BEST PRACTICES 363
Number (v) is especially important because it means that the
method allows the examiner to pick and choose stylemarkers
without any predictability. This fundamental methodological flaw
enables a host of problems, all rooted in subjectivity. On the one
hand, it is essentially impossible to replicate a forensic stylistics
analysis, while on the other hand, it is always possible to find an
alternative analysis and opposing conclusion. This is the
dilemma of any “pick and choose” method.
A. Litigation DependenceFinegan documented a case in which five linguists were
hired to conduct an authorship identification.^69 The five linguists
each offered an opinion; each opinion used forensic stylistics to
support the side which hired them. This is possible because each
linguist picked stylemarkers, and each stylemarker could be
deemed important or not by the linguist without any standard
reference set. Finegan’s report of this case demonstrates that
forensic stylistics suffers from a classic case of confirmation bias
being built in to a method without litigation-independent
validation testing.^70
Without litigation-independent testing, the expert battles
inside litigation are inevitable. Finegan predicted that this battle
of the experts would occur and that it may be a good thing:
The expectation of expert rebuttal witnesses should
contribute significantly to improvements in the quality of
linguistic opinion available within the judicial system—
and to justice.^71
I would suggest that a better practice is litigation-independent
validation testing, a controversial stance within the forensic
stylistics community. In a recent recorded interview prior to
deposition, Professor Leonard stated that he had “misgivings”
about testing the method.^72
(^69) See Edward Finegan, Variation in Linguists’ Analyses of Author
Identification, 65 AM. SPEECH 334 (1990).
(^70) Id. at 339.
(^71) Id. at 338.
(^72) Interview with Dr. Robert Leonard (Feb. 22, 2011).