Behaviour Research and Therapy 42 (2004) 1227–1255

www.elsevier.com/locate/brat

Fear of contamination
S. Rachman 
Department of Psychology, University of British Columbia, Vancouver, Canada, V6T 1Z4
Received 6 August 2003; accepted 6 October 2003

Abstract

Compulsive cleaning is an attempt to remove feelings of contamination that threaten one’s physical
health, mental health or ability to function socially. The fear of becoming contaminated can be complex,
powerful, persistent and easily spread. Contamination is defined, the main types of contaminants set out
and the characteristics of the fear are described. The distinction between normal and abnormal feelings of
contamination is considered, and abnormal beliefs about contamination are analysed. Attention is drawn
to the fact that contamination can occur without any physical contact, and the concept of mental pol-
lution is used to elucidate this process. The causes and consequences of contamination fears are
described, and some connections between fear and disgust are considered. The concept of cognitive co-
morbidity is applied to an analysis of associations between the fear of contamination and obsessions,
social fears and phobias. It is suggested that applying cognitive analyses and tactics may improve our
ability to treat these powerful and tenacious fears.
# 2004 Elsevier Ltd. All rights reserved.

Keywords: Compulsive cleaning; Contamination; OCD; Pollution

1. Introduction

Compulsive cleaning is driven by fears of contamination. It is an attempt to clean away a per-

ceived contaminant in order to reduce or remove a significant threat posed by the contaminant.
Contamination can threaten one’s physical health, mental health and social life. The con-
taminants fall into four broad classes: disease, dirt/pollution, mental pollution, harmful sub-
stances.
The fear of contamination is the most fascinating of all human fears. It is complex, powerful,
probably universal, easily provoked, intense, difficult to control, extraordinarily persistent, vari-


Tel.: +1-604-822-5861.
E-mail address: rachman@interchange.ubc.ca (S. Rachman).

0005-7967/$ - see front matter # 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.brat.2003.10.009
1228 S. Rachman / Behaviour Research and Therapy 42 (2004) 1227–1255

able in content, evident in all societies, often culturally accepted and even prescribed, tinged
with magical thinking and full of psychological twists and turns. The fear spreads rapidly and
widely, does not decay and is usually caused by physical contact with a contaminant. Surpris-
ingly, a fear of contamination can also be established without physical contact. Abnormally
strong fears of contamination are unyielding, expansive, persistent, commanding, contagious,
and resistant to ordinary cleaning.
From a clinical perspective, fears of contamination are important because they feature so
prominently in OCD. ‘‘Obsessional-compulsive cleaning disorders’’ are based on ‘‘a fear of con-
tamination, liberally defined to include ‘mental contamination’’’ (Rachman and Hodgson, 1980,
p.113). Because the compulsions are prominent and so bizarrely unadaptive they became almost
definitional of OCD. Over time, the compulsive cleaning is so well practised that it becomes
stereotyped and even robotic. The fears of contamination are challenging because the current
treatment, exposure and response prevention, can reduce the fear but it is demanding and many
patients find it exhausting. Too high a proportion decline the treatment. Moreover there are
indications that even when the treatment succeeds in reducing the fear, it is prone to return,
often in an intense form. What started out as the easiest type of OCD fear to treat no longer
holds that position. For example, Coelho and Whittal (2001) recently reported that patients
with compulsive cleaning rituals, virtually always associated with a fear of contamination, who
participated in a controlled trial, responded less well to treatment than did patients with other
forms of OCD. Evidently we need a fresh analysis of the fear of contamination, and hopefully
one from which an improved form of treatment can be deduced. In keeping with recent advan-
ces the most obvious and first choice is a cognitive analysis.
Of all the manifestations of OCD, the dread of contamination is the most obviously phobic in
nature (Rachman and Hodgson, 1980). The associated cleaning compulsions are the second
commonest form of OCD compulsion, exceeded only by compulsive checking. In a sample of
560 people with OCD, Rasmussen and Eisen (1992) found that 50% had fears of contamination,
very similar to an earlier figure of 55% compiled from a series of 82 patients seen at the Maudsley
and Bethlem Hospitals in London by Rachman and Hodgson (1980).
As with virtually all human fears there appears to be a continuum of fears of contamination
ranging from the mild and circumscribed to moderate fears and ultimately to those which are
abnormally intense, abnormally extensive and abnormally sustained by belief and conduct. The
present analysis goes beyond the conventional boundaries of contamination to include the con-
cept of mental pollution (Rachman, 1994), a phenomenon that like psychology itself can be said
to have a long past but short history. Centuries ago John Bunyan introduced the term ‘‘mental
pollution’’ in describing his horrifying blasphemous obsessions—‘‘tumultuous thoughts’’ . . .
‘‘masterless hellhounds (that) roar and bellow and make a hideous noise within me,’’ (Bunyan,
1947 edition, p.136). Unlike most types of contamination, mental pollution, a sense of internal
dirtiness, can be provoked by direct physical contact with a contaminant or by indirect contacts
such as insults, moral criticisms, objectionable intrusive thoughts, memories, symbolic associa-
tions. These feelings of internal dirtiness, mental pollution, are imperfectly connected to observ-
able, identifiable sources of pollution. ’’It looks clean but makes me feel dirty.’’ The feelings are
not properly responsive to cleaning.
 S. Rachman / Behaviour Research and Therapy 42 (2004) 1227–1255 1229

2. Contamination defined

Contamination is an intense and persisting feeling of having been polluted or infected or

endangered as a result of contact, direct or indirect, with a person/place/object that is perceived
to be soiled, impure, infectious or harmful. The feeling of contamination is accompanied by
negative emotions among which fear, disgust, dirtiness, moral impurity and shame are promi-
nent. Typical examples of pollutants are fecal matter, putrefying flesh, decaying vegetable mat-
ter. Infectious contaminants include contact with items or people carrying germs, contact with
bodily products such as blood/saliva, contact with people or places believed to be infected (e.g.
blood, hospitals, places/people thought to be associated with sudden acute respiratory syn-
drome, AIDS). Potentially harmful substances such as chemicals, pesticides and certain foods
can be sources of contamination. Mental pollution can be caused by associations with impurity,
dirtiness (rupophobia), immorality, accusations, nasty memories, direct or remote contacts with
people who are regarded as enemies or as untouchables (culturally or personally defined), and
by aversive events such as a sexual assault.
Contact with any of these four sources of contamination—soiled, infectious, harmful sub-
stances, mentally pollutant—produces fear usually accompanied by a sense of dirtiness and
instigates vigorous attempts promptly to remove the contaminant, most frequently by cleaning.
‘‘My hands feel aflame with contamination’’. The feeling of contamination also triggers avoid-
ance behaviour and attempts at prevention by removing potential sources of contamination.
The idea can be summed up in this way: ‘‘Avoid if you can, but escape if you can’t’’. The affec-
ted person attempts to prevent the spread of the contamination by ‘‘isolating’’ their hands, for
example by using their feet or elbows to open doors. If this is not practical they will resort to
protecting themselves by wearing gloves or holding tissues. A simple demonstration of this feel-
ing and its consequences can be carried out by asking people to insert their fingers into a jar
containing sticky jam. It makes them feel dirty, and so they isolate their hands, avoid touching
their clothing or face and hair, and have a powerful urge to wash away the offending jam.
In clinically significant fears of contamination it is believed that the infectious/polluted/
dangerous substances will cause serious harm to the person’s physical and/or mental health or
present a social threat. It is also believed that the contamination will persist until adequate
cleaning has been completed, but it is difficult for patients to achieve certainty about the suf-
ficiency of their cleaning. It is believed that unconstrained contamination will spread to other
parts of one’s body, clothing, possessions. It is almost always believed that the contamination
is transmissable to other people, places or objects. It comes as no surprise that affected
people complain of a lack of control over the waves of contamination. Compulsive cleaning
which over-rides the person’s rational appraisals is behaviour that is partly out of control and
abnormal.

3. Phenomena

Aspects of contamination that are worthy of close consideration include its easy trans-
missability, its failure to degrade and its proneness to return even after initially successful ther-
apy. One of the most remarkable properties of contamination is how easily and widely the sense
1230 S. Rachman / Behaviour Research and Therapy 42 (2004) 1227–1255

of contamination transfers from object to object, from person to person, from person to objects,
and from objects to persons. The contamination is generally transmitted at full strength. More-
over, a small amount of contamination goes a long way (in both senses). (Mental pollution is an
exception to the ease of transmission from person to person, or person to objects, see below).
The rapid and ever-widening spread of the feeling of contamination has no parallel in fear,
except of course in the fear of contamination itself. In fears of contamination the person’s con-
strual of the contamination component dominates the path and the shape of the fear.
The transmission of contamination is determined primarily by contact. In specifiable circum-
stances, some fears can be transmitted from person to person (see Rachman, 1990, 2004 and
below for a discussion of the vicarious acquisition of fears) but the large majority of significant
fears remain confined to the fearful person; transmission of an abnormal fear of contamination
to another person is not common. Transmission of a feeling of contamination from one object
to another can, however, take place if the person perceives a significant similarity between the
already contaminated item and a neutral but similar item; no physical contact is necessary. The
transmission of fears of contamination is often subject to two prominent laws of sympathetic
magic in a manner similar to the transmission of disgust as described by Rozin and Fallon
(1987). The laws of contagion and of similarity which operate in the induction or transmission
of disgust (see below for details), can also operate in the transmission of contamination fears.
There is a curious asymmetry in the spread of contamination. A teaspoonful of contaminated
fluid is sufficient to spoil an entire barrel of clean water. However, a teaspoonful of clean water
will do absolutely nothing to cleanse the contents of a barrel of contaminated water. The same
asymmetry, in which the transmission of a sullied substance overwhelms a clean substance, is
observed in the transmission of contamination from person to person, or even from group to
group. A person from a group thought to be sullied or polluted, such as an ‘‘untouchable’’ in
India, can ‘‘contaminate’’ a person of a higher and cleaner status even by mere proximity.The
reverse rarely occurs; a person of high status cannot ‘‘cleanse’’ an untouchable person by a
direct or indirect contact. The entry of a contaminated person into an unsullied group will con-
taminate the group but the entry of a clean person into a ‘‘contaminated’’ group will do nothing
to cleanse that group.
The prevalence of such beliefs about contamination makes it possible to deliberately contami-
nate a rival or enemy. Contamination can be created in the obvious way by spreading dirt,
germs, anthrax, etc., and also by less obvious but equally effective non-physical (’’mental’’)
manipulations. Passing on information, correct or incorrect, can produce feelings of contami-
nation—e.g. that dangerous chemicals have seeped into the water supply, or were used in food
preparation. It is also possible to create mental pollution by labelling a person or a group of
people (e.g.untouchables) as polluted or contaminated. This technique was used in the Soviet
Union (Applebaum, 2003, p.121) and in Nazi Germany (Gilbert, 1986, pp. 23–31) to exclude,
imprison or kill political or personal enemies. Describing people as vermin was commonly a
precursor of isolation and violence. According to Human Rights Watch Report (1999), this
chain persists in parts of modern India.
Feelings of contamination tend not to degrade spontaneously and even under treatment are
slow to degrade. As Tolin et al. (2004) demonstrated in an elegant study, contamination passes
from object to object with virtually no loss of intensity. A contaminated pencil was used to
touch neutral pencils, and the same level of contamination was thereby transferred from pencil
 S. Rachman / Behaviour Research and Therapy 42 (2004) 1227–1255 1231

