The following text has been automatically reproduced by an Optical Character Recognition (OCR) algorithm. It may not have been checked over by human eyes. For matters of precision please consult the original pdf.

Studying Child Sexual Abuse

Studying Child Sexual Abuse:

Morality or Science?

Sue Clegg

Child abuse has become a major topic of public polemic and
academic research. I Modem feminists, like their nineteenthcentury sisters, have singled out sexual abuse for special
attention. 2 After decades in which sexual abuse was the
concern of a limited number of professionals who dealt
directly with cases, it became a political issue in America in
the late 1970s and soon afterwards in Britain. 3 Much has
been written about the repeated ‘discovery’ of different
forms of child abuse. The ‘battered child’ serves as an
example of the way in which public alarm was orchestrated
as a way of calling attention to a social ‘ill’ and attracting
resources to the professionals dealing with it. 4 The history
of child protectionism offers other examples of the way in
which concepts of abuse have been shaped by the activities
of the professionals in the field, and radical feminist and
socially conservative pressure groups.s There has been less
focus in the literature on the scientific status of child sexual
abuse as an object of inquiry; this absence makes lan
Hacking’s recent comments in Critical Inquiry extremely
interesting. 6 Hacking ventured some provocative remarks
about the social construction of child abuse in relation to its
status as a real phenomenon, expressing his concern that
efforts to study abuse were marred by ‘a wrong idea of
knowledge and causation in human affairs’.7 In the case of
child sexual abuse he argues that that which is studied and
classified as abuse flows from preconceived
conceptualisations of its causation. So that: ‘Views on the
causes and prevention of child abuse have determined, to a
great extent, the class of events that are labelled abuse. ‘8
This article will attempt to both extend and criticise Hacking’s valuable in sights from the basis of a critical realist
understanding of explanation. 9
Hacking is right in his contention that the scientific
status of investigations into the causes and effects of sexual
abuse have proceeded on the basis that what is abusive is
already constituted through a discourse outside science.

Child protectionists have been extraordinarily successful in
fixing the moral and political parameters of the debate.

Especially in the case of sexual abuse the definitions employed by the child protectionists are transferred into scientific research on the basis of their moral authority. Hacking’s critique aptly demonstrates that the class of phenomena labelled sexual abuse do not constitute a natural kind.

Radical Philosophy 66, Spring 1994

However, his own nominalist assumptions lead him to a
general scepticism about the existence of natural kinds in
the social world. This weakens the specificity of the critique
since on Hacking’s view a similar argument can be adduced
for any set of social phenomena. However, I wish to argue
that the set of events studied as ‘child sexual abuse’ do not
constitute a natural kind, not because of a general absence
of such kinds in the social world, but because of the failure
of researchers properly to theorise their particular object of
inquiry. It is the combination of an empiricist research
programme alongside a particular moral agenda that makes
the work on sexual abuse unsatisfactory. Social contestation, and therefore the irreducibility of the value dimension,
in the social sciences does not in itself preclude scientificity.

The problem with Hacking’s sophisticated conventionalism is that it leaves no room for research programmes which
could provide us with valid know ledge of the social. Hacking analyses the moral normalisation of behaviour as a
result of concrete social activities of child protectionists,
but he appears to close the door on the theoretical
redescription of the objects thus normalised as part of
scientific inquiry. Yet such ‘relatively enduring’ objects are
the stuff of the social sciences. This article will attempt to
demonstrate the vitality of critical realism in mounting a
critique in a particular field of research, rather than offering
a general defence of its philosophical principles. If philosophy is to underlabour for science it must demonstrate its
capacity in relation to concrete research agendas. Child
sexual abuse provides a topical as well as emotive example
and I will argue that, as described by the protectionists,
sexual abuse does not constitute an object for science.

The most striking aspect of studies of sexual abuse is the
range and diversity of behaviours that are categorised as
abuse and the variation between researchers regarding the
basis for inclusion. IO Empirical definitions have been based
on the social normalisation of certain standards in relation
to children and the categorisation of behaviours which fall
outside this as abusive. The age under which activity is
deemed child abuse is roughly based on conventional legal
definitions of age of consent, often established as a result of
the activities of child protectionists and not theoretically
grounded in an assessment of the real psychological or
social capacity to assent to sexual activity. Behaviours

categorised as abusive are, from a scientific point of view,
arbitrarily determined since there is no attempt to specify
the common mechanisms at work outside the moral dimension. As a result vaginal and anal penetration of toddlers coexists with sexualised verbal suggestions from uncles to 17
year-olds as examples of abuse in the various surveys. Such
empirical diversity is of course of itself no reason to reject
the conceptualisation as flawed. On a critical realist model
complexity at the level of actual events may be generated by
the operation of real causal mechanisms. Abstracting from
this messy world of events is a stage in the scientific
process. This process is characterised by Roy Bhaskar as

Practical explanations, i.e. explanations of particular
concrete phenomena, are especially tailored to open
systems, the normal condition of things. They are
decompositively-retroductive in structure, exhibiting the RRRE schema: viz. resolution of a complete
event (situation, etc.) into its components;
redescription of these components into theoretically
significant terms; retrodiction, via independently
validated normic or tendency statements, to possible
antecedents of the components; and elimination of
alternative possible causes. ll
The diversity of phenomena under scrutiny does not of itself
preclude scientific study; only a naive positivism operates
with a model of uniform empirical outcomes. The criticism
is rather that child sexual abuse is a ‘chaotic conception’;
the first movement of a scientific argument, resolution and
redescription has not been achieved. This is reflected in the
diversity of operational definitions adopted by researchers
which do not reflect any clearly articulated theoretical
disagreements. 12 The events uncovered are real at the level
of the actual and the indi vidual suffering represented should
not be underestimated, but proper theorisation is lacking.

