She or he? Who is more violent?

Intimate-Partner Violence Is not Merely Non-‘Gendered’ but Predominantly and Apparently in Essence FemalePerpetrated: 

The Inverse of Ideological Constructs Now Debunked, and Congruent with Inter-Sexual Dynamics and the Basis of Human Pair-Bonding


Intimate-partner violence [IPV] portrayed as a ‘gendered’ phenomenon of male perpetration and female victimisation is a now comprehensively discredited wholly ideological counter-factual invention, but the now accepted conceptualisation of ‘non-gendered’ does not fully take account of evidence that in no respect is there male predominance in perpetration; in all or the most important respects this is female.

Furthermore, converging lines of recent profound research point to IPV being in essence a phenomenon of specifically female perpetration, and this is congruent with new understanding of the basis of human pair-bonding and inter-sexual dynamics.

The misrepresentation of this reality as its inverse had been through a series of bogus interpretations and non-parsimonious constructs recycling the notion of a supposed patriarchal [sic] control in other guises.

This was in the service of attempts to hide the failure of Marxian ‘theory’ in a face-saving exercise, which being in line with natural prejudice towards males to prevent illicit sexual access accounts for the tenaciousness of its emotive fervency.

With the scientific research into IPV for many years now comprehensively showing that the ideologically-driven (feminist) understanding of the phenomenon is entirely fallacious in its positing exclusively or predominantly male perpetration / female victimhood (very many studies and many reviews, perhaps most influentially by Archer (2000), and more recently, eg, Dutton (2012)), IPV is now considered not ‘gendered’ [sic] (as in what has been dubbed the ‘gender-paradigm’ model) (Dutton and Nichols 2005), but ‘non-gendered’ [sic] or sex-symmetrical (eg, Koller 2013).

Yet this characterisation does not redress the hitherto extreme misrepresentation sufficient for an accurate summation of the literature.

The data and all lines of evidence (as is here outlined) converge on the conclusion that perpetration is greater by females, and in important respects is distinctively female; not merely that it is similar to that by males.

This is rather hidden by the notion and expression ‘non-gendered’ [sic], understandable though its use is to counter ‘in a nutshell’ the ideological stance.

Equal perpetration / victimisation by the sexes is not exactly or usually what researchers mean when they use the term ‘non-gendered’ [sic]: the typical stated position is that female IPV perpetration is ‘at least as much as’ or ‘similar if not greater than’ or ‘as much or more than’ male (eg, Fiebert 2010).

Although some or much raw data reveals only a low sex-differentiality or a near sex-symmetry, researchers are well aware of very high levels of male under-reporting of victimisation (see below).

‘Non-gendered’ [sic] seems to be at least in part a ‘fudge’: an attempt not to fall foul of ‘groupthink’ political imperatives and extreme-feminist rhetorical ire without being obliged to deny greater female perpetration.

Some key recent quantitative findings are illustrative:

 * Women are by multiples the more frequently violent partner in both mutual (reciprocal) IPV – twice as much as men — and unilateral (non-reciprocal) IPV – more than three times (and interestingly, there is no unilaterality of perpetration in homosexual couples) (Whitaker et al. 2007). This replicates previous studies/reviews of a sex-differential in unilateral IPV of at least twofold (Williams et al. 2001, Anderson 2002, Ehrensaft and Vivian 1999), rising with the seriousness of the violence to fully threefold (Stets and Straus 1990) or sixfold (Magdol et al. 1997).

* Women are three times more likely to hit a partner in a hypothetical provocation scenario (30% of the female participants in the study compared with 10% of the males). Furthermore, the great majority of women would expect no disapproval in doing so (96%, compared to just 24% of men) (Bartholomew et al. 2013).

* A review of all studies examining unidirectional IPV show that this is always more by women in all sample types bar police reports (Langhinrichsen-Rohling, Selwyn and Rohling 2012), which last are invariably highly biased against male reporting (and may include much female-perpetrated IPV mis-recorded as female victimisation): in an hypothetical IPV scenario, when the victim is male, observer participants (both male and female) view the situation as less serious, with the victim as more responsible, and are more likely to ignore the situation (Sylaska and Walters 2014); and in a similar study, women observers of IPV scenarios attribute perpetration to males much more than to females (Rhatigan, Stewart and Moore 2011).

As has long been known, male IPV victims typically do not consider themselves to be victims, and nor do men view an assault on them by a woman as a crime; and consequently fail to report victimization (Straus and Gelles 1992).

Other major reasons men may not report their own victimisation are, obviously, stigma and ridicule, and realising that they are unlikely to be believed, if not mistakenly arrested on the presumption of being not the victim but the perpetrator: in general men have objective good reason to fear gross unfair treatment in the criminal justice response at every turn (Shernock and Russell 2012).

Within the same couples, self-report of perpetration of IPV by the female is more than twice that indicated by self-report of victimisation by the male; whereas for perpetration in the other direction the reporting disparities reverse (Stets and Straus 1990).

Male ‘chivalrous’ attitudes and systematic pro-female / anti-male prejudice would seem to be underpinning much lower reporting to police by males (by a sex-differential factor of three according to the (UK) 2012/13 Crime Survey (ONS 2014), or up to ten (Stets and Straus 1990)).

* A comprehensive review of studies of the sex differential in rates of IPV perpetration shows pooled prevalence of female perpetration (28%) greater than that by males (21%) (Desmarais et al. 2012b); this sex-difference holding across variation in sample and study characteristics, and consistent with previous reviews. Furthermore, the sex-differential here is heavily under-measured in there being no weighting according to the severity of the violence (see above), and that it likely reflects the above-discussed prejudice towards males.