to pencil without loss of intensity. This demonstration is consistent with reports made by OCD
patients experiencing feelings of contamination. Objects that were felt to be contaminated five,
ten, even twenty years previously retain their original level of contamination! It is not uncom-
mon for affected people to seal off parts of their homes and even lock up rooms that contain
contaminated material. The absence of perceived degradation of contaminated materials is often
associated with another remarkable feature of OCD contamination.
Most patients retain a precise memory of the nature and exact whereabouts of contaminated
material, even going back as far as thirty years or more. For example, a patient was able to
recall the exact spot in the hospital parking lot where he had seen a discarded stained band-aid
ten years earlier. He was still avoiding the affected area. Characteristically he was able to
describe in detail the original stained band-aid, its location, and the position of his own car.
Others will recall the exact spot on a shelf on which a container of pesticides was briefly placed
years earlier, and so forth. In these common instances one observes a familiar combination of
the non-degradable quality of the contaminant and an enhanced memory. The results of an
experiment by Radomsky and Rachman (1999) are consistent with the clinical observations in
showing that people with fears of contamination display superior recall of contaminated objects
relative to anxious controls without this particular fear. The participants with a fear of contami-
nation displayed a superior memory for those items which had been touched by a ‘‘contami-
nated’’ cloth in an array of 50 items, half of which were free of contact with the contaminant.
Incidentally the clinical observations and experimental findings which indicate superior memory
among people with OCD, under specifiable conditions, are difficult to reconcile with the idea
that these people suffer from a memory deficit probably attributable to biological abnormalities
(see reviews by Clark, 2004; Tallis, 1997).
It can be said that the contamination and the memory of the contamination are both non-
degradable. The patients require no reminders of what has been contaminated and where it lies
festering. Interestingly they all appear to know, without the benefit of curious psychological
experiments, that (OCD) contaminants are not degradable. If the contamination of places or
objects does not degrade, it remains threatening and the location of contaminated items, people
and places is therefore worth remembering. Presumably they will be less well remembered if by
one means or another the person begins to believe that the particular contamination has
decayed. Is it possible for OCD contamination to degrade?
Some interventions are capable of promoting the degradation of the contaminant and a
reduction of the fear of the contaminant. The best established method is exposure and response
prevention ( Abramovitz, 1997; Barlow, 2002; Clark, 2004; Rachman and Hodgson, 1980) in
which repeated, controlled exposures to the contaminants are followed by steady and progress-
ive reductions in fear. This process is facilitated by ensuring that the patient refrains from clean-
ing away the contaminants.The reductions in fear are accompanied or followed by a
degradation of both the original contaminant and the contamination induced during the
exposure sessions. The fear and the associated cleaning compulsions decline. Not infrequently,
this overall progress is punctuated by partial returns of the fear, especially between sessions.
Intra-session habituation of the fear generally exceeds inter-session habituation, especially in the
early stages of therapy. It is also known that the fear evoked by the exposures to the contami-
nant, and the accompanying urges to clean, decline ‘‘spontaneously’’, if somewhat slowly, even
when the patient refrains from their customary post-contact cleaning compulsions. In other
1232 S. Rachman / Behaviour Research and Therapy 42 (2004) 1227–1255

words, in a controlled planned series of exposures the fear induced by contamination will tend
to dissipate. The question ‘‘What is the degree of contamination now?’’, is rarely asked but it is
safe to assume that in these therapeutic circumstances the level of contamination does indeed
degrade. The behavioural changes observed during fear reduction treatment, especially the
decline in avoidance behaviour, strongly suggest that the intensity of the contamination has also
diminished.
Thus far no distinction has been made between OCD contamination and the accompanying
fear; traditionally the two terms have been used interchangeably. It is, however, possible to
make a distinction and it is not inconceivable that a reduction in an OCD fear may leave the
level of contamination itself unchanged but no longer regarded as a source of danger and no
longer fearful. Although contamination is almost always uncomfortable, even distressing, it is
not always fearful. Treatment may detach the fear component from the contaminant. It remains
to be determined whether the gradual decline of fear is accompanied, preceded or followed by a
degradation of the contamination.
Such a determination might throw light on the model set forward here, namely that the pros-
pect of becoming contaminated and therefore harmed, causes the fear. The prospect of being
contaminated can be broken down into the two factors of probability and seriousness: the prob-
ability of being contaminated,and secondly,the prospect of sustaining a serious degree of con-
tamination. It is predicted that during therapeutic exposures the severity factor begins to
diminish first and is followed by a decline in the probability factor. The other possibility men-
tioned above, of a detachment of the fear from the contaminant, seems less likely but may occur
in some circumstances. The cognitive mediation of contamination fears may produce detach-
ments (’’yes, my hands remain contaminated but I now recognise that it is not dangerous;
uncomfortable perhaps but not a danger’’).
As with other types of fears, contamination fears collapse downwards. For example, during
exposure treatment a reduction of the fear of a contaminated item that is high on the patient’s
hierarchy is generally followed by a spontaneous (i.e. non-treated) decline in contamination
fears lower down in the hierarchy. (A reduction in the high level of contamination of an item at
the top end of a hierarchy of contaminants may well collapse the items lower in the hierarchy.)
The exceptions to this downward collapse of the fears stem from categorical differences, that is
from the patient’s belief that the non-treated source of contamination is categorically different
from the treated source (e.g. a reduction in the fear of pesticides is most unlikely to generalize
to a reduction in the fear of being contaminated by AIDS). In circumstances in which the categ-
orical differences, or the similarities, between contamination fears are of experimental or clinical
importance, it is however prudent to bear in mind that the fearful person’s verbal report of the
similarities and differences may be misleading . The results of an experiment on people with
multiple fears showed that there was a large discrepancy between the respondents’ appraisals of
the similarity of their fears and the results of behavioural tests of the similarities of their various
fears (Rachman and Lopatka, 1986). Moreover, the respondents’ predictions of how the
reduction of one of their fears would affect their ‘‘untreated’’ fear/s were no better than chance.
A specific test of the ‘‘accuracy’’ of patients’ subjective appraisals of the similarities and differ-
ences between their contamination fears would prove useful.
 S. Rachman / Behaviour Research and Therapy 42 (2004) 1227–1255 1233

4. What causes a fear of contamination?

As there is no a priori reason to assume that the fear of contamination is caused by unique
factors or circumstances, the question of causation will be approached from one of the prevail-
ing theories of the development of human fears, namely the three pathways theory of fear
acquisition (Rachman, 1978, 1990, 2004). According to this theory, the three pathways consist
of conditioning, vicarious acquisition, the transmission of fear-inducing information.
There is a good deal of laboratory evidence that fears can be the product of conditioning, but
there are several reasons for discounting an exclusively conditioning explanation for the devel-
opment of fear (e.g. the skewed distribution of fears, failures to develop fears despite condition-
ing trials, failures to develop fears after repeated exposures to aerial bombing, significant fears
in the absence of relevant conditioning exposures, etc.). It became apparent during the shift
towards cognitive explanations of behaviour and experience that the conditioning theory must
be expanded to incorporate two additonal pathways to fear: vicarious acquisition by observa-
tional learning (Bandura, 1969) and by the transmission of threat-relevant information (e.g.
‘‘the water in your cup is dangerously contaminated!’’). A powerful illustration of the informa-
tional genesis of intense fear is provided by recurrent epidemics of koro in S.E. Asia (Rachman,
2004). Rumours of an outbreak of koro, a fear of male genital shrinkage and impending death,
caused thousands of people to panic. One epidemic of koro was set off in Singapore by a
rumour that the Vietcong had contaminated the food supply.
Each of the pathways can be illustrated by case histories. A highly responsible grandfather
was caring for his infant granddaughter,as promised, despite feeling ill. He had what seemed to
be a bad cold (actually flu) and was sneezing and coughing but persisted in carrying out his
obligations. Late that night the parents realised that the baby was struggling to breathe and
rushed her to emergency. She had such a serious fever and respiratory difficulties that the doc-
tors immediately admitted her to hospital where she received intensive treatment. The cause of
the infant’s illlness was medically unclear but the grandfather interpreted it as his fault, feeling
that he had transmitted his flu to the child. As a result he developed a strong fear of disease
contamination and took to strenuous, repeated cleaning with disinfectants. He became preoccu-
pied with the fear that he was at risk of becoming contaminated and possibly die, and unsurpis-
ingly was terrified of transmitting diseases to his family. Another patient developed a vicarious
fear of disgust contamination after witnessing a friend slip and fall into a deep puddle of pig
manure during a holiday in the countryside. The friend screamed as she fell into the puddle and
after climbing out was filthy and uncontrollably distressed for hours. Shortly after this event the
patient became highly sensitive to dirt and began washing vigorously and frequently, especially
before leaving his home.
A full account of the present status of the fear-acquisition theory is provided elsewhere
(Rachman, 2004) and for present purposes three points merit attention. Firstly, a fear of con-
tamination can be generated by the transmission of threatening information; physical contact
with a contaminant is not a necessary condition for the fear to emerge. Secondly, a fear of con-
tamination assuredly can be generated by observing the frightened reactions of other people to
actual or threatened contact with a notorious contaminant. Thirdly, conditioning processes can
establish disgust-reactions in a manner comparable to conditioned fear reactions. It remains to
1234 S. Rachman / Behaviour Research and Therapy 42 (2004) 1227–1255

be seen whether fear and disgust can be simultaneously conditioned. There is no obvious barrier
to this possibility.
The interesting question of whether people, some or all of us, are predisposed easily to
acquire ‘‘prepared’’ fears (very likely so) re-emerges now because it is suggested that we are also
predisposed to react with disgust to aversive/dangerous materials such as putrefying flesh,
decaying vegetable matter, foul-smelling water, etc. The fact that young children do not display
the expected reactions of disgust does not rule out the possibility of a biologically-based predis-
position any more than does the absence of certain fear reactions among the very young rule
out the likelihood of prepared phobias. Prepared fears, and presumably ‘‘prepared disgusts’’ as
well, can emerge during early maturation. The marked individual differences in prepared fears
may be attributable to learning experiences—probably we learn to overcome our prepared reac-
tions by increasing knowledge and self-efficacy. In a seeming paradox, the more experience one
has with a potentially fearsome animal or place,the lower the likelihood of of developing a sig-
nificant fear. We learn to ‘‘not fear’’ (Rachman, 1978, 1990). A telling demonstration of learn-
ing to not-fear was provided by Di Nardo et al. (1988) who showed that people with lots of
contact with dogs, even if it included some aversive experiences with them, were notably less
likely to have a phobia of dogs than were people who had few experiences with dogs. The pros-
pect of immunizing people to feelings of disgust by systematically exposing them to disgusting
material can safely be left in the realm of pointless academic speculation.
We also have to consider whether some fears of contamination might arise without any rel-
evant learning experiences. It has been suggested that at least a few fears arise in this manner;
they have always been present. Poulton and Menzies (2002) set out the case for some of these
‘‘non-associative’’ fears (see Craske, 2003, for a critical view). Subject to new information it
seems preferable at this stage to allow for the possible occurrence of prepared fears of contami-
nation.

