Cheshire-Cat Logic: The Recurring Theme of Terrorist
Abnormality in Psychological Research1
“In that direction,” the Cat said, waving its right paw round, “lives a Hatter: and in that
direction,” waving the other paw, “lives a March Hare. Visit either you like: they’re both mad.”
“But I don’t want to go among mad people,” Alice remarked.
“Oh, you ca’n’t help that,” said the Cat: “we’re all mad here. I’m mad. You’re mad.”
“How do you know I’m mad?” said Alice.
“You must be,” said the Cat, “or you wouldn’t have come here.”
Terrorism, probably more than any other form of organised violence, forces observors almost
immediately into questioning the mental state of the perpetrators. Terrorist violence is so unusual
and runs so contrary to the accepted standards of social behaviour, that it seems to suggest
psychological anomaly. The often extreme callousness and brutality of terrorist conflict leaves it
difficult for commentators to remain objective when considering the motivations and
personalities of those responsible.
Extreme violence seems to have a special ability to bias our perceptions of the perpetrator’s
psychology. In the immediate aftermath of World War II, an Allied psychologist examined the
Rorschach scores of 16 captured Nazi leaders, including the scores of Hermann Goering and
Rudolf Hess. The psychologist concluded that the Rorschach results showed the individuals were
“hostile, violent, and concerned with death, that they needed status, and that they lacked any
real human feeling.” This matched perfectly with the public expectation of what such men would
1 Published as: Silke, A. (1998). ‘Cheshire-Cat Logic: The recurring theme of terrorist abnormality in psychological
research’, Psychology, Crime, and Law, 4/1, pp.51-69.
be like. However, years later, these same scores were inserted among a selection of scores
belonging to normal, healthy subjects. The mix was then given to a panel of experts who were
unable to distinguish the Nazi scores from the rest, and found nothing whatsoever to support the
earlier derogatory diagnosis (Dworetzky, 1988, p.440).
This is an example of what I call Cheshire-Cat logic. Alice’s first encounter with the Cheshire-
Cat is one of the most memorable scenes of Lewis Carroll’s book. The cat believes only mad
people could inhabit Wonderland, so consequently anyone you meet there, must be mad.
Attribution theory has shown that we tend to view our own behaviour as stemming from
situational or environmental forces, but that we see the behaviour of other people as stemming
from internal forces, such as stable personality traits (Eisen, 1979; Quattrone, 1982). Cheshire-
Cat thinking can be considered a form of attribution error where observers develop expectations
about an individual’s personality based on what the individual does or, as in Alice’s case, where
the individual is located. Perhaps no study better demonstrates such a process in action than
Rosenhan’s (1973) classic work on the staff of psychiatric hospitals.
The major finding of the study was that the staff failed entirely to detect the presence of
normal individuals among the psychiatric patient population. A central theme was how in
many instances the staff “profoundly misinterpreted” normal behaviour, so that it would fit in
with preconceived expectations thay had of the patient’s supposed psychopathology.
Such Cheshire-Cat perceptions can be seen in the way terrorists are perceived in our
society. Terrorism is a topic which provokes extreme perceptions, perceptions which spill easily
into considerations of the actors behind the violence. Misconceptions and prejudices born in the
wake of the amorality of terrorist acts - the wanton destruction of property and the suffering of
victims - if pervasive enough will go on to influence the policies used to combat terrorism and
can have a powerful influence on offical attitudes on how to deal with the terrorists. Adopting
misguided policies, for example by refusing to negotiate in the belief that terrorists cannot be
expected to respond reasonably to concilatory approaches, can needlessly prolong campaigns of
violence and exacerbate the search for acceptable solutions.
While many terrorism researchers contend that terrorist psychopathology is a dead issue,
resolved over a decade ago, the reality is that it has continued to survive as a resilent source of
inspiration on which to base theories. That terrorists are somehow psychological different from
the rest of the population has become an underlying assumption of much, if not most,
psychological research on terrorists in the past 30 years. As Schmid and Jongman (1988) noted,
“The chief assumption underlying many psychological ‘theories’ ... is that the terrorist is in one
way or the other not normal and that the insights from psychology and psychiatry are adequate
keys to understanding.”
In an early review of evidence, Corrado (1981) found virtually no data to support the view that
terrorists were in some way psychologically abnormal and plenty of evidence to suggest just the
opposite. Nevertheless, modern authors (e.g. Johnson and Feldmann, 1992) continue to formulate
hypotheses based on the same unsubstantiated models Corrado disparaged years previously. Has
then, new evidence emerged to warrant the enduring interest in terrorist psychopathology?
The simple answer to this question is no, but then again the paucity of work that has been done
on the individual psychology of terrorists means that recent evidence of any kind is rare. Modern
research efforts are being channelled - productively - into a group and organisational focus (e.g.
Post, 1987a; Crenshaw, 1990). Combined with these more accessible avenues of research, most
general books on terrorism agree with Corrado that one “cannot accept the comforting and
unchallenging assumption that equates [terrorist] behaviour with mental illness” (Taylor and
Quayle, 1994, p.189). Most serious researchers in the field at least nominally agree with the
position that terrorists are essentially normal individuals.
However, there is some dissension in the ranks. Respected theorists such as Wilkinson (1977)
and Laqueur (1977) have made generalisations about terrorists that indicate a perception of them
as being abnormal. Other theorists such as Post (1987b) and Pearlstein (1991), are more
forthright about their beliefs and have built explanations of terrorists around models of abnormal
With regard to criminal behaviour in general, it is worthwhile to note that some authors (e.g.
Raine, 1993) believe that a psychopathological model has enormous explanatory power.
