15
St - Yves , M. & Tanguay , M. ( 2009 ). The psychology of interrogation: A quest for a
confession or a quest for the truth? In M. St - Yves & M. Tanguay (Eds), The
psychology of criminal investigations: The search for the truth (pp. 9 – 40 ). Toronto :
Thomson Carswell .
St - Yves , M. , Tanguay , M. & Cr é pault , D. ( 2004 ). La psychologie de la relation: cinq
r è gles de base . In M. St - Yves & J. Landry (Eds.), Psychologie des entrevues d ’ enqu ê te:
de la recherche à la pratique (pp. 135 – 153 ). Cowansville, Qu é bec : Yvon Blais .
Ury , W. ( 1991 ). Getting past no: Negotiating with diffi cult people . New York : Bantam
Books .
Wald ,
M. ,
Ayres ,
R. ,
Hess ,
D. W.
,
Schantz ,
M. & Whitebread ,
C. H.
(
1967 ).
Interrogations in New Haven. The impacts of Miranda . Yale Law Journal , 76 ,
1519 – 1648 .
Walsh , D. P. J. ( 1982 ). Arrest and interrogation: Northern Ireland 1981 . Journal of
Law and Society , 9 , 37 – 62 .
Walters , S. B. ( 1995 ). Principles of kinesic interview and interrogation . New York :
CRC .
Williamson , T. M. ( 1990 ). Strategic changes in police interrogation: an examination of
police and suspect behaviour in the Metropolitan Police in order to determine the
effects of new legislation, technology and organizational policies. Unpublished PhD
thesis, University of Kent.
Williamson , T. M. ( 1993 ). From interrogation to investigative interviewing: Strategic
trends in police questioning
.
Journal of Community and Social Psychology ,
3 ,
89 – 99 .
Williamson , T. M. ( 2006 ). Investigative interviewing and human rights in the war on
terrorism . T. Williamson (Ed.), Investigative interviewing: Rights, research, regula-
tion (pp. 3 – 22 ). Cullompton : Willan Publishing .
Zulawski , D. & Wicklander , D. ( 1992 ). Practical aspects of interview and interrogation .
New York : CRC .
Cases
R v Oickle [ 2000 ] 2 R.C.S.3 .
R v Rothman [ 1981 ] 1 R.C.S. 640 .
Chapter Two
A Typology of Denial Strategies by
Suspects in Criminal Investigations
Stephen Moston
Department of Psychology, James Cook University
and
Geoffrey M. Stephenson
Department of Psychology, London South Bank University
In this chapter we propose a typology of denials made by suspects in police
interviews. This is an important forensic topic, fi rst, because police offi cers
need to consider in advance of an interview the likely strategies that might be
adopted by those who intend to deceive them concerning their involvement
in a criminal offence. Can offi cers nullify such attempts before the suspect
commits to such a strategy? How should they respond if such a denial is not
averted? Secondly, and perhaps more important still, is the ability to distinguish
between the denials of those intent on deceit and those who are genuinely
innocent of any involvement. Should all denials be treated with scepticism? Or
should all denials be taken at face value pending discovery of contradictory
evidence? Should denials be welcomed on the grounds that evidence will sub-
sequently be used to prove the intention to deceive? Or should denials best
be circumvented by adroit questioning, in the hope that a confession will
subsequently be elicited?
We shall see that experts are not at one on these issues (e.g. Gudjonsson,
2006 ) and, moreover, that there is a dearth of evidence about the form that
denials take generally. Before examining the questions further, we shall describe
Handbook of Psychology of Investigative Interviewing: Current Developments and Future Directions
Edited by Ray Bull, Tim Valentine and Tom Williamson
© 2009 John Wiley & Sons, Ltd.
18
Handbook of Psychology of Investigative Interviewing
briefl y what is known about the opposite stance taken by suspects: confession
to the crime in connection with which they are being interviewed.
Confessions in the p olice s tation
Since the introduction of the Police and Criminal Evidence Act 1984 (PACE)
in the United Kingdom there have been several major observational studies,
with sample sizes in excess of 500 cases, on police interviewing styles and
responses to accusations of criminal wrongdoing by suspects in criminal inqui-
ries by police offi cers (e.g., Moston, Stephenson & Williamson, 1992 ; Baldwin,
1993 ; Moston & Stephenson, 1993 ; Bucke & Brown, 1997 ; Phillips & Brown,
1998 ; Bucke, Street & Brown, 2000 ). In addition, there have been several
smaller - scale studies focusing on specifi c populations (e.g., juveniles; Evans,
1993 ) or focused research issues (e.g., the impact of psychological vulnerabili-
ties on confessions; Pearse, Gudjonsson, Clare & Rutter, 1998 ).
