Extending the Confusion About Bayes
| Author | G. A. Vignaux,Bernard Robertson,Charles E. H. Berger |
| DOI | http://doi.org/10.1111/j.1468-2230.2011.00857.x |
| Date | 01 May 2011 |
| Published date | 01 May 2011 |
understood either as a form of sharing (of the sort described above) or as an
instance of ‘compensation’,
116
alineofargumentthatwasrejectedbythemajority
on the basis that his decision to move to academia was not motivated by the
family’s needs but was simply Mr Granatino’s own choice.
117
What is clear is that this issue will provide fertile ground for argument in
future cases, as spouses whose agreements fall within the majority’s principle in
favour of upholding pre-nuptial agreements, at least to the exte nt of precluding
equal sharing, seek tomaximise their residual claim that need- or compensation-
based provision would be‘fair’. Determining, as a matterof principle, how broad
such residual claims should be will be a keyconsideration for the Law Commis-
sion as it deliberates on its consultation responses. Indeed, the Law Commission
will need to decidewhether to adopt that model of reform at all, or instead to base
the court’s power to set an agreement aside on some broader basis, such as‘signi¢-
cant injustice’.
118
In the meantime, whatever its scopemay be, the continued avail-
ability of relief on need ^ or compensation-based grounds ^ despite an
agreement, serves toensure that marriagedoes remain distinct from cohabitation.
Marriage is not dead. And there was no need on the facts of this case to make
further provision for Mr Granatino in order to secure its continued vitality.
Extending the Confusion About Bayes
Bernard Robertson
n
,G.A.Vignaux
nn
and Charles E. H. Berger
nnn
In RvT[2010] EWCACr im 2439, [2011] 1 CrApp Rep 85, the Court of Appeal indicated that
‘mathematical formulae’, such as likelihood ratios, should not be used by forensic scientists to
analyse data where ¢rm statistical evidence did not exist. Unfortunately, when considering the
forensic scientist’sevidence, the judgment consistently commits a basic logical error,the ‘transpo-
sition of the conditional’ which indicates that the Bayesian argument has not been understood
and extends the confusion surrounding it. The judgment also fails to distinguish between the
validity of the relationships in a formulaand the precision of the data.We explain whythe Baye-
sian method is the correct logicalmethod for analysing forensic scienti¢c evidence, howit works
and why ‘mathematical formulae’ can be useful even where ¢rm statistical data is lacki ng.
‘‘Theprobability that an animal with four legs is a cow is not the probability that a
cow has four legs.’’
At the scene ofa crime, a footwearmark is found. A suspect’s shoes are examined.
They are of the same brand as the footwear that left the mark and one shoe has
116 Lady Hale principally refers to‘need’ in this section of her judgment, but does also talk about
‘compensating’the hypotheticalparent, ibid at [193].
117 ibid at [121].
118 For the full range of options on o¡er, see Law Commission, n 4.above para 8.19.
n
Barrister,Wellington.
nn
VictoriaUniversity of Wellington.
nnn
Netherlands Forensic Institute.
Extending the Confusion About Bayes
444 r2011The Authors.The ModernLaw Reviewr2011The Modern Law ReviewLimited.
(2011)74(3) 430 ^455
To continue reading
Request your trial