Certainly, as far as violent criminals are concerned there does seem to be more than a little
justification to believe this to be the case (Hodgins, 1995). In the past decade, psychopathy in
particular has attracted enormous interest as an explanation of persistent crime and violence
(Hare, 1996). While such interest may be of value in considering “ordinary decent” criminals, its
usefulness with regard to terrorists is far less certain. While, government agencies may be quick
to view terrorists as simply violent criminals, very few researchers would share this perception
(Schmid and Jongman, 1988).
However, ambiguities about who exactly is a terrorist and who is not, have helped to spread the
government-style perception into wider society. There is still no precise and agreed definition of
terrorism, and some have concluded mordantly that “it is unlikely that any definition will ever be
generally agreed upon” (Shafritz, Gibbons and Scott, 1991). A refreshingly concise definition is
provided by Martha Crenshaw, probably the most prominient psychologist involved in terrorism
research, when she described terrorism as “a particular style of political violence, involving
attacks on a small number of victims in order to influence a wider audience” (Crenshaw, 1992).
The claims as to what behaviours fits into this definition varies enourmously1 but the focus of
this paper is very much that of ‘insurgent’ terrorism. Insurgent terrorism is essentially a strategy
of the weak, adopted “by groups with little numerical, physical or direct political power in order
to effect political or social change” (Friedland, 1992).
In practical terms, ‘insurgent’ terrorists are members of small covert groups engaged in an
organised campaign of violence. This violence is often extreme and frequently indiscriminate.
The terrorists themselves tend to live isolated and stressful lives and enjoy varying levels of
wider support. Nationalist/seperatist groups such as the IRA or ETA have a considerable degree
of support from their ethnic groups where new recruits are relatively easy to find. The
ideological groups such as the Red Army Faction or the Italian Red Brigades have more limited
support and more difficulty in finding dedicated recruits.
1 Interested readers are referred to Silke (1996) for a recent account of the definitional debate.
However, even the ‘popular’ terrorist groups remain a violent and extreme minority within what
is still a minority section of the larger society. While the terrorists, and in particular the larger
ethnic groups, may be largely tolerated within their communities, the number of individuals
actively involved in the campaign of violence is always relatively low. Ultimately, very few
individuals of aggreived minorities go on to become active terrorists. The question has always
been, why did these particular individuals engage in terrorism when most of their compatriots
THE CHESHIRE-CAT THEORY OF TERRORISTS
A concern with this recurring question perhaps explains why theorists have persistently
returned to the concept of psychological abnormality as a basis on which to build an
understanding of terrorists. One example of the resilence of such thinking is provided by Hassel,
who in 1977 proposed a typology of terrorists based on psychopathology. Fifteen years later,
Johnson and Feldmann (1992) drew upon this profile when developing their own model, which
again saw terrorists as essentially possessing pathological personalities.
The new version described terrorist leaders as being “fanatical” and as having “a paranoid and
narcissistic pathology.” The paranoid personality was not reserved for leaders alone, but was
also claimed to be a frequent hallmark of other members. Individuals who actually carried out
attacks, possessed an additional “underlying antisocial matrix.” Finally, it was noted that most
members of terrorist groups possessed what DSM-II referred to as the “inadequate personality.”1
Such typologies have persistently cropped up in the research literature. Time and again, the
importance of one personality type or another has been stressed and linked to certain individuals
1 It is interesting to note that while Johnson and Feldmann’s work was completed in the 1990s, their model did not
look to the available DSM-III-R for reference but instead drew on older DSM editions. The reader will find that the
early editions of DSM are quoted extensively in this article. The reason for this is ultimately a disheartening one.
The peak period of activity in researching the abnormality of terrorists occurred in the late 70s and 80s when DSM-
II and DSM-III were in use. Not surprisingly, researchers drew upon those editions to hypothesise about terrorist
personalities. The more recent research and theorising, such as Johnson and Feldmann’s, has cemented itself around
the earlier work, in the process taking on board the older DSM terms and definitions. To better illustrate these
theorists’ reasoning the same older versions are used in this article.
or roles within terrorist groups. Turco (1987) noted that this mode of thinking was particularly
rampant in the reasoning of law enforcement agencies, who unerringly focused in on the
psychopathic personality. Their claim was, and generally still is, that psychopaths are the foot
soldiers of terrorist groups, providing the core around which the group is built.
However, it is not simply clinically naive police officials who have devoted much effort in
espousing terrorist pathology, a steady flow of researchers have proposed similar, if somewhat
more sophisticated models. For example, Suellwold (1981) proposed that there were two
personality types particularly attracted to terrorism. The first, displayed the characteristics of the
narcissistic personality, while the second could be described as the typical paranoid personality.
Suellwold’s typology provided the basis for a more complex model developed by Turco in 1987.
Turco saw the following personality types as relevant:
“(1) the inadequate personality with an excessive, exaggerated demand and “grand play”; (2) the antisocial
personality with a criminal outlook and direct rational expectations; (3) the paranoid with bizarre demands,
frequent religious overtones and underlying homosexual conflicts; and (4) the hypomaniac or depressive as
whom one must consider a “suicide to be.”” (Turco, 1987)
The temptation to view terrorism in terms of abnormality, particularly such forms of it as the
antisocial, narcissistic and paranoid personality types becomes very understandable once one
realises how appropriately terrorists can be viewed in these terms. The repeated attempts to view
terrorists as possessing abnormal personalities is not simply the result of academic observers
succumbing to the popular myth of the terrorist as madman (though that may have some role).
Rather, observers cannot but fail to notice the striking parallels between the two.