In each of these studies the percentage of suspects who confessed during
interviews has been reasonably consistent, ranging from a low of 55% (Moston
et al. , 1992 ; Phillips & Brown, 1998 ; Bucke et al. , 2000 ) to a high of 62%
(Baldwin,
1993 ), with other studies falling within this narrow range (e.g.
Bucke
& Brown,
1997 : 58%; Pearse
et al. ,
1998 : also 58%; Moston
&
Stephenson, 1993 : 59%; Evans, 1993 : also 59%).
Surprisingly, there have been few comparable studies in other countries. In
the USA there have only been two comparable attempts to quantify confession
rates. Cassell & Hayman (1996) reported that 42% of suspects made confes-
sions or admissions, while Leo (1996) suggested the confession/admission
rate reached 64%. In Australia, Dixon & Travis (2007) provide data on two
separate samples of cases, with one sample giving a confession/admission rate
of 76%, while in the second sample the rate was lower, at 46%.
International comparisons are fraught with diffi culties, as the sample sizes
are often relatively small (e.g., 87 cases in Dixon & Travis ’ s second sample)
and different sampling techniques are used (e.g., Cassell and Haymann ’ s study
used a non - random sample). We must also recognize that police offi cers in
different jurisdictions may be more or less scrupulous about whom they accuse
of a particular crime. A similar variation may also occur between different forces
in the UK. Indeed, there have been well - documented cases of remarkably high
clear - up rates in some forces, suggesting that crimes ‘ taken into consideration ’
(TICs), to which suspects have obligingly confessed, have boosted the percent-
age of successful prosecutions (Gill, 1987 ).
Denials in the p olice s tation
Most suspects who do not confess, in one way or another deny their culpabil-
ity. Others may remain silent or otherwise evade answering directly, although
A Typology of Denial Strategies by Suspects in Criminal Investigations 19
in the UK the numbers so doing may have decreased since a revised caution
used to indicate guilty concealment. Existing data on the percentage of sus-
pects who deny are harder to interpret, with some studies either failing to
mention the denial rate altogether, or collapsing data (including denials,
evasion and use of the right to silence) into a single category, perhaps better
described as a ‘ non - confession ’ rate. Such problems notwithstanding, studies
suggest that in the UK between 35% (Moston & Stephenson, 1993 ) and 45%
(Phillips & Brown, 1998 ) of suspects deny involvement or do not confess.
Data from pre - PACE studies in the UK show a similar pattern, with Softley,
Brown, Forde, Mair & Moxon (1980) reporting a denial rate of 35%, and
Irving (1980) a rate of 42%, suggesting that changes in police questioning
practices following the introduction of PACE and the mandatory tape
-
recording of interviews had little impact on this particular metric of
interviewing effectiveness.
Why do s uspects c onfess?
Moston et al. (1992) analysed over 1,000 cases in the UK in which suspects
were interviewed by police offi cers. In addition to collecting detailed records
of the cases through a questionnaire attached to custody records, they also
analysed over 500 taped records of interviews. Their research found clear links
between three case characteristics (strength of evidence, offence severity and
legal advice) and the outcome of interviews. It was suggested that police
interviewing techniques played a relatively minor role in infl uencing confes-
sions. There were two reasons for this conclusion.
First, most admissions were freely volunteered at the outset of interviews
and those suspects who denied an accusation at the outset of an interview
typically maintained this denial throughout, even in the face of seemingly
incontrovertible proof of their guilt. A similar pattern of results was found by
Baldwin (1993) , with only nine suspects from a sample of 600 showing a
response shift from denial to admission during the course of questioning.
The second and perhaps most surprising observation was that police inter-
viewing skills were not greatly in evidence. Only a limited range of questioning
techniques was employed, and interviewers often appeared to be more nervous
than the suspects they were questioning. Many offi cers concluded interviews
at the fi rst sign of resistance from suspects, whilst others doggedly continued
with repetitive, stereotyped questioning or assertions.
Such fi ndings run contrary to the popular myth concerning police interviews
with suspects. It is probably the case that most people believe that a typical
interrogation begins with an initial denial from the suspect, but that by skilled
questioning the interviewer reveals the inconsistencies in that person ’ s state-
ment or its confl ict with other evidence. When confronted with the high-
lighted contradictions, the suspect recants their denial and makes a full
20
Handbook of Psychology of Investigative Interviewing
confession. Such a dazzling sequence does occur in rare instances, but it is
probably not that common.
