Within the last decade in particular there has been nothing short of an explosion of cultural concern, a veritable incitement to discourse, regarding the problem of paedophilia. Television, radio, and countless press reports detail the activities of paedophiles and paedophile networks almost on a daily basis. 

A recent Newsweek cover story, 'The Darkest Corner of the Internet', typifies the media's representation of cultural anxiety. 

• 'Before the Internet came along, pedophiles were lonely and hunted individuals. Authorities had child pornography under control', began the 'special report', as it worked toward the ominous crescendo: 'Today networks of child abusers are proliferating worldwide' (2001: 46).

Time Magazine and myriad national and international syndicates throughout the Western world routinely pump out similarly alarming reports. Judging by the avalanche of paedophilia reportage there would certainly appear to have been a dramatic rise in paedophile numbers and systematised paedophile networks. This perception has been given credence by various national police organisations. 

The Australian Federal Police was reported not only to have estimated in 1998 that 'up to 5000 known or suspected paedophiles were in Australia and were responsible for molesting up to 40,000 children each year', but that they were adding suspects to its 'secret paedophile database' at an unprecedented annual rate (The Age, 1998: 6). The US Federal Bureau of Investigation has similarly reported an inflation of paedophile activity, releasing a statement asserting that internet paedophiles are 'a growing threat' (CNN News, 1997).

• What is happening in Western societies right now? [*1] 
• Are paedophiles really, as these reports would have us believe, a proliferating threat?
• Have we been ignorant in the past to the numbers of paedophiles in our populations and to the depth and breadth of paedophile activity? 
• Or are our criteria expanding, such that many more behaviours are being pulled into the definitional fields of child sexual abuse and paedophilia? 
• Are we inventing more and more paedophiles? 
• Is it even possible to define the paedophile? 

 This article is not concerned with adjudicating the question of whether or not the cultural incidence of paedophilia has increased. [*2] It aims instead to interrogate the conceptual ground upon which recent efforts to identify the paedophile and paedophilic activity have pivoted.

• [*1] For a psychoanalytic cultural analysis of the escalating panic with regard to paedophilia, see Angelides (2004a).
• [*2] For a useful discussion of the question of the shifting conceptual boundaries of 'child abuse' and the effect of this on how we quantify child abusers and victims, see Hacking (1991).
  

The hegemonic domain for the propagation of paedophilia research has been the field of psychopathology. I argue that this field has profoundly 'misrecognised' paedophilia. In outlining this, I propose that the study of abnormal psychology must engage psychoanalytic, feminist, and deconstructive critiques of identity, and that it must resist the temptation to affix an ontological essence to the 'paedophile'. I conclude with the suggestion that only when research methodologies take seriously the question of the prevalence of intergenerational sexual desire in the general population can we even begin to understand paedophilia.

Un/Knowing the Enemy

Definitional consensus continues to elude psychologists and psychiatrists with respect to the meaning of the very terms 'paedophile' and 'paedophilia'. These terms are highly unstable, ambiguous, and often contradictory. Research that takes such terms as their point of reference is often dubiously circular, presuming the existence of that which is unable to be defined consistently. 

As adjustments to diagnostic categories in the American Psychiatric Association's Diagnostic and Statistical Manual of Mental Disorders (DSM) between 1980 and 1994 reveal, if clinicians alter their diagnostic practices to match changing definitions, the number of paedophiles subsequently identified may vary enormously. 

For example, in the DSM-III-R the diagnosis of paedophilia was dependent upon the subject admitting he experienced 'recurrent intense sexual urges and sexually arousing fantasies involving sexual activity with a prepubescent child or children' (American Psychiatric Association, 1987: 285). Notwithstanding the fact that he may have had sex with many children, so long as he denied recurrent urges or fantasies of sex with children would he be spared the diagnostic label of paedophile. The DSM-IV (American Psychiatric Association, 1994) sought to redress this problem by including recurrent 'behaviours' in the definitional criteria. Under the revised definition it is clear that many more people can be identified as paedophiles.

There has also been 'poor consensus' in delimiting the age difference between adults and children constitutive of paedophilic acts. This, too, problematises efforts to define the 'paedophile'. DSM-III (American Psychiatric Association, 1980) set the age difference at ten years or more. This was later revised to five years, with the offender having to be a least 16 in order to indicate paedophilia (American Psychiatric Association, 1987). 

However, the problems such definitional issues raise, as clinician W. L. Marshall (1997: 154) points out, is that past and future research—both quantitative and qualitative—will be incommensurable and incomparable. [*3] 

• [*3] And this says nothing of the highly dubious means by which social scientists elicit information regarding an individual's urges and behaviours, recurrent or otherwise. I am thinking here of phallometry. See Barbaree and Peacock (1995) for a summary of some of the problems associated with this approach. 
  

Compounding the issue of definition further is the fact that the precise meaning of paedophilia relies upon definitions of what it means to be a 'child'. Yet definitions of the category of the 'child' vary dramatically according to historical period, culture, and society, not to mention the ways in which the term is differentially and contradictorily defined in legal statutes within a particular nation, society, or historical period. And in a context within which children reach puberty at different ages — the average of which has been decreasing steadily in the West with each decade — the category of the 'paedophile' is even more difficult to pin down.

One of the effects of the drive to construct ever more precise definitions and profiles of paedophile character types is the creation of countless subspecies of paedophile. According to Marshall, the increasing heterogeneity of offender subtypes is 'the major stumbling block to continued progress in this field' (1997: 168). The solution, he argues, is 'reducing this heterogeneity to manageable proportions' (1997: 168). 