The Terrorist as Psychopath
Corrado (1981) carried out one of the first reviews of the effort to view terrorists as
psychologically abnormal. In his critique, Corrado gave particular emphasis to the personality
disorder of sociopathy or psychopathy (now known as the antisocial personality). He noted that
this was “the most prominent mental disorder linked to political terrorism and terrorism in
He quotes Pearce (1977), a strong advocate of terrorist sociopathy, to illustrate how the terrorist
was viewed. Pearce regarded the terrorist as “an aggressive psychopath, who has espoused some
particular cause because extremist causes can provide an external focal point for all the things
that have gone wrong in his life.”
The following describes the coping style of the typical antisocial personality, and illustrates very
effectively why this pathology holds such attraction for theorists:
“They are driven by a desire to dominate and humiliate others, to wreck vengeance upon those whom they
see as having mistreated them
“People are used as means to an end , often subordinated and demeaned so that they can
vindicate themselves for the grievances, misery, and humiliation they experienced in the past. By provoking
fear and intimidating others, they seek to undo the lowly caste into which they feel they were thrust in
childhood. Their search for power, therefore, is not benign; it springs from a deep well of hate and the
desire for retribution and vindication.” (Millon, 1981)
The parallels between this description of the antisocial personality and anyone engaged in
terrorism is striking. The terrorist is challenging the dominant forces in society with violence,
attempting to replace the established power structure with one of their own choosing. There is a
disregard for life, to the extent that totally innocent individuals may be callously used, for
example, taken as hostages or deliberately targeted in random attacks.
The media rarely goes beyond the sociopath in any discussion of terrorist abnormality, but
among academics other disorders are a source of considerable interest. Pre-eminent among these
lesser known pathologies is the narcissistic personality disorder.
The Terrorist as Narcissist
There is a strong relationship between the antisocial personality and the narcissistic personality.
Millon (1981, p.157) noted broad clinical similarities between the narcissist and the psychopath,
and that both types focused angrily on maintaining their independence. For these personalities,
“... self-determination is a protective manoeuvre; it is a means of countering , with their own
power and prestige, the hostility, deception, and victimization they anticipate from others.”
Such strong similarities with the antisocial personality underlie the attraction the narcissistic
personality has for those speculating about terrorists, and researchers have focused on this
disorder with considerable energy. Both Lasch (1979) and Pearlstein (1991) have said that a
narcissistic disorder lies at the heart of the terrorist personality, and others, such as Post (1987b)
and Suellwold (1981), have indicated that traits characteristic of the narcissist, are common
features of terrorist personalities.
DSM-III (1980) noted that narcissists are likely to disregard “the personal integrity and rights of
others.” Expanding further, Millon (1981) commented that “... narcissists have learned to
devalue others, not to trust their judgements, and to think of them as naive and simpleminded.
Thus, rather than question the correctness of their own beliefs, they assume that it is the views of
others that are at fault. Hence, the more disagreement they have with others, the more convinced
they are of their own superiority and the more isolated and alienated they are likely to become.”
This lack of regard for the welfare of others, fits very well with the nature of terrorist attacks,
where innocent civilians, frequently totally uninvolved with the terrorists’ real enemies and
causes, are deliberately targeted. Women and children, normally shielded from conventional war,
are placed in the front-line by terrorists. Such a profound lack of empathy is a hallmark of both
the antisocial personality and the narcissistic personality.
The fact that most terrorists belong to minority groups, and generally minorities within
minorities, finds an echo in the narcissistic tendency to become increasingly alienated and
isolated. The terrorist becomes segregated from mainstream society, first, by merging with that
segment which condones the terrorist actions, then by moving into the faction which provides
more tangible support for the terrorists, and finally by becoming an active member of the
terrorist group itself. At each step, the individual becomes more secluded and alienated. One
cannot help but believe that a narcissistic personality would slip from the majority to the final
terrorist minority very easily due to their interpersonal attitudes.
Of further interest is the tendency for some commentators, (e.g. Hassel, 1977; Johnson and
Feldmann, 1992), to suggest that the narcissistic personality is particularly common among
terrorist leaders. If true, this would not be especially surprising, for as Millon (1981) points out:
“... the sheer presumptuousness and confidence exuded by the narcissist often elicits admiration and
obedience from others.”
Continuing with the narcissistic personality, it is significant to note that one of the more common
concurrent types involving the narcissist is found in combination with the anti-social personality
type. This narcissistic-antisocial mixed personality (Millon, 1981) seems on surface overview
ideally suited to a potential terrorist. Here we have an individual who takes from the narcissist a
pronounced lack of regard and empathy for others, as well as a tendency to become marginalised
and alienated from society, and melds these characteristics with the antisocial’s willingness to
use aggression and violence in order to maintain their own independence.
The Terrorist as Paranoid
The remaining personality of any significant interest to researchers is the paranoid personality
disorder. DSM-III describes the paranoid as follows:
“The essential feature is a personality disorder in which there is a pervasive and unwarranted
suspiciousness and mistrust of people, hypersensitivity, and restricted affectivity ...
“Individuals with this disorder are ... viewed as hostile, stubborn and defensive. They tend to be rigid and
unwilling to compromise. They often generate uneasiness and fear in others. Often there is an inordinate fear
of losing independence or the power to shape events in accord with their own wishes.” (DSM-III, 1980,
Millon (1981) goes on to give a fuller picture of the typical paranoid:
“To assure their security, they go to great pains to avoid any weakening of their resolve and to develop new
and superior powers to control others. One of their major steps in this quest is the desensitization of tender
and affectionate feelings. They become hard, obdurate, immune, and insensitive to the sufferings of others ...
“Fearful of domination, these personalities watch carefully to ensure that no one robs them of their will.