Most feminists have adopted patriarchy theory as providing
an adequate explanatory framework, but rather than providing an explanation of abuse it has served to naturalise a manblaming morality. 13 It also labels those critical of its approach as collusive in disguising abuse out of their own
(conscious or unconscious) inability to face the existence of
the awful events described. This is not my position. I do not
wish to deny that acts of abuse take place, nor that these may
be more frequent than previously believed. My argument is
that the mass of research currently being conducted will
yield scant insight into the underlying causes of abuse, and
that protectionism is a socially conservative programme
which is at best palliative, at worst punitive.

Before embarking on the main arguments a rider is
perhaps in order. Ian Hacking in his article felt obliged to
lodge a disclaimer dissociating himself from the backlash
against child abuse legislation. In particular he wished to
put a distance between his argument and those who have
claimed that the child protectionists have launched a witch
hunt against parents. I would offer a similar, but partial,
disclaimer. Both in America and Britain conservative social
critics have attempted to deny the existence of sexual abuse

and to condemn professionals for interference in family life.

One example is the MP Stuart Bell’s defence of the Cleveland parents – an ideologically charged intervention based
on a populist support for families against the State in the
hope of electoral advantage. Neither socialists nor feminists
can condone this style of argument which denies a priori
that children could have been sexually abused inside ‘respectable’ families. The issue oflegislation is more contentious. If patriarchy theory does not offer a scientific explanation of sexual abuse then the case for punishing individual men, on the basis that their acts are exemplary of the
exercise of male power, is significantly undermined. A
naturalistic morality ties moral conclusions to scientific
explanation through the operation of critique. Casting doubts
on patriarchy as an adequate explanatory model carries
within it a criticism of the naturalised morality which argues
for punishment. A similar argument would hold in the case
of battering, where, if it can be shown that the causal
mechanisms at work do not involve individual wickedness,
the case for legal redress is substantially weakened; the
difference of course is that punishing women who are cruel
to their children is rarely proposed in radical circles whereas
punishing men who sexually abuse is. My caveat therefore
involves dissociating myself from the conservative backlash but at the same time engaging critically with feminist

Extended childhoods beyond the dependency of infancy are
a relatively recent phenomenon, and ideas about sexually
appropriate behaviour towards the young have undergone
profound changes. 14 In Britain the age of consent for intercourse was raised from 13 to 16 in 1885 after concerted
agitation by child protectionists and feminists. 15 Consent for
other sexual acts however remained at 13 until 1922. The
current law is blatantly discriminatory, consent for male
homosexuals is 21, there is no specific age for lesbian sex.

There are Europe-wide variations in ages and the Dutch
have recently reduced theirs in order to bring the law into
closer accordance with the realities of teenage sexuality.

Consent is of course a crucial moral category; the extension
of women’s social, economic and political independence
has been fundamental in extending sexual consent, and
children should also be free to exercise control over their
own bodies. Whether age of consent laws make this easier
or more difficult is of course open to debate. However, I
wish to focus on a narrower argument and explore the
implications of adopting a legal, moral and political category deemed to apply to a whole group on the basis of a
conventionally agreed age limit as the basis for scientific
work. Child abuse researchers have explicitly argued for a
moral category as the basis for their operational definitions.

David Finklehor, probably the most influential writer in the
field, argues
adult-child sex is wrong because the fundamental
conditions of consent cannot prevail in the relationship between an adult and a child. The proposition
Radical Philosophy 66, Spring 1994

seems to be a great improvement over other arguments, particularly the argument that such acts are
wrong solely because they harm the child. It puts the
argument on a moral, rather than empirical, footing. 16



Consent here is not individually given or refused, rather a
whole group of people are deemed to be outside the legitimate discourse of consent; the issue is not that children give
or refuse consent but that they do not possess that capacity
in relation to adult-child sexuality. The argument is posed
in moral terms and therefore any empirical evidence suggesting that in some circumstances older children may be
both psychologically and socially capable of giving informed consent is rendered inadmissible. For abuse researchers, for example, the American Psychological Asso-


ciation’s judgement that a child of 14 can meaningfully
consent to abortion is irrelevant to the issue of whether that
same young person could have consented to the sex which
made her pregnant. l ? Any sexual act with a child on
Finklehor’s definition is abusive. This highlights the very
broad range of acts which can come under the definition of
abuse; what these acts have in common is that they are
morally unacceptable, not that they have a common source
or produce common harmful effects. The major thrust of the
research programme that has flowed from this definition is
to establish how many empirical instances of the behaviour
occur. Much effort has gone into prevalence studies. Social
surveys using different definitions, differing sampling strategies’ and different question-asking techniques (face to
face interview, questionnaires) have produced a range of
estimates of sexual abuse prevalence from 6 to 54 per cent
for girls and 5 to 30 per cent for boys.18 Arguments about
which figure is more reliable are not only influenced by
methodological considerations but also by the child protectionist and feminist agenda, which correctly claims that
there is a greater prevalence of sexual abuse than is detected
by social workers (this is true even if the lowest figures are
accepted). These fuel the argument that more should be
‘done’ to protect children from abuse, and also the contention that there is an ‘epidemic’ of child sexual abuse. These
studies therefore meet the moral agenda implied by
Finklehor’ s definition. However, the classic empiricist strategy of counting what may well be unrelated events renders
their status as scientific findings dubious.