5. Appraisals, persistence and predispositions

Whatever the proximal cause of the fear, the persistence and elaboration of the fear will be
determined by the person’s interpretation of the personal significance of the instigating event/s.
If the events are interpreted as a continuing threat to one’s physical and/or mental health (e.g. I
remain at risk of developing AIDS) the fear will persist and become elaborated (e.g. the threats
spread ever wider). However, if the events are interpreted as being insignificant and/or circum-
scribed, then the fear will fade (e.g. it was a nasty event but a one-off, or, it was a nasty event
and we were all equally upset at the time). The maintenance of a fear of contamination is facili-
tated by maladaptive cognitions and by the adoption of self-defeating safety behaviour, notably
avoidance and complusive cleaning.
Are there any factors that might predispose people to develop fear and or disgust reactions,
and in particular are there any factors that might predispose people to develop a fear of con-
tamination? In addition to the predisposing factors that are thought to increase one’s vulner-
ability to significant fears (Barlow, 2002; Craske, 2003), there are some indications that a
specific sensitivity to disgust might play a role in fears of small/slimy creatures, with snakes and
spiders attracting most interest (see Woody and Teachman, 2000). It has also been suggested
 S. Rachman / Behaviour Research and Therapy 42 (2004) 1227–1255 1235

that disgust might contribute to blood-injury sensitivity (Koch et al., 2002). The several findings
of a relation between disgust sensitivity and animal phobias, and blood-injury reactivity, are
interesting but can be said to beg the question. What is the basis of the sensitivity in the first
place?
The accumulating evidence of a correlation between the commonly-used measures of disgust
sensitivity and anxiety sensitivity (Woody and Teachman, 2000) inevitably raises the possibility
of a generally elevated sensitivity (neuroticism perhaps?), and one that is bound to be a predic-
tor of anxiety problems. As far as the fear of contamination is concerned, elevations of both
measures of sensitivity are to be expected; in cases of predominantly disease-contamination the
anxiety scale is likely to be particularly high, and in cases of predominantly dirt/disgust con-
tamination the disgust scale should prevail. Presumably the scores on the scales of sensitivity
should decline after successful treament. As to the origin of the sensitivities it is difficult to
avoid speculation about the influence of a biologically-based sensitivity that is manifested dur-
ing exposures to the threat of contamination and is then consolidated or not, depending on the
outcome of the exposures and the person’s interpretation of these events. In keeping with devel-
opments in the subject of psychopathology, it is advisable to consider the role of cognitions in
the appearance and persistence of fears of contamination.
The demonstration of individual differences in sensitivity to disgust and to anxiety is a useful
step but there is a gap in our knowledge of the cognitions involved in contamination. We need
to collect information about the beliefs involved in contamination, and the appraisals that
people make about their actual or anticipated contamination, with or without direct contact.
It is suggested that a fear of contamination will arise if the affected person interprets the feel-
ings of contamination as a significant danger to their physical and/or mental health, or as pre-
senting a significant social threat.

6. Normal and abnormal feelings of contamination

Of course not all feelings of contamination are excessive, irrational and unadaptive. Contami-
nation by contact with dangerous or disgusting material is a common, probably universal
experience. However, the sense of contamination does not arise until the person passes through
the earliest years of childhood. Young children attempt to touch or even eat matter that is
known by everyone else to be dangerous or disgusting. Naturally they are ignorant of possible
sources of infection, and indeed the entire conception of infection. They do not avoid infectious
people or materials and display no disgust even in contact with excrement. Further, people are
normally tolerant of their own bodily products (and those of their infants) but are disgusted by
those of other people. Contact with the bodily products of other people or animals usually pro-
duces feelings of disgust contamination and of consequent urges to clean oneself. Beliefs govern-
ing the nature of dangerous/disgusting contacts, how to avoid them and how to remove the
unavoided contamination, are prevalent in all societies. Davey (1993) suggested that reactions of
disgust have biological utility and protect people from eating or touching diseased/dangerous
substances. Presumably this folk knowledge is then used to protect the entire community, for
example by proscribing the eating of particular substances. His interesting division of animals
1236 S. Rachman / Behaviour Research and Therapy 42 (2004) 1227–1255

into those that evoke fear or revulsion has not been fully confirmed (Woody and Teachman,
2000).
In any community the construal of contamination is shaped by cultural and religious beliefs
and by the knowledge prevailing in the particular society. In some communities folk beliefs pre-
vail and in others scientific information is dominant. The beliefs are activated when the person
comes into contact with a culturally denoted potential source of danger; although some beliefs
about pollution appear to be extremely widespread (e.g. regarding excrement) there are many
and major cultural differences in the perception of pollution.
In some societies, such as India, contact with people of a lower caste is considered to be con-
taminating and assiduously avoided. If contact nevertheless occurs, the affected person has to
engage in a ritualized cleansing process. The untouchables are required to avoid contact with
people of higher status. Despite the fact that untouchability was legally abolished in India in
1950, according to a Human Rights Watch report in 1999, the beliefs and practices persist.
‘‘Untouchables may not cross the line dividing their part of the village from that occupied by
higher castes.They may not use the same wells, visit the same temples, drink from the same cups
in tea stalls. . . (and) are relegated to the most menial tasks, as manual scavengers, removers of
human waste and dead animals...’’. Before handling the money they receive from untouchables,
merchants from a higher caste clean it in water (Anand, 1940). Temples can be polluted if an
untouchable person approaches, and Anand describes a priest’s assertion that if an untouchable
person comes within 69 yards of the temple it will become contaminated. Acts of violence
against the degraded and ostracised untouchables are common (Human Rights Watch, 1999).
Beliefs about contamination are universal and culturally shaped, and they determine the
experience and manifestation of feelings of contamination and their consquences. In their con-
text the beliefs and behaviour are normal. However, these beliefs and behaviour can become
extreme and even abnormal.

7. The illusion of vulnerability: abnormal beliefs about contamination

1. The illusion of vulnerability: A belief that one has a special personal vulnerability to the
harmful effects of contamination; this unique vulnerability is not shared at all, or if shared,
then one’s own vulnerability is extremely elevated (for example, ‘‘other people will not con-
tract AIDS from using public telephones but for me the risk is very high’’).
2. Beliefs about contamination not uncommonly have a bizarre quality. Some clinical examples
include: a belief that one is vulnerable to contamination from mind germs, or from the sight
of physically handicapped people, or can develop gangrene from touching (non-gangrenous)
patients in hospital—in these cases as in others, the bizarreness of the belief is all the more
remarkable because many of the people holding such beliefs are well-educated and acknowl-
edge that their beliefs are strange and restricted to themselves.
3. An extreme over-estimation of the probability that harmful effects, dangerous and/or dis-
gusting, will be caused by the contamination.
4. An extreme over-estimation of the seriousness of the consequences of the contamination.
5. An extreme over-estimation of the ease of transmission of contamination to one’s self or
others.
 S. Rachman / Behaviour Research and Therapy 42 (2004) 1227–1255 1237

6. An extreme over-estimation of the range of potentially contaminating people/places/items.

7. In calm moments the affected person recognises the irrationality of the beliefs.
8. In not so calm moments, particularly when exposed to a possible contaminant, the intensity
of, and the conviction in the truth of the beliefs are elevated.
9. The beliefs are not responsive to contradictory evidence.
10. The beliefs and feelings are inclined to persist and may even become permanent.

As is evident from this account of the qualities of abnormal beliefs and feelings about con-
tamination, they can reach delusional levels. The affected people believe with unyielding convic-
tion that they are specifically and personally vulnerable to particular threats of contamination
which will produce seriously harmful effects.The beliefs often have a bizarre quality, are imper-
vious to contradictory evidence, are long-lasting and tend towards permanence. The presence of
such beliefs will predispose the person to the acquisition of fears of contamination. A compre-
hensive account of how these beliefs are formed and consolidated is not yet available.
Clinicians are well aware of the difficulties involved in trying to determine whether intensely
held abnormal beliefs are delusional or not (Chadwick and Lowe, 1990). There is no simple div-
ider and the tendency is to base one’s assessment on the presence and intensity of a few criteria,
notably the uniqueness of the belief, bizarre qualities and its resistance to disconfirmatory evi-
dence. From a clinician’s point of view it is the last-named quality that presents the greatest
obstacle to treatment. In tackling a patient’s delusional beliefs about contamination and con-
taminants some guidance can be found in the gradual but encouraging progress made by clini-
cal researchers who are using psychological methods for reducing the abnormal cognitions
and beliefs of patients with psychotic disorders (Chadwick and Lowe, 1990; Garety et al., 1994;
Tarrier et al., 1993, among others).

8. The consquences of a fear of contamination

The depth and complexity of contamination fears is confirmed by the wide-ranging con-
sequences which follow the emergence of such fears. The consequences are cognitive, emotional,
perceptual, social and behavioural.
Once the fear is established the affected person construes the world and himself/herself in a
changed fashion. The parameters of danger/distress are expanded and the areas of safety are
newly constrained. The person becomes highly sensitive to possible threats of contamination
and hypervigilance is the result. The person believes that they are especially vulnerable to con-
tamination and its anticipated effects, such as infection or the distress of feeling polluted. It is
believed that contact with the personally defined contaminants will cause harm and may well be
dangerous as well as distressing.
The elevated and directed hypervigilance is concentrated predominantly on external cues,
such as dirty bandages, but includes the scanning for internal cues of contamination, dirt/infec-
tion. ‘‘Am I now entirely, certainly, safely clean? Does my body feel absolutely clean?’’ As with
other fears, it gives rise to consistent over-predictions of both the likelihood of experiencing fear
and of the intensity of the expected fear. ‘‘If I visit my relative in hospital I will certainly experi-
ence fears of contamination and they will be intense.’’ It remains to be determined whether, in
1238 S. Rachman / Behaviour Research and Therapy 42 (2004) 1227–1255

cases of contamination fear, the pattern of over-predictions extends also to predictions about
the likelihood of becoming contaminated in specifiable situations. It seems probable, and may
even occur when the person is sensitive to possible contamination, but not necessarily fright-
ened. A fastidious person who places great importance on personal cleanliness may over-predict
the probability of contamination by dirt, with resulting hypervigilance.
The fear of contamination can lead to intense social anxiety and avoidance. Patients who fear
their own bodily pollution (e.g. offensive bowel smells) can become acutely sensitive to the
effects of their pollution on other people. Given their beliefs about the pollution, it is not
unreasonable for them to dread how people will react to it, and they anticipate rejection.
Another social threat, seen most strongly among people with an inflated sense of responsibility,
is the dread of passing the contamination on to other people and therefore endandering them.
In these instances the usual fear and avoidance is accompanied by guilt.