Circumstances that prompt feelings of helplessness and incompetence, or decrease their freedom of
movement, or place them in a vulnerable position subject to the powers of others, may precipitate a sudden
and ferocious “counterattack.” Feeling trapped by the dangers of dependency, struggling to regain their status
and dreading deceit and betrayal, they may strike out aggressively and accuse others of seeking to persecute
Yet again the attraction for theorists is obvious. The paranoid’s compulsion to strike at higher
power seems to explain the terrorist’s need to attack the incumbent state surrounding them. It
certainly finds a comfortable home when considering nationalist/ separatist terrorists who wish to
become independent from a larger force, but paranoids are also associated with the ideologically
motivated groups, take for example, Suellwold (1981) who claimed there was a high proportion
of angry paranoids among the left-wing Red Army Faction.
The hostility and aggression of paranoids when they are in the weaker position, also seems to
account for the violence of terrorists, who are a disempowered minority within a larger
more powerful majority. Finally, the suggestion put forward by theorists such as Hassel (I977)
and Turco (1987), that terrorist leaders tend to possess this personality type, finds support in the
DSM-III (1980, p.308) statement that paranoids “avoid participation in group activities unless
they are in a dominant position,” and even stronger support from the DSM-III-R claim that
“people with this disorder may be overrepresented among leaders of cults and fringe groups”
Hassel’s (1977) suggestion that it is paranoid and/or narcissistic individuals who tend to become
terrorist leaders is of interest, as the largest concurrent form of the paranoid disorder is the
paranoid-narcissistic mixed personality. Millon (1981) notes that this personality type is much
more prone to psychotic delusions than the more mildly dysfunctional narcissists. Millon goes on
to say of the mixed variant, that “they may propose grandiose schemes for “saving the world,”
for solving insurmountable scientific problems, for creating new societies and so on. These
schemes may be worked out in minute detail and are formulated often with sufficient logic to
draw at least momentary attention and recognition from others” (Millon, 1981, p. 387).
Such a description seems to have utility in describing terrorist leaders whose stated aim is to
bring about a better society (as they interpret it), and often have detailed manifestos as to how
this “new order” will be achieved.
Another mixed personality of great relevance is the paranoid-antisocial mixed personality.
Millon’s (1981) comments on these individuals serves to highlight their significance to the
“These paranoids are characterised best by their power orientation, their mistrust and resentment of others,
and by their belligerent and intimidating manner. There is a ruthless desire to triumph over others, to
vindicate themselves for past wrongs by cunning revenge or callous force ...
“Essential to these paranoids is their need to retain their independence from the malice and power of
others ... Paranoid persecutory themes are filled with dread of being forced to submit to authority, of being
made soft and pliant, and of being tricked to surrender self-determination.” (Millon, 1981, p. 387)
A revulsion of submitting to a higher authority, and a willingness to use “callous force” to
maintain one’s independence and “self-determination,” seem to capture the heart of terrorist
motivation and the ruthless nature of much terrorist violence.
For all of these personalities, we have seen time and again, their applied and implied connections
to the terrorist. Every classification system of terrorists seems to have a place for the antisocial
personality, most have roles for paranoids (especially among leaders), and roughly around half
involve narcissistic personalities in some form. The reasons why theorists have returned again
and again to these personalities has been described above - their relevance is self-apparent. That
may be so, but does the theory correspond to the reality?
EVIDENCE FOR THE CHESHIRE-CAT THEORY
It is surprising to discover, that despite the admittedly attractive theoretical reasons for linking
terrorism and psychopathology, the evidence to support such coupling is neither plentiful nor of
Largely, the theorists who advocate the psychopathology model provide only anecdotal evidence
to support their assertions. For example, Pearce (1977) stressed the importance of sociopathy in
terrorism, but based his conclusions purely on secondary sources such as terrorist
autobiographies, biographies and media interviews. In one case, Pearce made a diagnosis of
psychopathy based mainly on an individual having tattoos on his torso.
Cooper (1978) adopts a similar approach to Pearce’s, when he uses exposés of the career of
German terrorist, Andreas Baader, to justify labelling Baader as a sociopathic personality.
Another relevant case, is provided by Christopher Lasch (1979), who advocated a narcissistic
personality explanation of terrorism. Again, the conclusions were based entirely on secondary
sources, such as the public statements and autobiographical accounts provided by the terrorists.
Lasch claimed there was sufficient evidence in these accounts to justify a diagnosis of
narcissistic personality disorder.
Efforts such as Pearce (1977), Cooper (1978) and Lasch (1979), are typical of the type of
research used to support the notion of terrorist abnormality. Very few studies have attempted to
provide stronger, more direct evidence, but one exception to this general trend, is work carried
out by Hubbard (1978). Together with another psychiatrist, Hubbard examined 80 imprisoned
terrorists in 11 countries. They found that nearly 90% of them had defective vestibular functions
of the middle ear. In addition to causing poor balance and co-ordination, Hubbard claimed this
impairment was linked with antisocial behaviour designed to gain attention and an inability to
relate to other people. In essence, Hubbard seems to be suggesting that people had become
terrorists because of an ear problem.
