At 13:58 20/01/04 +0000, Robert Moore wrote:
>I have not followed the minutiae of the case decided yesterday - but did it
>depend solely on a misunderstanding of probability? Or have I (to coin a
>phrase) misunderestimated the complexity of the case?
Robert, I think this issue goes far deeper than the statistics.
To my mind, the main issue is that in this case (and, I presume, many
others like it) there is absolutely no 'physical' evidence that any crime
has been committed by anyone, let alone the person accused. Autopsy not
only failed to determine a cause of death, but also failed to reveal any
signs of any sort of 'foul play'. Available medical science is therefore
completely unable to assign any probability to, let alone draw any
'confident conclusion' about, the question of whether the deaths were
'natural' or 'unnatural'.
From that consideration alone (and, of course, in the absence of any other
direct evidence {such as a witness'!} of any 'foul play' having been
involved), I cannot see how any conviction could possibly be 'safe'
('beyond a reasonable doubt'). If there is considerable doubt as to
whether any crime has been committed at all, there surely must be at least
as much doubt as to whether a particular individual has committed the
alleged (quite possibly non-existent) crime.
There is obviously a desire to avoid a situation in which a person can 'get
away with' committing 'a perfect crime' (i.e. one that leaves absolutely no
evidence, even evidence that a crime has been committed) - but, in the
final analysis, it is clearly (I would have thought!) 'unsafe', hence
unacceptable, to merely 'guess' when there is absolutely no evidence. This
is, to my mind, in a different league from, say, convicting someone of
murder when no body has been found. In the SIDS cases, we DO have the
bodies, and cannot derive from them any direct evidence that death was due
to unnatural causes. I doubt very much whether anyone would be convicted
of murdering an adult if the victims body was available and showed no
evidence of having died from unnatural causes, even if the actual cause of
death was not apparent.
As for the actual 'statistical issue', I think it goes a lot further than a
mere misunderstanding/misinterpretation of the probabilities. The stupid
statistical claims (about the allegedly very low probability of one family
suffering multiple cases of SIDS) were, of course, based on the
very-probably-incorrect assumption that SIDS events are necessarily
independent. On the contrary, although we understand very little about
this syndrome, there is every reason to suspect that there will probably be
at least some families in which multiple episodes of sudden infant death
are anything but independent - indeed, in the extreme case, some (hitherto
undiscovered) genetic factor which renders it extremely probably that there
will be further infant deaths after the first has occurred.
However, that's merely a statement in statistical language of the
problem. What perhaps worries me most is that one really does not (should
not) need to know anything about statistics to see the flaw in the
argument; it's just 'common sense'. Even if the jury somehow managed not
to notice that, the judge surely should have done.
One can think of countless possible analogies. Using similar reasoning
(and incorrect statistics), in the total absence of any direct evidence,
one could convict someone of fraud/arson if their firework factory burned
down two or three times over a period of time!
The more I see and here of jury verdicts, the more I come to the conclusion
that I would be extremely nervous if my fate was ever in the hands of '12
good men and true'. Great though the idea of juries is in theory, I think
I would be far happier if my fate were in the hands of people who had been
appointed on the basis of their ability and training to make sensible and
rational judgements/ decisions.
Kind Regards
John
----------------------------------------------------------------
Dr John Whittington, Voice: +44 (0) 1296 730225
Mediscience Services Fax: +44 (0) 1296 738893
Twyford Manor, Twyford, E-mail: [log in to unmask]
Buckingham MK18 4EL, UK [log in to unmask]
----------------------------------------------------------------
******************************************************
Please note that if you press the 'Reply' button your
message will go only to the sender of this message.
If you want to reply to the whole list, use your mailer's
'Reply-to-All' button to send your message automatically
to [log in to unmask]
*******************************************************
|