This becomes clear when the same evidence is used as a
basis for theorisation. Child abuse researchers have looked
for empirical regularities in relation to the perpetrators of
abuse as a basis for explanation and/or prediction. The data
collected however refer to remembered incidents in the
victim’s/survivor’s past which they would categorise as
Radical Philosophy 66, Spring 1994

abusive or which fall under the researcher’s definition of
abuse. Information about the perpetrators of abuse is therefore extremely sketchy. Surveys produce information about
the relationship with the abused, and the age difference
between them, and the types of acts; some efforts have also
been made to identify ‘risk factors’ such as social isolation
and the mother’s absence, alongside information about the
victim’s social class, race and educational background.

However, the mass of data on risk factors is a statistical
chimera since the predictive value for any particular case
can be as low as 3 per cent. 19 More importantly, the idea of
risk factors rests on a reductive account of explanation
which owes more to survey logic and the spurious desire for
prediction than to the scientific attempt to identify real

The major finding from these studies is that most perpetrators are male, whether the abuse is of boys or girls.

However, the status of this as a ‘finding’ is open to doubt
since on theoretical grounds alone systematic resolution
and redescription of the event could have retroduced that
most ~exual abuse is male. There are instances of women
sexually abusing children, but women’s abuse in the main
takes different forms involving physical violence and emotional pressure or neglect. 20 This can be retroduced from the
theoretical work on gender roles and the ways women’s and
men’s sexualities are differentially constituted, as well as
from the available data. The application of the category of
sexual abuse to women is interesting, however, as it carries
within it the possibility of a destabilisation of the category
‘sexual’ . A similar destabilisation could also be extended to
nurturing behaviour on the part of men and the possibility
that practices regarded as good have now become defined as
bad. Bathing with a child, sexual openness and lack of
inhibition about the sexual act with a partner are open to
reinterpretation as abusive. Certainly a general retrospecti ve reinterpretation of the sexual liberalisation of the 1960s
and 1970s as bad for women is now commonplace among
radical feminists.21 The irony is that feminists in the past
have attempted to extend the concept of the sexual by
acknowledging, for example, the sexual dimension to acti vities such as breast feeding and other nurturant acts. 22 The
idea that the ‘sexual’ can be unambiguously determined is
another example of the empiricism underpinning both the
surveys and social work practice, where it is not uncommon
to find assertions that nurturing and sexuality are unmistakably distinct. Most of the survey work operates in a theoretical vacuum regarding definitions of sexuality, since it is
presumed to be obvious to respondents. Women’s memories and interpretations are of course experientially valid to
the women themselves and may have inter-subjective validity for other survivors and social scientists, but they do not
necessarily describe common events or form a basis from
which to establish empirical tendencies. There are theoretical disagreements within feminism about the nature of
consent and sexuality which are relevant to considering
how women’s remembered experiences can be understood. 23
Some radical feminists have claimed that the institution of
compulsory heterosexuality means that all cross-gender

sexuality is potentially abusive, whereas socialist feminists
have pointed to the tensions within masculinity and questioned the association between male sexuality and domination.24 These different perspectives yield potentially diverse
interpretive stances, but in the survey work on child sexual
abuse the hermeneutic moment is short-circuited in the
interests of the production of prevalence statistics.

The accepted definition of child sexual abuse has frequently been based on an explicit rejection of Freudian
insights into child sexuality. 25 Freud stands condemned of
not believing women’s accounts of incestuous abuse and of
founding a misogynist therapeutic practice which denies
women’s reality. The rejection of Freud involves a critique
of the scientific status of the unconscious. Without entering
into a general debate about the status of Freud it is important
to note that the Freudian model is capable of distinguishing
real events and their representation in the mind. The Freudian model however rests on a rejection of the empiricist
view that there is a simple and direct connection between
experience and thought. This is relevant both to survey
work and ‘disclosure’ work with victims, as the moral
dictum to believe an utterance as a statement about real
events rests on a model of the psyche as capable of being
read in a straightforward unmediated way. While many
practitioners recognise resistance when a child does not
disclose, disclosure work has often been governed by a
simplistic logic of belief. 26 It is however possible to accept
the legitimacy of the therapeutic principle of accepting a
child’s statement as ‘true’ from an anti-empiricist Freudian
perspective. These issues are not capable of empirical
resolution but require serious theoretical work.

Alternative approaches to survey work are available
which might suggest that the redescription of child abuse in
theoretically significant terms could more usefully proceed
on the basis of a different involvement with available data.