An unorthodox hypothesis to begin with, Hubbard’s contention is weakened by concerns with the
validity and reliability of his work. He never released detailed descriptions of the data he
gathered or of his analysis procedures, and there have been no replications of his unusual
Stronger evidence for abnormality is provided by Ferracuti and Bruno (1981). They analysed the
few available case histories of individual left-wing Italian terrorists, and found that they rarely
suffered from any serious personality defects. However, a similar analysis showed that among
Italian “right-wing terrorists ... disturbed, borderline, or even psychotic personalities have a
much higher incidence.” They went on to comment that, “In right-wing terrorism, the individual
terrorists are frequently psychopathological ...” and “Even when they do not suffer from a
psychopathological condition, their basic psychological traits reflect an authoritarian-extremist
personality”, which was composed of a mix of various unsavoury features (Ferracuti and Bruno,
At first look, this result seems to be a godsend for the hard-pressed advocates of terrorist
abnormality. However, the support is more ephemeral than it appears. Ferracuti and Bruno did
not actually quote the figures or percentages behind their conclusions. Later, they would say that
the incidence of abnormality among the right-wing terrorists was “limited” (Ferracuti and Bruno,
1983, p.307). Deeper examination revealed that the numbers involved were very small. True, the
incidence of abnormality was higher among the right-wing terrorists than among the left-wing
terrorists, (where abnormality was virtually unknown), but the numbers involved still composed
just a tiny minority of the right-wing’s overall membership.
Weakening the Italian’s position further, was the conclusion of a similar, but more extensive
West German study, which found that right-wing terrorists were no more unbalanced than their
left-wing counterparts (Crenshaw, 1983, p.386).
TERRORISTS IN WONDERLAND: EVIDENCE FROM ALICE’S PERSPECTIVE
Overall then, the research supporting terrorist abnormality has been sparse and of questionably
validity. In contrast, the research suggesting terrorist normality has been both more plentiful, and
in general, of much greater scientific validity. On this latter point, take for example the case of
Morf (1970), who failed to find any clinical evidence for the presence of narcissistic personality
disorder among political terrorists, a finding in stark contrast to that of Lasch (1979).
Significantly, Lasch based his conclusion solely on a study of secondary sources, whereas Morf’s
judgement was the result of personal interviews with the terrorists.
This is an important point - those who suggest abnormality are by and large making this
inference from research on secondary sources. Rasch (1979) said this widespread habit was a
scientific travesty, and that any explanation not backed by direct examination of terrorists,
amounted to little more than “idle speculation.”
On the other hand, those who say terrorists are not abnormal tend to be those who have direct
contact and experience with actual terrorists. The reality of close contact has displaced any
comfortable notions of aberration that may have been harboured. They have learned that
terrorism cannot be dismissed so easily. Yet, this seems to have been a remarkably slow lesson
for others to learn.
Another early example of the value of direct-contact research is provided by Paine (1975), who
reported that the psychiatrists who examined Koza Okamoto, the sole surviving Japanese
terrorist of the Lydda airport massacre, found him to be absolutely sane and rational. In a similar
vein, Wilfried Rasch has provided some excellent research dealing with German terrorists, and
his 1979 paper is perhaps the most cited article in the debate on terrorist abnormality.
A psychiatry professor in Berlin, Rasch examined 11 captured terrorists, including the infamous
Baader, Meinhof, Ensslin and Raspe. After extensive examination, Rasch wrote “nothing was
found which could justify their classification as psychotics, neurotics, fanatics or psychopaths.”
Such a conclusion entirely contradicts Cooper’s (1977) claim that Baader was a sociopath. Rasch
also noted that the terrorists could not be diagnosed as “paranoid,” and he mentions that this was
particularly true in the case of Baader, and the other three already mentioned, which is notable as
these individuals went on to commit suicide while in prison. Rasch’s examination of 40 other
suspected terrorists, also showed no sign of psychological disturbances.
Rasch’s findings were supported by the work of Ferracuti and Bruno (1983) who in their study
on captured terrorists could not find a “psychiatric explanation” for why these people were
engaged in terrorism. Further work by Ken Heskin in Northern Ireland had similar results,
concluding that IRA members could not be diagnosed as psychopathic or mentally abnormal
(Heskin, 1980, 1984, 1994).
Continuing on this theme, Lyons and Harbinson (1986) provide even more compelling evidence.
Taking their sample from the Northern Ireland prison population, they compared terrorist
murderers with non-political murderers. They found that the politically motivated killers were
generally more stable, showed a lower incidence of mental illness, and came from more stable
family backgrounds than their non-political counterparts.
This work gains in significance when one realises the bias which existed in the sample. While
representative of the non-political murderers, the sample was skewed for the political murderers.
In Northern Ireland, murderers are routinely sent for psychiatric assessment, unless the killers are
terrorists. Consequently, the vast majority of terrorists are never psychiatrically assessed. The
only ones included in Lyons and Harbinson’s study are those terrorists whose behaviour in
custody was so abnormal that the authorities felt motivated to have them assessed. The majority
of ‘normal’ terrorists were thus never included. Even so, the ‘abnormal’ terrorists still emerged
as more normal and more mentally stable than the average non-political murderer.
The study found that 16% of the terrorists were mentally ill, but the researchers noted that this
16% was composed mainly of individuals “who seemed to be operating on the fringe of a para-
military organization and who were devoid of discipline. They killed in a most sadistic way while
heavily intoxicated. This small group was by no means typical of the rest and raised the figures
for those [political murderers] under the influence of alcohol. It included three who used a knife,
which is a very rare method of political killing” (Lyons and Harbinson, 1986, p. l97).
Even so, an incidence among this sub-group of only 16% is incredibly low, especially when
compared with an incidence of 58% among the non-political offenders. The actual figure for
terrorists on a whole is almost certainly far less than 16% but clarification of this point must wait
until the authorities are prepared to allow random sampling of the terrorist population.
It is important to remember that the above research isn’t saying that mentally unbalanced or
pathological personalities are never present in terrorist organisations. On the contrary, both
Rasch (1979) and Lyons and Harbinson (1986) did find such individuals in their samples.
