332 Child Homicides

This is a minor revision to my earlier post “330 Child Homicides” published in October 2016. A revision under the revised title “332 Child Homicides” was first published in February 2016, motivated by the discovery of an additional serious case review (SCR) from 2009 which appeared in my rough notes but did not get transcribed to the original post. This additional SCR is discussed in a new Section 10.2, and all statistics herein have been updated accordingly (though the changes are very slight).

This latest revision, being published in July 2017, does not include any further additional SCRs, so all statistics are unchanged from the February 2017 version. This further revision is to bring out in the text all identified cases of spite killings by mothers in the period considered (SRCs issued 2009 to 2015). A case which was included in the original statistics, but not specifically identified as a spite killing, is now discussed in a new Section 10.3. 

This brings the number of identified cases of spite killings, or ‘family annihilations’, by mothers to 6 (involving 10 children’s deaths) in the period 2009 to 2015, two of which occurred whilst the mother was ‘on contact’ – see Sections 9, 10 and 11.1. 

In addition I have also brought out in the text, in new Sections 10.4 to 10.8, three cases of murder-suicide by mothers, plus two cases of murder and attempted suicide by mothers (in total involving 7 child killings).

Combining spite killings by mothers with murder-suicide cases by mothers, I have now identified 9 such cases, involving 15 child killings, in the 7 year period studied. The significance of this is the comparison with Women’s Aid’s “19 Child Homicides” report which, in the case of killings by fathers, identified 12 cases with 19 child deaths over a ten year period. 

Rather than edit the original post, which has been widely copied elsewhere, I have opted for a re-post. Changes from the original are in blue text (mostly Sections 10.2 to 10.8 and additional conclusions). 

I take this opportunity to reiterate the purpose of this, and the earlier, post. The purpose is emphatically not to reverse the ‘reasoning’ of Women’s Aid’s 19 Child Homicides report, by swapping the genders and proposing that mothers are unfit parents. This would be as preposterous as the reverse. Rather the purpose is to expose the fraudulence of the ‘argument’ presented in 19 Child Homicides. It is fraudulent for two distinct reasons. Firstly the cherry-picking of the 12 cases presents a grossly (and deliberately) unbalanced picture. The more thorough examination of all serious case reviews involving child killing over a 7 year period, presented below, is explicitly for the purpose of restoring balance. However, there is another, more fundamental, reason why 19 Child Homicides is fraudulent which applies irrespective of whether the numbers are 19 or 332. This is the implicit assumption that these appalling cases of human misery have anything to tell us about the vastly greater number of typical private law cases in the family courts. They do not. They are exceptional – exceptionally awful, that is. They are no basis for deciding action in the vast bulk of family law cases, contrary to the claims of Women’s Aid’s “Child First” lobbying. 

**********************

In the 7 years 2009 to 2015 there were 332 children culpably killed in the UK whose deaths have been subject to Serious Case Reviews by the child protection authorities (excluding suicides). Below I present a review of information gleaned from these Serious Case Reviews (SCRs). This serves two purposes. Firstly, as I and others have noted elsewhere, identification of the perpetrators of child homicides in the UK is subject to considerable obfuscation. This review of the SCRs serves to provide some information on perpetration by mothers, fathers, the mother’s partner, and other people.

The second purpose, and the initial motivation for this review, is to provide a response to the recent Women’s Aid report “Child First: Nineteen Child Homicides“. This report is discussed further in the text below. In brief it trawled 10 years of SCR data in order to identify a carefully selected dozen cases of men killing their children whilst on ‘contact’ visits. These carefully selected cases are used in Nineteen Child Homicides to spuriously justify calls for further restrictions on fathers being granted contact with their children. This Women’s Aid report is staggeringly dishonest. This statement is justified by the findings of the more complete review of SCRs presented below.

Women’s Aid are dedicated to promulgating a partisan account. For example, just two months before the publication of Nineteen Child Homicides a mother who was separating from her partner stabbed to death her two young daughters whilst resident in one of their refuges (see Section 10, Samira Lupidi, below). There was no mention of this in Nineteen Child Homicides. Nor was there any mention of the two cases (identified in the review below) of mothers killing their children while they were on ‘contact’ visits, the father having been granted residency – the gender-reverse of the specially selected cases in Nineteen Child Homicides. Given the relative rarity of men being granted a residence order, these cases alone are sufficient to discredit the entire narrative of Nineteen Child Homicides. But the review below takes a much more comprehensive look at the SCRs.

On the basis of their Child First campaign, Women’s Aid won parliamentary time for a debate in the House of Commons on Thursday 15 September 2016. It was not a debate, as reading the Hansard record testifies. A debate has two sides. This was a monoculture of opinion. Every MP reiterated the same perspective – the perspective put in their mouths by Women’s Aid. The organisation used their considerable resources to lobby individual MPs in their constituencies. MPs surgeries were targeted by perfectly valid cases of abused women and children – and nothing else. But half the truth is a whole lie.

The blame here lies with the MPs. A lobby group cannot be expected to be balanced. But MPs are under an obligation to represent everyone fairly, and are assumed to be intelligent enough to seek balance. Instead it appears that our parliament can very easily be led by the nose by a lobby group which presses the right emotional buttons.