Although not explicitly realist in her research agenda, Linda
Gordon’s work is suggestive. 27 She bases her study on a

detailed examination of records of American child protection agencies. The usual objection to this strategy is that
they are an unrepresentative sample. This is of course true,
but the purpose of such an examination is not to estimate
prevalence but to provide some insight into the nature of
abuse thus allowing the resolution of the phenomenon into
at least some of its parts. On the basis of the historical record

she reconstructs a picture of abuse of older girls where the
mother is absent (through death, social incapacity or illness)
and where the child is forced to take over the running of the
household and the father uses the child sexually as well as
becoming dependent on her domestic labour. Here sexuality is understood in a particular domestic context with the
man being seen as having legitimate sexual rights to his
wife’s body (rights not challenged in English law until very
recently). Gordon points to the illegitimate extension of the
‘right’ of the man to sex inside the family to girls trapped
into taking over households. This is a specific form of abuse
found in poor and immigrant households where access to
other resources were low and demands on girls to fulfill
domestic obligations were high. The mechanisms at work
relate to the nature of famil y formation in working -class and
immigrant urban communities in the late nineteenth and
early twentieth centuries. Gordon’s work is interesting
because it points to the multi-determined conjunctural
nature of specific types of abuse. Explanation is therefore
rooted in the operation of the mechanisms implicated in the
reconstruction of the family under capitalism. A class
dynamic, and one of racism in the case of immigrant
families, operates through women’s oppression inside the
famil y in ensuring that child rearing remains pri vatised. The
dangers of this form of privatised family life to women as
well as girls are obvious and well documented. The associated public morality of ‘fallen’ or ‘pure’ trapped girls, and
Gordon convincingly argues that prostitution and incest had
interrelated not separate dynamics. Here Gordon makes a
case for regarding divergent indicators of child abuse,
incest and prostitution, as in fact causally connected. Empirical diversity is condensed to theoretically significant
commonalities through the operation of common mechanisms. In making this example I have strayed considerably
from Gordon’ s own claims for her work, but I have done so
in order to demonstrate that there are ways of making
connections between diverse phenomena through theoretical rather than empiricist reasoning.

In contrast to the survey work which is unconvincing in
its ability to provide explanations, Gordon’ s work does
suggest that events can be redescribed to provide significant
advances in understanding. The material she uses is more
usually cited by child abuse researchers alongside surveys
and other historical and cross-cultural examples as evidence for an a-historical category ‘child sexual abuse’; the
historical specificity of Gordon’s work is thus lost. The
mechanisms at work in Gordon’s example shed little light
on, for example, modem sexual molestation of infants.

Theoretical redescription may reveal the need to posit
different causal mechanisms for these other categories of
events which surveys have counted but not resolved into
their component parts. This is the work of science rather
than philosophical critique. Key researchers in the area at
present seem to see no reason to go beyond studying events
grouped on moral not scientific grounds. I want to suggest
that there are two related reasons for this reluctance. One is
that probing the scientific credentials for child sexual abuse
is deemed unnecessary. An adequate explanatory frameRadical Philosophy 66, Spring 1994

work is assumed to exist in the form of patriarchy theory
which justifies putting together disparate events on the basis
oftheirreducibility to the excesses of male power. Secondly
to question the basis of prevalence research is to stand
condemned of denying the realities of male sexuality and
‘blocking’ the knowledge that abuse exists.

Patriarchy – when a theory is not a theory
Patriarchy theorists claim that the power of their analysis
derives from its ability to redescribe the diverse acts covered by the term ‘child sexual abuse’ together with other
acts directed against women – rape, violence, harassmentand to theorise them all as evidence of the general tendency
whereby all men keep all women in a state of subjugation.

Children’s oppression is thus understood as part of the
broader picture of women’s oppression. This approach
links the inappropriate sexual use of children across cultures and throughout history to contemporary abuse. An
early and influential example of this argument is found in
Florence Rush’s work on child sexual abuse which provides
a broad historical panorama of men’s sexual brutality towards predominantly female children. 28 The argument that
there is now an epidemic of child sexual abuse comes as no
surprise, since sexual abuse has always been there but due
to the operation of patriarchal ideology it has remained
hidden. Patriarchy theory not only accounts for the existence of the phenomenon, but also for the existence of other
ideological theories which either redescribe abuse as the
legitimate expression of child sexuality or deny its existence. It therefore operates as critique and grounds an
emancipatory politics in so far as feminists work for the
abolition of patriarchy in order to end abuse. More pragmatically it grounds feminist interventions supporting chil~
dren and their mothers and the removal of perpetrators
through punitive judicial measures or ‘social work’ .29
Although rarely explicitly justified in critical realist
terms, the apparent power of patriarchy theory rests on its



seeming ability to provide scientific explanation, critique,
and an associated naturalised political practice. It has gained
the moral high ground in relation to child sexual abuse
because of its unambiguous defence of the child against
men. However, a case can be made that patriarchy theory
does not provide a scientific explanation and that, rather
than grounding its moral claims on the basis of an explanation of the concrete, it works the other way round moving
Radical Philosophy 66, Spring 1994

directly from moral censure of men to theoretical blaming
in the form of patriarchy theory. This objection would of
course not discompose many feminists since they would
argue that my critique rests on a false ideological, i.e.

patriarchal, model of science. They insist that science
should properly start with its political commitments and
that feminist science represents a different standpoint, in the
case of child sexual abuse that of the child victim/survivor.