What many researchers have tended to do is to lump paedophilic offenders into two broad groupings: 'regressed' versus 'fixated' types 

• (Groth and Birnbaum, 1978; Howells, 1981; Finkelhor, 1984: 49; McConaghy, 1993: 312; Cossins, 2000: 59–60). 

• Fixated offenders are thought to be those who exhibit an exclusive sexual preference for young boys, while 
• regressed offenders are thought to exhibit 'normal' sexual preference but be 'situationally induced' to have sex with children. 

McConaghy sums up the general rule of thumb:

•  'men who have a history of offending against girl children could all be considered as regressed, and 
• homosexual pedophiles and hebephiles are fixated'
  (1993: 312; see also Howells, 1981: 78). [*4]
  
• [*4] A 'hebephile' usually refers to someone with an erotic preference for adolescents between the ages of 13 and 16.
  

There are countless problems with this approach. 

While Marshall (1997) is right to argue that it is not useful to continue expanding the subtypes of paedophilia, I would argue that the major obstacle in the field of paedophilia research is itself a methodological reliance on an epistemology of identity and a politics of identification ('I am this' / 's/he is that'). 

The least productive way of approaching the problem is to identify the quintessential paedophile character profile or essentialised 'type'. 

If the historiography of sexuality has taught us anything, it is that human desire and behaviour are not best understood by making further refinements to our identity categories. There are no subspecies of the human animal that can be catalogued according to the various recurrent behaviours or desires exhibited. 

Behaviour and desire do not equal identity. 

Identity categories are the product of culturally constructed and value-laden designations and socio-political and intersubjective relationships. Yet as trite as such statements may sound to those of us studying in the social sciences and humanities, the preponderance of paedophilia research has remained almost totally unaffected by the postmodern deconstruction of identity, or the supposed 'qualitative revolution' (Denzin and Lincoln, 1998: vii).

This is particularly evident in the 'regressed' versus 'fixated' model. 

Child sex offenders have too easily been mapped onto a homo/heterosexual distinction. In the course of reviewing the research of Groth and Birnbaum (1978), Howells (1981: 78) did note that the regressed-fixated dimension might not be entirely orthogonal to the heterosexual-homosexual dimension. Despite this qualification, however, the tendency has been to assume a hierarchy of offenders and normative masculinities. 

• The 'fixated' is usually associated with the pathological homosexual predator (or 'true' paedophile) and 
• the 'regressed' is rendered the more harmless, somewhat normative, heterosexual male suffering from stressful life circumstances such as unemployment or marriage breakdown. [*5]
  
• [*5] Summarising Groth's account of the causal factors involved in regressed offender behaviour, Howells notes 'the precipitating events as physical, social, sexual, marital, financial and vocational crises to which the offender fails to adapt' (1981: 78). See also Freund (1972).
  

Shortly, I will discuss how such an approach is challenged by both quantitative and qualitative research. 

The point I wish to make here is that the fixated 'paedophile' becomes an enduring psychological profile, or identity category, metonymically associated with that of the 'homosexual' and instated in contradistinction to normative masculinities — masculinities that in some instances are seen to have gone only temporarily astray. The heterosexual offender is often spared the socially demonised identity label of true paedophile. The net effect of the implicit call to 'identity' is a dizzying incitement to discourse, where the focus of analysis becomes one of identifying a normal/deviant opposition, and fitting individuals into it accordingly. 

Accurate knowledge often becomes inextricable from projected fears and emotions, as parents, psychologists, lawyers, criminologists, and law enforcers, among others, attempt to draw boundaries around 'us' and 'them', and thus race to put a profile to the 'paedophile': 

• to identify the quintessential paedophile character type and sexuality, 
• to compile data on paedophile populations, 
• to calculate percentages of offender rates, 
• to estimate probabilities of potential paedophiles and recidivism rates, 
• to ensure adequate screening of child care workers, and generally 
• to protect children from all manner of potential threats. 

Pivotal to these responses are the notions that to recognise and dissect 'the paedophile' is to control or transform him, to protect society from him, and even to extirpate him from the social body. In fact, such has been, and continues to be, the thrust of the vast bulk of decades of social science research in the area of psychopathology. 

This is an unremarkable fact, of course, given we have just left behind what can only be described as the 'century of identity', a century characterised by the persistent multiplication of categories and subcategories of the human personality. The problem with much of this social science research into deviance is that it has tended to take the symptom for a disorder, and worse still, the symptom for an identity (Fink, 1997: 75). I argue, however, that when figured through an epistemology of identity and a politics of identification paedophilia is profoundly misrecognised.

Notwithstanding the underlying urge to identify the culprits, one of the clearest things to have come out of decades of research is that any notion of the quintessential paedophile profile is no closer in view. In recent years, in fact, the typical image of the child molester or paedophile has changed dramatically, and in ways that only hamper efforts at identification. Thanks in large part to the feminist discourse of child sexual abuse and clinical and statistical psychological data, almost gone is the profile of the child molester as an asocial stranger or dirty old man in a trench-coat — although newspapers often use this image to accompany sensationalised stories. [*6] 

• [*6] For a critique of the normative framing of the feminist discourse of child sexual abuse, see Angelides (2004b).
  

Now the converse seems to be the case. One of the most consistent research findings is that the overwhelming majority of child sexual offenders are men, and men acquainted with the child (Cossins, 1999: 51). Furthermore, the child molester or paedophile is increasingly depicted as a warm, caring friend or relative. 

As Peter Pallot warns in an article entitled, 'Doctors uncover how molesters target children', these men 'are very nice to meet', they will often 'gain the trust of families', offer 'to play sports, games or to teach the children how to play musical instruments', provide children with 'understanding and affection', 'compliments and positive attention' (1995). This process of 'manipulative' friendship building is what sociologists call 'grooming'. 