However, these individuals were a rarity, being the exception rather than the rule, a finding
supported by Ferracuti and Bruno’s (1981) survey. Further, the research also indicated that when
they do appear, such personalities tend to be fringe members of the terrorist group, rather than
THE CHESHIRE-CAT RETURNS: A NEW INTERPRETATION OF TERRORIST
Even with the best empirical studies rejecting the idea that terrorists are clinically abnormal,
theorists are nevertheless still prone to the prevailing trend of believing terrorists possess many
of the traits of pathological personalities (Bollinger, 1982; Ferracuti, 1983; Post, 1990). The
current predilection seems to be one of placing terrorists in a twilight region bordering on
Take for example, Post (1987b) who seemed to accept the thrust of the empirical findings when
he said “most terrorists would be considered to fit within the spectrum of normality,” and even
commented later that his own “comparative research on the psychology of terrorists does not
reveal major psychopathology” (Post, 1990). So while on the one hand freely acknowledging the
lack of “major” psychopathology, Post and many others, have then been quick to switch the
search towards finding some form of minor psychopathology. Note how Schmid and Jongman
(1988) perceive this trend:
Some authors see little prospect in the search for the terrorist personality and question whether a profile
analogous to the “authoritarian personality” is possible at all. Walter Laqueur, for instance, holds that the
search for a “terrorist personality” is a fruitless one, but a few pages earlier he notes that “Terrorists are
fanatics and fanaticism frequently makes for cruelty and sadism.” Paul Wilkinson is also ambiguous. On
the one hand, he maintains that “We already know enough about terrorist behaviour to discount the crude
hypothesis of a ‘terrorist personality’ or ‘phenotype’,” but at the same time he admits that “I do not believe
we really understand much about the inner motivations of those who readily enunciated terrorist
Post (1987b) is more certain that some form of abnormality exists, and he confidently states his
belief that “individuals with particular personality dispositions are drawn to the path of
Such “dispositions” as described by Post, are largely variants of the three major personality types
already discussed. For example, Shaw (1986) notes that a narcissistic injury is commonly seen as
a central feature of the terrorist’s psychology, and he cites Kozo Okamoto as an example of such
an individual. Yet, earlier we read how Okamoto had been found to be absolutely sane and
rational by the team of psychiatrists who assessed him!
Terrorists are ascribed many pathological characteristics, yet in the new approach they
somehow manage to avoid the actual diagnosis. Such fine distinctions are the hallmark of this
entire trend. Consider the following from Post (1987b, p.24):
Without being frankly paranoid, there is an overreliance on the ego defence of projection. ... Bollinger
(1982) found psychological dynamics resembling those found in narcissistic borderlines ... The terrorists he
interviewed demonstrated a feature characteristic of individuals with narcissistic and borderline
personalities - splitting. [italics mine]
In effect, Post is dancing around the psychopathology issue. Aware that terrorists cannot be
clinically diagnosed, he has gone around this obstacle, and relentlessly ascribes to them many of
the characteristics associated with such diagnoses. He does, however, appear aware of the
weakness of such speculation. As he comments himself:
It is not my intent to suggest that all terrorists suffer from borderline or narcissistic personality disorders or
that the psychological mechanisms [related to such disorders] ... are used by every terrorist. It is my distinct
impression, however, that these mechanisms are found with extremely high frequency in the population of
terrorists, and contribute significantly to the uniformity of terrorists’ rhetorical style and their special
psycho-logic. (Post, 1990, pp. 27-28)
However, Post is somewhat overstating the issue when he talks about an “extremely high
frequency.” In discussions of the personality flaws and life-style problems of terrorists, the
figures involved rarely go higher than 33% of the populations considered. A large series of
studies on 250 West German terrorists (Jager, Schmidtchen and Suellwold, 1981; von
Baeyer-Kaette et al., 1982), supplies much of the hard data behind this line of speculation.
These studies revealed among other things, that 25% of left-wing terrorists had lost one or both
parents by age 14, that 33% reported severe conflict with parents, and that 33% had been
convicted in juvenile court. Facts like this, encouraged the German researchers to view terrorists
as “advancement orientated and failure prone.” Further, other researchers used this German
work to form the basis of their contention that terrorists are “marginal, isolated, and inadequate
individuals from troubled families” (Post, 1986, p.211).
While of interest and suggestive, it is unwise to read too much into these figures. For a start, the
statistics rarely composed a majority of the terrorist population. It would be one thing if 92% had
lost a parent before 14, but 33% is of uncertain significance. It still leaves 67% with both parents,
but who nevertheless went on to become terrorists. More important are the methodological flaws
in this research, and in particular, the lack of valid control groups in virtually all of the studies.
One exception is provided by Ferracuti (1983), who carried out a similar study to the above West
German one, but used politically active youths as controls. Vitally, he found that the life patterns
and backgrounds of the terrorists and the non-terrorists showed no significant differences. This
finding is of considerable importance. If there is no meaningful difference with controls, it brings
into question the whole validity of the line of reasoning adopted by Post and the others. Or at
least it would, except that Ferracuti also reported that there existed some personality flaws in the
terrorists which supported some of Post’s earlier hypothesises.
Stronger evidence for those attempting to find some aberration in the psychology or upbringing
of the terrorist, is provided by a suggestive piece of work looking at the membership of ETA.
Clark (1983) found that although Basques with mixed parents (one Spanish - one Basque) form
only 8% of the total Basque population, 40% of ETA is composed of such individuals. This
intriguing finding has been a rich source of speculation for those looking for an aberration
explanation of terrorists, but it is an isolated result. While it is possible to find some supporting
anecdotal cases dealing with mixed parentage in other terrorist groups, there are no other large-
scale studies with similar findings.