At the level of the general characterisation of science it
is of course true that value neutrality is a myth. In the
transitive dimension the standpoint of the scientist affects
the type of science produced and it is easy to find examples
of misogynist, racist and homophobic science. 3o Epistemic
relativism is a condition of all human inquiry. The problem
however is not at the epistemic level but rather at the
judgemental, of how truth claims are to be adjudicated. The
question therefore is whether the standpoint of the scientist
guarantees superior truth claims. Some contributors to the
debate about a distinctively feminist methodology, notably
Liz Stanley and Sue Wise, eschew the problem by locating
different epistemologies in the different ontologies of the
oppressed; they therefore embrace many feminist discourses. 3l In this version ontology collapses into epistemology, the site of knowing determines what is known. The
problem of adjudicating superior truth claims is therefore
suspended through the acceptance of a radical perspectivism.

Dale Spender even cedes the grounds for a male science to
exist alongside its feminist equivalent; women simply ‘know’

differently.32 This leaves any discussion of sexual abuse
outside the terms of scientific explanation since radically
different accounts could co-exist and only be judged in
terms of their interpretative adequacy.33
There is however another position which makes more
substantial claims for feminist science as a ‘successor
science’. All sorts of valid historical arguments can be
adduced to support the contingent hypothesis that scientists
who represent the position of the oppressed are more likely
to be able to grasp the totality of social relations and
penetrate beneath surface appearances. In an argument
which draws conscious parallels with Lukacs’ position
feminists have developed the idea of a distinctive feminist
standpoint which parallels that of the proletariat. Sandra
Harding explains:

The feminist standpoint originates in Hegel’ s thinking about the relationship between the master and the
slave and in the elaboration of the writings of Marx,
Engels, and the Hungarian Marxist theorist, G. Lukacs.

Briefly, this proposal argues that men’s dominating
position in social life results in partial and perverse
understandings, whereas women’s subjugated position provides the possibility of more complete and
less perverse understandings. Feminism and the
women’s movement provide the theory and motivation for inquiry and political struggle that can transform the perspective of women into a ‘standpoint’ a morally and scientifically preferable grounding for
our interpretations and explanations of nature and
social life. 34

Harding thus shares the radical feminist reduction of ontology to epistemology but claims to ground truth claims in the
historicised commitments of feminism. The evidence for a
standpoint cannot be grounded in the real, since the beliefs
of women (or workers) will not necessarily reflect their true
interests. Historical claims about the emancipatory potential of social groups rest on a scientific understanding of
their role; Marx concentrated on the working class not
because he believed they were morally superior but because
his analysis indicated that they were the social group who
could transform society by virtue of their position in the
existing social system. The parallel argument from patriarchy is that women share a common oppression and therefore have a common interest in transforming the system of
male domination; all feminist politics is rooted in this
analysis. The argument for standpoint therefore rests on the
truth claims made by Marxists and feminists. These claims
can be validated by demonstrating the superiority of their
explanatory schema; the triple critique of Capital is the
much cited model in this respect, since it provides an
analysis of capitalist social relations, a critique of political
economy, and ceteris paribus the case and means for
changing it. Standpoint does not form an a priori guarantor
of truth but indicates the group whose transformational
practices play a part in validating the theory. Feminist
claims are therefore tied to the truth claims of patriarchy
theory which identifies women as having a common and
emancipatory bond. Arguments which dent the validity of
patriarchy theory therefore rebound on the argument that
feminist science has epistemic privilege on the basis of the
membership of a particular social group. Patriarchy theory
must be independently validated in terms of its explanatory
adequacy for it to act as a guarantor of the feminist standpoint, as well as its capacity to elucidate specific concrete
phenomena such as child sexual abuse. Moral commitment
to the victims of oppression does not of itself provide an
adequate definitional starting point for scientific theory.

How successful is patriarchy theory in explaining the
concrete events brought together under the label child
sexual abuse? General criticism has been levelled against
patriarchy theory as a-historical and incapable of theorising
change. 36 The underlying hypothesis of all men benefitting
by controlling all women belies the fact that men appear
extraordinarily flexible in their control strategies, and patriarchy theory suggests no independent dynamic to explain
why these changes take place. Radical feminists have on the
whole been untroubled by this problem since men are
presumed to act consciously as a class to protect their
interests whatever the circumstances. Socialist feminist
theorists have tended to look to an independent dynamic of
history based loosely on Marxist analytical categories while
arguing that men nonetheless manoeuvered to hold onto
power. There is a wealth of socialist feminist scholarship
describing contingent mechanisms operating in a particular
periodY Explanation of the transformation of one period
into another relies on ‘Marxist’ categories because patriarchy theory itself suggests no internal contradiction and
therefore no dynamic. A closer examination of the role

which patriarchy theory plays in historical research usually
reveals it to be a complex typology, as for example in Sylvia
Walby’s work, and/or an a priori assumption. 38 ‘Class
struggle’ between men and women cannot serve as the
necessary dynamic because, although male dominance is
assumed to explain women’s oppression, it does not simultaneously create either the conditions or necessity of resistance. Patriarchy involves no dialectic, it is a unitary concept
which functions to explain constancy not change. The
difficulties of patriarchy theory in dealing with women’s
history are magnified in relation to the oppression of children where the connections are more tenuous.

The suppression of history is particularly evident in
some radical feminist analyses of child sexual abuse. Florence Rush’s cross-cultural and historical survey provides
some exceptionally horrible accounts of the defloration of
girls at ages when gross physical pain and threat to the life
of the child must have been the consequences of early
marriage. Such a catalogue of horrors cannot fail to shock.