9. Methods of coping

When the fear is evoked, usually by direct contact with a perceived contaminant, it immedi-
ately generates a powerful, even overwhelming, urge to clean. The urge is generally so strong
that it over-rides other considerations. Whenever possible the person will avoid touching any-
thing until they have had an opportunity to clean themselves. It temporarily freezes other
behaviour. The attempts to clean oneself, and one’s possessions such as vehicles, clothing etc.,
are compulsive in that they are driven by powerful urges, very hard to resist, repetitive, and rec-
ognised by the affected person to be extreme and at least partly irrational. The most common
form of compulsive cleaning is repeated hand-washing, which typically is meticulous, ritualistic,
unchanging, very difficult to control and so thorough that it will be repeated again and again
even though it abraids the skin. There are instances in which the patient continues washing
despite the reddening of the water in the basin caused by their bleeding hands. Paradoxically the
compulsive washing causes dryness of the skin because it removes the natural oils and the per-
son’s skin becomes blotchy, dry and cracked, especially between the fingers. If the core fear is
that one’s health might be endangered by contact with contamination material, it is common to
over-use disinfectants, supposedly anti-bacterial soaps, very hot water. When the basic fear is
one of becoming distressed by the feeling of pollution, the use of hot water and much soap is
the order of the day and disinfectants are seldom employed. The details of the methods used to
achieve a sense of safety and/or non-pollution can provide clues to the nature of the underlying
fear.
In addition to the need to remove a present contaminating threat by cleaning it away, com-
pulsive cleaning is carried out in order to prevent the spread of the contamination. ‘‘If I do not
clean my hands thoroughly I will spread the contamination throughout the house.’’ Other
attempts to prevent contamination include the use of protective clothing (e.g. gloves, keeping
outdoor clothing and indoor clothing separated), using tissues to handle faucets, doorhandles,
toilet handles, and taking care to remove sources of potential contamination (e.g. removing pes-
ticides, anti-freeze fluid). In the process of avoiding contamination the person steadily sculpts a
secure environment, establishing some sanctuaries. One’s own room tends to develop into the
safest place and great care is taken to ensure that it remains unpolluted. The home as a whole is
 S. Rachman / Behaviour Research and Therapy 42 (2004) 1227–1255 1239

safe but less so because other people do not share the super-sensitivity to contamination and are
not as careful about taking precautions. At the other extreme from these places of sanctuary lie
a number of highly contaminating places, such as public lavatories, clinics for the care of people
with sexually transmitted diseases, and so forth. People are also ‘‘ranked’’ in terms of their
safety from contamination, and those who are low in the ranking are vigorously avoided.
When they feel that they have been contaminated, people clean themselves and also attempt
to escape from the source of the contamination. The fear of contamination generates elaborate
and vigorous attempts to avoid coming into contact with perceived contaminants. In extreme
cases this can take the form of avoiding entire cities! An otherwise well-adjusted woman
developed an intense fear of being contaminated by any bodily waste matter, animal or human.
She became hypervigilant and avoidant but on one fateful day she woke up to find that a dog
had defecated on the lawn directly outside her front door. She was shocked and felt thoroughly
contaminated. Repeated showers relieved her not, and within days she dreaded leaving or
returning to her house (now using only the rear door). The fear became so intense that she sold
her house and moved into a rented home in another suburb. As this failed to help her she
decided to move to another city, and forever avoided going anywhere near the city in which the
trigger event had occurred; she regarded the entire city as contaminated. This example of extra-
ordinary avoidance also illustrates the rapid and uncontrolled spread of imagined contami-
nation. This example is extreme but not different in form or consequences from other attempts
at avoidance. It can take the form of avoiding entire groups of potentially contaminating people
such as the homeless, cancer patients, and so on.
Among people who have an inflated sense of responsibility, a major factor in many instances
of OCD (Salkovskis, 1985), their fear of contamination is manifested in the usual compulsive
cleaning but they also devote special efforts to prevent the occurrence or spread of contami-
nation. They are strongly motivated to protect other people from the dangers of contamination
and therefore strive to maintain a contamination-free environment. They try to ensure that the
kitchen and all eating implements are totally free of germs, dirt, tainted food. One father
insisted on sterilizing his baby daughter’s feeding bottles at least ten times before re-use. Affec-
ted people try to ensure that their hands are free of contamination before touching other people
or their possessions. If they feel that they have not been sufficiently careful, anxiety and guilt
arise. They try to recruit the cooperation of relatives and friends in preventing and avoiding
contamination, but rarely succeed in persuading adults to comply with their excessive and
irrational requests
If the affected people feel that their cleaning and avoidance behaviour have not ruled out the
threat from contamination they may resort to neutralizing behaviour and/or a compulsive
search for re-assurance. The main form of neutralizing, cleaning, is overt; the internal neutraliz-
ing that is prominent in other types of OCD is used infrequently, presumably because it is
thought to be ineffective in removing the contaminant.
It is likely that internal neutralizing is more often resorted to in cases of mental pollution.
Repeated requests for reassurance are a common feature and can be as intense and insistent as
in other types of OCD. It the fundamental fear is of a threat to one’s health because of con-
tamination, attempts to obtain reassurance,especially from doctors, are common but as transi-
ently relieving as in cases of health anxiety.
1240 S. Rachman / Behaviour Research and Therapy 42 (2004) 1227–1255

In sum, there are multiple consequences of contamination fear—cognitive, perceptual,

emotional, social and behavioural—and they serve to remind us of the complexity and wide
ramifications of what might at first sight appear to be merely another fear to add to the cata-
logue. The methods of coping are attempts to maintain safety or to regain safety. The fear of
contamination can be powerful and come to shape, restrict and even dominate a person’s life.

10. Disgust and the DSM

Fear and disgust share some common features and at times overlap but most often they are
independent of each other. Fear/anxiety is the central and necessary component of the several
clinical problems that comprise the DSM category of Anxiety Disorders. Disgust is a common
and often intense emotion that frequently defies control, but has no pre-ordained place in the
DSM. There are no clinics for the treatment of disgust and the ever-vigilant drug manufacturers
show no interest in developing and marketing anti-disgust medications. The intense, uncontrol-
lable emotion of fear is considered to be a clinical problem but disgust is not. Why not? This
question is best tackled by analysing their common and distinct qualities.
Fear and disgust are intense and unpleasant emotions. With a few exceptions in the case of
fear, these emotions are aversive and people exert considerable efforts to escape from or avoid
them. A clinical overlap between fear and disgust is observed in selected cases of OCD and in
animal phobias. The connecting link between the two emotions is a dread of contamination. In
both fear and disgust the emotion is provoked by direct or indirect contact with a perceived
contaminant. In both instances the observed consequences—cognitive, behavioural and percep-
tual—are similar and most prominently, both disgust and contamination fear generate compul-
sive cleaning. If it is disgust contamination then soap and hot water will do, but if there is a
threat of infection by contamination, disinfectants may be added. However, the attempt in both
instances is to remove the contaminant. In both instances it is believed that after contact the
contamination can be spread, and in both of them attempts are made to limit or prevent this
contagion. Some stimuli (e.g. dirty bandages, decaying food) can provoke both disgust and a
fear of contamination. Others can provoke one or the other but not both. The feeling of con-
tamination is the main connecting link between fear and disgust.
There are innumerable stimuli or situations capable of provoking disgust that convey no
threat and produce no fear; there is far more disgust than contamination fear. Likewise, in the
large majority of fears there is no element of disgust. The cues for disgust generally are visual
and olfactory and include putrefaction, the stench from bodily waste, decaying vegetable mat-
ter. Smell plays little part in fear. In instances of disgust the distress is readily removed by clean-
ing and once it is completed, no threat or discomfort persists. The successful removal of the
contaminant can be confirmed visually and by the disappearance of the smell. In those instances
of contamination which threaten one’s health, and they probably are the majority, the problem
and the fear are relieved but not removed even after full cleaning.The possibility that one might
have been infected by contact with a disease contaminant cannot be adequately resolved by
cleaning, as in fears of AIDS. Disgust contamination and fear contamination run different time
courses. Unlike disgust contamination, the triggers for the fear of being infected by contami-
nation, are not always identifiable. The suspect viruses or germs are invisible and hence difficult
 S. Rachman / Behaviour Research and Therapy 42 (2004) 1227–1255 1241

to remove with certainty. Disease contamination is accompanied and followed by considerable

doubting in a way that is seen infrequently in disgust contamination.
The facial expressions associated with disgust and fear differ as do the physiological reactions
that accompany the two emotions. The physical reactions to stimuli that evoke disgust include
an array of gastric sensations such as nausea, gagging, vomiting. Fear reactions include a
pounding heart, sweating, trembling, shortness of breath.
Health-endangering contamination resembles the disorder formerly called hypochondria and
now preferably described as health anxiety (Salkovskis and Warwick, 1986), in a number of
ways. The threats are similar, temporally remote, often ambiguous, felt to be very serious and
difficult to resolve. The threats give rise to comparable behaviour, including repeated attempts
to gain reassurance, an unrealistic striving for total certainty. Health anxiety and fears of dis-
ease/infection contamination are characterised by episodic spikes in fear that leave a residue of
lingering anxiety.
The subject of disgust was re-introduced in a most stimulating analysis by Rozin and Fallon
(1987). Their original paper was limited to oral disgust (subsequently expanded by Rozin et al.,
1993), but nevertheless their remarkably creative ideas inspired a considerable amount of
research, and in recent years not a small part of it has dwelt on the relations between fear and
disgust (McKay, 2002; Olatunji et al., 2004; Sawchuk et al., 2000; Woody and Teachman, 2000,
among others).
Rozin and Fallon discerned the operation of some laws of sympathetic magic in the evocation
and transmission of disgust. The first of these, the law of contagion, transmission by contact, is
straightforward and occupies central stage in virtually all discussions of disgust, including the
present one; it is summarized as ‘‘once in contact, always in contact’’ (Rozin and Fallon, 1987,
p. 30). The second of the laws, similarity, is less obviously relevant but Rozin and Fallon con-
vincingly argued for the importance of the concept. In short, contagion can pass from object to
object, or indeed from person to person, on the basis of a perceived similarity; ‘‘like produces
like, that is, resemblance in some properties indicates a fundamental similarity or identity,’’
(p. 30). They bolstered their argument by a series of imaginative demonstrations of the play of
disgust (e.g. they provoked disgust by inviting people to eat a piece of fudge shaped as dog
faeces). Rozin and Fallon also noted the important role of interpersonal factors in disgust. In
fears of contamination, contagion by contact is of course paramount but contamination can
also be evoked or spread without physical contact. In these instances the operation of the law of
similarity is particularly evident and can involve magical thinking. It is likely that similarity also
plays a significant part in the phenomenon of mental pollution.
Although the law of similarity is most evident in the transmission of contamination, and in
the resulting fear and avoidance, in rare cases the same law appears magically to negate the
contamination. An intriguing example of this reversal of the usual pattern was encountered in
the treatment of an in-patient who suffered from an intense fear of contamination and conse-
quent handwashing. She received daily sessions of exposure and response prevention, and was
as ever, strongly encouraged to refrain from ‘‘illegitimate’’ washing between sessions. The first
four sessions were slow and difficult for her because the contaminants evoked such high levels
of fear that she was able only to touch a few items and those swiftly and minimally. She would
touch an item with the tip of one finger and then rapidly withdraw her hand. When the thera-
pists arrived for the fifth session they were pleasantly surprised to discover that the patient’s
1242 S. Rachman / Behaviour Research and Therapy 42 (2004) 1227–1255