With Clark’s study standing alone, and Ferracuti’s controls having very similar backgrounds to
the terrorists, all that’s really left are the borderline characteristics which have received so much
attention. However, such a focus must ultimately be unfair to the terrorists. Quite simply, the best
of the empirical work does not suggest, and never has suggested, that such a borderline
diagnoses is warranted. As Crenshaw (1981) notes, “what limited data we have on individual
terrorists ... suggest that the outstanding common characteristic of terrorists is their normality.”
Speculation of the type offered by Post (1990) and others, including some like Ferracuti (1983)
who supply mixed results, amounts to the second major attempt to set terrorists psychologically
apart from the rest of the population. The first attempt in the 1970s was more direct, and because
of its directness, its tenets were easily tested, and the concept was refuted. This second wave
however, is more insidious. Its tenets are more subtle and less tangible, and it does not leave
itself open to such direct testing. As such it is dangerously misleading. It encourages the notion
that terrorists are in some respects psychologically abnormal, but it balks at going so far as to
advocate clinical personalities. In the end, the trend has done little except taint terrorism with a
The issue of terrorist psychology is one of vital importance. Ultimately, the individuals involved
in terrorism require a more complex response from society than simply a quest for their
apprehension. Sooner or later, terrorists have to be dealt with in a different milieau, such as
prisoners serving long sentences or as political opponents seated across a negotiating table.
Believing inaccurate and misleading characterisations leads inevitably to damaging policies and
deficient outcomes. Campaigns of violence are needlessly prolonged, property is wastefully
destroyed and people are unnecessarily maimed and killed. It is from this perspective that the
concern arises with how researchers - and the policy makers guided by them - perceive the
psychology of terrorists.
Unfortunately, theorists have left the psychology of the individual corrupted in their wake.
Blatant abnormality is rejected by most commentators, (though as we have seen, there is a
steady stream of speculators who return to this notion). Instead, a pervasive perception
exists that terrorists are abnormal in more subtle ways.
We cannot believe normal people could commit terrorist acts, so we look for abnormality. As
efforts to find gross abnormality fail again and again, attention has been turned to more subtle
shades of aberration, in a seemingly never-ending effort to realise latent preconceptions.
The situation today is somewhat similar to that of the 1970’s when the possibility that terrorists
could be clinically abnormal was taken seriously. Eventually, empirical work, published in the
late 70s and early 80s, largely - but not entirely - put paid to that extreme view. Now we must
await the empirical research that can adequately assess the truth of the modern refined version of
those original claims. However, with the research focus now firmly shifted to group psychology,
a final rebuttal may be a long time in the coming.
The Cheshire-Cat theory of terrorists does not hold up to close scrutiny. While the attraction to
view terrorists as being abnormal is understandable, for now, the evidence allows only one
conclusion: terrorists are normal people. Without doubt, while the individual actors may be
generally normal, the activity itself most certainly is not. This is the heart of the whole Alice in
Wonderland argument, that normal people can do abnormal things.
Those who are content to adopt a Cheshire-Cat perception of terrorists, would do well to
remember Alice’s reaction to the grinning feline. To her credit, Alice was not convinced by the
cat’s reasoning and came up with her own theory to explain her predicament. Her answering
logic was that she was still a normal, rational person, but that she was trapped in a very
abnormal, bizarre place (i.e. Wonderland). As far as the Cat was concerned, simply being in
Wonderland was enough to guarantee that an individual was mad, but Alice knew better. It did
not guarantee abnormality. It only guaranteed that others would prefer to see you as abnormal.
American Psychiatric Association, (1980). Diagnostic and statistical manual of mental disorders (3rd
edition), Washington, D.C.: American Psychiatric Association.
American Psychiatric Association, (1987). Diagnostic and statistical manual of mental disorders (3rd
edition, revised), Washington, D.C.: American Psychiatric Association.
Bollinger, L. (1982). Analysen zum terrorismus 3: Gruppenprozesse. Cited in Post, J.M. (1987).
Carroll, L. (1865/1971). “Alice’s Adventures in Wonderland.” In D.J. Gray (Ed.), Alice in
Wonderland, London: Norton.
Clark, R.P. (1983). Patterns in the lives of ETA members. Terrorism, 6, 423-454.
Cooper, H.H.A. (1977). What is a terrorist: A psychological perspective. Legal Medical Quarterly, 1,
Cooper, H.H.A. (1978). Psychopath as terrorist. Legal Medical Quarterly, 2, 253-262.
Corrado, R.R. (1981). A critique of the mental disorder perspective of political terrorism.
International Journal of Law and Psychiatry, 4, 293 - 309.
Crenshaw, M. (1981). The causes of terrorism. Comparative Politics, 13, 379-399.
Crenshaw, M. (1983). The psychology of political terrorism. In M.G. Hermann (Ed.), Political
Psychology: Contemporary Problems and Issues, London: Jossey-Bass.
Crenshaw, M. (1990). The logic of terrorism: terrorist behavior as a product of strategic choice. In W.
Reich (Ed.), Origins of Terrorism, Cambridge: Woodrow Wilson Center.
Crenshaw, M. (1992). How terrorists think: what psychology can contribute to understanding terrorism. In
L. Howard (Ed.), Terrorism: Roots, Impact, Responses (pp. 71-80), London: Praeger.
Dworetzky, J.P. (1988). Psychology (3rd Edition), St. Paul: West Publishing.
Eisen, S.V. (1979). Actor-observer differences in information inferences and causal attribution. Journal of
Personality and Social Psychology, 37, 261-272.
Ferracuti, F. (1983). Psychiatric aspects of Italian left wing and right wing terrorists. Paper presented to
VIIth World Congress of Psychiatry, Vienna, Austria, July, 1983.