The problem is that many of the instances she cites were
clearly not culturally regarded as abuse. The nature of
consent had different and historically specific meanings.

Putting together all this material and relabelling it as child
sexual abuse is a dubious historical practice, reinterpreting
the past through modern moral categories. This is quite
legitimate as a polemical device for pointing out the horrors
of previous societies based on the oppression and exploitation of the vast majority of humanity , but it cannot scientifically grasp the changing fate of children. The moral category child sexual abuse does not help explain early marriage or the ‘white slavery’ of nineteenth-century child
prostitution; instead it defines our ethical stance in relation
to a diverse set of harmful practices.

Children’s changed role in capitalism is most adequately
understood with reference to change in the mode of production, the separation of factory from home and private
reproduction of labour power alongside demands for an
increasingly sophisticated and literate workforce. Childhoods were clearly class demarcated and early concern
about child abuse, whatever the rhetoric, in practice addressed the cruelty and neglect of teeming working-class
urban life where poverty also fuelled mass prostitution. The
Radical Philosophy 66, Spring 1994

NSPCC eschewed charity in favour of forcing parents into
their responsibilities for their children and the type of
philanthropic intervention was influenced by general social
anxieties about working-class life. Child protectionism
provides an archetypal example of conservative social
control. Many nineteenth-century feminists and moral purity activists who campaigned against sexual abuse advocated conservative solutions, stronger sentencing for sex
crimes and continence for men, even though they were
forced to acknowledge the poverty that led to child prostitution.

More immediate changes have affected the forms sexual
abuse takes. In advanced industrial countries child prostitution is no longer the predominant form of sexual abuse. 39
This is not due to changed patriarchal practices but to
women’s economic position, and to greater financial provision for the young. This in turn has affected male and female
sexualities and the definitions of sex with young girls. The
main form of sexual abuse with girls is now ‘in the family’,
fathers, stepfathers, uncles or other close family friends often babysitters. Labelling acts as patriarchal does not
explain either why they occur, or the changes in the pattern
of abuse. Patriarchy theory provides little indication of the
sorts of research required to identify the mechanisms at
work in producing different forms of child abuse. The
structure of the theoretical argument functions as a tautology since the reason child abuse is a function of patriarchal
power is because sex with children is an example of male
dominance. As a descriptive term patriarchy is more useful
in its narrow more literal sense of ‘power of the father’ and
as such does gi ve some insight into fathers’ rights to dispose
of their children’s labour and sexuality in pre-industrial
societies. Used in this way the term draws attention to an
important distinction since the disposal of children’s sexuality was regarded as legitimate, whereas all forms of sexual
contact with children are now regarded as illegitimate and
termed abuse.

A further drawback of the category of patriarchy as the
starting point for theoretical redescription is that it focuses
exclusively on the gender of the perpetrators and on the
sexual nature of the acts. The distinctive element of childhood is lost. Many of the factors in advanced capitalist
societies which make children vulnerable to sexual abuse
also render them open to physical abuse. This appears
especially true for boys where sexual abuse is often accompanied by physical abuse. Feminists have on the whole not
considered the possible connections between these different forms of child abuse, as women are perpetrators in a
large number of cases of non-sexual abuse, suggesting that
male violence as a mechanism has little explanatory purchase on the dynamics behind child abuse. 40 Since the major
theoretical reason for not looking at the two phenomena
together has been the different gender of perpetrators,
useful theoretical links have not been drawn, yet these may
be significant for certain classes of sexual abuse. Insights
into the sorts of pressures that produce cruelty may be useful
in understanding sexual abuse since most men and women
do not abuse, yet patriarchy theory normalises child sexual
Radical Philosophy 66, Spring 1994

abuse alongside other male abuse of women. Understanding why some men and women abuse their children could
provoke more work on the way in which pressures inside
families are mediated through differently constituted
sexualities to produce different forms of response in men
and women.

Sexual abuse is regarded by feminists as abuse of power,
providing a moral basis for the definition of abuse. Sexual
abuse of children is conceptualised as a sub-species of
men’s general abuse of power. A major criticism of this
position is that much of the admittedly tentative evidence
suggests that abusing men are often relatively powerless
and inadequate in other aspects of their lives. 41 There is
nothing particularly surprising about this, but it does illustrate the dangers of mixing up moral categories with attempts at explanation. No one supposes that when a mother
batters a child she is exercising real social power – rather her
behaviour is more usefully understood as a lack of control,
a consequence of the lack of both personal and social power.