mood and engagement in the exposure exercises had greatly improved. She readily touched the
full range of contaminated items, using her hands in a normal manner, and reported only slight
fear. On the following day her fear had returned to its original high level but once again she
completed the exposure tasks without difficulty. This pattern continued for a few more days
until the therapists stumbled on the explanation for her inconsistent behaviour. A nurse hap-
pened to mention in passing that the patient no longer dreaded the treatment sessions because
she now had a special cloth that she used to get over the exposure to the contaminated material.
The patient had imbued a piece of cloth with ‘‘supernatural powers’’ that instantly cleaned
away all contaminations. Interestingly the magical power of the piece of cloth was confined to
contaminated items used in the sessions; the patient was unable to extend its powers to remove
naturally-occurring contaminations. A pity.
The analysis of disgust introduced by Rozin and Fallon in 1987 was widened by Rozin, Haidt
and McCauley in 1993 and provided the basis for a scale to measure sensitivity to disgust (see
the revised version by Haidt et al., 1994). It has been used successfully in various studies
(McKay, 2002; Sawchuk et al., 2000; Woody and Tolin, 2002). In clinical studies,especially
those pertaining to the fear of contamination, it is best accompanied by measurements of anxi-
ety sensitivity. Presumably, in cases of fear of contamination the scores on both scales of sensi-
tivity will be elevated. It will be of interest to determine whether the use of these scales can help
to elucidate any factors that pre-dispose people to develop these fears; conceivably there is a
pre-existing sensitivity to disgust and to anxiety.
We return to the original question of why fear but not disgust features in the DSM. The stim-
uli that provoke disgust are easily identifiable, circumscribed and can be dealt with promptly by
straightforward cleaning or by removal of the cues or the person. The removal of the cues is
easily confirmed by visual inspection and the disappearance of any offending smells. Disgust is
therefore manageable and hence transient. It leaves no residue of problems and does not inter-
fere with one’s life. Fear, and the associated anxiety, on the other hand are less manageable.
The cues are less easily identified and all too often have irrational qualities (that tend to be
unmanageable). Fears and especially anxiety linger and often are disruptive and persistingly dis-
tressing. They warrant inclusion in the DSM.
However, even disgust can develop irrational properties. As described above, some people
develop irrational fears of disgust contamination and clinicians encounter these in the context of
OCD. A fear of contamination from bodily products, especially one’s own, occurs in some cases
of OCD. Patients become preoccupied with the fear that they might pollute others and/or
exude offensive smells because of bodily pollution, and spend many long hours cleaning them-
selves before going out. Dirt/disgust contamination is unsurprisingly associated with social
anxiety. ‘‘People are generally careful to avoid evoking disgust in others, and failing to be prop-
erly fastidious can do just that,’’ (Woody and Teachman, 2000, p.307). The belief that one is
bodily polluted, and that the olfactory and visual signs of one’s polluted state are evident to
other people, can generate serious anxiety and an ultimate fear of rejection.
Non-physical stimuli, including accusations, memories and even one’s unwanted intrusive
thoughts, can provoke a sense of disgust and of internal dirtiness, a sense of mental pollution
(see below).
 S. Rachman / Behaviour Research and Therapy 42 (2004) 1227–1255 1243

11. Similarities between fear and disgust

1. Contact with fear-evoking or disgust-evoking stimuli produces feelings of contamination.

2. Fear and disgust can co-occur and even overlap; putrefying meat can be fearful and disgust-
ing.
3. Contamination is associated with negative emotions.
4. It is believed that the contamination spreads.
5. Contamination provokes urges to avoid or escape, and to undertake preventive actions.
6. Contamination provokes attempts to clean away the contamination; these attempts are prone
to become repetitive.

12. Some differences between fear and disgust

1. Disgust contamination is seldom perceived to constitute a threat to one’s health or well-

being; in fear contamination this is usual.
2. Feelings of disgust contamination tend to present a social threat, even a fear of rejection.
3. Sight and smell are prominent triggers for disgust; the triggers for fear are predominantly
visual and auditory, and a sense of smell rarely features.
4. In disgust the contaminant is tangible and the site of the contamination is evident.
5. In fear the contaminant is tangible but the site of the threat is wider and has uncertain
boundaries.
6. As a result, cleaning away or avoiding a fear contaminant is less manageable (e.g. the germs
are not visible).
7. Attempts to avoid or escape from fear-evoking contaminants, especially by cleaning, are sel-
dom wholly satisfactory because the boundaries of the contaminants are wide and unclear,
and it is therefore difficult to achieve certainty about the efficacy of one’s efforts (invisible
germs again). Disgust contamination tends to be circumscribed.
8. It is easier to achieve certainty that one has removed a disgusting contaminant.
9. Hence disgust contamination tends to be relatively short-lived; fear of a contaminated threat
to one’s health tends to persist.
10. The predominant physical reactions of disgust are nausea, gagging, vomiting, faintness; the
predominant reactions in fear are pounding heart, tremor, dry mouth, shortness of breath.
11. A feeling of dirtiness is more common in disgust than in fear.
12. These distinctions between disgust and fear are blurred in cases of mental pollution.

One reason to pursue the overlap between fear and disgust arises from the possibility that the
presence of a significant disgust factor may impede the successful treatment of anxiety disorders.
In their thorough review of the connections between disgust and fear, Woody and Teachman
(2000) found little indication that this is a danger (e.g. Merckelbach et al., 1993). However,
most of the information pertains to small animal phobias and to blood-injury sensitivity (some-
times called blood-injury phobia, perhaps misleadingly). Fortunately, the main method for
reducing fear, repeated exposures, appears also to reduce any circumscribed and associated dis-
gust; broad and unrelated feelings of disgust however, were not altered when a fear of snakes,
which is associated with disgust, was reduced (de Jong et al., 1997, 2000). There is no reason to
1244 S. Rachman / Behaviour Research and Therapy 42 (2004) 1227–1255

expect a broad reduction of disgust after a fear tinged with disgust is reduced. However, it is
possible that a reduction of mental pollution may be followed by a broad decline in feelings of
disgust. As early as 1987 Rozin and Fallon suggested that disgusts can be unmade by a process
of ‘‘extinction by frequent exposure’’, but that overall, ‘‘conceptual reorientation might be a
more effective method’’ (p. 38). If it is confirmed that the exposure method is indeed capable of
reducing disgust, and to reduce the fear simultaneously in most cases, then from a therapeutic
standpoint the distinction between fear and disgust may be one of those differences that make
no difference.
Given the role of disgust contamination in many cases of OCD, it is however necessary to
give serious consideration to the therapeutic implications of this fact. It has not been studied
directly but the overall results accomplished by treatment suggest that the disgust component is
responsive to exposure and response prevention. The emphasis throughout the therapeutic
research on OCD has been on the fear component while the disgust component has been
ignored or regarded as part of the fear or simply secondary to it. Nevertheless the feelings of
disgust do decline after exposure treatment. It has not been asserted that patients with a fear of
disgust (e.g. fear of touching their own bodily products) are more or less responsive to the treat-
ment than are patients with other types of contamination fear. Is this still a difference that
makes no difference? In recognition of the role of mental pollution in some cases of OCD, and
following the opinion of Rozin and Fallon (1987), perhaps a cognitive form of therapy can
tackle the disgust component, and hence make that difference? To modify the examples given by
Rozin and Fallon, can the disgust be undone or reduced by cognitive reorientations such as by
informing the person that what they thought was rotten milk was actually yogurt, or that the
forbidden pork was actually lamb after all? (Cognitive reorientation can almost certainly act in
the opposite direction; it can induce feelings of pollution. To use the lamb example again—the
meat you ate was not lamb but forbidden pork.)

13. Mental pollution: this disease is beyond my practice

The concept of a peculiar sense of dirtiness, mental pollution, was introduced because some
patients with OCD express great frustration about their inability ever to feel wholly clean
(Rachman, 1994). They continue to feel dirty despite strenuous attempts to clean themselves.
Taking four, five or six hot showers in rapid succession fails to produce the desired state of
cleanliness. How is it possible to wash repeatedly and yet remain dirty?
Lady Macbeth is the supreme example of mental pollution. Incessant washing failed to give
her peace or even relief. Her nurse observed Lady Macbeth persistently rubbing her hands, ‘‘It
is an accustom’d action with her, to seem thus washing her hands: I have known her continue in
this a quarter of an hour’’, (Macbeth, Act 5, scene 1). Lady Macbeth’s repeated attempts to
clean herself were futile, ‘‘What will these hands ne’er be clean ?’’ and later, ‘‘Here’s the smell of
blood still: all the perfumes of Arabia will not sweeten this little hand. Oh,oh,oh.’’ Her doctor
reacts, ‘‘What a sigh is there!’’ and concedes that ‘‘This disease is beyond my practice’’.
A distinction has to be made between ordinary dirtiness and pollution. Mental pollution is
described as ‘‘a sense of internal un-cleanness which can and usually does arise and persist
regardless of the presence or absence of external, observable dirt’’ (Rachman, 1994, p. 311).
 S. Rachman / Behaviour Research and Therapy 42 (2004) 1227–1255 1245

Feelings of dirtiness generally arise after direct physical contact with soiled material or objec-
tively unclean items, such as bodily excretions, animals or decaying matter. The source of the
feelings of dirtiness is known. The feelings are easily attributable to such contact and the site of
the dirt is identifiable. Anyone who comes into contact with such items will feel dirty and a
straightforward wash is sufficient promptly to remove the source of the dirt and the feelings of
dirtiness. Contact with the soiled material will produce feelings of dirtiness in anyone; it is not
exclusive to one person, not unique.
In contrast, ‘‘mental pollution has a slight or indirect connection’’ with soiled material and
generally emerges in ‘‘the absence of physical contact with soiled items’’ (Rachman, ibid). The
source and the site of the dirtiness are unclear—and no doubt this intangible quality of mental
pollution is part of the explanation for the Macbethian futility of trying to clean it in the cus-
tomary way. Applying soap and water to the outer surfaces of one’s body is misdirected; it fails
to address the internal dirtiness that characterises mental pollution. Moreover, the feelings of
pollution persist long after any initiating contact, if any, has been removed.
The pollution can be induced or exacerbated by ‘‘mental events’’ such as accusations, insults,
threats, humiliations, assaults, memories and even by unwanted and acceptable thoughts and
images (e.g incestuous images, impulses to molest children etc). The occurrence of intrusive and
repugnant images/impulses/thoughts is of course the central feature of obsessions. Mental pol-
lution is specific and unique to the affected person. As mentioned above it is not easily trans-
missable, and the most characteristic way in which ‘‘ordinary’’ pollution is transmitted, by
physical contact, does not occur. Ordinary pollution is transmissable from object to object,
from person–person and person–object. Mental pollution is not transmissable from person to
person, objects to objects, or person to objects, probably because it is uniquely personal. It is
likely that deliberate efforts would need to be undertaken in order to overcome the barrier to
easy transmission. The magical practice of deliberately transmitting harm, not contamination,
to an enemy by being unkind to a model of the enemy is an exploitation of the law of similarity
but is not applicable to mental pollution. Any sign that mental pollution can be conveyed by
modelled similarity is bound to unsettle the futures market in the supply of wax.
A detailed list of the properties of mental pollution and how it differs from contamination by
contact with dirt and contamination by contact with infectious materials is set out in Rachman
(1994) and a refined table was provided by Fairbrother et al. (2004). An abbreviated version of
these two tables is shown in Table 1 below, and the quality of mental pollution is conveyed by
case illustrations.
In summary, contamination is caused by physical contact with dirt/disease, spreads widely
and is easily transmissable to others. The source is known and the site is identifiable. It lacks a
moral element. Other people are vulnerable to contamination by contact with similar dirt/dis-
ease stimuli. The predominant emotions are fear and revulsion, and the main consequences are
escape and avoidance. Cleaning can be transiently effective. Mental pollution is a sense of
mainly internal dirtiness of uncertain site and is seldom traceable to a specific source. It has per-
plexing qualities and the affected person feels uniquely vulnerable. Some stimulus generalization
occurs but the pollution is rarely transmissable to others. The associated emotions are anxiety
and moral uneasiness. Cleaning is ineffective. Avoidance is broad and sometimes puzzling to the
person and others.
1246 S. Rachman / Behaviour Research and Therapy 42 (2004) 1227–1255