The corresponding data re victimisation (Desmarais et al. 2012a) is inconsistent even as to direction according to sample type and country, with a predominance of male victimisation in all studies investigating recent IPV (over just the previous year, when it would be more easily recalled); and the data is of much greater variation (right from 0% to 99%). All this reflects still more profound methodological issues in respect of data re victimisation compared to data re perpetration.

As well as hiding the predominance of female perpetration and male victimisation, the term ‘non-gendered’ [sic] reflects, just as it drives, the assumption that with IPV being female-on-male as well as male-on-female, then IPV has to be considered within general theory of aggression.

However, the quantitative findings re IPV of sex-difference call this into question, and even more so when account also is taken of factors which obscure what otherwise would be the large or very large actual size of these sex-differentials in the female-to-male direction.

Although these are not easy to quantify, the scale of it can be gauged by considering the disparity between predicted and recorded sex-differential rates of injury resulting from IPV. Here there is an enormous mismatch, indicating at least an order-of-magnitude preponderance of female IPV perpetration. There now follows an outline of how this is inferred.

Surprisingly, there is at most only slightly greater (less than a 3:2 ratio) female than male injury as a result of IPV (Archer 2000, Mirrlees-Black et al. 1998), a non-significant difference (Capaldi and Owen 2001), or parity (George 2003). Even more surprisingly, considering only serious injury, there are either substantially more male than female victims (Felson and Cares 2005) or males constitute over 40% of victims (Hoff 2012).

Note that the proportion of male injured would be considerably higher still if account was taken of the very high levels of under-reporting by males re IPV even in anonymous survey (here not ascribing any injury to IPV) – and in some types of study there is also the factor of pro-female sex-discriminatory recording methods (such as routinely in hospitals being pro-active in enquiring if injuries were IPV-related only with women).

Intuitively, this near sex-symmetry would not be even remotely expected, and objectively what is so unexpected about these figures is that they are despite the combined factors of female relative body-frame and facial-bone fragility, rendering a comparatively major susceptibility to injury; and much superior male upper-body strength (double the female (eg, Miller et al. 1993, Janssen et al. 2000)) and throwing velocity (with no overlap between the sexes – already an effect size of four by age twelve (Thomas and French 1985)), facilitating comparatively far more powerful hitting.

Many factors contribute to considerably greater male punching power, such as the male’s ability to shift weight (inhibited in females because of a different hip configuration); the extra mass and muscularity of the shoulder, arm, and the body overall; the greater performance of male muscle; shorter male neuro-muscular delay; different coordination of muscles in males; likely multiple sex differences in how muscles are supplied with oxygen and nutrients (producing more intense short-term muscle activation in males); all of these factors multiplying together to tend to push the sex-differential to several-fold.

Even if it were hypothesised that the sexes were equally responsible for IPV, as in the ‘non-gendered’ [sic] model, then it would be predicted that of all injuries sustained through IPV, 95% would be female (Dixon 2012).

In other words, female injury rates would be anticipated to be as much as twenty times – twice an order-of-magnitude – those for males.

If it were hypothesised instead that males perpetrate far more IPV than females, as in the discredited ‘gender-paradigm’ model, then this differential would be far greater still than 20:1 – a multiple of that ratio, to two orders of magnitude and beyond.

Therefore, with the enormous disparity between prediction on either model and the available data, then the only plausible inference is that IPV perpetration in reality is overwhelmingly by women.

Consequently, IPV may have to be considered a phenomenon essentially of female perpetration and male victimisation, with IPV in the other direction an aberration.

This would indicate a likely qualitative difference in underlying mechanism, with female IPV being IPV per se, and the relatively small amount of male seemingly equivalent behaviour being residual from a mode of violence with a different aetiology.

As regards the small subset of serious IPV that is spousal homicide: far from contradicting this possibility, the recorded crime data further suggest it.

Not only is there similarly almost no sex-differential – more than 40% of officially recorded spousal homicide victims are male (Ferguson 2003) – and, therefore, as with injury, many times below the 20:1 ratio that would be expected; but the data is well understood as failing to include the bulk of mariticide (husband-murder); this being largely undetectable.

Uxoricide (wife-murder) is typically overt, and often extremely so, with the husband also killing himself, and therefore unavoidably detectable. By contrast, mariticide typically is either indirect via third parties (by proxy at the hands of a lover, male friend, male relative or hired ‘hit-man’) or by subtle methods at some remove (classically, a poison which is not obvious in its effects and/or difficult to detect post-mortem, or a staged ‘accident’).

Either way, this leads to much or most mariticide being not thus recorded. This is additionally through pro-female / anti-male prejudice in respect of perpetration of violence, especially where a female is the victim. Instead, such murders necessarily will be recorded either as being both perpetrated and instigated by a male, or as not murders at all but death as the result of accident or sudden illness.

Both typical modes of mariticide reflect female indirect aggression styles, serving female goals to maintain the family otherwise intact (the mother wishing to stay with her children), complete with the murdered husband’s assets.

It is apparent, then, that a new theoretical understanding may be needed. In other words, with IPV ‘gendered’ [sic] in the female-on-male direction and in at least some key respects overwhelmingly so, then an aetiology (cause of condition) may be required in terms of the sexes and how they interact – but one that this time is scientific, not ideological – rather than considering the phenomenon as having the same aetiology as other forms of aggression.

IPV data may be best regarded as confounded with factors extra to IPV per se, that if removed would reveal not merely sex-difference but sex-dichotomy: that IPV per se – IPV specifically directed against the partner and with intent to cause significant harm – is in essence a female-perpetrated phenomenon, leaving the minority data showing male perpetration in effect as artefactual, bar that by psychopathic or seriously mentally ill / seriously personality-disordered males.