Directions in r esearch on p olice i nterviewing
Much psychological research on the process of police interviews with suspects
has focused on how the inappropriate use of interviewing tactics can result in
false confessions from innocent suspects (e.g., Kassin & Gudjonsson, 2004 ;
Kassin, 2005 ). The impact of the same interviewing tactics on the decision -
making of guilty suspects has largely been neglected. One reason for this stands
out above all others: a number of those who deny their involvement in criminal
activity are likely to be telling the truth, with the consequence that routine
employment of deception, ‘ minimization ’ , threats, false promises and other
devious practices may result in an unacceptable number of erroneous convic-
tions; this is especially problematic given the weight that has normally been
accorded to confession evidence. Interviewing (with the aim of securing a
confession) can all too easily be used as a shortcut to a search for independent
evidence. Police offi cers are not often in the position of having undeniable
proof of guilt before suspects are interviewed. If they have such evidence, then
a confession is not essential but merely convenient. In fact, by the same token,
interviews will sometimes be used by police in order to eliminate from further
inquiry those suspects against whom there is little evidence or about whom
there is, rightly or wrongly, a presumption of innocence.
The current emphasis on instructions to police offi cers in the UK is to
interview to establish ‘ the truth ’ and to make no prior assumptions regarding
guilt or innocence. In principle, suspects are to be treated as witnesses, albeit
potentially, but not necessarily, as witnesses to their own guilt. The truth or
falsity of denials, whatever form these denials may take, will be evaluated
according to their plausibility when considered in relation to other evidence.
If rigorously followed, this ethical, ideal route to justice will ensure a minimum
of false convictions. However, police time is not unlimited, and investigations
to prove the truth of a confession obtained in response to skilled questioning
will undoubtedly be more speedily concluded than those made without assis-
tance from the suspect. Research into the veracity or evidence value of different
categories of denial may be of considerable value to the effectiveness and
overall integrity of police interviewing.
The emphasis on avoiding false confessions has such clear importance for
civil liberties that the current emphasis on the need to avoid prior assumptions
of guilt needs little justifi cation. The complexity of establishing the effective-
ness of specifi c interviewing tactics on suspects, be they guilty or innocent,
undoubtedly exerts a powerful infl uence on research design. Farrington ( 1981 :
100) offered the following reasons why it was almost impossible to identify
the effects of varying interrogation styles on suspect behaviour because of
variations in other variables.
A Typology of Denial Strategies by Suspects in Criminal Investigations 21
The relationship between style of interrogation and likelihood of confessing
could be investigated in real life in a correlational study. Such a study could show
that one style was associated with a signifi cantly higher likelihood of confessing
than another. However, it would be impossible to attribute this difference to the
style variable in such a study, because of all the uncontrolled variables. For
example, one style might be preferred by certain police offi cers who were more
effective in inducing confessions for some reason unconnected with style; or one
style might be used with certain suspects who might be more likely to confess
than others (again for reasons unconnected with style). Only an experiment can
isolate the variable of style and demonstrate unambiguously that it infl uences
confessing.
Experiments have indeed established that ‘ confessing ’ can be in
creased by
interviewing tactics, although, unfortunately, at the risk of increasing the false
confession rate. For example, subsequent experimental research on police
interviewing has shown that certain well - used tactics, such as minimization
(Kassin & McNall, 1991 ), can increase the number of people who will falsely
confess to a range of minor transgressions they have not committed. While
such research is interesting, it often neglects to consider the potential impact
of such tactics on guilty suspects. Further, an unwritten assumption seems to
be that guilty suspects are relatively homogeneous and that a given tactic will
be equally effective no matter what variations in offence (e.g., property offences
vs. person offences), suspect (e.g., presence or absence of prior criminal
history), interviewer (e.g., male vs. female) or evidence (e.g., eyewitness vs.
fi ngerprint evidence) might exist. This is an important oversight as such vari-
ables can have a powerful infl uence on the attitudes and assumptions of inter-
viewing offi cers. For example, Stephenson & Moston (1993) found that prior
criminal history and strength of evidence predicted whether or not suspects
were assumed to be guilty prior to questioning in the police station. However,
an experimental study by Weber (2007) involving retired police offi cers found
that only strength of evidence, and not criminal history, predicted presump-
tions of guilt.
It follows that differing assumptions about guilt or innocence may have a
bearing on the choice of interviewing strategy. Experimental research by
Kassin, Goldstein & Savitsky (2003) , using mock interrogators and suspects,
has shown that the presumption of guilt has a strong infl uence on the types
of questions interrogators choose, with presumed guilt prompting more
aggressive interrogations, which constrained the responses of the suspects.
Similarly, Weber (2007) found that presumption of guilt was linked to the
selection of confession - oriented tactics in mock interview plans developed by
(retired) police offi cers.
A similar argument can be advanced for understanding a suspect ’ s strategy
during an interview. Variations in offence, suspect, interviewer or evidence
might also be expected to infl uence how a suspect responds to an accusation,
Handbook of Psychology of Investigative Interviewing: Current Developments and Future Directions Page 6