246 Advances in spoken discourse analysis
approaches open to the discourse analyst. Firstly, he can appeal to the
experimental evidence about linguistic memory quoted above and argue
that even an honestly produced after-the-event ‘verbatim’ transcript must
be totally unreliable as a record of the exact words that were uttered and
only partially reliable as an account of the content of the reported interview.
Secondly, he can compare authentic with disputed texts in the knowledge
that, although any forgery sets out to look like the real thing, the forger
will make mistakes.
Pursuing this line the first thing to examine in a text which is claimed to
be a ‘verbatim’ record produced after the event is length. Average speaking
rates are 8–15 times faster than average writing speeds, and thus an accurate
after-the-event record of a 30-minute interview which, by definition, had
not been slowed down by note-taking, should be many times longer than a
contemporaneous record of a 30-minute interview—indeed solicitors and
barristers are now finding themselves overwhelmed by the length of transcriptions
of tape-recorded interviews.
Usually we can demonstrate that the disputed texts are simply too short.
For example, in one case currently going to the Court of Appeal the claimed
after-the-event ‘verbatim’ record of a 38-minute interview, when read aloud
at normal speaking rate, lasts a mere 11 minutes, which fits very well with
Hjelmquist’s claim that participants remember 25–30 per cent.
The question of verbal memory was of vital importance in the evaluation
of claims made in the case of one of the ‘Birmingham Six’, William
Power. Parts of the interview record and the subsequently produced statement
are verbally identical. The police officers claimed that this was because
Power had simply retold the same story in the same words; Power’s explanation
was that the second text, the statement, was in fact in part copied word for
word by the police officers from their written record of the first, the
interview.
In order to support Power’s claim I needed to draw on the evidence that
people do not remember even their own utterances verbatim and that, when
they need to say the same thing on a different occasion, they encode anew.
Power himself, interestingly, kept reminding the prosecution barrister of
this at his trial, when he was asked to confirm that he had actually said
certain things attributed to him:
Power: Yes sir, some words to that effect...
I don’t know the exact words, sir.
(TBJ, p. 4)By great good fortune part of the Birmingham Six trial was ‘replayed’
because, at one point, the defence disputed the admissibility of part of
the prosecution case. So, for three days there was a ‘trial within a trial’
(TWT), at the end of which the judge accepted the admissibility of the
evidence and then the same content was covered again for the benefit of
the jury (TBJ). As a consequence there are many examples of Power