Ferracuti, F. and Bruno, F. (1981). Psychiatric aspects of terrorism in Italy. In I.L. Barak-Glantz and
C.R. Huff (Eds.), The Mad, the Bad and the Different: Essays in Honor of Simon Dinitz (pp. 199-213),
Lexington, Mass: Heath.
Ferracuti, F. and Bruno, F. (1983). Italy: a systems perspective. In A.P. Goldstein, and M.H. Segall
(Eds.), Aggression in Global Perspective (pp. 285-312), New York: Pergamon.
Friedland, N. (1992). Becoming a terrorist: social and individual antecedents. In L. Howard (Ed.),
Terrorism: Roots, Impact, Responses (pp. 81-93), London: Praeger.
Hare, R.D. (1996). Psychopathy: theory, research, and implications for society. In D.J. Cooke, A.E. Forth,
J. Newman, and R.D. Hare (Eds.), International Perspectives on Psychopathy (pp. 4-5), Leicester: The
British Psychological Society.
Hassel, C. (1977). Terror: The crime of the privileged - an examination and prognosis. Terrorism : An
International Journal, 1, 1-16.
Heskin, K. (1980). Northern Ireland: A Psychological Analysis, Dublin: Gill and Macmillan.
Heskin, K. (1984). The psychology of terrorism in Northern Ireland. In Y. Alexander and A. O’Day
(Eds.), Terrorism in Ireland (pp. 88-105), Beckenham, Kent: Croom Helm.
Heskin, K. (1994). Terrorism in Ireland: The past and the future. The Irish Journal of Psychology, 15,
Hodgins, S. (1995). Major mental disorder and crime: An overview. Psychology, Crime and Law,
Hubbard, D. (1978). Terrorism and protest. Legal Medical Quaterly, 2, 188-197.
Jager, H., Schmidtchen, G. and Suellwold, L. (Eds.) (1981). Analysen Zum Terrorismus 2: Lebenslauf-
Analysen, Darmstadt: Deutscher Verlag.
Johnson, P.W. and Feldmann, T.B. (1992). Personality types and terrorism: Self-psychology
perspectives. Forensic Reports, 5, 293 - 303
Laqueur, W. (1977). Terrorism, London: Weidefeld and Nicolson.
Lasch, C. (1979). The Culture of Narcissism, New York: W.W. Norton.
Lyons, H.A. and Harbinson, H.J. (1986). A comparison of political and non-political murderers in
Northern Ireland, 1974-84. Medicine, Science and the Law, 26, 193-198.
Millon, T. (1981). Disorders of Personality: DSM-III: Axis II, Chichester: John Wiley and Sons.
Morf, G. (1970). Terror in Quebec - Case Studies of the F.L.Q,., Toronto: Clark, Irwin.
Paine, L. (1975). The Terrorists, London: Robert Hale.
Pearce, K.I. (1977). Police negotiations. Canadian Psychiatric Association Journal, 22, 171-174.
Pearlstein, R.M. (1991). The Mind of the Political Terrorist, Wilmington: Scholarly Resources.
Post, J.M. (1987a). Group and organisational dynamics of political terrorism: implications for
counterterrorist policy. In P. Wilkinson, and A.M. Stewart (Eds.), Contemporary Research on Terrorism,
Aberdeen: Aberdeen University Press.
Post, J.M. (1987b). Rewarding fire with fire: effects of retaliation on terrorist group dynamics. Terrorism,
Post, J.M. (1990). Terrorist psycho-logic: Terrorist behaviour as a product of psychological forces. In
W. Reich (Ed.), Origins of Terrorism: Psychologies, Ideologies, States of Mind, New York: Cambridge
Quattrone, G.A. (1982). Overattribution and unit formation: When behaviour engulfs the person. Journal
of Personality and Social Psychology, 36, 247-256.
Raine, A. (1993). The Psychopathology of Crime, London: Academic Press.
Rasch, W. (1979). Psychological dimensions of political terrorism in the Federal Republic of
Germany. International Journal of Law and Psychiatry, 2, 79 - 85.
Rosenhan, D.L. (1973). On being sane in insane places. Science, 179, 250-258.
Schmid, A.P. and Jongman, A.J. (1988). Political Terrorism (2nd Edition), Oxford: North-Holland
Shafritz, J.M., Gibbons, E.F. Jr. and Scott, G.E.J. (1991). Almanac of Modern Terrorism, Oxford: Facts
Shaw, E.D. (1986). Political terrorists: dangers of diagnosis and an alternative to the psychopathology
model. International Journal of Law and Psychiatry, 8, 359-368.
Silke, A.P. (1996). Terrorism and the blind men’s elephant. Terrorism and Political Violence, 8, 12-28.
Suellwold, L. (1981). Cited in J.M. Post, (1987b). Rewarding fire with fire: effects of retaliation on
terrorist group dynamics. Terrorism, 10, p.24.
Taylor, M. and Quayle, E. (1994). Terrorist Lives, London: Brassey’s.
Turco, R.M. (1987). Psychiatric contributions to the understanding of international terrorism.
International Journal of Offender Therapy and Comparative Criminology, 31, 153 - 161
von Baeyer-Kaette, W., Classens, D. Feger, H. and Neidhardt, F. (1982). Cited in J.M. Post, (1990),
Terrorist psycho-logic: Terrorist behaviour as a product of psychological forces. In W. Reich (Ed.), Origins
of Terrorism: Psychologies, Ideologies, States of Mind, New York: Cambridge University Press. p. 28.
Wilkinson, P. (1977). Terrorism and the Liberal State, London: Macmillan.