Employing this concept to explain how 'the victim is seduced into a friendship', the State Government of New South Wales Royal Commission's Paedophile Inquiry in Australia reported that an essential element in the grooming process is 'the ability' of the paedophile 'to provide the victim with material and emotional support that was lacking in his or her home life' (State Government NSW, 1997: 635, 629). Ironically, of course, these are the very things also identified as signs of good parenting.

Clinical studies thus paint a disturbing picture of ignorance regarding paedophilic offenders. The only thing psychologists and psychiatrists can be said to have ascertained, as Marshall reveals, is that paedophiles 'have few distinctive features, and on those few features that distinguish them as a group there is considerable overlap with non-child molesters' (1997: 166). 

The garden-variety paedophile is sounding more and more like a surrogate parent figure. Despite painstaking efforts at identification, 'paedophiles', as James Kincaid observes, have not yet been fully ''othered', or marginalized; they have been removed from the species, rendered unknowable' (1998: 88). Perhaps the reason for this failure to 'other' the paedophile is precisely because he is 'un-otherable'. 

I suggest that it is on the question of paedophilia that we are confronted with the limits of our epistemology of identity. Our efforts at 'othering', as we are about to see, boomerang right back onto us.

Un/Knowing Ourselves

In order to understand paedophilia I propose that we resist the knee-jerk tendency to 'other', in identitarian terms, those offenders that fall within its rather murky definitional terrain. We ought instead to problematise the category of the 'paedophile' and interrogate its structuring assumptions.

Just as psychoanalysis has long recognised that in the realm of human behaviour such binary distinctions as good/bad, normal/pathological are not immutable and mutually exclusive categories (Fink, 1997), so too has deconstruction driven home the point that 'alterity' (otherness, difference) is epistemologically dependent for its meaning upon 'sameness' (Johnson, 1987). Individuals are not permanently located at either one or the other pole of these oppositions.

.Normative definitions and identities are historically, socio-culturally, and politically specific. They are also in a continual state of transformation and contestation. Most of us throughout our lives will find ourselves simultaneously embracing, or ricocheting back and forth between, each oppositional pole. 

What this means is that in order to understand the supposed 'other' we must first understand 'ourselves'. Unfortunately, the modern history of identity has demonstrated that we have persistently failed to analyse the mutually constitutive dynamic of sameness and difference. 

• Difference, or otherness, has been the focus of inquiry, while
• 'normality' or sameness has been generally presumed or taken for granted. 

The case of homosexuality is an obvious example. The monumental initiative to detect, delimit, and define the deviant homosexual 'other' throughout the last century or so has been in large measure an attempt, by default, to naturalise heterosexuality as the norm, the right, and the good. This continues, even in the face of compelling historical, behavioural, and cross-cultural evidence suggesting the malleability of sexual desire/practice and the overlapping nature of our categories of homo-, hetero-, and bisexuality (Angelides, 2001).

The case of paedophilia is not dissimilar. Most studies of paedophilia do not clarify whether samples used comprise first or repeat offenders (Marshall, 1997: 161). Nor do they always distinguish between paedophilia and hebephilia (Freund et al., 1984: 194; Marshall, 1997). Even where research has resisted the category of the 'paedophile' and employed the more behaviourally-defined one of 'child sex offender', the focus has still been only on convicted offender populations, which have been estimated to constitute only 1 to 3 percent of child sexual offenders (Finkelhor, 1984: 17). As Finkelhor points out, convicted offenders are most probably only 'the most compulsive, repetitive, blatant, and extreme in their offending' (1984: 35). 

Add to this the greater likelihood, as Cossins (2000: 61) argues, that convicted offenders might have fallen foul of police because of committing the least socially tolerated forms of child abuse (homosexual, stranger) or of coming under greater scrutiny due to their class, race, or ethnicity, and it becomes evident that paedophilia research is almost certainly distorted due to unrepresentative population samples. 

The 'urges, fantasies and behaviours' of both undetected offenders and non-offenders are left unaccounted for. However, even when offenders are analysed in isolation, Marshall found 'no clear evidence of recurrent urges or fantasies in almost 60% of our nonfamilial child molesters and in over 75% of incest offenders' (1997: 153).

The problem with any examination of 'recurrent urges and fantasies' is that it operates only at the level of the ego or conscious thought. Not only is this approach dependent upon subjects' honest reporting — and the problem of falsifying one's urges and fantasies is a well-documented one — but also an ego-centred approach is unable to analyse unconscious mental events. [*7] 

• [*7] On the myriad problems associated with ascertaining truthful statements of offenders, see Barbaree and Peacock (1995); French (1988); Hindman (1988); Salter (1995).
  

It has clearly demonstrated how sexuality persists at the level of the unconscious, leaving all of us unaware of many of the motivations and conflicts behind our desires and actions. Not only does the erasure of unconscious dynamics further weaken our understanding of the psychology of child sexual offending, but it also has serious implications for therapeutic intervention.

Aside from the use of chemical methods, a common focus of therapeutic treatment programs is behaviour-modification techniques and educational training. These include ego-based methods aimed at empathy enhancement, 

• 'decreasing inappropriate sexual arousal by behavioural . . . methods, or increasing socially appropriate sexual arousal through guided fantasy and masturbatory conditioning' (Kosky, 1989: 177). 

However, if many offenders are motivated by unconscious conflicts, ego-based behaviour-modification techniques and education may simply be misdirected.

A focus on conscious thought also undermines the effectiveness of clinical analysis that aims at identifying paedophiles by measuring empathy levels. Frequently studies on paedophilia assume that offenders exhibit a generalised lack of empathy, a deficiency associated with operative concepts of psychopathy (Finkelhor, 1984).