Table 1
A comparison of contamination and mental pollution
Contamination Mental pollution
Dirt/disease contact Physical contact not necessary
Feeling dirty/infected Internal dirtiness predominantly
Spreads widely Some generalization, but
Easily transmissable to others Rarely transmissable to others
Others considered vulnerable Uniquely vulnerable
Source known Source obscure
Site identifiable Site inaccessible
Tangible Intangible
Re-evocable by contact Re-evocable by mental events, inc. memory, images
Lacks a moral element Moral element common
Revulsion, nausea, fear Anxiety, revulsion, shame
Transiently responsive to cleaning Cleaning is ineffective

14. Case Illustrations

An intelligent and otherwise rational young man was utterly preoccupied with avoiding and
compulsively washing away his ‘‘mind germs’’, but treated ordinary dirt in the usual manner.
He tried to avoid all places that had for him particular verbal connotations, and if he failed, as
he occasionally he did, he was driven to clean away the resulting contamination.The intensive
de-contaminating cleaning took hours to complete and on one occasion he was so desperate to
achieve a feeling of clean-ness that he washed all of his dollars bills in a washing machine and
tried, unsuccessfully, to dry them in a microwave machine. After this exotic venture into money-
laundering, he had to make an embarrassing trip to the bank to explain the mauled state of his
money, and it was only after his therapist confirmed to the bank the veracity of his story that he
was compensated. On another occasion he made a desperate attempt to clean his contaminated
computer by placing it in a bath of hot water. Another patient tried to reduce her feelings of
dirtiness by repeatedly showering in very hot water. Despite using strong soaps and stiff
brushes, she felt just as dirty at the end of each shower as she had before she began. ‘‘No matter
how many showers I take, and how hard I try, I can’t get clean!’’ The feelings of dirtiness
emerged after she was sharply accused of being sexually immoral. Another patient who was
similarly accused by her family and ostracised was overcome by such intense feelings of dirtiness
that she repeatedly tried to clean herself with abrasive materials which ultimately damaged her
skin. Her feelings of dirtiness could be triggered by telephone calls, by letters from ‘‘contami-
nated’’ relatives and other remote stimuli. The feelings of pollution were not altered by the
cleaning. These two patients dealt with ordinary dirt in a normal fashion, and neither of them
had elevated fears of disease. Their feelings of pollution were particular, personal, not originally
provoked by physical contact with a contaminant, and unresponsive to straightforward clean-
ing. Other clinical examples are described by de Silva and Marks (1998).
Fairbrother et al. (2004) conducted an experimental test of the idea that mental pollution can
be induced without physical contact with a contamination. A sample of 121 students were asked
to imagine experiencing a non-consensual kiss at a party described on an audiotape, or a con-
 S. Rachman / Behaviour Research and Therapy 42 (2004) 1227–1255 1247

sensual kiss described on a comparable audio recording. The experimental manipulation suc-
ceeded and the participants in the non-consensual condition reported experiencing mental pol-
lution that included feelings of internal dirtiness (5 times as strong as did the participants in the
consensual group), and feelings of a non-physical type of dirtiness. In addition, eight of the
participants in the non-consensual group tried rinsing/washing to rid themselves of feelings of
pollution. This demonstration of pollution induced without contact recalls Rozin and Fallon’s
(1987) observation that disgust too can be induced by what they describe as ideational means,
such as contact with items that are clean but resemble dirty items (e.g. eating off a dish that
resembles a toilet bowl).
Feelings of mental pollution can have a moral, rather immoral, element (Rachman, 1994) and
some of the participants in the experiment by Fairbrother et al. (2004) said that the images
which they formed while listening to the non-consensual audiotape made them feel sleazy,
cheap, ashamed. In a study of 50 women who had endured a sexual assault Fairbrother and
Rachman (2004) found evidence of mental pollution in 60% of them. Twenty-seven said they
felt internally dirty, 17 could not clean to their full satisfaction,13 felt dirty long after contact
with a contaminant and 7 felt morally tainted. Two-thirds of the victims reported that after the
assault they had carried out extraordinary washing (e.g. 51% had a longer/hotter shower, 9%
carried out extra washing of genitals, 6% used special cleaning products). Sixty percent reported
that the feelings of pollution could be re-evoked by memories, images or information. When
specifically asked to recall the event the participants reported a 33.84% rise in feelings of dirti-
ness (versus 3.49% after a neutral memory) and a 24.19% increase in the urge to wash (versus
4.28% after a neutral memory). Nine of the women washed their hands after the deliberate,
requested recall of the event.
Formal assessments of mental pollution, by questionnaire or standardized interviews, can
incorporate the sensations,beliefs and feelings described above (e.g. internal dirtiness), supple-
mented by items dealing with behaviour (especially cleaning, avoidance, prevention, unwanted
intrusive thoughts and concealment; Newth and Rachman, 2001), and appraisals of the prob-
ability/seriousness of harm arising from contamination. In short, it will likely include five com-
ponents of mental pollution: feelings of contamination, associated beliefs, appraisals, unwanted
intrusions and concealment, plus items pertaining to sensitivity to the opinions of others and
especially to comments made about one’s character, morality, behaviour. The interview schedule
reproduced in Fairbrother and Rachman (2004) was constructed for the study of victims of sex-
ual assault but can be modified for other purposes.

15. Cognitive co-morbidity

It has been suggested that the study of cognitive co-morbidity is likely to prove more fruitful
than the traditional approach to co-morbidity which is based on statistical associations between
disorders (Rachman, 1991). Analyses of cognitive co-morbidity search for psychological links
between disorders and begin with a consideration of the affected person’s primary cognitions
pertaining to the disorder. The basic cognition in the fear of disease contamination is that con-
tact with specifiable contaminants will threaten one’s health. The underlying, and overlying,
cognition in health anxiety (hypochondria) also is a threat to one’s health. In the light of this
1248 S. Rachman / Behaviour Research and Therapy 42 (2004) 1227–1255

central common cognition we should observe common beliefs and behaviour in these two dis-
orders. In both of them the threats are perceived to be serious, even life-threatening, are tempo-
rally remote, coloured by uncertainty, difficult to cope with, and hard to eradicate. In both, the
affected person feels specially vulnerable and at significantly greater risk of illness than are other
people. However, the risks perceived by a sufferer from health anxiety range more widely than
do those who suffer from a fear of contamination. Another difference is that in the fear of dis-
ease contamination the perceived source of the danger is predominantly external and identifi-
able; in health anxiety most of the sources of perceived danger are internal and can be
ambiguous. Both disorders are notable for persistent hypervigilance. In the fear of contami-
nation the elevated vigilance is directed towards the identification and avoidance of external
threats in the form of contaminants. In health anxiety the hypervigilance generally takes the
form of internal scanning for signs of trouble, such as perceived lumps, irregular heart beats,
etc. So the common cognition is a perceived threat to one’s health and the consequent behav-
iour is hypervigilance and safety-seeking. The direction of the vigilance is determined by the
perceived location of the threat.
In both of these disorders steps are taken to protect one’s health—by avoidance, escape and
preventive measures. The exact form of the safety-seeking is determined by the perceived
location of the threat, external or internal. Uncertainty and doubt reign and these lead to repeti-
tive checking, medical testing, and repeated requests for reassurance. Although the energetic
search for safety is common to both, compulsive cleaning is far more prominent in the fear of
disease contamination than in health anxiety—because the source is usually identifiable and can
be dealt with by removing it. In both disorders the elevated level of background anxiety is sub-
ject to episodic surges in reaction to specific threats that leave behind a residue of uneasiness.
The physiological and behavioural similarities between phobias and a fear of contamination
were noted a while ago (Rachman and Hodgson, 1980). The linking cognition appears to be an
exaggerated prediction of serious harm coming from the phobic object, person or place. Both
disorders are disorders of fear but with differing threat content; the simplest course is to fold the
fear of contamination into the general category of phobia. For most purposes this is a satisfac-
tory step, especially from a therapeutic standpoint (‘‘exposure treatment rules’’), but it runs the
risk of losing some subtle variations. In the present analysis for example, a workable distinction
is made between the fear of disease contamination and the fear of dirt contamination.
Next, there appears to be a cognitive link between obsessions and mental pollution. It has
been proposed that obsessions are caused by a catastrophic misinterpretation of the personal
significance of one’s unwanted intrusive thoughts (Rachman, 1994, 2004). Unwanted thoughts/
images/impulses are extremely common but can be transformed into obsessions by misinter-
pretations of their significance. Additionally, obsessions are accompanied or even promoted by
negative self-appraisals. Mental pollution can also be promoted by pre-existing negative
appraisals, and comparable to obsessions, can be precipitated by unwanted intrusive images or
thoughts or impulses. An example that covers both disorders can be found in incestuous images
and impulses. If the incestuous images are interpreted by the person as evidence of lurking,
unacceptable, repulsive tendencies they can turn into recurrent, uncontrollable obsessions. They
can also produce feelings of mental pollution, a sense of internal, inaccessible, and moral dirti-
ness. The primary but not the sole cognitive link between the two disorders is the catastrophic
misinterpretation of one’s unwanted, intrusive thoughts. However, there are several routes to
 S. Rachman / Behaviour Research and Therapy 42 (2004) 1227–1255 1249

mental pollution and the misinterpretation of one’s intrusive thoughts is one of them. Other
precipitants include sexual assaults (Fairbrother and Rachman, 2004), assaultive images, mem-
ories, humiliations, insults, accusations. The relative frequency and seriousness of the various
precipitants can not be estimated at present. Probably mental pollution is most often associated
with blasphemous and sexual obsessions and less often with harm obsessions. As mentioned
earlier, mental pollution is likely associated with a fear of dirt contamination; they share the
behavioural manifestations of disgust, cleaning and avoidance.
A probable cognitive link between a fear of dirt contamination and social phobia is the dread
and expectation of social disapproval or outright rejection and isolation.The troublesome cogni-
tions involved in social anxiety are described by Clark and Wells (1995). Fears of contamination
by contact with dirt have at least two elements, a dread of the unpleasantness of the feelings of
disgust induced by the contact (nausea, vomiting etc), and a dread of the social consquences of
one’s state of pollution. People who fear social rejection feel compelled to avert this awful possi-
bility by making concentrated afforts to convey a good impression of themselves in company.
The beliefs associated with social anxiety include the conviction that other people tend to scruti-
nise one intensively and to do so in a critical manner. In addition it is commonly believed that
the scrutineers can detect one’s anxiety and uneasiness easily and accurately, and that they
therefore downgrade one as weak, boring, stupid, inadequate. Among people who fear contami-
nation by dirt similar beliefs and cognitions can be observed. ‘‘If people find out that I am
dirty/polluted they will be disgusted and reject and avoid me.’’ ‘‘Other people can readily detect
my dirtiness/pollution, my bad breath/flatulence/body odour/smelly socks, they can tell.’’ And
it follows that I must ensure that all possible signs of my dirtiness are eliminated or masked
before entering the company of others—or social disaster will occur. Typically this takes the
form of compulsive cleaning, especially after bowel movements, masturbation, handling dirty
products and so forth. Some patients are reluctant to venture out unless and until they empty
their bowels. Repeated checking to ensure that one is free of pollution is common. There are of
course differences between these two disorders but they do appear to share important cognitions
about the likelihood of being subjected to critical scrutiny in company, and the threat of a
social disaster.