Self-evidently this clearly is IPV, but it is by aberrant – non-normal – males.

Considering normal males, a major confound (though still awaiting research) is behaviour by males which though included as IPV actually is merely the attempt to restrain IPV by the female partner; that is, male behaviour which is not retaliation but likely to be misconstrued as such (if not deliberately misrepresented as unilateral violence).

Of what would then remain of male-perpetrated IPV, conceivably all may be collateral from the very high level of male intra-sexual violence (so very much greater than female-female violence): by displacement rather than directed and intentional (and presumably mainly alcohol-related).

This is not to say that this is not male-perpetrated IPV, but that it is de facto rather than as a result of a specifically IPV aetiology.

IPV conceived of as essentially a female-perpetrated phenomenon is supported by several compelling converging new lines of evidence, making this a cogent conceptualisation worthy of advancing as a hypothesis.

These are:

(1) the discovery of a completely different pattern of perpetrating violence in a couple context by women, revealing a female-specific implicit motivational basis;

(2) neurophysiological research uncovering a likely cross-species evolutionarily high-conserved male-specific bespoke neural pathway producing profound self-inhibition in aggression in response to close physical interaction with a female;

(3) striking data about the effect of oxytocin to greatly increase violent behaviour by women in a couple context compared to men;

(4) new findings reinforcing old that ‘control’ in couples is not sex-symmetric but mainly by the female partner; and

(5) a strongly inferred entrenched absence of risk to females of male aggression back into the evolutionary past from consideration of the content of women’s delusionary ideation.

These are here outlined in turn.

(1) Re motivation: it had long been popularly supposed that males ‘hold back’ when aggressing towards females, but this had not been researched until a decade ago, when it was found that men – but not women — were self-inhibited from being violent towards their partners though uninhibited towards strangers (Felson, Ackerman and Yeon 2003).

An even more striking sex-dichotomy has since emerged: that men indeed are self-inhibited from using physical violence where a female would be the target; but women, not merely uninhibited, actively prefer physical violence as the mode of aggression in an intimate-partner context.

This was found in studies using, first, hypothetical scenarios (Cross, Tee and Campbell 2011) and then self-report (Cross and Campbell 2012).

Building on these findings, subsequently it has been shown that the sexes are a mirror-image, so that whereas men are far less violent to intimate partners than to other men, females are more violent to intimate-partners than to other women (and a similar contrast according to sex applies to verbal aggression), with three times as many women as men perpetrating IPV in the absence of showing violence to same-sex non-intimates (Bates, Graham-Kevan and Archer 2014).

That the disparity here between the sexes is so great is reflected in the data re inter-sexual violence as a whole.

Surprisingly, although females appear not to actively choose to physically aggress against males in other (than intimate-partner) contexts, and males in general are by far the more violent sex; nevertheless the sex-differential in the perpetration of inter-sexual violence overall is threefold female:male (Morse 1995).

Another revealing comparison is in review of the data from same-sex couples: that IPV rates are significantly higher, with good evidence that rates are highest of all in not male but female same-sex couples (West 2012).

The data entirely contradicts the ‘gender paradigm’ model (Canon and Buttell in press), and has prompted a book specifically on the problem (Kaschak 2002).

IPV in lesbian couples was earlier researched (when it was less taboo to do so, and therefore there was less imperative to disguise and misrepresent data) as from two to three times the rates for heterosexuals (Coleman 1990, Bologna, Waterman and Dawson 1987, Lie et al. 1991).

And this differential is regarding incidence – multiplicity of incidents per relationship — as well as prevalence, with three-quarters of lesbians in couples experiencing six or more incidents; half, ten or more (Renzetti 1992).

This shows that female disinhibition from being violent within a couple context is not according to the sex of the partner (that the other partner is male), but because of the context of an intimate partnership.

That the rate of IPV in ‘gay’ couples is intermediate between those for heterosexuals and lesbians shows that males cease to be inhibited from perpetrating violence when the partner is not female.

It appears, then, that for both sexes in same-sex couples, the usual intra-sexual aggressiveness is apparent, but whereas this is an additional factor for women, it seems for men it is the only explanation required. In other words, for male same-sex couples, any violence is similar to that between males in other contexts, whereas this is not the case for females.

From these findings re motivation alone — even without considering other lines of evidence — it seems that IPV is a female-specific form of perpetrated violence, whereas for men not only is violence against intimate-partners not a form distinguishable from other forms of violence, but in the heterosexual case it is special only in its being usually avoided or at least diluted for the same reason males avoid violence in other scenarios where what would be in common with a couple situation – a female target — is salient.

This suggests an aetiology specific to IPV for female perpetration, and by contrast an understanding of male perpetration as being residual from parallel forms of violence. Thus is explained the highly unexpected findings regarding injury in IPV: a result of a combination of females being especially ready in this particular context to resort to violence, whereas males very much back off from using it.

(2) Re a male-specific self-inhibition of violence: what is strongly suspected to be an evolutionarily well-conserved three-level neural pathway common across species to include humans, has been discovered which produces a general near elimination of aggression, triggered by close physical contact with a female (Yuan et al. 2014).

The pathway features male-specific ‘GABA-ergic’ neurons, with male-specificity applying to all three levels of the mechanism.

For the function to be engaged, mere close proximity is insufficient, but the close physical contact required does not have to be copulation; and the mechanism still works even if the male’s courtship has been rejected. The neural pathway is a dedicated one, not involving previously identified machineries for learning and memory, so it appears to be a specific major adaptation.