However, after a review of the literature Marshall (1997: 163) found 'little support' for this idea. He argues that lower empathy levels were not an indicator of psychopathology, and were therefore not a marker of a 'paedophile'. Offenders presented appropriate empathy toward other child victims of abuse and other children in general. Only toward their own victims did they display deficient empathy. According to Marshall, then, this is likely the result of a strenuous effort to restrain or distort feelings of empathy toward victims in order that they may molest the child. 

This is consistent with psychoanalytic research that has established how repression and other defence mechanisms are employed by all individuals as a means of coping with feelings that are associated with prohibited behaviours (Moore and Fine, 1995). 

Empathy is not a fixed trait or property of an individual, but a capacity that is mediated by social and intersubjective processes. As such, it must be analysed in specific social and interpersonal contexts. This requires comparative studies of other criminal offenders' empathy levels toward victims, as well as empathy levels of non-criminal offenders in their everyday encounters with people. Doing little, if anything, to tackle underlying unconscious dynamics and conflicts that are associated with child sexual abuse, such ego-based methods reinforce the unhelpful view that paedophilia is a problem of an aberrant personal identity structure.

Even the designation of paedophilia as a form of personality disturbance contradicts current research. 

Although much of it is ambiguous, Marshall (1997: 163) says the evidence shows that very few child molesters exhibit any form of psychopathology, and few studies show personality responses outside normal limits. 

The widely touted 'cycle of abuse' theory, 'practically an article of faith among clinicians' (Williams and Finkelhor, 1990: 236), is another of the conceptual tools that has only mystified our understanding of the problem of psychopathology. Held up as a possible way of identifying the factors determining child sex offending — and thus identifying the 'paedophile' or potential 'paedophiles' — the theory is based on the idea that abuse begets abuse. An abused child, it is argued, has a greater likelihood of becoming an offending adult. 

Comparative data seems to contradict even this theory however (Hanson and Slater, 1988; Freund et al., 1990; Finkelhor, 1984). The proportion of sex offenders who reported being abused in childhood are similar to those found in non-offender groups (Freund et al., 1990: 558; Williams and Finkelhor, 1990: 236). [*8] 

• [*8] Again, of course, the problem of falsifying personal information is an insurmountable factor. 
  

Confirming this observation is the fact that female children are more frequently victims of child sexual abuse, yet in adult females 'pedophilia is very rare' (Freund and Kuban, 1993: 315). Centred on the 'identification of high-risk offenders' and personality profiles, the cycle of abuse model has served only to 'other' a certain 'type' of person and to obscure the similarities with non-abused populations.

An identitarian model of sexuality informing the hegemonic discourse of paedophilia also seriously undermines our thinking on the subject. It is one that is biased toward heteronormative notions of fixed or core sexualities, and thus an essentialist construction of the 'paedophile'. As Howells describes it:

A distinction is made between offenders whose deviant behavior is a product of a deviant sexual preference for children, and those whose deviant behavior is situationally induced and occurs in the context of a normal sexual preference structure (1981: 78; my emphases).

In the fixated/regressed classification scheme,

• fixated abusers are often referred to as 'sexual preference mediated' or ' 'pedophilic' persons' (Swanson, 1968; Howells, 1981: 76–77; Finkelhor, 1984: 49). Their behaviour, or sexual desire, equals their identity or character structure. 
• The regressed, or situational offenders, on the other hand, are not defined by their behaviour. 'Not every sex offender against a child is a true pedophile', asserts a widely cited expert, 'sometimes the child serves only as a substitute for a mature partner' (Freund et al., 1990: 559). 

Curiously, it is incestuous fathers who are spared the label of psychopathic paedophile. 

• 'Most incestuous fathers are not pedophiles', argues Seng, 'but many, if not all, child molesters are pedophiles' (1986: 52). [*9] As he goes on to explain, 'it should be remembered that child molesting describes a behavior, while pedophilia more appropriately is a diagnostic term referring to a psychosexual condition' (1986: 52). 

Within this model a naturalised deviant sexuality becomes the marker separating 'true' paedophiles from so-called 'normal' men. [*10] 

• [*9] For similar statements, see also Quinsey et al. (1979); Araji and Finkelhor (1985: 24).
  
• [*10] Marshall argues that as 'long as the official diagnostic manuals attempt to maintain the fiction that some child molesters suffer from a mental disorder called 'pedophilia' while the rest do not, such manuals will not be very helpful to practitioners working with these offenders' (1997: 156).
  

Once again, the identity paradigm triumphs, and 'paedophobia', as Cossins (2000: 133) dubs it, functions like (or I would suggest as) homophobia to demarcate so-called normal from pathological masculinities and sexualities. This heteronormative structure then provides a basis for the perception that homosexual offenders are a greater social threat than heterosexual offenders (Abel et al., 1985). Even this fails to accord with research, however, which reveals that the vast majority of paedophilic offences are of a heterosexual nature (Bradford, 1988: 225, 218; Cossins, 2000).

Constituted through an epistemology of identity and a politics of identification, the fixated/regressed model is calibrated in such a way as to identify the differences between the two broad groupings of offenders and less often their similarities. The emphasis on identity also obscures the overlapping causal factors between both offender groups. 

A further complication is the fact that the entire fixated/regressed schema is premised on the self-reports of convicted offenders. Such reports are notoriously unreliable (French, 1988; Hindman, 1988; Salter, 1995: 6–7), and offenders often report according to the strictures of personal identity categories rather than the reality of the behaviours or fantasies experienced.

This is reminiscent of the way in which the identity/behaviour opposition has functioned in research into homosexuality. There has been a long history of categorising individuals who engage in homosexual activities according to a similar fixated versus situational structure. 

• 'True' homosexuals are seen to have an enduring sexual desire for members of their own sex, and are thus ontologically defined by their sexual desire. 
• Situational homosexuals, on the other hand, are seen to be 'true' heterosexuals who report having merely resorted to homosexual sex under certain circumstances, such as in prison or when access to women is made difficult (Angelides, 2001). 

Again, myriad problems abound regarding the focus on self-reported conscious thoughts, not the least of which is the disregard for unconscious fantasies, identifications, and conflicts. Yet these are crucial constituents that go toward explaining behavioural motivation. 

I argue that only when we accept the psychoanalytic theory of human sexual desire as unconscious, non-predictable and perverse — or polymorphous, if we are to use Freudian terms — can we even begin to challenge the tendency to pathologise and 'other' certain manifestations of desire, and thereby affix behaviours or desires to identities. [*11]

• [*11] In suggesting that sexual desire is perverse I mean simply that it has no preordained object or aim but is that which emerges as the effect of an individual's unique psychical negotiation of bodily experience and the world of object relations. For a useful post-Freudian theorisation of the perversity of sexual desire, see Laplanche (1989).
  

Such an argument does not require an associated claim that sexual desire is an autonomous biological entity that is simply shaped by familial and cultural imperatives (and thus in its fixed forms reflective of particular cultural identities such as homosexual, heterosexual, paedophile). 

Rather, it is to make an argument about: 

• the object relational nature of sexual desire and its ability to solder itself to anything from men, women and children to animals or underpants; 
• the non-predictability of sexual object choice; 
• the role of the unconscious in both sexual desire and object choice; and 
• the arbitrariness of the relationship between desire and object choice or identity. [*12] 
• [*12] See Chodorow (1999) and (2004) for excellent psychoanalytic accounts of sexuality. 
  

As psychoanalysis has demonstrated, contradictory sexual desires are the norm rather than the exception for any given human subject. Therefore while an individual may affirm the singularity of his or her conscious sexual desire — e.g. one hundred percent heterosexual or homosexual or even paedophilic — this says nothing about the individual's unconscious desires and identifications, which may contradict conscious ones. 

What this means is that paedophilic desire and other normative manifestations of desire are not necessarily mutually exclusive within the one person. Sexual desire is meaningless outside the context of an individual's cultural location along with his or her own personal history of object relations, unconscious fantasies, conflicts, identifications, and so on. 

With regard to paedophilia, it is extremely unhelpful — theoretically and empirically — to construct the phenomenon in terms of 'types' or categories of person who exhibit a singular form of perverted sexual desire (for children) and therefore a set or sets of character traits. We ought instead to view it as an issue of the multiplicity of desire and its possible movements, and of how specific sets of individual, social, and historical conditions give rise to certain sexual desires and behaviours. 

I am not suggesting that we are all 'paedophiles' or 'child molesters' in the sense of those who are prosecuted for such crimes, only that at the level of desire there are not fixed or essential boundaries separating 'them' from 'us', or, 'deviance' and perversity from 'normality'. [*13] 

• [*13] I would suggest something similar with regard to the distinction between 'criminals' and 'non-criminals'. We are not well served by essentialising the differences between these groups. 
  

The point I wish to underscore, then, is that we will only improve our understanding of paedophilia by examining the so-called 'normal' and the 'deviant' together within a single analytic frame.

Situating the Individual 

With the advent of the feminist discourse of child sex abuse in the 1970s an important dimension has been added to our understanding of male sex offending. [*14] Instead of isolating the individual offender in order to understand the supposed peculiarities of his condition, feminist research has attempted to understand the condition by situating the individual offender in a social context (Rush, 1980; Herman and Hirschman, 1981; Finkelhor, 1984).

• [*14] I say 'male' sex offending for the simple fact that the overwhelming majority of sex offenders are men. See Cossins (2000).

The focus has been on the historical, social, and political forces and relations of power that condition gender and sexual dynamics and modes of behaviour of whole cultures and societies. Turning their attention to socially constructed forms of masculinity, moreover, feminists have eschewed the dominant psycho-medical tendency of highlighting the differences between offending and non-offending male behaviour. 

For, as Cossins, drawing on this tradition of radical feminist scholarship points out, the psycho-medical reification of the aberrant 'pervert' or 'paedophile' is merely a way of 'preserving the normality of normative masculinity at the same time as obscuring what the offender shares with other men' (2000: xx). 

In mounting a critique of discourses of psychology and psychiatry, Cossins makes two very important points.

In reflecting a now familiar feminist line of reasoning, she argues, firstly, that, far from representing a deviation from normative masculinity, 'child sex offending is congruent with normative masculine sexual practices' (2000: 74). By this she means that there are contradictory messages in Western societies regarding the notion of intergenerational sex. Therefore, in spite of the social and legal prohibition of sex between adults and children, there are simultaneously circulating messages that condone adult sexual of desire of children. Obvious examples are child pornography, the sexualisation of children in advertising and film, and the valorisation of forms of masculine sexuality that involve the eroticisation of power over the sexual partner. 

However, Cossins couples this analysis with recent theoretical innovations in the area of masculinity studies to argue, secondly, that the category of the paedophile functions as a way of establishing power relations between men. Could it be, she asks, 'that the label of 'deviant' has more to say about the power relations between child sex offenders and the people who label them as deviant?' (2000: 133). 

Drawing on the work of R. W. Connell, Cossins argues that child sex offending ought to be seen as a 'masculinising practice' in the context of competing masculinities. What she means is that power and sexuality are inextricably entwined in the production of masculine identities and practices, and child sex offending 

• 'can be seen, both as an expression of a man's socially sanctioned power over a child and an expression of his lack of power' in relation to other men (2000: 127).

It must be said that there are significant qualitative differences between the various kinds of sexual offences committed, and clinical research has uncovered a range of causal factors involved. Nonetheless, one of the more consistent findings of this body of research is that, underlying this multiplicity of factors, the issue of power is indeed ever present in child sexual abuse. 

One of the benefits of Cossins' feminist framework is that it goes some way toward incorporating an analysis at both the macro-level of social organisation and intersubjective relationships, as well as at the micro-level of individual behavioural motivation. 

The other significant achievement of this approach is its ability to analyse the dynamic of sameness/difference within a culture or society, and thereby incorporate all men of a particular culture or society into a single analytic frame. 

To reiterate the earlier deconstructive point, it is only possible to understand 'deviance' if first we understand what something is deviating from. For Cossins and many feminists, the recognition that adult male sexual desire for children is consistent rather than incongruent with the cultural construction of forms of masculine sexuality effects an important empirical and theoretical shift; a shift that has consequences for how we theorise deviance, normative masculinity, therapeutic intervention, and social education with regard to sexuality in general. 

It also suggests, disturbingly for some, that the 'paedophile', or rather paedophilia, is much closer in view than we care to recognise. As Briere and Runtz have demonstrated, in one of the few studies to interrogate the sexual desires of both child sexual offenders and non-offenders, 

• 'there is strong support for the notion that male sexual response to children is relatively common in our society, even among normal (non-incarcerated and nonclinical) males' (1989: 71). 

This accords with anthropological and historical data on the 'ubiquity of sexual behaviors with children' (Quinsey in Briere and Runtz, 1989: 71). [*15] 

• [*15] See also Angelides (2004b).
  

Not only reinforcing the founding psychoanalytic premise that human sexuality is unconscious, non-predictable and perverse, this data also underscores the limitations of focussing on sexual desire as a means of distinguishing between so-called 'normal' and 'deviant' masculine behaviour. 

However, it is on the point of psychoanalysis' implication in understanding human behaviour and practice that I would like to register a reservation about feminist models such as Cossins' model. 

As a social constructionist-inspired account, it is somewhat deficient with regard to the theorisation of subjectivity and sexuality. For the most part, individual subjects emerge as rather two-dimensional figures, which are interpellated and spat out, in rather crude structuralist-like terms, by social discourses. Although Cossins (2000: 99–100) argues that gender practices and modes of behaviour are 'internalised' by individuals and not imposed by socialisation, what remains unexplained is the dynamic two-way relationship between the interlocking structures of psyche and society. 

As Tim Dean argues, these two structures are mutually informing, even though 'never directly or homologously' (2000: 96). Neither can assume the position of first cause, and nor can one been seen as a mirror image of the other. The question of how individual subjects affect discourse, how discourse affects subjectivity, and why within this dynamic certain subjects take up particular (dis)positions is pivotal to any account that attempts to explain behavioural motivation and subject positioning. [*16] 

•  [*16] I have borrowed this formulation of '(dis)positions' from Mark Bracher (1993: 19).
  

Cossins' model is unable to explain how certain practices and meanings are internalised. This is a particularly glaring omission when it comes to the question of why some men engage in certain practices — such as child abuse — and others do not. [*17]

• [*17] Cossins appears to work hard to avoid psychoanalysis. In a footnote to the issue of the internalisation of power relations and their possible resistance, she cites the work of Jefferson, who argues for the necessity of a psychoanalytic theory of subjectivity. However, no sooner does she register this call for a psychoanalytic theory of subjectivity does she go on to evade its implications for her work. She concludes the footnote by saying that 'the use of psychoanalytic theory embodies the danger of reducing cultural practices to the individual psyche, thus losing sight of the social context of a person's life, the dynamic impact of that context on the person in question and his or her engagement with, and relationship to, that context' (2000: 100). By failing to account for how social practices impact on the individual and how the individual engages with them, Cossins ends up at the other extreme, and reduces psychological dynamics to social practices.
  

If we are to have any hope of understanding paedophilia and child sexual abuse a psychoanalytic theory of subjectivity must be engaged. Apart from some legitimate concerns about psychoanalysis' track record of biologism, masculinism, and heterosexism, I suspect that one of the reasons why so many constructionist feminist scholars are so circumspect about psychoanalysis is because of what it tells us about human sexual desire: that it is unconscious, non-predictable, and perverse. Cossins is working within a feminist tradition that is extremely resistant to this idea, perhaps for the reason that from this it seems but a slippery slide to the argument that paedophilia is a 'natural' form of sexuality. [*18] 

• [*18] In fact, many scholars and activists have made this argument. See Angelides (2004b). 
  

However, it is not enough to reduce human sexuality to manifestations of discourse, as though adult sexual desire for children is merely the product of aberrant social messages conveyed to men. Even though Cossins recognises the current reality of male sexual desire for children, her model suggests just such a conclusion. The implication is that adult sexual desire for children will disappear once we correct the contradictory social scripts. I suspect, on the contrary, that not only is sexuality marked by a polymorphous mobility with regard to gender of object choice and type of object chosen, but also that sexuality is also polymorphously mobile with regard to age of object choice. [*19]

• [*19] See Angelides (2004b) for a discussion of the need to theorise age as a distinct axis of social analysis within discourses of sexuality.
  

Conclusion 

In my view we will continue to misrecognise paedophilia as long as we remained wedded to an epistemology of identity and a politics of identification. Such an approach too often leads to unhelpful and emotive projections and scapegoating, and deflects attention away from the broader social and intersubjective contexts that condition abusive outcomes. [*20] 

• [*20] See Angelides (2004a) for a psychoanalytic reading of the mutually constituting dynamics of identification, projection, and repression with regard to contemporary discussions of paedophilia, and the highly potent and emotive socio-discursive scenario this has engendered in many Western societies. 
  

Instead, we must take seriously psychoanalytic insights about the unconscious, non-predictable and perverse formation of human sexual desire, the deconstructive insight that sameness and difference (or, 'normality' and 'deviance') require one another for their self-definitions, and the feminist application of these by situating male sexual offending in a broader social context of human behaviour. 

It is in this way that I have suggested that knowing the 'enemy' is not so much about knowing the pathological 'paedophile other' as it is about knowing ourselves. One way of assisting the development of greater self-knowledge more attuned to these insights is by conducting research that is more representative of all segments of society. 

We ought to be conducting studies that are not driven solely by a search for 'difference' or 'deviant' identities, but by the desire to appreciate the diversity (and perversity) of that which goes by the name of 'sameness' or 'normality'. This is not to suggest that we abandon the study of psychopathology, only that such research needs to be supplemented and situated in a much broader context. 

For instance, I argue that the subject of intergenerational sexual desire is but one particularly crucial issue that requires contextualisation and further research. For in spite of the voluminous social science literature on a range of adult sexual practices, comparatively we know very little about adult sexual desire for children. 

Briere and Runtz's (1989) data of the prevalence of adult male sexual desire for children desperately needs to be tested. So too must we quantitatively interrogate the question of the extent of adult female sexual desire for children. A gender comparison might lead to important insights into the issue of male sex offending, not to mention child sex offending more generally. 

Surveying non-offending populations might also enable us to examine the factors that inhibit adults from acting on various urges and fantasies. This would seem to be a particularly pressing task in view of Briere and Runtz's (1989) findings. Of their college male sample, not only did 

• 21 percent disclose some level of sexual attraction to children and 
• 9 percent confess to having fantasies of sex with a child (with over half this latter group having masturbated to such fantasies), but disturbingly, 
• 7 percent admitted that they might consider having sex with a child if they could escape detection. 

As the authors of the study (1989: 71) conclude, given the socially repellent nature of such confessions, it is highly probable that actual rates are greater. Ascertaining the degree of prevalence with regard to adult sexual desire for children will doubtless contribute to our understanding of the diversity of sexual desire, as well as the various individual, intersubjective, and social factors that inhibit or engender child sexual abuse. 

Perhaps, rather than begin with the assumption that sexual desire or sexual abuse is the mark of pathological 'otherness', we might be better served by beginning with the psychoanalytic premise that sexuality is itself perverse and marked by an inassimilable otherness, and that all adults are capable of sexually desiring and abusing children. 

This is by no means an attempt to normalise intergenerational sex — although on the other hand it is not an attempt to demonise all of its manifestations either. Rather, it is an attempt to foreground the reality of desire's perversity and the theoretical lines of inquiry opened up thereafter. The subject of intergenerational sexual desire is an enduringly controversial one. However, as a society we must not let volatile and emotive issues of sexuality compromise our attempts to improve knowledge about ourselves; and we must also not let them stand in the way of our ability to acknowledge and to own our imperfections and failings. For only when we do this will we honestly be able to learn how best to ensure that our social and intersubjective relations do not produce abusive outcomes for our children and ourselves.

Works Cited 

Abel, G. G. et al. 'Sexual Offenders: Results of Assessments and Recommendations for Treatments,' Clinical Criminology: Current Concepts. Eds. M. H. Ben-Aron et al. Toronto: M and M Graphics, 1985, 191–205. 

American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders: DSM-III. Washington, DC: American Psychiatric Association, 1980.
——. Diagnostic and Statistical Manual of Mental Disorders: DSM-III-R. Washington, DC: American Psychiatric Association, 1987.
——. Diagnostic and Statistical Manual of Mental Disorders: DSM-IV. Washington, DC: American Psychiatric Association, 1994

Angelides, Steven. A History of Bisexuality. Chicago: University of Chicago Press, 2001.
——. 'Historicizing Affect, Psychoanalyzing History: Pedophilia and the Discourse of Child Sexuality,' Journal of Homosexuality 46.1/2 (2004a): 79–109.
——. 'Feminism, Child Sexual Abuse, and the Erasure of Child Sexuality,' GLQ 10.2 (2004b): 141–177. 

Araji, Sharon and Finkelhor, David. 'Explanations of Pedophilia: Review of Empirical Research,' Bulletin of the American Academy of Psychiatry and Law 13.1 (1985): 17–37. 

Barbaree, Howard E. and Peacock, Edward J. 'Assessment of Sexual Preferences in Cases of Alleged Child Sexual Abuse,' True and False Allegations of Child Sexual Abuse: Assessment and Case Management . ed. Tara Ney. New York: Brunner/Mazel, 1995. 242–259. 

Bracher, Mark. Lacan, Discourse, and Social Change: A Psychoanalytic Cultural Criticism. Ithaca, NY: Cornell University Press, 1993. 

Bradford, John M. W. 'Organic Treatment for the Male Sexual Offender,' Annals of the New York Academy of Sciences 528 (1988): 193–202. 

Briere, John and Runtz, Marsha. 'University Males' Sexual Interest in Children: Predicting Potential Indices of 'Pedophilia' in a Nonforensic Sample,' Child Abuse and Neglect 13.1 (1989): 65–75. Chodorow, Nancy J. Femininities, Masculinities, Sexualities: Freud and Beyond. Lexington, KY: University Press of Kentucky, 1994.
——. Thinking Psychoanalytically about Sexual Orientation: Theoretical and Clinical Complexities, The Freud Conference, Treacy Centre, Melbourne (15 May 2004). 

Cossins, Anne. 'A Reply to the NSW Royal Commission Inquiry into Paedophilia: Victim Report Studies and Child Sex Offender Profiles — A Bad Match?,' The Australian and New Zealand Journal of Criminology 32.1 (1999): 42–60.
——. Masculinities, Sexualities and Child Sexual Abuse. The Hague: Kluwer Law International, 2000. 'The Darkest Corner of the Internet,' Newsweek. 19 Mar. 2001. 46. 

Dean, Tim. Beyond Sexuality. Chicago: University of Chicago Press, 2000. 

Denzin, Norman K. and Lincoln, Yvonna S. Eds. The Landscape of Qualitative Research: Theories and Issues. Thousand Oaks, CA and London: Sage, 1998. '1800 Names on Secret Paedophile Database.' Age 3 Apr. 1998: 6.

 'FBI: Internet Pedophiles a Growing Threat,' CNN News. 8 Apr. 1997. <http://cnn.com/US/9704/08/kiddie.porn.fbi/index.html>. 

Fink, Bruce. A Clinical Introduction to Lacanian Psychoanalysis: Theory and Technique. Cambridge: Harvard University Press, 1997. 

Finkelhor, David. Child Sexual Abuse: New Theory and Research. New York: Free, 1984. 

French, D. D. 'Distortion and Lying as Defense Processes in the Adolescent Child Molester,' Journal of Offender Counselling, Services and Rehabilitation 13.1 (1988): 27–37. 

Freund, Kurt et al. 'The Female Child as a Surrogate Object,' Archives of Sexual Behavior 2.2 (1972): 119–133.
——. 'Pedophilia and Heterosexuality vs. Homosexuality,' Journal of Sex and Marital Therapy 10.3 (1984): 193–200.
——. 'Does Sexual Abuse in Childhood Cause Pedophilia: An Exploratory Study,' Archives of Sexual Behavior 19.6 (1990): 557–568. 

Freund, Kurt and Kuban, Michael. 'Deficient Erotic Gender Differentiation in Pedophilia: A Follow-Up,' Archives of Sexual Behavior 22.6 (1993): 619–628. 

Groth, A. N. and Birnbaum, H. J. 'Adult Sexual Orientation and Attraction to Underage Persons,' Archives of Sexual Behavior 7.3 (1978): 175–181. 

Hacking, Ian. 'The Making and Molding of Child Abuse,' Critical Inquiry 17.2 (1991): 253–288. 

Hanson, R. K. and Slater, S. 'Sexual Victimization in the History of Child Sexual Abusers: A Review,' Annals of Sex Research 1 (1988): 485–499. 

Herman, Judith Lewis and Hirschman, Lisa. Father-Daughter Incest. Cambridge, MA: Harvard University Press, 1981. 

Hindman, J. 'Research Disputes Assumptions about Child Molesters', NDAA Bulletin 7 (1988): 1–3. 

Howells, K. 'Adult Sexual Interest in Children: Considerations Relevant to Theories of Etiology,' Adult Sexual Interest in Children. Eds. M. Cook and K. Howells. New York: Academic, 1981: 55–94. 

Johnson, Barbara. A World of Difference. Baltimore: Johns Hopkins University Press, 1987. 

Kincaid, James R. Erotic Innocence: The Culture of Child Molesting. Durham: Duke University Press, 199

8. Kosky, Robert John. 'Should Sex Offenders be Treated?,' Australian and New Zealand Journal of Psychiatry 23 (1989): 176–180. 

Laplanche, Jean. New Foundations for Psychoanalysis. trans. David Macey. New York: Basil Blackwell, 1989; 1987. 

Marshall, W. L. 'Pedophilia: Psychopathology and Theory,' Adult Sexual Interest in Children. Eds. D. Richard Laus and William O'Donohue. New York: Guilford, 1997: 152–17

4. McConaghy, N. Sexual Behaviour: Problems and Management. New York: Plenum, 1993. 

Moore, Burness E. and Fine, Bernard D. Eds. Psychoanalysis: The Major Concepts. New Haven: Yale University Press, 1995. 

Pallot, Peter. 'Doctors uncover how molesters target children,' Electronic Telegraph. 28 Apr. 1995. <http://www.telegraph.co.uk>. 

Quinsey, V. L. et al. 'Sexual Preferences Among Incestuous and Nonincestuous Child Molesters,' Behavior Therapy 10 (1979): 562–565. 

Rush, Florence. The Best Kept Secret: Sexual Abuse of Children. Englewood Cliffs, NJ: Prentice-Hall, 1980. 

Salter, Anna C. Transforming Trauma: A Guide to Understanding and Treating Adult Survivors of Child Sexual Abuse. Thousand Oaks, CA and London: Sage, 1995. 

Seng, Magnus J. 'Sexual Behavior Between Adults and Children: Some Issues of Definition,' Journal of Offender Counselling, Services and Rehabilitation 11.1 (1986): 47–61. 

State Government of New South Wales. Royal Commission into the New South Wales Police Service: Final Report Volume IV: The Paedophile Inquiry. Sydney: State Government of New South Wales, 1997. 

Swanson, David W. 'Adult Sexual Abuses of Children,' Diseases of the Nervous System 29.10 (1968): 677–683. 

Williams, L. M. and Finkelhor, D. 'The Characteristics of Incestuous Fathers: A Review of Recent Studies,' Handbook of Sexual Assault: Issues, Theories, and Treatment of the Offender. Eds. W. L. Marshall et al.