The claim that parents are abusing their power should be
recognised as a moral category, and not a particularly
progressive one, which serves to blame an individual and
frequently to justify punishment. The moralism of patriarchal theory makes for bad analysis: it also makes for bad

One of the consequences of feminist analysis and the
insistence on moral blame has been the suppression of the
psychological. There is a general antagonism to psychological theory partly based on a suspicion of family therapy,
which is seen as upholding the status quo with mothers cast
in the role of ‘natural’ protectors of children.~2 There is some
validity in this criticism; existing psychological theory is










n?torious for normalising existing sex roles and for empiriCIst research programmes that uncritically enumerate sexdifferentiated behaviour. The idea that the dynamics inside
the family can have deleterious effects on its members, and
that these may be capable of some modification, is not of
itself a conservative hypothesis, since the underlying causes
of the distorted relationships inside the family can in turn be
systematically related to women’s oppression and pressures outside the family. Criticism of particular conservative psychologies does not preclude recognition of reality of
the psychological as an appropriate, although clearly not the
only, level of theorisation. It would seem likely that adequate theories dealing with different types of abuse will
require some redescription at the level of the psychological
a.s well as positing social explanations. In a properly theonsed explanatory model these should not be alternative
accounts but adequately grasp the nature of determination
appropriate to each level. In the absence of such work the
apparent finding that men who are abusers are likely to have
been abused is unlikely to prompt research aimed at understanding the impact of sexual abuse on boys. While it is
possible to hypothesise that sexual humiliation in men may
produce a callousness to others in its turn, it is difficult to
justify the description of this as the exercise of genderbased power except on the most over-extended definition of

Patriarchy theory commits its advocates to a research
programme which is designed to demonstrate how much
child sexual abuse there is and how much there has been. It
is resistant to any redescription of events since any
redescription will involve incorporating other factors into
the theorisation and disturb the categorisation of child
sexual abuse as simply another example of timeless male
power. Whether such redescription can yield positive results cannot be decided a priori and is a matter of science not
philosophy. Patriarchy theory, however, represents not so


much a scientific theory of child sexual abuse as its incorporation into a particular moral and political framework.

There are good reasons for preferring this to moral stances
which blame the child, but a commitment to helping the
victims of oppression does not constitute adequate grounds
for accepting the validity of a theory.

This article has not attempted a new theoretical understanding of child sexual abuse. My argument is that’ child abuse’

as currently understood is a chaotic conception which
requires theoretical redescription in terms which can resolve abuse into its component parts as meaningful theoretic~l entities. I have tentatively suggested some examples;
mneteenth-century prostitution and incest in working-class
and poor immigrant communities might provide the basis
for one such redescription. More work could reveal whether
this resolution has much to offer in terms of understanding
large-scale prostitution in less developed countries, but it
seems likely that racism and class are key components in
understanding prostitution and that these cannot be reduced
to patriarchal control. The major conclusion is the need for
properly theoretically directed research; not more surveys
directed at establishing how much ‘sexual abuse’ there is.

Scientific analysis does not operate in a vacuum; arguments about the treatment of victims and offenders are
closely related. If moral commitments foreclose scientific
investigation, then research will remain at an impasse. This
is precisely what is happening in the case of child abuse.

Any theory that appears to ‘excuse’ men from their personal
responsibility for sexual crimes against children is regarded
as unacceptable to feminists. Understanding abusers is,
however, perfectly compatible with a condemnation of the
circumstances that create the conditions under which men
abuse; feminism does not provide the only grounds for an
emancipatory practice. More worrying is that patriarchy
theory as a guide to morality leads to illiberal conclusions.

These have been defended on the grounds that liberalism is
bad for women and children and the state should act to
protect both. A cursory glance at reports into cases where
children have been ‘protected’, however, suggests another
story.43 Children suffer from child protectionism as well as
perpetrators, and secondary abuse by professionals is now
recognised as a problem. 44 The effects of state intervention
and supervision fall disproportionately on working-class
men and women. 45
A critical realist theory of science is optimistic in its
conceptualisation of the relationship of science to action.

But bad science can produce bad morality. In the case of
child sexual abuse the illiberal conclusion of many feminists is tied to the inadequacies of patriarchy theory. The
moral starting point for the research programme into child
sexual abuse fixes it, and the appeal to moral not scientific
definitions prevents adequate reconceptualisation. All acts
with children under a certain age are bad and therefore all
should be counted together. My argument for better science
is not based on an indifference to children’s suffering but on

Radical Philosophy 66, Spring 1994

a view of science that connects understanding to change; the
contribution of philosophy to the process is to provide the
basis for critique. lan Hacking rightly showed that child
sexual abuse is not a natural kind, but the challenge is to
explore what kinds there are in the social world and to show
that scientific understanding has more to offer than moral


R. Dingwall, ‘Some problems about predicting child abuse and
neglect’ in D. Stevenson (ed.) Child Abuse Public Policy and
Professional Practice, Harvester Wheatsheaf, New York, 1989.


Violence Against Children Study Group, op. cit.; S. J. Creighton,
Child Abuse Trends in England and Wales 1988-1990. An
overview from 1973-1990, NSPCC, London, 1992.


S. Jeffreys, Anticlimax, Women’s Press, London, 1990.


Evident from accounts of women’s health groups, NCT reunions
and in comments from writers such as Sheila Kitzinger on the
joys of childbirth and breastfeeding.


This is not just true for other interpreters’ accounts. Survivors’

groups drew heavily on existing feminist understandings of
male violence. See Feminist Review, op. cit.


Jeffreys, op. cit. For a different view L. Segal, Slow Motion:

Changing Masculinities, Changing Men, Virago, London, 1990.


For an extremely useful account of these debates see A. Scott,
‘Feminism and the seductiveness of the “real event”, Feminist
Review, No. 28, 1988. For an anti-positivist defence of the
emancipatory potential of Freud, see R. Keat, The Politics of
Social Theory, BIackwell, Oxford, 1981.


For critical discussion of disclosure work see e. Wattam, J.