This is a particularly profound discovery, of what seems clearly to be a mechanism functioning to obviate the risk of displacement from male-male violence impacting on a female sexual partner. The series of studies are meticulous in controlling for any conceivable confounds, and though the research is not in a mammal but an insect model (as usual, drosophila), this is a standard mode of investigation of principal, evolutionarily highly-conserved neural pathways; the purpose of the research team being explicitly to gain understanding of human aggression.

(3) Re oxytocin: it is now found that this hormone, the very one underpinning pair-bonding, prompts women (though not, or much less so, men) to perpetrate IPV. And this is notwithstanding that the effect depends on high trait aggressiveness (which generally is a quality of males much more than of females).

Oxytocin in males has (or mainly has) a different function in respect of aggression: it underpins the form directed against an out-group (DeWall et al. 2014). ‘Gender’ [sic] priming produces a much stronger effect than priming for attachment. All this indicates or strongly suggests a neuro-hormonal pathway underpinning IPV perpetration specific to women.

Animal research demonstrates that oxytocin underpins maternal aggression, which, in being the phylogentically-ancient key form of female violence appears to be homologous to IPV perpetration. Maternal aggression has a completely different neuro-hormonal basis than intra-sexual aggression, and is fearless (Gammie et al. 2008), just as female-perpetrated IPV is fearless in being actively preferred over employing other modes of aggression.

With no male equivalent of maternal aggression, then here is a neat explanation for why there would not be a male-equivalent aetiology specific to IPV perpetration: why it would be mistaken to consider the much lower level of male-on-female violence within a couple context to be the same or a similar phenomenon to female-on-male violence in this same context.

Evolutionary theorists have posited that oxytocin is associated with mate-retention behaviour, and maternal aggression in defence of offspring obviously is not far removed functionally from female aggression in defence of the means of producing offspring (that is, attempting to retain the male partner). Its co-option in the evolutionary process would be a simple, minor instance of this common sort of development.

(4) Re ‘control’: women themselves report perpetrating significantly more of such behaviour than do men (Bates, Graham-Kevan and Archer 2014), echoing previous findings that women are typically the ‘controlling’ partner (Vogel et al. 2007), and that the female is thus in 90% of couples (Coleman and Strauss 1986).

It is a well-researched finding that such ‘control’ in some circumstances manifests as physical violence (eg, Dasgupta 1999, Felson and Outlaw 2007).

The Bates, Graham-Kevan and Archer study reveals a telling contradiction between self-reported perpetration and victimisation: whereas there was a considerable sex differential in respect of self-reported perpetration, self-reported victimisation was sex-symmetric.

This is likely due to a combination of males being far less likely than females to report any victimisation they sustain (see above) and ‘control’ being understood only in how it typically manifests as practised by males.

What is measured in studies is the occurrence of male forms of ‘control’ irrespective of whether this was by the male or the female partner, thereby excluding any forms that are specifically female, as would be expected to be exhibited by females (and not by males).

Some studies explicitly measure ‘control’ in this way (eg, Graham-Kevan and Archer 2009), and the suspicion must be that most other studies do so implicitly. This would explain why several studies hitherto have contradicted other research in reporting a sex-symmetry in ‘control’: such a finding is likely to be artefactual.

(5) Re female persecutory delusional ideation: with the content of nightmares and the persecutory delusions of the mentally ill providing strong insight into what people have long evolved to fear; then, if IPV was ever a salient risk for women, they would be expected to feature violent men.

Yet research into nightmares fails to find this (Schredl 2010); ditto in respect of the persecutory delusions experienced by women mental health patients.

Instead of threatening males (either familiar or strangers), familiar females feature as ogres (Walston, David and Charlton 1998).

Male violence towards women – IPV or otherwise – apparently was not a significant phenomenon throughout evolutionary history, indicating that it is not part of (normal) male nature.

Taken together, these several lines of evidence converge to strongly indicate that IPV is a phenomenon essentially of female perpetration rather than being ‘non-gendered’ [sic]; that is, it is ‘gendered’ [sic] but in the diametrically opposite way to that envisaged in the IPV ‘gender paradigm’ of ideological imagination.

There is no difficulty explaining why IPV essentially would be a phenomenon of female perpetration, nor what underpins male reluctance to physically aggress against a sexual partner, and female lack of inhibition from doing so — if not, indeed a preference to aggress inter-sexually.

On a fundamental biological level of analysis, the female is the limiting factor in reproduction, and therefore within the reproductive group it would be expected that there would evolve mechanism to prevent violence against females; notably male-on-female violence, because of the upper-body strength of males and the body-frame weakness of females.

[Note that this does not at all imply ‘group’ / ‘multi-level’ selection: apparently group-level phenomena, such as dominance hierarchy, can now be explained in a number of mathematically equivalent models, though how to decide between them is meta to the present discussion.]

By contrast, the male in effect is expendable, so no corresponding evolution would be anticipated re males.

In any case, males have evolved to deal with the violence from other males, and females would have little to fear in the way of male retaliation given male inhibition to express violence towards the female.

Through the distinct and complementary functions of the sexes, they do not cross-compete: inter-sexual conflict in the form of dominance interaction does not occur — dominance hierarchy is always same-sex.

Consequently, again, inter-sexual violence is unlikely. [Examination of supposed instances of species exhibiting female dominance shows that it is male deference – signalled non-engagement in dominance interaction – in female feeding priority.] (Moxon 2009, 2012).

The imperative of non-violence towards the female would apply most obviously within the context of the pair-bond, where the risk violence would pose to female fecundity is particularly self-evident, and the male partner has a major direct interest in the female partner’s fertility.

