
Crown's expert pathologists, Dr Alan Williams and Professor Michael Green.
They had opined that the deaths of the children were due to their being
smothered, having initially suggested shaking as the cause of death. The
retraction and revision of opinion concerning the cause of death came in the
face of a challenge by a defence pathologist. There were numerous other
flaws identified in the Crown's pathological evidence and, though attempts
were made to counter them, the evidence was seemingly sufficient to be
deemed beyond reasonable doubt by the jury and, later, passed muster with
the Court of Appeal which held the medical evidence incriminating Mrs Clark
to be 'overwhelming'.
There the matter might have rested for these doubts could be
plausibly explained as being due to genuine divergences in medical opinion
that need not detain a court. Indeed, only a few months ago, Mrs Clark's
husband Stephen, dedicated to clearing his wife's name, sounded notably
despondent. Then, fortune, which seemed to have deserted the Clarks, skulked
back to their side for with the assistance of Mr Martin Bell, the journalist
and former independent MP for Tatton, they succeeded recently in persuading
the General Medical Council to investigate the evidence given in the case by
the two pathologists.
At one level these matters can be seen as a classic illustration of
the clash of cultures involving science and the law. Science operates in a
world of the provisional and the tentative; the law demands the closures of
certainty. If these medical experts had submitted their interpretations and
opinions to a learned journal, their errors would have been pointed out to
them by their peers with a view to revision, a process that could take
months and years. Such luxuries of editorial amendment are not open to a
court; stark choices have to be laid before lay judge and jury. Result--
possible injustice (see N.L.J., 16 November 2001 p 1686).
On the other hand it seems odd-- and disturbing-- that the correction of
possible injustice should wait upon the serendipities of scientific
discovery and the caprices of investigations into professional competence
by a regulatory body. But, be that as it may, let us be thankful for
slender mercies. There now has arrived through diverse channels information
which suggests even greater doubt about the Crown's case against Mrs Clark.
Quite simply there exists reasonable doubt regarding the safety of the
verdict pronounced in that case. The least the Criminal Cases Review
Commission could do is to refer this case back to the Court of Appeal.
Copyright (c). Reproduced by kind permission of New Law Journal
and B. Mahendra.