16. Cognitions and contamination

The starting premise for a cognitive analysis of the fear of contamination is the same premise
that precedes the analysis of any and all fears. Fear is a response to a perceived threat. The
physiological reactions to the threat, and the behavioural consequences, are similar across all
fears. The distinguishing characteristics of the different fears arise out of the nature of the parti-
cular threat, and how the person interprets the perceived threat. Broadly, all threats come down
to three possibilities,or combinations of the three: a fear of physical harm, a fear of mental
harm,a fear of social harm. ‘‘I fear that I will get seriously ill, be injured or killed. I fear that I
will lose control of my mind/behaviour and become insane. I fear that I will be rejected and iso-
lated from other people’’. All three of the broad threats can arise in the fear of contamination,
but one or other of the three is predominant in response to particular contaminants. The threat
in mental pollution is one of losing control of one’s thoughts and behaviour and hence going
1250 S. Rachman / Behaviour Research and Therapy 42 (2004) 1227–1255

insane, especially as the feelings of internal dirtiness/pollution appear to have a meaningless

quality, are intangible and puzzling, and often repugnant. As discussed above unwanted, intrus-
ive, repugnant thoughts can induce feelings of pollution (e.g. incestuous images). The thoughts
are contradicted by the available facts but remain unyielding, elusive and uncontrollable. They
are uncontrollable, perplexing, frightening and morally repellant. The inability to control these
objectionable thoughts raises self-doubts about one’s character, misgivings about the ability to
control one’s behaviour, and ultimately to a fear of having a breakdown or of going insane.
Feelings of mental pollution can also be provoked by external events that leave one sullied, for
example after a sexual assault. The cognitions associated with external provocations of mental
pollution will bear a resemblance to the cognitions that are common in PTSD (see Ehlers and
Clark, 2000). The victims of an assault, humiliation, accusation, may believe that they are per-
manently damaged/polluted by the event and that their future life is blighted. Feelings of inad-
equacy and hopelessness can be expected. The cognitions associated with mental pollution are
bound to be more complex than those associated with the fear of contamination by dirt or dis-
ease.
The threat in fears of disease contamination is clearly enunciated; contact with the contami-
nant will infect the person and cause serious bodily damage, even death. A secondary threat
cognition in disease contamination, sometimes even more intense than the fear of harm to one-
self, is the threat of being responsible for passing the danger on to other people. The threat of
being contaminated by dirt may have the two elements discussed earlier, an anticipation of dis-
gust and distress, but also a secondary dread of the negative social consequences of being dirty
or polluted. ‘‘I feel polluted/dirty, others can detect it, they will reject and avoid me’’.

17. Treatment

As the prevailing treatment for the fear of contamination, exposure and response prevention
is well-established and reasonably effective in many cases (Barlow, 2002; Clark, 2004), a
renewed consideration of therapy may appear to be redundant. There are however, several good
reasons for undertaking a re-examination. As mentioned earlier, recent evidence indicates that
patients with OCD who suffer from fears of contamination do less well than do patients with
other forms of OCD (Coelho and Whittal, 2001). In addition, too many patients with a fear of
contamination experience a return of fear after initially successful treatment. Moreover, the fear
that returns has a nasty tendency to be intense. A third reason to justify a reconsideration arises
from the recent elaboration of the concept of mental pollution, which is sometimes manifested
as a fear of contamination. Do we need to take special steps to deal with this newly recognised
form of the fear of contamination, or will exposure and response prevention be sufficient?
Fourthly, it is acknowledged that the prevailing therapy needs to be improved. Too many
patients are unable/unwilling to carry out the requirements of treatment which certainly can be
demanding and unsettling, and the extent and depth of the therapeutic changes need to be
enhanced. Fifthly, it is timely to consider whether the progress that has been made by adding
cognitive analyses and techniques to behaviour therapy (see Clark and Fairburn, 1997, among
others) has anything to contribute to improving therapy for people suffering from the often ten-
acious fears of contamination.
 S. Rachman / Behaviour Research and Therapy 42 (2004) 1227–1255 1251

The fundamental questions about the nature of the processes involved in the reduction of fear
by exposure will not be re-opened here because it would be a digression from the main topic,
but it is worth noticing that the gradual and progressive changes that generally take place dur-
ing treatment, plus the differences between intra- and inter-session changes in fear, are sugges-
tive of a process of steadily increasing habituation. There are problems with this construal
however, including an inconvenient circularity (Rachman, 1990). The occurrence of habituation
is assessed by a decline in responsiveness to a repetitive stimulus; however,as there is no inde-
pendent measure of habituation, the theory that fears are reduced by a process of habituation
cannot be tested. A competing explanation is that the therapeutic exposures are best seen as
repeated disconfirmations of the person’s expectation of harm. In cases of contamination fear
however, especially those pertaining to threats of infection, the threat is not regarded as immi-
nent and hence the repeated exposures provide no convincing disconfirmation of expected
harm(say falling ill in three year’s time). ‘‘The fact that I have not fallen ill despite omitting my
compulsive cleaning does not convince me that the long-term danger to my health has dimin-
ished’’.
The inability and/or unwillingness of patients to undertake exposure and response prevention
treatment is attributable to over-riding fear or to their scepticism about the likelihood of the
treatment helping them, or both of these obstacles. It is evident from the failures that encour-
agement, support and persuasion is not sufficient. Drawing a lesson from the positive results
achieved by cognitive behaviour therapy in managing other anxiety problems, such as panic,
and also from clinical experiences in treating obsessions with cognitive behaviour therapy
(CBT), it is desirable to precede the taxing exposure exercises with therapeutic sessions devoted
to cognitive analysis and therapy. In difficult cases the initial use of cognitive therapy (CT) can
modify the maladaptive beliefs and cognitions sufficiently to smooth the way for behavioural
experiments and cognitively structured ERP exercises. The behavioural experiments, and the
cognitively shaped ERP exercises (see Rachman, 2004), not only modify the unadaptive cogni-
tions but also have a way of exposing previously unidentified cognitions. When the cognitive
therapy begins to reduce the fear of contamination it becomes a little easier to introduce the
ERP exercises which formerly were blocked by the presence of over-riding fear. Behavioural
experiments are proving to be a valuable tactic for gaining new information about the fear while
simultaneously engineering successive disconfirmations of the inevitably over-predicted fear
(Rachman, 2004). Patients with an over-riding fear of contamination are better able to
approach and then touch a contaminant if the task is presented as a limited behavioural experi-
ment designed to collect information rather than as a first exercise that commits them to touch-
ing a lengthy list of very disturbing contaminants. If and when the patient’s maladaptive beliefs
and misinterpretations are substantially reduced the ERP exercises proceed at a good pace and
the fear and avoidance decline accordingly. In the absence of a clear cognitive analysis, progress
in ERP can be slow and erratic as well as puzzling. For example, the progress of ERP in treat-
ing a patient with strong fears of contamination and severe compulsive cleaning was indeed
slow, erratic and puzzling until it emerged that she had feelings of mental pollution (brought
on by unjustified but serious accusations of repugnant immoral conduct). It turned out that
these feelings were responsible for the puzzling jumps and shifts in her fear hierarchy. When the
mental pollution was reduced her progress in ERP accelerated and her fears declined to an
insignificant level.
1252 S. Rachman / Behaviour Research and Therapy 42 (2004) 1227–1255

At this stage we lack clear guidelines and tactics for identifying and treating mental pollution
but there are some clues. When the pattern of contamination seems irregular, unusual and puz-
zling and/or therapeutic progress is erratic, it is worth considering the possibility of mental pol-
lution, especially if the fears and cognitions have a moral element (e.g. ‘‘whenever a
contaminated person comes near me I feel frightened and guilty,as if I’ve done something really
awful,but I know that I haven’t’’). A patient who suffered from blasphemous, sexual obsessions
developed a fear of being contaminated by some items of her own clothing and found it so per-
plexing that she wondered if she was losing her mind. It transpired that she had worn most of
the contaminated items to church services and if she had experienced repulsive and blasphem-
ous sexual images during prayers, the clothes became contaminated. She was tormented by
these uncontrollable thoughts and concluded that she was a hypocrite and secretly subservient
to the devil rather than to God. When her feelings of religious pollution were tackled she made
gratifying progress in overcoming her fears of contamination. In view of the evidence that many
victims of sexual assault experience mental pollution (Fairbrother and Rachman, 2004) clini-
cians should be alert to this possibility when attempting to assist such victims.
In their discussion of disgust Rozin and Fallon (1987) considered two methods of overcoming
such feelings: extinction by exposure and ‘‘conceptual reorientation’’, and presciently antici-
pated that the later might be more effective. Disgust can be reversed by simple reorientations, as
in the examples given earlier (the rotten milk was actually yogurt, the forbidden pork that you
inadvertently ate was actually lamb, the brown spot on your hand is chocolate not fecal matter,
etc.). As mentioned earlier, interpersonal factors can be important, as in this example; ‘‘The
dry-cleaned sweater that you refused to wear because it would pollute you does not belong to to
your despised enemy, but to a close friend.’’ It remains to be seen whether conceptual reor-
ientations that can be so effective in reversing ordinary feelings of disgust are equally successful
in dealing with abnormal feelings of disgust and ultimately, whether such reorientations can be
deployed to overcome mental pollution. Our clinical experiences in dealing with mental pol-
lution, limited though they are, give grounds for optimism about the potential therapeutic
power of cognitive reorientations, even when masquerading as CBT. Research into the nature
and treatment of mental pollution, especially in connection with fears of contamination, is in
the early stages.
Some of the abnormal beliefs that underpin contamination fears are strongly held and so defy
contradictory evidence that no change seems possible. In these circumstances some of the meth-
ods developed for treating obsessions can be helpful (Rachman, 2004), particularly the collec-
tion of personal, direct information by means of behavioural experiments (the evidential value
of personally collected information greatly outweighs stacks of pallid statistical information).
The planned collection of ‘‘survey’’ information in which the patient systematically records the
views and experiences of friends and relations can be helpful, especially in weakening the abnor-
mal belief that one is specially, even uniquely, vulnerable to contracting diseases by contami-
nation. At a more strategic level it sometimes is more effective to tackle the patient’s anxiety
and fear before addressing the abnormal beliefs. As Arntz et al. (1995) pointed out, it is com-
mon for people to misinterpret their fear reactions as confirmation of the presence of danger. ‘‘I
feel anxious therefore there must be a danger’’. In this way the abnormal beliefs are inadver-
tently reinforced. If the patient expresses this type of cognition it might be advisable to reverse
the order of proceeding and begin by concentrating on the fear before approaching the abnor-
 S. Rachman / Behaviour Research and Therapy 42 (2004) 1227–1255 1253