Even so, this imperative applies throughout the reproductive group in any context; hence the finding (see above) that males are self-inhibited from physically aggressing where a female would be a target, whether or not the female is an intimate-parter – it would apply also to a scenario involving a female stranger in the street.

Note that although this self-inhibition by males in turn explains why females are not thus self-inhibited, it does not explain why females would prefer violence as the mode of aggression specifically within the pair-bond context.

The proximal basis of this is now known in that (as outlined above) oxytocin, the hormone underpinning pair-bonding, promotes physical violence for mate-retention purposes most especially in women, despite it being dependent on high trait aggressiveness, which generally is far greater for men.

This indicates that this is likely a female-specific mechanism. As for not the proximal but the ultimate basis of the phenomenon, insight comes from the nature of human pair-bonding; and therefore what follows is an exposition and how this informs an understanding of IPV.

It has recently become clear that human pair-bonding (which is an evolved human universal (Fisher 1994, 1989 – and see Saxon 2012 and Ellsworth 2011 as ripostes to the recent counter-claim of Ryan and Jetha 2011)) evolved in the female interest.

This is apparent in the much greater importance to women of pair-bonding, revealed by their much greater investment in it than by men. Women desire to commence pair-bonding several years earlier and to maintain a pair-bond for twice as long as do men (on average for fourteen years as against seven) (Palchykov et al. 2012).

They seek close relationship generally and in particular intimate partnership much more than do men (Schmitt et al. 2003), and are twice as likely as men to develop the extremely persistent pathological courtship behaviour known as de Clérambault’s Syndrome (Brune 2001) — the attempt to insist on pair-bonding irrespective of whatever resistance there may be by the individual targeted.

Women are much more concerned than are men by the prospect of partner infidelity (Shattuck et al. 2012), which is the opposite of what would be predicted from the consideration that only the male partner is at risk of being reproductively usurped through the importation of a replacement for his genetic complement in offspring from another individual outside of the pair-bond (this being, then, an objective male fear without a parallel for women).

The female concern, as evidenced by what in particular arouses feelings of jealousy in women, is not with extra-pair sex but an emotional attachment with another woman (eg, Sagarin et al. 2012), because it is this rather than sex which is likely to herald the male partner’s defection from the pair-bond.

Comprehensive review work has shown that pair-bonding evolved neither as male proprietorial control of female fertility nor to facilitate male provisioning of partner/offspring (Chapais 2008, 2011; Winking 2007, Moxon 2013), but as a means of maximising / optimising female fertility (Winking 2007), and through mate-guarding (Chapais 2008); but just what form of mate-guarding had been unclear.

Mate-guarding had been thought to be simply male proprietorial control of female reproduction to ensure confidence in paternity, but this is contradicted by findings that in many ostensibly ‘monogamous’ species there is a high incidence of extra-pair paternity (eg, Colombelli-Négrel et al. 2009, Le�alová-Piálková 2010), with humans no exception (variably estimated, with a median worldwide of about 10% (Baker and Bellis 1995)); and tellingly this applies even where it might least be expected, in ‘traditional’ community (Scelza 2011).

According to a review across species (Manson 1997), whereas the courtship element of pair-bonding is predictive of reproductive outcome, this is not true of the element of mate-guarding. The fact that despite mate-guarding, females easily obtain extra-pair sex of their choice has prompted a challenge to the conceptualisation of mate-guarding as male proprietorial control (Kokko and Morrell 2005).

As Dunbar (2012) points out, the evidence shows that it should be envisaged in terms of a ‘body-guard’ (‘hired gun’) to assist the female in keeping at bay undesirable males. This previously has been suggested as the basis of primate pair-bonding generally (Norscia and Borgognini-Tarli 2008); and deterrence of unwanted suitors has been recognised (Lumpkin 1983), but assumed to be a development of male mate-guarding as usually understood, through its exploitation by females.

However, the ‘body-guard’ (‘hired gun’) understanding is to envisage mate-guarding as a defence against sexual aggression – serious harassment and/or assault – but ancestrally (well back down not just the hominid but the ape phylogeny) behaviour of that kind would be strongly sanctioned within the social group, and such sanctioning must have been internalised in males over evolutionary time to implicitly inhibit males from sexually aggressing in the first place; and, therefore, a female hardly requires a pair-bond partner to dissuade other males from sexual aggression.

Simple (non-aggressive, non-mating) excessive male attentiveness is another matter, and ‘body-guarding’ to reduce this would have clear benefits. Simple male (over-)attentiveness is recognised to have costs for attractive females even in an insect model (Partridge and Fowler 1990, Long et al. 2009).

In higher animals, self-evidently, displacing social and sexual attentiveness by low mate-value males would not merely eliminate a nuisance but in effect facilitate access by / to males of still higher mate-value (higher than that possessed by the ‘body-guarding’ male himself); who (potentially) would be desirable to the female as extra-pair sex partners.

Ancestrally, the ‘body-guarding’ male would not have been in a position to thwart a male of higher rank than himself (because of the risk of escalation to agonistic contest, which surely he would lose), and, therefore, usual dominance interaction would ensue – the lower-ranking male would not mount a challenge – and, in being a recurring dynamic, this would become fixed as evolved implicit psychology. This is the standard machination of male dominance hierarchy, which cannot but be foundational to human male psychology.

The upshot is that mate-guarding by males functions neither in the male’s interest to ensure paternity nor in the female’s interest to keep all males at bay, but in the female’s interest to selectively keep at bay low-mate-value males whilst thereby in effect actually facilitating extra-pair sex with high-mate-value males; this arrangement being much more in the female’s interest than a blanket exclusion of all (other) males, as would be the case with former conceptualisations of mate-guarding.