Hughes and H. Blagg (eds) Child Sexual Abuse: Listening,
Hearing and Validating the Experiences of Children, Longman,
Harlow, 1989.


L. Gordon, Heroes of Their Own Lives. The Politics and History
of Family Violence, Boston 1880-1960, Virago, London, 1989.




An American survey on sexual abuse carried out in 1981
reported that 90% of respondents claimed to have seen a discussion on TV, while 85% claimed to have read something about
sexual abuse (see D. Finklehor, Child Sexual Abuse, Free Press,
New York, 1984). In Britain an estimated 16.5 million people
watched BBC Childwatch screened on 30 October 1985 (see
Violence Against Children Study Group, Taking Child Abuse
Seriously, Unwin Hyman, London, 1990).

On 19th-century feminist campaigns see S. Jeffreys, The Spinster and Her Enemies, Feminism and Sexuality 1880-1930,
Pandora, London, 1985. For modern feminist debates see Feminist Review Special Issue: ‘Family Secrets: Child Sexual Abuse’ ,
Feminist Review, No. 28, Spring 1988.

The debate on sexual abuse in Cleveland resulted in two highly
polemical books and media speculation as well as a major report.

See Lord Justice Butler-Sloss, Report of the lnquiry into Child
Abuse in Cleveland 1987, HMSO, London, 1988; also S. Bell,
When Salem came to the Boro. The True Story of the Cleveland
Child Abuse Case, Pan, London, 1988; andB. Campbell, Unofficial Secrets: Child Abuse – The Cleveland Case, Virago,
London, 1988.


N. Parton, ‘The natural history of child abuse: a case study in
social problem definition’, British Journal of Social Work, No.

94, 1979.


See for example Behlmer’s detailed account of Victorian anticruelty campaigns and the formation of the NSPCC, G. K.

Behlmer, Child Abuse and Moral Reform in England 18701908, Stanford University Press, 1982.


I. Hacking, ‘The making and moulding of child abuse’, Critical
1nquiry, Winter 1991.


Ibid., p. 259.

Ibid., p. 261.

See in particular R. Bhaskar, The Possibility of Naturalism,
Harvester, Sussex, 1979.

See D. Finklehor, A Sourcebook on Child Sexual Abuse, Sage,
London, 1986.

R. Bhaskar, Scientific Realism and Human Emancipation, Verso,
London, p. 68.

Finklehor, op. cit.

Feminist Review, op. cit.. Other examples include E. Driver and
A. Droisen (eds.) Child Sexual Abuse. Feminist Perspectives,
Macmillan, Houndmills, 1988, and an early radical feminist
statement, F. Rush, The Best Kept Secret, McGraw-Hill, 1980.

P. Aries, Centuries of Childhood, Penguin, Middlesex, 1962.

Jeffreys, op. cit.

D. Finklehor, ‘What’s wrong with sex between adults and
children’ ,American Journal of Orthopsychiatry, Vol. 49, No. 4,

J. Haugaard, N. Reppucci and Dicken, The Sexual Abuse of
Children, Jossey Bass, San Francisco, 1988.

Finklehor, 1986, op. cit. The highest estimates come from Diana
Russell’s study. Russell chose San Franciscan women as she
believed that they would have a greater awareness of the issues;
she also used specially trained interviewers. See D. Russell,
Sexual Exploitation: Rape, Child Sexual Abuse and Workplace
Harassment, Sage, Beverley Hills, 1984.









Radical Philosophy 66, Spring 1994


Rush, op. cit.


M. MacLeo and E. Saraga, ‘Challenging the orthodoxy. Towards a feminist theory and practice’, Feminist Review, No. 28,


For a discussion of some examples see H. Rose and S. Rose (eds)
The Political Economy of Science, MacmilIan, London, 1976.


See L. Stanley (ed.) Feminist Praxis Research, Theory and
Epistemology in Feminist Sociology, Routledge, London, 1990.


D. Spender, For the Record, Women’s Press, London, 1985.


Great insight can of course be gained through these accounts and
the hermeneutic moment is indispensable in the social sciences.

The strengths of feminist work in the area of child sexual abuse
lay in the close attention to the accounts of victims/survivors.

My argument is with the move to hypothesising that the causal
mechanism at work is male domination.


S. Harding, The Science Question in Feminism, Cornell University Press, Ithaca, 1986, p. 23; also S. Harding (ed.) Feminism
and Methodology, Open University Press, Milton Keynes, 1987.


Bhaskar, op. cit.


For examples see L. German, Sex, Class and Socialism, Bookmarks, London, 1989; L. Segal, Is the Future Female?, Virago,
London, 1987.


One synoptic account can be found in S. Walby, Patriarchy at
Work, Polity, Cambridge, 1986.



Although as Judith Ennew makes clear no such sanguine conclusions apply internationally. See J. Ennew, The Sexual Exploitation of Children, Polity, Cambridge, 1986.

There are of course exceptions. See Violence Against Children
Study Group, Taking Child Abuse Seriously, Unwin Hyman,
London, 1990.


e. Hooper, ‘Child sexual abuse and the regulation of women:

variations on a theme’, in e. Smart (ed.) Regulating Womanhood, Routledge, London, 1992.

Butler-Sloss,op. cit.; Stevenson, op. cit.

Wattam, Hughes and BIagg, op. cit.

Creighton, op. cit.









Download the PDFBuy the latest issue