mal beliefs. This sequence has proved successful in helping some patients suffering from health
anxiety to reduce their fears and then their stubbornly abnormal beliefs.
Do fears of contamination ever disappear permanently ? The evidence from randomised con-
trol trials of therapy for OCD (Abramovitz, 1997; Barlow, 2002; Clark, 2004) shows that in
many cases the answer is affirmative, in the limited sense that no recurrence occurred during the
follow-up period. On the other hand we are able to affirm that the return of fear phenomenon
(Rachman, 1990; Craske, 1999) certainly occurs in some cases after the initially successful treat-
ment of OCD. Recurrences can emerge after minimal re-contacts and in the absence of major
stressors, and worryingly, the fear that returns can be intense, even exceeding the original level
of fear. This raises the possibility that some fears of contamination never disappear. The fear
can be suppressed by treatment for a period at least, but continues to lurk. It is exceedingly dif-
ficult to find a satisfactory answer, for how can one ever prove that a fear, in this instance of
contamination, has disappeared permanently? How much time must elapse before the absence
of the fear qualifies as a disappearance? The fact that large numbers of patients with OCD,
including those with a fear of contamination, do benefit from treatment, lastingly, should not be
ignored.
Re-appearing fear (e.g. of contamination) sometimes exceeds the intensity of the original fear;
if it is a permanent fear, it is one of a changed intensity. Furthermore, the returned fear is not
always an exact replica of the original fear; sometimes the fears have new features, even new
contaminants. This interesting development, in which a ‘‘dormant’’ fear acquires new tentacles,
is thought-provoking. Using a conditioning model, the initially successful treatment and the
subsequent return of fear, can be construed as extinction followed by spontaneous recovery.
However, if the returning fear is not a replica of the original fear, we need to ask how it is poss-
ible for a fear to evolve during a period of ‘‘dormancy’’. In the present instance how does a fear
of contamination evolve during a period of dormancy? A prime candidate is the operation of
the law of similarity—the fear is suppressed but during this period, connections are formed
between between the original contaminants and previously neutral stimuli. When the fear
returns it comprises the main features of the original fear of contamination plus some new fea-
tures. In one case the original fear of blood returned some years after treatment but had spread
to a new range of objects, anything that contained brown spots, even though the patient
asserted that she knew full well that the spots were not blood. In addition to the possible oper-
ation of the law of similarity, the presence of mental pollution may well have an impact on dor-
mant fears of contamination.
Ideally the deliberate and fuller infusion of cognitive analyses and tactics into the prevailing
exposure and response prevention method of treating fear of contamination will increase the
effectiveness and efficacy of therapy. We need to reduce treatment refusals, make the treatment
more tolerable for tentative patients, accelerate the process of treatment, reduce the occurrence
of returning fear, develop methods for treating mental pollution, and improve our grasp of the
disorder. The point of departure for a cognitive analysis and treatment is the premise that the
fear of contamination comes down to three broad threats—a threat of physical harm, of mental
harm, of social harm. These threatening cognitions are bolstered by abnormal beliefs about con-
tamination and its dangers. Most forms of safety behaviour, especially compulsive cleaning and
avoidance, conserve the maladaptive cognitions and abnormal beliefs. (Early in treatment the
discreet and circumscribed use of safety tactics can facilitate treatment, for example by enabling
1254 S. Rachman / Behaviour Research and Therapy 42 (2004) 1227–1255

and prolonging exposure to the fear-evoking contaminants.) The full details of a cognitive treat-
ment, flowing from this point of departure, are under development and embrace assessment, the
modification of the abnormal cognitions and beliefs, some exposure work, behavioural experi-
ments, and the discouragement of safety behaviour. Plans for the treatment of mental pollution
are at an early stage and are likely to concentrate on the modification of the maladaptive cogni-
tions, abnormal beliefs and perplexing feelings, supplemented by behavioural experiments and
exposures. The precise strategies and tactics await advances in our understanding of this
intriguing disorder.

References
Abramovitz, J. (1997). Effectiveness of psychological and pharmacological treatments for OCD: A quantitative
review. Journal of Consulting and Clinical Psychology, 65, 44–52.
Anand, M. (1940). Untouchable. London: Penguin Books.
Applebaum, A. (2003). Gulag: A History. New York: Doubleday.
Arntz, A., Rauner, M., & van den Hout, M. (1995). ‘‘If I feel anxious, there must be danger’’: ex-consequential
reasoning in inferring danger in anxiety disorders. Behaviour Research and Therapy, 33, 917–925.
Bandura, A. (1969). The Principles of Behavior Modification. New York: Holt.
Barlow, D. (2002). Anxiety and its Disorders. (2nd ed.). New York: Guilford Press.
Bunyan, J. (1947). In A. Stanley (Ed.), An Anthology. London: Eyre and Spottiswoode.
Chadwick, P., & Lowe, C. (1990). The measurement and modification of delusional beliefs. Journal of Consulting and
Clinical Psychology, 58, 225–232.
Clark, D. A. (2004). Cognitive-behavior therapy for OCD. New York: Guilford.
Clark, D. M., & Fairburn, C. (Eds.), (1997). Science and Practice of Cognitive Behaviour Therapy. Oxford: Oxford
University Press.
Clark, D. M., & Wells, A. (1995). A cognitive model of social phobia. In R. G. Heimberg, M. Liebowitz, & D. Hope
(Eds.), Social Phobia: Diagnosis, Assessment, Treatment. New York: Guilford Press.
Coelho, J., & Whittal, M. (2001). Are sub-types of OCD differentially responsive to treatment? World Congress of
Behavioural and Cognitive Therapies. Canada: Vancouver.
Craske, M. (1999). Anxiety Disorders. Boulder: Westview Press.
Craske, M. (2003). Origin of Phobias and Anxiety Disorders: Why More Women than Men? Oxford: Elsevier Press.
Davey, G. (1993). Factors influencing self-rated fear to a novel animal. Cognition and Emotion, 7, 461–471.
de Jong, O., Andrea, H., & Muris, P. (1997). Spider phobia in children: disgust and fear before and after treatment.
Behaviour Research and Therapy, 35, 559–562.
de Jong, P., Vorage, I., & van den Hout, M. (2000). Counterconditioning in the treatment of spider phobia: Effects
on disgust, fear and valence. Behaviour Research and Therapy, 38, 1055–1069.
de Silva, P., & Marks, M. (1999). The role of traumatic experiences in the genesis of OCD. Behaviour Research and
Therapy, 37, 941–951.
Di Nardo, P., Guzy, L., & Bak, R. (1988). Anxiety response patterns and etiological factors in dog-fearful and non-
fearful subjects. Behaviour Research and Therapy, 21, 245–252.
Ehlers, A., & Clark, D. M. (2000). A cognitive model of PTSD. Behaviour Research and Therapy, 115, 319–345.
Fairbrother, N., & Rachman, S. (2004). Feelings of mental pollution subsequent to sexual assault. Behaviour
Research and Therapy, 42, 173–190.
Fairbrother, N., Newth, S., Rachman, S., 2004. Mental pollution: Feelings of dirtiness without physical contact.
Behaviour Research and Therapy, In press.
Garety, P., Fowler, D., Kuipers, E., Freeman, D., Dunn, G., Bebbington, P., Hadley, C., & Jones, S. (1994). Cogni-
tive behavioral therapy for drug-resistant psychosis. British Journal of Medical Psychology, 67, 259–271.
Gilbert, M. (1986). The Holocaust. London: Collins.
 S. Rachman / Behaviour Research and Therapy 42 (2004) 1227–1255 1255

Haidt, J., McCauley, C., & Rozin, P. (1994). Individual differences in sensitivity to disgust. Personality and Individual
Differences, 16, 701–713.
Human Rights Watch Report. (1999). Broken People: Caste violence against India’s ‘‘Untouchables’’. New York:
Human Rights Watch Publication.
Koch, M., O’Neill, K., Sawchuk, C., & Connolly, K. (2002). Domain-specific and generalized disgust sensitivity in
blood-injection-injury phobia. Journal of Anxiety Disorders, 16, 511–527.
McKay, D. (Ed.), 2002. The role of disgust in anxiety disorders. J. Anxiety Dis., 16, 475–566.
Merckelbach, M., de Jong, P., & Arntz, A. (1993). The role of evaluative learning and disgust sensitivity in the treat-
ment of spider phobia. Advances in Behaviour Research and Therapy, 15, 243–255.
Newth, S., & Rachman, S. (2001). The concealment of obsessions. Behaviour Research and Therapy, 39, 457–464.
Olatunji, B., Sawchuck, C., Lohr, J., & de Jong, P. (2004). Disgust domains in the prediction of contamination fear.
Behaviour Research and Therapy, 42, 93–104.
Poulton, R., & Menzies, R. (2002). Non-associative fear acquisition: A review of the retrospective and longitudinal
research. Behaviour Research and Therapy, 40, 127–150.
Rachman, S. (1978). Fear and Courage. San Francisco: W.H. Freeman.
Rachman, S. (1990). Fear and Courage. (2nd ed.). New York: W.H. Freeman.
Rachman, S. (1991). A psychological approach to the study of co-morbidity. Clinical Psychology Review, 11,
461–465.
Rachman, S. (1994). Pollution of the mind. Behaviour Research and Therapy, 32, 311–314.
Rachman, S. (2004). Anxiety. (2nd ed.). East Sussex: Psychology Press.
Rachman, S., & Hodgson, R. (1980). Obsessions and Compulsions. Englewood, New Jersey: Prentice Hall.
Rachman, S., & Lopatka, C. (1986). Do fears summate? Behaviour Research and Therapy, 24, 653–660.
Radomsky, A., & Rachman, S. (1999). Memory bias in OCD. Behaviour Research and Therapy, 37, 605–618.
Rasmussen, S., & Eisen, J. (1992). The epidemiology and clinical features of OCD. Psychiatric Clinics of N. America,
15, 743–758.
Rozin, P., & Fallon, A. (1987). A perspective on disgust. Psychological Review, 94, 23–41.
Rozin, P., Haidt, J., & McCauley, C. (1993). Disgust. In M. Lewis, & J. Haviland (Eds.), Handbook of Emotions
(pp. 574–594). New York: Guilford Press.
Salkovskis, P., & Warwick, H. (1986). Morbid preoccupations,health anxiety and reassurance: A cognitive behav-
ioural approach to approach to hypochondriasis. Behaviour Research and Therapy, 24, 597–602.
Salkovskis, P. (1985). Obsessional compulsive problems: A cognitive behavioural analysis. Behaviour Research and
Therapy, 23, 571–583.
Sawchuk, C., et al. (2000). Disgust sensitivity and contamination fears in spider and blood-injection-injury phobias.
Behaviour Research and Therapy, 38, 753–762.
Tallis, F. (1997). The neuropsychology of OCD. British Journal of Clinical Psychology, 36, 3–20.
Tarrier, N., Beckett, R., Harwood, S., Baker, A., Yusupoff, L., & Legarteburu, I. (1993). A trial of two cognitive-
behavioral methods of treating drug-resistant residual psychotic symptoms in schizophrenic patients. British Jour-
nal of Psychiatry, 162, 524–532.
Tolin, D., Worhunsky, P., & Maltby, N. (2004). Sympathetic magic in contamination-related OCD. Journal of
Behaviour Therapy and Experimental Psychiatry, 35, 193–206.
Woody, S., & Tolin, D. (2002). The relationship between disgust sensitivity and avoidant behavior. Journal of Anxi-
ety Disorder, 16, 542–559.
Woody, S., & Teachman, B. (2000). Intersection of disgust and fear: Normative and pathological views. Clinical Psy-
chology, Science and Practice, 7, 291–311.