Pair-bonding and extra-pair sex are not only fully complementary from a female perspective, but seemingly are together inherent in how pair-bonding functions, allowing the female the flexibility to maximise her overall fertility according to circumstance.

Note that this complements the evolution in humans of hidden oestrus, which in preventing males from assessing when a female is in her monthly six-day fertile window, renders much less worthwhile time-consuming if not risky (costly) efforts at sexual access by low-mate-value males, but leaves the female free to be proceptive and/or receptive to high-mate-value males during her six-day fertile window, as well as to have sex with her pair-bond partner. [The former notion that hidden oestrus evolved to entice a male to pair-bond (so as to get around the problem of never knowing when the female is fertile) seems to be a misconception (Moxon 2013).]

There is a further and possibly more important way in which pair-bonding benefits the female in maximising fertility: it allows her, in effect, to project forwards in time the peak in her mate-value, thereby circumventing the major problem that female mate-value declines rapidly with age (and the impact of childbirth).

A female at or close to her peak in fertility in her late teens or early twenties is at an age when correspondingly she is at her most attractive, and therefore best placed to secure a high-mate-value male. By retaining him for subsequent conceptions she can produce collectively higher quality offspring than through the alternative of promiscuous sex, where the quality of the male genes she could obtain likely would decline with each successive conception as she assortatively mates with males of mate-values corresponding with her own rapidly falling mate-value.

Given that males, by contrast, typically rise in mate-value with age (as their intra-sexual competitiveness is more fully realised in status, and thereby is revealed the true quality of their genetic complement); then in effect the female also brings forward, as it were, the peak mate-value of the male she chooses.

She thus retains for herself as an exclusive sexual partner a male who would become more and more attractive to her still more fertile rivals; such that if, instead of forming early a pair-bond with that particular male, she were instead later to enter the mating market to try to acquire this same male, then she would lose out to these rivals and never acquire him.

It is conceivable that the implications of this ‘sex-dichotomous mate-value trajectory’ may be as or more important a basis of a benefit to the female of pair-bonding than is mate-guarding.

The male can exploit the female benefits from pair-bonding only through being able to offer it in exchange for higher fertility than he could expect to obtain through promiscuous sex – in other words, he acquires regular sex with the same attractive female, rather than more uncertain, occasional opportunities for sex with various less attractive – less fertile – females (which in any case he can continue to obtain as extra-pair sex).

This indeed is a benefit, though an extension of the costs males incur in obtaining sex.

It’s an additional dimension of ‘good genes’ the male is required to exhibit in order to sufficiently impress a prospective female partner, which is hardly as clearly a benefit as those accruing to the female.

With the pair-bond being of much more importance to the female than to the male partner, then it would be expected that the female would be keener than the male to exert ‘control’ over both the relationship as a whole and the other partner; which is what is found (see above).

Mate-guarding as formerly conceived actually is how the female behaves; not the male. In other words, the female is the sex which mate-guards in the classical sense: the sexes both mate-guard but in different ways. The male protects his mate from the attentions of lowly males, whereas the female tries to prevent her mate from straying.

Although extra-pair sex by the female would be a risk factor for her of IPV, it is much less so than is usually supposed, in that with paternity-certainty not the root of pair-bonding, then extra-pair sex by the female simply provokes the male to abandon the pair-bond because of the high risk that the extra-pair sex had resulted in conception; which, of course, for the male would render subsequent sex with this particular female a waste of time.

The male has not sustained any cost, and would sustain an opportunity-cost only if he remained within the pair-bond. It is simply that he should henceforth engage in either promiscuous sex and/or sex within a replacement pair-bond. And here is a potential actual benefit for the male in the dissolution of the pair-bond, in that with the typically rising status (mate-value) of a male with age, likely he could obtain as a replacement pair-bond partner a still more fertile female than his current (now ex-) partner.

The reverse is the case for the female, who would have difficulty replacing her partner – certainly with a male of an equivalent mate-value – given her declining mate-value with both age and the impacts of childbirth. In any case, she would have especial difficulty at first, owing to her likely pregnancy, with ensuing gestation and lactation, throughout all of which she would not be fertile, making it pointless for any male to pair-bond with her, even without the consideration of the imminent arrival of another man’s child.

All this serves further to make the female keen to retain her pair-bond partner.

Given the surprising lack of costs to the male in dissolving a pair-bond, together with the ferocity of male intra-sexual competitiveness, then the ire a male feels in being cuckolded will be principally directed towards the cuckolding rival male, for making the ultimate transgression in male-male competition.

Rather than the (ex-) partner, it is the cuckolding rival male whom the male will tend to blame for initiating the extra-pair sex. Consequently, the partner is at risk at most of merely displacement aggression (and even then usually a form of aggression other than physical violence).

From this fuller understanding of pair-bonding, sense can be made of the typical pattern within couples of female infidelity leading to the departure of the male (simply walking out) whilst male infidelity leads to the female remaining to ‘stand by her man’.

The male usually takes himself out of the relationship and its domestic setting, thereby greatly reducing what likelihood there was of his physically aggressing against his (now ex-) partner; but by contrast the female most often stays within the relationship and its domestic setting, thus maintaining the close proximity that may precipitate into IPV the anger she very likely feels against her wayward partner.

Couched within a wider understanding of human inter-sexual interaction, then, it becomes clearer still that IPV is best considered in essence a female-perpetrated phenomenon, and that the notion of IPV as ‘non-gendered’ [sic] is a far too conservative position which had been adopted as taking a ‘moderate’ stance against the bogus ‘gendered’ [sic] notion of feminist ideology.

[There is the question of the degree to which this would be confirmed cross-culturally, but comparison here is difficult if not (on current data) impossible, because in more ‘traditional’, family-centred and communal societies, much of what might otherwise be female-perpetrated violence is likely to be carried out on behalf of (whether or not instigated by) the female partner by third parties; this not being recorded as IPV despite being equivalent to IPV in a ‘Western’ nuclear-family private domestic setting.

This is akin to the problem with data in ‘Western’ countries re spousal homicide (as discussed above). For this reason there is no review here of IPV literature from non-‘Western’ cultures.]

With there being nothing remaining of the ideological posture that withstands any scrutiny, it is time to more accurately discuss IPV being ‘gendered’ [sic] as a female-on-male phenomenon without regard to political sensibilities and in the interests of a proper scientific understanding not just of IPV but more widely of how the sexes interact.


It would be instructive to consider the development of the now discredited ideological ‘theory’ of IPV, as it still persists in the advocacy-research lobby, to see how it was obliged to modify when confronted by evidence from scientific research.

‘Theory’ aside, the crude tactics employed by the ideologues have been discussed by Straus (2007) and added to by Graham-Kevan (2007) as eight categories.

They are to:

(1) suppress evidence,

(2) avoid obtaining data inconsistent with the ‘patriarchal dominance theory’,

(3) cite only studies that show male perpetration,

(4) conclude that results support feminist beliefs when they do not,

(5) create “evidence” by citation,

(6) obstruct publication of articles and obstruct funding research that might contradict the idea that male dominance is the cause of IPV,

(7) harass, threaten, and penalize researchers who produce evidence that contradicts feminist beliefs, and

(8) play with numbers (misuse of statistics).

These tactics are evident most obviously in restricting studies to utilising only the inevitably highly skewed clinical samples of women in IPV ‘shelters’, ignoring in particular the actual research examining IPV in both the male-on-female and female-on-male directions simultaneously in the same sample.

The use of these common ruses the ideologues have attempted to justify by spuriously eschewing empirical study as being supposedly “act based” (see Dutton 2007); refusing to accept that proper measurement of violence is admissible, on the ideological grounds that by (their) definition violence against women is somehow qualitatively different from that towards men.

In the ideologically-based view, males essentially are perpetrating ‘patriarchal’ [sic] ‘power’ when they employ IPV, whereas females are not; but there is no cogent theory of ‘patriarchy’ [sic], nor of a ‘power’ [sic] relation between the sexes (Moxon 2012): merely a bizarre distortion of Marxist ‘theory’ (see below).

The ideological IPV ‘theory’ is not a scientific theory in its being unfalsifiable: the premise dictates all interpretation of data in a tautological loop.

‘Theory’ developed as an unedifying staged retreat as ever more implausible and unparsimonious defences were erected to be knocked down in turn as research directly addressed them. This has been previously outlined (Moxon 2011) and here is revisited in an updated and more concise review (not least through some of the relevant evidence already having been discussed above).

There was an obligation to develop the ‘theory’ when the most common ruse, of restricting data to crime statistics and surveys, hardly could be sustained given the very high levels of male under-reporting in crime or victimisation survey being the most well-known feature of IPV data (see above).

The explanation of disproportionately male under-reporting is not difficult to find.

Signalling victim status in any context would seriously compromise the key male need to show male intra-sexual competitive ability in order to acquire status, and thereby sex. To then do so outside even the context of male intra-sexual competition would be particularly compromising for a male; literally to render him an object of derision.

Additionally, that from a civic perspective a couple is regarded as being headed by the male partner, this entails men viewing violence by their partner in effect as a failure of their own making.

It is impossible to eliminate from research the confound that these considerations introduce. Even if the relevant ‘demand characteristics’ are removed from crime and personal safety surveys, comparatively greater male under-reporting of their victimisation remains (Archer 1999).

Notwithstanding this, the annual crime figures for the UK collected by the Home Office consistently reveal males constituting four in ten of IPV victims (eg, Thomson 2010).

As discussed above, this sort of statistic of near parity between the sexes can only conceal a very large preponderance of female perpetration and male victimisation when account is taken of the distortion in the statistics of male under-reporting by multiples of the female level.

With the persistent exposure of these elementary ruses, the IPV activists continued their misrepresentation by a focus on the misleading headline statistic of a sex-differential injury rate (discussed above), and to deem all data as somehow beside the point in the character of IPV supposedly being different according to the perpetrator’s sex: that women merely retaliate. This was more than refuted by the fact that IPV is fairly evenly split between mutual and unilateral manifestation, and that women are far more likely to initiate (see above).

In reaction to this contradiction of the bogus ideological model, female perpetration in mutual IPV was deemed self-defence, but this in turn was contradicted by research showing not only that self-defence is cited by men as often as it is by women (Carrado et al. 1996, Harned 2001), but that it is cited by only a minority of women (Foo and Margolin 1995, Sommer 1994). There is no evidence of self-defence by women known to be IPV perpetrators in the testimony of the women’s own relatives – notably their mothers (Sarantakos 2004).

These ideological implausible sex-dichotomisations at last yielded to conceding that women indeed are perpetrators, but a new ruse was introduced that it was typologically distinct: supposedly, women are ‘expressive’ in their perpetration of violence rather than ‘instrumental’; this of course being merely a re-hash of the rhetoric about ‘patriarchal’ [sic] ‘power’ [sic].

No sooner had this new attempt to defend the feminist ‘gender paradigm’ model been erected than it fell with the studies revealing women to have ‘instrumental’ beliefs at the root of their IPV no less than do men (Archer and Haigh 1999). Furthermore, much ‘instrumental’ violence by women is not apparent in data, in that women pass it off as being by their victim (Lewis and Sarantakos 2001); this being easy through misrepresenting male attempts to restrain a partner’s IPV as itself IPV.

The activist community has even gone so far as ostensibly to concede a sex-symmetry of perpetration / victimisation, in the typology of IPV by Johnson (Johnson 1995, 2005).

This is superficial, however. The sex-symmetry is admitted only in respect of what Johnson regards as a minority and non-serious form of IPV: “situational couple violence”, which he identifies as ‘non-patriarchal’ [sic]. The bulk of IPV Johnson does, of course, regard as ‘patriarchal’ [sic]: a supposed unilateral male “intimate terrorism”, to which females respond with merely “violent resistance”.

This schema is disingenuous, to say the least, in that the data was already well in place to dismiss it (see above), and it has now been specifically debunked by studies showing that women perpetrate what would be considered in Johnson’s terms “intimate terrorism”, and correspondingly men respond with “violent resistance” (eg, Hines and Douglas 2010).

It’s yet another re-statement of tired ideological notions in a slightly new form, albeit actually the most extreme to date in that aggression is considered to be serious according only to whether or not it is ‘patriarchal oppression’ [sic].

Absurdly, this leads to the assertion that even violence causing major injury is deemed to have less impact than mere verbal assertion of ‘patriarchal’ [sic] norms. Even by the standards of the ‘gender paradigm’ ideological stance, this is a new bid to render it completely data proof. The posture is in vain, of course, for the self-evident reason that no-one outside of the extreme ideological ‘groupthink’ would ever accept the notion that verbal assertion supposedly of ‘patriarchal’ [sic] norms is in any way equivalent to inflicting serious physical injury, let alone trumping it.

There has been a further bid for unfalsifiability in the attempt to replace objective measurement by subjective experience, in a consideration of IPV in terms of the fear experienced by actual or potential victims.

This is an unjustifiable pathologisation of female but not male reaction to IPV, made in wilful ignorance of there being little if any basis for females to fear violence from males (as discussed above).

The claim is that given an equivalent act of IPV, the consequence for a female is subjectively far more severe, but the experience of fear is measurable empirically, not least as clinical-psychological sequelae; and the data shows that anxiety, depression and PTSD result from IPV for males just as for females.

There are no differences according to sex (Randle and Graham 2011, Hines and Douglas 2011, Pimlott-Kibjak and Cortina 2003; Coker, Davis and Arias 2002), even with a full set of controls (Callahan, Tolman and Saunders 2003).

The ‘fear’ ploy appears superficially plausible because it keys into the inter-sexual dynamic of the female signaling her vulnerability in order to elicit male response; this stemming, once again, from the evolutionary logic of the female being the limiting factor in reproduction.

Very likely it also relates to an actual courtship ploy of female signaling (in a way unconnected to actual dominance behaviour) ‘low-dominance’, corresponding to the male signaling (likewise in a way unconnected to actual dominance behaviour) ‘high-dominance’, thereby enabling the male to faux-exhibit his dominance in a feedback loop with the female to produce a courtship routine. [This is an evolved co-opting, apparent in many species, of merely the signaling element of male-male behaviour for the very different context of courtship, where male dominance is relevant in indicating male mate-value. This is behind the apparent paradox that is the subject of the book Fifty Shades of Gray (and the film, Secretary), which feminists deplore for what they consider to be a portrayal of female complicity in male violence, when in fact there is not violence but merely symbolic ‘disembodied’ dominance for erotic ends.]

While it might be popularly supposed that females by their nature express greater fearfulness and/or to be more fearful, from the considerations above of the actual very low risks to females of violence, and the evidence from female delusional ideation, then it should be no surprise that an especial female fear is not borne out by the findings regarding fear of crime.

Many women are non-fearful of crime, and many men the opposite, such that overall there is a near absence of any sex difference (Gilchrist et al. 1989), with the small sex differences that there are revealed to be artefacts of social pressure (Sutton and Farrall 2005). The same almost-nil or nil sex-difference is evident throughout childhood and adolescence (Maccoby and Jacklin 1974, Else-Quest et al. 2006).

Fear specifically re IPV has been found to be as significant for men as for women (Hines, Brown and Dunning 2007). The hitherto assumed absence of fear in men is shown to be profoundly false: instead it is present to the extent that it can precipitate psychosomatic symptoms, and manifests as various fears; specifically a sharp sense of impotence to protect their children and a high vulnerability to being separated from them (and from their family as a whole, which would take away what is often for men their sole source of emotional support) (Lewis and Sarantokos 2001).

It remains to briefly consider, following the exposition of how the reality of IPV has been persistently so completely denied and misrepresented, quite why this was so; not least because to understand the why would provide more confidence in the outline of the how.

The extreme anti-scientific ideological posturing requires an explanation as to the motivation behind the tenacity with which it is held. This previously has been given a brief account (Moxon 2011, 2008), and here follows a fuller outline.

IPV & Marxism

Commissioned as a book chapter in Partner Violence: Risk Factors, Therapeutic Interventions and Psychological Impact. Nova Science Publishing 2014.

Author, Steve Moxon, granted permission to publish:
One of the 2014 papers was commissioned but not actually published, so you can have that one:
[It’s on my site:
“Intimate-Partner Violence Is not Merely Non-‘Gendered’
Steve Moxon, researcher & writer re the biological roots of human sociality, with especial interest in the sexes; and PC-fascism.