244 Advances in spoken discourse analysis
Looked at from one point of view, what the analyst is faced with in a
fabricated text is the work of an amateur dramatist, someone trying to create
a convincing record of a fictional interaction. The analyst, from his knowledge
of how real interactions, in particular authentic interviews and statements,
are structured, and by comparison with other texts, both fabricated and
fictional, sets out to detect non-authentic features. However, in most cases
the evidence is not as clear as in the extract below where the incriminating
utterances (7) and (8), totally lacking in coherence and cohesion both with
what has gone before and with what follows, were added later to an otherwise
authentic and apparently faithfully recorded interview.
1 Have you got a brother named Roy?
2 Yes.
3 On the 31st October 1986 you deposited £1,000 into the T.S.B. Where
did that come from?
4 The sale of the GTI with ‘Rabbit Injection’ written on the back.
5 Will you sign an authority for us to look at your bank account?
6 No.
7 I take it from your earlier reply that you are admitting been (sic) involved
in the robbery at the M.E.B.?
8 You’re good, Thursday, Friday, Saturday, Sunday and you’ve caught
me, now you’ve got to prove it.
9 Do you want to read over the notes and caption and sign them.
10 I’ll initial the mistakes, but I won’t sign them. (end of interview)
In this particular case the linguistic evidence of fabrication could be confirmed
by ESDA, a machine for Electro-Static Deposition Analysis, which enables
the analyst to read, from indentations, what was written on the sheet(s)
above. ESDA analysis clearly showed that an earlier version of this text had
been written on the page above, an earlier version that was identical, except
for the absence of the challenged utterances (7) and (8).^2
THE EVIDENCE
The kinds of mistake which a fabricator of interviews and statements makes
can usefully be grouped under three headings which I will label psycholinguistic
considerations, quantity and discourse structure.
Psycholinguistic considerations
There are several misconceptions about how the brain deals linguistically
with the decoding, storing and receding of information which can lead to
identifiable mistakes in fabricated texts.
Firstly, there is a commonly held belief that people are able to remember
verbatim what was said to them—everyday story-telling depends crucially
for its vividness on this assumption: