Abstract

This paper critically discusses an extremely influential multi-factorial theory of child molestation, Finkelhor’s Precondition Model. This model was one of the first comprehensive theories of the sexual abuse of children and represents a significant achievement. It provides a clear framework for approaching the study of men who have sexually abused children and has lead to both clear treatment goals and clinical innovations. It has rarely, however, been systematically critically examined and the cogency of its core constructs evaluated.

Our analysis suggests that alongside its strengths, Finkelhor’s Precondition Model has some conceptual problems.

• It suffers from vagueness;
• contains overlapping constructs;
• and a rich array of vulnerability factors that require teasing out and clarification.

The model’s attempts to provide a taxononiy highlight the diversity inherent in child sexual abuse, but it has not yet provided a structure to adequately inform treatment.

Introduction

• What causes an adult to sexually molest a prepubescent child?
• Why would a man in a stable intimate adult relationship want to have a sexual relationship with a teenage girl?
• How is it possible for someone to become sexually aroused to a child who is neither physically mature nor sexually motivated?

The answers to these questions matter. They are not simply abstract questions with no discernible impact on individual everyday lives. Understanding why child molestation occurs, and how it develops and changes over time, is of the utmost importance to help us prevent it.

Moreover, such understanding helps in the design of intervention programs to stop men from re-offending. Theory generation is not a luxury to be indulged in after the task of detecting basic phenomena and their relationships has taken place.

Treatment programs for sex offenders are typically based on theoretical assumptions concerning the psychological, biological and socio-cultural mechanisms that result in child molestation. There are usually clinical theories underlying the selection of the modules contained in treatment manuals.

For example, the inclusion of relapse prevention strategies in most state-of-the-art intervention programs for sexual offenders follows from the belief that relapse is a sequential process influenced by different cognitive, affective and contextual factors (Marshall, 1999).

The major task of a good etiological theory of rape, child molestation, or any other form of sexual offending, is to account for its onset, development and maintenance. Ideally this would be a comprehensive theory that integrated all the relevant phenomena into a coherent and rich theoretical structure.

However, an all embracing theory of this kind is still some way off, and in the current environment the most rational strategy is to embrace theoretical pluralism (Hooker, 1987). This involves the development of different kinds of theories focusing on different factors

• (Lanyon, 1991; Schwartz, 1998; Ward and Hudson, 1998).

In a recent paper we outlined a meta-theoretical framework for classifying theories based on

• their level of generality of focus, and
• the extent to which the relevant factors were anchored in either developmental or contemporary experiences and processes; the distal-proximal distinction.

In this framework, we distinguished between

• level I (multi-factorial),
• level II (single factor), and
• level III (micro-level or offense process models) theories
  (Ward and Hudson, 1998).

Level I theories represent comprehensive or multi-factorial accounts of sexual offending (e.g., Marshall and Barbaree, 1990). At this level the initial goal has been to construct a loosely associated set of constructs with which to interpret or explain empirical phenomena. In the long run, however, the aim is to develop a rich, comprehensive, integrated theory.

Level II or middle level theories have been proposed to explain single factors thought to be particularly important in the generation of sexual crimes, for example, the presence of empathy deficits (Marshall, Hudson, Jones and Fernandez, 1995). In this approach the various structures and processes constituting the variable of interest are clearly described, and their relationship with each other spelled out.

Level III theories are descriptive models of the offense chain or relapse process (e.g., Pithers, 1990; Ward, Louden, Hudson and Marshall, 1995). These micro-models typically specify the cognitive, behavioral, motivational and social factors associated with the commission of a sexual offense.

In addition to the distinction between levels of theory, we also emphasized the importance of taking into account the distal-proximal distinction.

This distinction is probably best construed as a dimension, with the terms “distal” and “proximal” anchoring each end of the scale. Therefore, whether a causal factor is distal or proximal in nature is essentially a matter of degree.

• Distal factors constitute pre-dispositional or vulnerability causal factors that emerge from both developmental experiences and genetic inheritance. These predisposing factors make a person vulnerable to offending sexually once triggering or precipitating and situational factors are present.

• Proximal factors are triggering processes or events, and emerge from the functioning of vulnerability factors.

In our use of this distinction, these are usually state variables that are the result of underlying psychological mechanisms, or contextual factors that trigger underlying vulnerabilities.

Proximal factors function to disinhibit the self-regulation of behavior and thereby
erode an individual’s capacity to control strong internal states such as deviant sexual fantasies, strong affect or negative cognitions. The failure to adequately deal with these states increases the chances of a sexual offense, particularly once the opportunity arises.

The evaluation of a theory or model involves a number of different critical dimensions (Hooker, 1987).

• The ability of a theory to account for research findings and to survive hypothesis testing is certainly a necessary requirement for scientific acceptance.
• Of equal or even greater importance is its ability to extend the scope of existing models and to integrate competing or diverse approaches to the study of the relevant phenomena.
• In addition, logical consistency, simplicity, and heuristic worth represent important epistemic values against which a theory can be evaluated.
• Ambiguity, inconsistency, vagueness, and undue complexity may restrict the overall value of a theory and should be noted whenever they are evident.

Of course theory evaluation is a comparative process and the fact that a theory contains gaps or logical inconsistencies does not mean that it should necessarily be abandoned or rejected. It’s value depends on how it compares with its competitors, and its overall explanatory value (Hooker, 1987).

In this paper, we critically discuss one extremely influential multi-factorial (i.e.. level I) theory of child molestation, Finkelhor’s Precondition Model. This groundbreaking model was the first multi-factorial explanation developed to account for child sexual abuse and has proved to be of inestimable value to researchers and clinicians alike.

To our knowledge its basic ideas have never been systematically evaluated; Howell’s ( 1994) excellent paper is essentially a review of its empirical base and clinical utility. The lack of critical attention is somewhat puzzling in view of the Precondition Model’s immense influence on the treatment of sexual offenders; a theory’s value depends in part on its conceptual integrity.

Theory construction and development tends to be neglected in psychology in general, and in the area of sexual offending in particular (Ward and Hudson, 1998). Psychological science involves both the detection of phenomena and the construction of theories to explain their occurrence.

Logical positivism, with its disdain for theory and the postulation of underlying causal factors, is no longer as dominant as it once was. Most philosophers of science now accept the crucial role that theory generation plays in the practice of science (Hooker, 1987). The critique of theory is an important component of this process and it is in this spirit that we systematically examine Finkelhor’s model of child sexual abuse.

Finkelhor's Four Preconditions Model

Finkelhor (1984) argues that sexual offending against children is a multifaceted phenomenon and is related to a variety of the man’s needs, as well as important situational and contextual variables. This requirement means that an adequate theory should be comprehensive and incorporate both psychological and sociological variables.

Finkelhor persuasively argues that it is a mistake to neglect the contribution of sexual motivation to the occurrence of child molestation; otherwise it is difficult to explain why an offense was sexual in nature. He also suggests that even if the needs for intimacy and power are related to children (emotional congruence) these can still be expressed in socially more benign ways such as teaching or coaching children’s sport. It is the theoretician’s task to account for the fact that they become sexualized to some degree.

He is also careful to point out that quite different combinations of needs may motivate different offenders. That is, an adequate theory needs to be able to account for different types of child molesters, for example situational versus preferential offenders. Indeed there are strong suggestions that the Precondition Model can function as a typology.

Another important assumption underlying his model is that individual psychopathology, related to history of abuse, is only likely to explain some forms of child sexual abuse. It is necessary to take into account pervasive socialization patterns and cultural norms and values. For example, the way children are frequently portrayed as sexual beings in the media or the fact that males are socialized to view women and children as personal possessions.

Therefore, Finkelhor argues that a complete theory of child sexual abuse needs to explain

• why some adults are capable of being sexually aroused by children,
• why sexual desires and impulses are directed toward a child, and
• why the individuals involved fail to be inhibited from acting on their impulses.
• It is necessary to explain why individuals who are unable to have their emotional needs (e.g., for intimacy) met by adults turn to children for sex and not just support or friendship.

Finkelhor states that a close examination of the literature on child sexual abuse reveals that four underlying factors have typically been used to explain the occurrence of abuse; usually in the form of, in our terms, single factor or level II theories. These factors are as follows:

• Sex with children is emotionally satisfying to the offender (Emotional Congruence);
• Men who offend are sexually aroused by a child (Sexual arousal);
• Men have sex with children because they are unable (Blockage) to meet their sexual needs in more socially appropriate ways; and finally, that
• these men become disinhibited and behave in ways they would not normally behave (Disinhibition).

He suggests that the first three factors explain why some individuals develop sexual interest in children and the fourth why this interest is manifested in sexually abusive behavior. These factors may be complementary or work antagonistically or even synergistically.

The key idea in Finkelhor’s theory is that these four groups of factors can be grouped into four preconditions that must be satisfied before the sexual abuse of a child occurs. These preconditions comprise motivation to sexually abuse a child; overcoming of internal inhibitions; overcoming of external inhibitions; and dealing with a child’s possible resistance to the abuse.

Factors 1 , 2, and 3, emotional congruence, sexual arousal to children, and blockage are all linked to precondition 1, that is, the motivation to behave sexually with a child. The final factor, disinhibition, relates to precondition 2, overcoming internal inhibitions. The remaining preconditions are more to do with the how of the offense process and seem not to relate to the causal factors. These preconditions are hypothesized to occur in a temporal sequence and each is necessary for the next to occur. Additionally, Finkelhor proposes that the model provides a typology. We will now describe each precondition in turn.

Precondition I: Factors Related to Motivation to Sexual Abuse

Finkelhor hypothesizes that there are three possible motivational factors underlying the sexual abuse of children,

• emotional congruence,
• sexual arousal, and
• the fact that some individuals are blocked from meeting their emotional and sexual needs in prosocial ways.

Emotional congruence involves a fit between an adult’s emotional needs and the characteristics of a child. For example, the need to feel powerful and in control or the identification with young children due to arrested emotional development. From
a socio-cultural perspective, the conditioning of males to be dominant and powerful in sexual relationships may facilitate the need to relate sexually to children. Similarly, the fact that men are socialized to prefer sexual experiences with partners who are smaller, younger and weaker may also increase the likelihood of this factor motivating sexual offending in some individuals.

The second factor of sexual arousal to children, or more accurately deviant sexual arousal, is hypothesized to be the result of inappropriate early conditioning experiences, traumatic childhood sexual experience or the modeling of sexual interest in children. The fact that most individuals who report early sexual experiences do not later sexually abuse children suggests that there must be something unique about this early sexual experience. For example, particularly powerful sexual encounters or the undue influence of biological factors such as abnormal hormonal levels. Finkelhor also suggests that exposure to child pornography and the tendency for males to sexualize emotional needs may result in deviant sexual arousal to children.

The final factor hypothesized to motivate some men to sexually abuse children is what Finkelhor refers to as blockage. Essentially this involves the failure of individuals to meet their sexual and emotional needs in prosocial ways. There may be a variety of different factors causing this blockage, for example,

• fear of adult females,
• castration anxiety,
• inadequate social skills, and
• marital problems.

On a socio-cultural level, repressive sexual norms about sexuality may leave some individuals feeling guilty about sex with adults and therefore unlikely to meet their needs in sexual relationships with other adults. Finkelhor makes a distinction between

• developmental blockages (e.g., fear of intimacy) and
• situational blockages (e.g., marital problems).

These factors may work in combination or independently.

Precondition II: Overcoming Internal Inhibitors

According to Finkelhor even though an individual may be motivated to have sex with a child, he may not translate this desire into a sexually abusive act. To do so requires that such an individual overcome his internal inhibitions against engaging in sexual acts with children. Thus a key question for any theory of child sexual abuse is why are conventional inhibitions about having sex with children overcome?

He suggests that

• alcohol,
• impulse disorder,
• senility,
• psychosis,
• failure of the incestinhibition mechanism in the family, and
• the presence of severe stress(e.g.. loss of a job, death of a relative)

may all erode the man’s ability control his deviant desire to have sex with a child.

In addition, socially entrenched attitudes supporting patriarchal rights for fathers and the social toleration of sexual interest in children may undermine individuals’ attempts to regulate their behavior. These attitudes and beliefs function as cognitive distortions and cause men to interpret potential sexual situations with children in self-serving ways.

Furthermore, child pornography and the inability of males to identify needs of children are additional social mechanisms contributing to this failure of control. A glance at Finkelhor’s discussion of this precondition indicates that some of these disinhibiting factors may be temporary (e.g., alcohol) and some more enduring (e.g., distorted beliefs).

Precondition III: Factors Predisposing to Overcoming External Inhibitors

Following the establishment of a motive to commit a sexual offense against a child and the overcoming of internal inhibitions, offenders are hypothesized to overcome external inhibitors. This essentially involves creating the possibility for an offense to occur and may involve elaborate planning or simply opportunistic behavior.

Finkelhor comments that there are a number of conditions that increase the possibility of this occurring. They include:

• a mother who is absent or ill,
• a mother who is not close to her child,
• social isolation of the family,
• a lack of supervision of the child,
• unusual sleeping conditions and
• the opportunity for abuser and child to be together.

In addition, broader social factors such as

• the lack of social supports for mothers,
• erosion of family networks,
• ideology of family sanctity, and
• lack of equality for women

provide the structural context for abuse to occur.

Precondition IV: Factors Predisposing to Overcoming Child’s Resistance

Once the above three preconditions have been met, the man typically has secured access to a child. He may use a number of strategies to gain and maintain sexual access to a child ranging from

• giving gifts,
• desensitizing a child to sex,
• establishing emotional dependence and
• using threats or violence.

Children who are emotionally insecure or deprived and who lack knowledge about sexual abuse may be particularly vulnerable to sexual abuse.

In addition, situations where there exists an unusual degree of trust between a child and an offender, for example, the relationship between a child and a priest, may facilitate sexual offending. The unavailability of sex education for children and the social powerlessness of children are socio-cultural factors that contribute to sexual abuse.

The basic idea behind the Precondition Model is that sexual offenses against children occur in a temporal sequence and that once the motivation to sexually abuse a child is present each successive precondition depends causally on the previous one.

The fact that precondition I addresses the role of multiple factors suggests that the model is a level I theory. However, each of the factors is specified in some detail (level II) and the temporal nature of the model also indicates that it addresses the dynamics of the offense process (level III). It also attempts to account for the heterogeneity of offenders and consistently accepts that within each precondition there are multiple ways of the major goals being achieved as well as the possibility
of multiple goals.

For example, Finkelhor argues that his model can explain why some offenders prefer boys or girls via consideration of the four factors; early conditioning experiences with a girl may lead to an entrenched preference for girls or Oedipal conflicts may lead to preference for a boy.

He supplements the Precondition Model with the constructs of exclusivity and strength of interest in children. He argues that these constructs represent distinct dimensions and it is therefore best to use them to further refine and enrich his theory.

For example, some men have strong interests in children but they are not exclusive, while others may be exclusively interested in sexual relationships with children but experience a relatively low sex drive.

Critical Comments

General

A notable feature of Finkelhor’s Precondition Model is that it contains a mixture of psychological theories from markedly different traditions, for example, psychoanalytic, attributional, and learning theories (Howells, 1994). While this strategy has the advantages of flexibility and inclusiveness, it runs the risk of inconsistency and incoherency. To speak of castration anxiety alongside skill deficits and classical conditioning is to engage very different theories and competing causal mechanisms.

It could be that Finkelhor has simply developed a theoretical framework rather than a substantive theory; his model is essentially theory neutral (see Howells, 1994). Therefore, it could be argued that his aim is to indicate the form a theory might take rather than to outline a consistent and coherent theory of sexual offending. However, Finkelhor is very explicit that his aim is to provide an explanation of child sexual abuse, and a typology, rather than construct a meta-theoretical framework to guide substantive theory generation in the field. Thus the problem of incompatible theories and causal mechanisms remains a problem for the Precondition Model.

Finkelhor commits the very mistake he perceptively notes in other theories: an adequate theory should demonstrate why psychological and social factors result in sexual offenses rather than some other type of behavior. He states that while the primary motives driving sexual offenses against children may be psychological, for example, a need for intimacy or emotional congruence, this does not suffice as an explanation on its own. There has to be some account of why such needs are expressed sexually.

However, in his model each of the three motivational factors may operate independently or in combination. If they can function as independent factors, then it is hard to see why the emotional congruence and blockage motives result in a sexual offense. What is missing is an account of why in such circumstances non-sexual needs are expressed sexually.

A possible rejoinder to this criticism is to state that while the broad category offers no clue as to why non-sexual motives should result in sexual offenses, specific examples may do. Thus castration anxiety or a lack of heterosocial skills explains why some men use children as sexual surrogates.

The inability to establish intimate relationships with women may result in men turning to children to meet their sexual and emotional needs. However, this is insufficient as it only really accounts for the fact that these men lack appropriate intimate relationships with adults. The question still remains as to why such individuals chose to have sex with children rather than use some other avenue to meet their frustrated needs such as prostitutes or pornography, or indeed why they do not simply remain isolated as far as intimate relationships are concerned. What is needed is the specification of additional mechanisms or processes that are capable of describing the links between unacceptable needs and goals, or need frustration, and the sexual processes that lead to sexual offenses against children.

In Finkelhor's model there is a lack of detail concerning the way the various psychological vulnerability factors are connected to the motive in question.

• For example, how are social skill deficits translated into developmental or situational blockages?
• Or, what is the exact nature of the relationship between a sense of powerlessness and emotional congruence?

This requires more theoretical work in order to provide mechanisms that could plausibly account for these links. For it is conceivable that low self-esteem and a sense of powerlessness might result in someone having some other form of pathology or indulging in other antisocial acts rather than child sexual abuse. The development of a substance abuse disorder or the infliction of physical violence against children could possibly fall into this category. This problem really highlights the provisional nature of Finkelhor's model and the necessity for further theoretical development.

Another weakness is the model's lack of attention to developmental factors and the tendency to focus on proximal causes of sexual offending. While the motives for sexual offending are outlined in precondition I, there is no clear sense as to their developmental trajectory or how different factors converge to create the vulnerability to commit a sexual offense.

This is in contrast to Marshall and Barbaree's (1990) integrated theory where developmental adversity interacts with socio-cultural factors, the onset of puberty, and opportunistic events to result in the initiation of sexual offending.

The Precondition Model is strongest in the way it links motives and the process of offending with environmental conditions. It is arguably the only existing multi-factorial theory to attempt to relate a broad range of causal factors to the offense chain and in doing so provides a useful framework for therapists. However, this is at the cost of failing to be explicit concerning the distal causal factors and constitutes
a weakness of the model.

A final general criticism concerns whether or not the Preconditions Model is a genuine multi-factorial theory of child sexual abuse. Finkelhor hypothesizes that any one of emotional congruence, blockage, or sexual arousal can motivate an individual to sexually abuse a child.

It is argued above that psychological vulnerability to commit an offense resides entirely within this precondition and that the second and subsequent phases represent the overcoming of control and obtaining sexual access to a child.

Therefore, Finkelhor seems to be in the curious position of arguing that some sexual offenses are essentially caused by a single (psychological) factor, contradicting his earlier claim that this problem is typically caused by diverse interacting factors. There appear to be multiple causal models embedded in the one theory.

The predisposition of an individual to sexually abuse a child could reside in three primary motives or combinations of motives

• (a) emotional congruence on its own with the other preconditions precipitating and facilitating offending;
• (b) sexual arousal on its own with the other preconditions precipitating and facilitating offending:
• (c) blockage on its own with the other preconditions precipitating and facilitating offending;
•  (d) congruence interacting in some way with sexual arousal with the other preconditions precipitating and facilitating offending;
• (e) emotional congruence interacting in some way with blockage with the other preconditions precipitating and facilitating offending,
• and so on.

The possibility of complex interactions is not itself a problem, research in many areas of psychopathology suggests this is a common Occurrence (Davison and Neale, 1998).

It is a weakness, however, that Finkelhor’s theory entails a multitude of single factor; two factor; three factor; and mutually exclusive explanatory models. For one thing, it points to a lack of conceptual clarity concerning the basic theoretical entities in the theory and their relationship with one another and the environment. It is also difficult to conceive of direct ways to test the theory in order to determine its overall adequacy as an explanation of child sexual abuse.

Precondition I

A puzzling feature of Finkelhor’s model is that the psychological vulnerabilities
outlined in precondition I, are all motives of various kinds:

• emotional congruence,
• sexual arousal, and
• blockage.

There is virtually no attention paid to the role of cognitive factors such as implicit theories, beliefs or attitudes.

Most formal and informal (or folk) psychological theories hypothesize that everyday behavior is the product of two types of mental states: beliefs and desires. Both beliefs and desires refer to associated groups of mental states, with

• beliefs to be understood broadly as including knowledge, convictions, suppositions, ideas, opinions, and 
• desires to be understood broadly as including all pro and con attitudes such as lusts, wants, wishes, preferences, goals, obligations, and values, (Wellman, 1990).

Beliefs are thought to be acquired from the world and in combination with basic needs and emotions, result in the formation of desires (Gopnik and Wellman,
1994; Wellman, 1990). Both beliefs and desires lead to action and form the framework within which other peoples’ behavior is interpreted (Wellman, 1990).

Thus emotional congruence depends on beliefs concerning children’s
abilities to satisfy an offender’s emotional needs while both sexual arousal
and blockage are influenced by beliefs concerning the nature and desires of
children. It is clear that Finkelhor tacitly assumes that cognitive factors do
causally interact with drives, needs and emotions, however he does not
identify them nor clarify their role sufficiently.

In Finkelhor’s model the psychological predisposition to sexually abuse children appears to reside entirely in the presence of motives to offend. There is little description of the way early events impact on cognitive, behavioral, affective and behavioral factors or an account of the nature of these mechanisms. As stated above, there is insufficient detail on the role of knowledge structures such as beliefs and attitudes, and their connection to desires, needs, affect and behavior.

An example of a model that does attempt to describe such interrelationships has been developed recently by Ward (in press). It is argued that child molesters’ cognitive distortions emerge from underlying causal theories about the nature of their victims, the world, and themselves. These implicit theories function like scientific theories and are used to explain empirical regularities (e.g., other people’s actions) and to make predictions about the world. They are relatively coherent and constituted by a number of interlocking ideas and their component concepts and categories.

These theories determine in part the goals offenders adopt and the strategies that are selected to pursue these goals and their subsequent revision or entrenchment. Ward argued that implicit theories are the result of early developmental experiences and provide children with a way of understanding their interpersonal world. This perspective links research and theory in a number of diverse areas of psychology and attempts to create a common framework for understanding the relationship between cognition, affect and behavior.

There appears to be some overlap in meaning in the Finkelhor model between the constructs of developmental blockage and emotional congruence. Both occur in precondition I and are postulated to be factors that motivate individuals to commit a sexual offense against a child.

The problem is that both refer to developmental contlicts and vulnerabilities. The fact that some individuals feel emotionally attracted to or at ease with children suggests that they have experienced early developmental disruptions that have left them ill equipped for intinlate adult relationships.

For example, the disruption of early attachment relationships may result in an individual fearing rejection from adults and seeking to establish emotional contacts with children who are perceived as trustworthy (Ward, Hudson and Marshall, 1996). Relatedly, developmental blockage may have its origin in early experiences with attachment figures such as the mother and leave an individual experiencing castration anxiety with potential female partners. Finkelhor also states that poor social skills in sex offenders might occur as a result of developmental blockage or in fact represent this factor.

In view of the above comments it is difficult to make a clear distinction between the two constructs and to understand exactly how they make separate contributions to the Finkelhor theory. The notion of blockage is arguably best reserved for situational factors where stressors cause an offender to revert to less adaptive ways of functioning and therefore make sex with children more probable.

This said, however, it is still hard to discern why situational stresses would constitute motives to sexually offend.

Additionally, the difference between situational blockage and factors that contribute to disinhibition becomes hard to appreciate. If the developmental strand of the blockage motive is absorbed into emotional congruence (perhaps under a different label) it makes sense to delete the construct of blockage completely. The situational form of the construct could then be placed into precondition II and facilitate the overcoming of internal inhibitions, when they exist.

A related point is that it is possible to account for many of the examples Finkelhor describes as instances of developmental blockage in terms of emotional congruence. Emotional congruence refers to the notion of a “fit” or match between an individual’s emotional needs and the characteristics of a child. A lack of social skills arguably leaves an individual unable to establish intimate and satisfactory relationships with adults. He is thus likely to feel more at ease and comfortable with children, emotional congruence.

From a psychodynamic perspective, castration anxiety may mean that an adult male may feel extremely uncomfortable with women and see children as more accepting and less threatening, emotional congruence.

Again, if an individual has experienced rejection by adult women he might attempt to avoid the further trauma by choosing children as his preferred sexual partners. The fact that he feels secure and at ease with children, and anxious and uncomfortable around adults, suggests some degree of comfort or a match between his needs and the characteristics of children, emotional congruence.

The only difficult example is when an offender turns to a child rather than another adult (or masturbation) for sex due to “repressive sexual norms”.

Finally, there are problems with the use of the term congruence. As we note above congruence really refers to the notion of the “fit” between characteristics of the child and the man’s needs. As such it seems equivalent to suitability, or the match between the child as sexual object and the man’s offending goals. The use of emotional congruence implies a similarity in emotional need that is not required.

The processes Finkelhor describes serve best to illustrate the diversity of goals, almost exclusively approach goals in our terms (Ward and Hudson, in press). It is not clear that this description is even an adequate typology of goal-derived requirements or motivational consequences to these goals. It does, however, hint at the diversity of processes involved.

Precondition II

We suggest that there is a problem concerning the role of precondition II, overcoming internal inhibitions, in light of the fact that offenders are already hypothesized to be motivated to commit a sexual offense. If individuals are motivated to have sex with a child, for whatever reason, then why do they need to overcome their internal inhibitions?

This suggests that they have coriflicting motives and are struggling to form a coherent and clear set of intentions; either to offend or to restrain their behavior. To speak of motives suggests that someone has the desire to satisfy a want or a need and a belief that a specitic set of actions will make this possible; the motive constitutes a reason for engaging in a specific action. To speak of overcoming internal inhibitions indicates that this conflict has not been resolved and arguably does not constitute a separate element or precondition. At the very least, the presence of internal inhibitions points to resistance and a contlict of goals that Finkelhor should have made clear.

Additionally, we argue that precondition II is not necessary for those offenders who do not experience such conflict. Theoretically, an absence of conflict suggests that there is no need to overcome internal resistance; the individual desires to have sex with a child and formulates a plan of action to make this possible.

In our work on the relapse and offense process in sex offenders we have presented data that suggests that once a desire to have sex with a child is present many offenders do not experience any conflict or attempt to inhibit this desire (Ward and Hudson, in press).

In our self-regulation model there are four offense pathways

• (avoidant-passive,
• avoidant-active,
• approach-automatic, and
• approach-explicit)

which are related to two broad classes of goals concerning sexual offending.

There are two relapse pathways associated with avoidance or inhibitory goals where the aim is to not sexually offend. There are also two pathways associated with approach or acquisitional goals where there is no conflict over the decision to sexually abuse a child.

For example, the fourth pathway, approach-explicit, involves conscious, explicit planning and well-crafted strategies that result in a sexual offense. Thus, there is intact self-regulation but inappropriate, harmful goals as, for example, where there are inappropriate standards concerning sex with children or attitudes toward women. These represent maladaptive ways of meeting basic human needs, such as the need to be intinlate with someone, or to establish a sense of personal power or control. Therefore, higher level goals have become mistakenly linked to sexually abusive behavior.

We suggested that these associations reflect early learning experiences (e.g., sexual abuse as a child) and the resultant belief that aggressive or sexually exploitative behavior are acceptable means to valued ends. These offenders may have excellent strategic and self-regulation skills but utilize them for socially pernicious ends.

The notion of disinhibition does not apply to such individuals; they do not lose control and do not use sex to escape from or reduce powerful negative mood states. Rather, the reverse might very well be true; these offenders may have the goal of maintaining or heightening positive emotions through the offending act. Data from two studies supports the basic assumptions of this model (see Hudson, Ward and McCormack, 1999).

Concerning the second precondition (i.e., overcoming internal inhibitions) we argue that combining both state factors and trait factors in one category is confusing. While the role of alcohol or stress in overcoming internal inhibitions is relatively straightforward, that of values or beliefs is not. The former factors adversely affect individuals’ ability to control or regulate their actions and may lead to impulsive and destructive consequences.

However, beliefs and values represent enduring psychological structures and underlay individuals life plans and projects. The fact that some may result in antisocial behaviors does not entail that they are simply disinhibitors.

We suggest that beliefs and values (via goals) are related to the control of behavior in a number of ways

• (Baumeister and Heatherton, 1996; Carver and Scheier, 1990; Ward and Hudson, 1998).

First, individuals can fail to control their behavior or emotions, and behave in a disinhibited manner, that is, they lose control. This type of self-regulation failure is usually associated with negative emotional states.

Second, the use of ineffective strategies to achieve goals can backfire and ultimately result in a loss of control; a misregulation pattern.
An example of this pattern would be a person who used alcohol or sexual fantasies to modulate negative mood states, or thought suppression to eradicate intrusive thoughts. The paradox is that the faulty use of strategies to achieve certain goals can backfire and lead to the emergence, or re-emergence, of problematic behaviors and
emotions.

The third pattern has been somewhat neglected, and paradoxically involves effective self-regulation. The major problem resides in the choice of goals, rather than a breakdown in the components of self-regulation.

For example, the setting of goals and their subsequent planning and implementation by a preferential child molester may be impeccable. The difficulty resides in his initial goals and associated values and beliefs. The emotional state associated with this type of problematic self-regulation is likely to be positive; the person is achieving their goals and does not regard their lifestyle as particularly problematic.

It is misleading therefore, to refer to the misregulation and intact selfregulation patterns as examples of disinhibition. Arguably, the notion of disinhibition only applies to the first group and it is a mistake to view values and beliefs as disinhibitors if they represent enduring cognitive factors.

Of course, distorted cognitions can function as disinhibitors if they do not reflect longstanding cognitive factors and are used to persuade the individual in question that sex with children is permissible in a certain set of circumstances.

For example, an individual might rationalize his offending by stating that he is only showing a child affection, teaching him or her about sex, or is not really engaging in sex at all. The key difference between this situation and one where an offender has longstanding belief systems that predispose him to commit a sexual offense, is that these rationalizations are simply that. They do not reflect his enduring beliefs and views concerning sex with children.

Precondition III arid IV

Concerning precondition’s III and IV, a real strength of Finkelhor’s model is that he addresses the role of socio-cultural and environmental factors in the etiology of child sexual abuse. Different types of offenders presumably use correspondingly distinct strategies to create sexual access to a child and to overcome his or her resistance. These individual differences are not, however, spelt out in the Precondition Model and their relationships to distinct motives, or combinations of motives, are unclear.

It is to be expected  that offenders motivated by emotional congruence, in the term’s more obvious sense, might be more inclined to attempt to establish an “intimate” relationship with a child due to their tendency to feel more at ease and secure (at least, in some forms of emotional congruence). While offenders who are sexually attracted to children might simply view them as sexual objects to exploit for their own satisfaction.

Recent theoretical and empirical research in the sexual offending domain has found that different offenders offense style and the degree to which they utilize violence or grooming strategies, appears to be a function of their basic implicit theories and needs (Ward and Hudson, in press). It would be useful if these underlying ideas in Finkelhor’s model could be made more explicit and their relationship to treatment clarified.

Part of the explicit reason Finkelhor gives for developing the model is to provide a classification system or typology. Certainly a strength of his work is to at least illustrate the diversity of issues and processes involved.

Yet there remains a confusion of levels, ranging from a surface description of offender behavior to putative mechanisms derived from psychodynamic theory. Thus not only is a functional typology absent, one is unlikely to result from the model as it stands.

Additionally,  we have argued that offense process, in terms of goals and strategies (essentially level III theories) is the most useful manner in which to proceed. Adequate classification of offending process then needs to be linked to pre-crime factors and the relevant situational determinants, in order to provide a rich and comprehensive system that will, ultimately inform treatment.

Related to the issues of taxonomic difficulties is the problem that there are many possible causal sequences. Moreover, as the factors and processes identified by Finkelhor are typically illustrative rather than exhaustive, the potential list is even greater. A useful typology simplifies, without being simplistic, using coherent themes. Given this it remains unlikely that the empirical adequacy of the model can be easily tested.

Conclusions

Finkelhor’s Preconditions Model was one of the first comprehensive theories of the sexual abuse of children and represents an impressive achievement. It provided a clear framework for approaching the study of men who have sexually abused children, and led to clear treatment goals and clinical innovations.

For example, targeting deviant sexual arousal, strengthening emotional regulation skills, working on intimacy issues, focusing on sociocultural factors, and teaching offenders how to identify and manage high-risk situations.

These undoubted strengths are offset, however, by conceptual difficulties which point to a need to reformulate the model in light of current theory and empirical research. For example, accounting for the different pathways and offense styles offenders exhibit.

Ironically, the richness and fertility so evident in the Precondition Model actually constitutes its Achille’s heel. It is replete with theoretical possibilities, overlapping constructs, and a rich array of vulnerability factors that require teasing out and clarification. This will also help to improve its clinical utility and research potential.

Additionally, its relationship to the other two major multi-factorial theories of child sexual abuse should be ascertained and an attempt made to integrate their strengths in order to compensate for each theory’s weaknesses.

Marshall and Barbaree’s (1990) Integrated Theory, as well as Hall and Hirschman’s (1992) Quadripartite Model, are also landmark contributions to the sexual offending area and merit serious examination, a task we are currently working on.

It is our conviction that theory construction and appraisal is a crucial part of the scientific enterprise and can lead to useful clinical innovations. To quote an old saying

• "There is nothing as practical as a good theory” (Lewin, 1951)!

Etiological theories are assumed by every practitioner and provide a map to chart the difficult and complex process of working with sex offenders. We struggle to live without such maps nor can we afford to deny ourselves the opportunity to improve their quality and scope.

Acknowledgements

We would like to thank Joseph Lee, Patrick Tidmarsh, Dion Gee, Claire
Stewart, and Chris Drake for their helpful comments on an earlier draft of
this paper.

References

Baumeister. R. F. and Heatherton, T. F. ( 1996). Self-regulation failure: An overview. Psychological Inquiry, 7, 1-15.

Carver, C. S. and Scheier, M. F. (1990). Principles of self-regulation: Action and emotion. In E. T. Higgins and R. M. Sorrentino (Eds.), Handbook  of motivation and social behavior (pp. 3-52). New York: Guilford.

Davison, G. C. and Neale. 1. M. (1998). Abnormal psychology, 7th edition. New  York: Wiley.

Finkelhor. D. (1984). Child sexual abuse: New theory and research. New York: The Free Press.

Gopnik, A. and Wellinan, H. M. (1994). The theory theory. In L. A. Hirschfeld and S. A. Gelman (Eds.), Mapping the mind: Domain specificity in cognition and culture (pp. 257-293). New York: Cambridge University Press.

Hall, G. C. N. and Hirschman, R. (1992). Sexual aggression against children: A conceptual perspective of etiology. Criminal Justice and Behavior 19, 8-23.

Hooker, C. A. (1987). A realistic theory of science. Albany, NY.: State University of New York.

Howells. K. (1994). Child sexual abuse: Finkelhor’s precondition model revisited. Psygology, Crima & Law, 1, 201-214.

Hudson. S. M.. Ward, T. and McConnack ( 1999). Offense pathways in sexual offenders. Journal of Interpersonal Violence, 14. 779-798.

Lanyon, R. I. (1991). Theories of sex offending. In C. R. Hollin and K. Howells (Eds.), Clinical approaches to sex offenders and their victiims  (pp. 35-54). Chichester: John Wiley & Sons.

Lewin, K. (1951). Field theroy in social science. New York: Harper Row.

Marshall. W. L. (1999). Current status of North American assessinent and treatment programs for sexual offenders. Journal of Interpersonal Violence, 14, 22 1-239.

Marshall, W. L. and Barbaree, H. E. ( 1990). An integrated theory of the etiology of sexual offending. In W. L. Marshall. D. R. Laws and H. E. Barbaree (Eds.), Handbook of sexual assauly: Issues, theories and treatment of the offender (pp. 257-275). New York: Plenum.

Marshall, W. L., Hudson. S. M.. Jones. R. and Fernandez. Y. M. (1995). Empathy in sex offenders. Clinical Paygology Revuw, 15. 99- 113.

Pithers. W. D. (1990). Relapse prevention with sexual aggressors: A method for maintaining therapeutic gains and enhancing external supervision. In W. L. Marshall, D. R. Laws and H. E. Barbaree (Eds.). Handbook of sexual assault: Issues, theories and treatment of the offender (pp. 343-362). New York: Plenum.

Schwartz, B. K. (1995). Theories of sexual offenses. In B. K. Schwaitz and H. R. Cellini (Eds.), The sex offender: Corrections, treatment and legal practice (p p. 1-32). NJ: Civic Research Institute.

Ward, T. (in press). Sexual offenders’ cognitive distortions as implicit theories. Aggression and Violent Behavior. 

Ward, T. and Hudson. S. M. ( 1998). The construction and development of theory in the sexual offending area: A meta-theoretical franiework. Sexual Abuse: A Journal of Research and Treatment, 10. 47-63.

Ward, T. and Hudson, S. M. (in press). A self-regulation model of relapse prevention. In D. R. Laws, S. M. Hudson and T. Ward (Eds.), Remaking relapse prevenbtion with sex offenders. Thousand Oaks. CA: Sage.

Ward, T., Hudson. S. M. and Marshall, W. L. (1996). Attachment style in sex offenders:A Preliminary study. Journal of Sex Research, 33, 17-26.

Ward, T., Louden, K., Hudson. S. M. and Marshall. W. L. (1995). A descriptive model of the offense chain for child molesters. Journal of Interpersonal Violence, 10, 452-472.

Wellinan, H. M. (1990). The child’s theory of mind. Cambridge, MA: MIT Press.