I emphasise that the review presented below is not intended to suggest that mothers in general are dangerous to their children. That would be ridiculous. The purpose of the review is to provide a balance which is totally absent in Nineteen Child Homicides. Women’s Aid is very willing to use a small number of carefully selected cases to suggest that all fathers should be considered dangerous, but I am not about to commit the same mendacious folly with the genders reversed. Indeed, after reading through hundreds of SCR Abstracts it becomes clear that these terrible cases have little or nothing to tell us about the bulk of society. In these SCRs we are almost always dealing with the sump of society: addicts, alcoholics, the seriously mentally ill, parents with severe learning difficulties, and people damaged by their own background.

330 Child Homicides – Contents

  1. Child First: Nineteen Child Homicides” Report
  2. Selection of Cases in Nineteen Child Homicides
  3. Child First” Cases where contact was informal
  4. Review of the Serious Case Reviews
  5. Do the SCRs cover all child homicides?
  6. Identification of Perpetrators
  7. Lone-living mothers
  8. Review of the SCRs: Statistics on Perpetrators
  9. Cases of mothers killing while on contact
  10. Other Spite Killings or Murder-Suicides by Mothers (Samira Lupidi Case and others)
  11. Mothers killing children after the father was subject to a non-molestation order or an arrest
  12. Cases of lone fathers killing children
  13. Informal Observations on the SCRs
  14. Conclusions
  15. Appendix: Categorisation Methodology

1. “Child First: Nineteen Child Homicides” Report

In January 2016, Women’s Aid launched a campaign called “Child First“. This campaign is underpinned by a report produced by Women’s Aid entitled “Child First: Nineteen Child Homicides“. The organisation writes,

We have launched the Child First campaign to stop avoidable child deaths as a result of unsafe child contact with dangerous perpetrators of domestic violence.

They request that the public should,

Help us stop avoidable child deaths and make sure children are put first in the family courts.

These objectives are entirely laudable. However, the thesis of this review is that these statements are actually a smoke screen concealing the true objectives of the Child First campaign. The pernicious nature of this subterfuge is that the above objectives will not be well served by the campaign’s recommendations.

Women’s Aid claim that their campaign report, Nineteen Child Homicides, 

“Highlights the tragic stories of 19 children and 2 women in 12 families that were killed by perpetrators of domestic abuse in circumstances related to unsafe child contact within a ten year period. It’s possible that these deaths could have been prevented if the domestic abuse had been considered as an ongoing risk factor.”

The campaign web site includes a petition and a request for supporters to contact their MP to express their support for the campaign. This was by way of preparation for a parliamentary debate on the Child First Campaign which took place in the House of Commons on Thursday 15 September 2016. This was a crucial opportunity, we were told, to press the government to ensure that there are no further avoidable child deaths as a result of unsafe child contact, and to make the courts a safe place for all children and survivors of domestic abuse. Our heart-strings are tugged by observations like,

No parent should have to hold their children and comfort them as they die or be told that their child has been harmed in an act of revenge or rage.”

The message is further projected by highly emotional accounts of particular cases such as “Claire’s story”. Without doubt these dreadful cases are all too distressingly real. But that does not mean that it is sensible to change court policy based upon a dozen carefully selected cases, as the Child First campaign attempts to do.

We shall show that the dozen cases forming the basis of Nineteen Child Homicides have been carefully selected to support the desired outcome and are not indicative of the general run of child killings. To call Nineteen Child Homicides unbalanced would be an understatement.

Polly Neate, CEO of Women’s Aid, in the Foreword to Nineteen Child Homicides writes, “whatever the stated requirements on the family courts, there is a deeply embedded culture that pushes for contact with fathers at all costs“. And later in the report we read,

The ‘pro-contact’ approach taken by the family justice system has seemingly overtaken the need for any contact orders to put the child’s best interests first.

The cultural assumption in the family justice system that contact with both parents is the most beneficial outcome for a child is perpetuated by a public conception of the family courts as being biased against fathers applying for contact. There is no evidence to suggest this is the case. Research shows that the majority of non-resident parents achieve the type of contact and the amount of contact they seek.

A refutation of these reality-reversing remarks would take us too far from our present purpose and must be a task for another day. However, these remarks usefully expose the true intent of the Child First campaign. It is not to put the children first, as claimed, but to further assist mothers in their custody battles against fathers, and to encourage the courts to prohibit all contact by fathers in more cases.

Polly Neate further writes,

While it is impossible to prevent every killing of a child, when the risks are known no other consideration should be more important – yet there is evidence here that other considerations were rated more highly.

True, but this admirable sentiment can take on a very different interpretation from that intended by Polly Neate. A more complete examination of the same evidence base as used by Women’s Aid shows that mothers are responsible for more deaths of children than are fathers and male partners combined. (This will be demonstrated herein). And it is worth asking whether there are indeed other considerations which are rated more highly than child safety, namely a concentration by the protection agencies on supporting the mother and being wary only of the danger of male partners, to the exclusion of the potential danger which mothers themselves may pose to their children.

Nineteen Child Homicides refers to various Key Themes, including,

  • The importance of recognising domestic abuse as harm to children.
  • Professional understanding of the power and control dynamics of domestic abuse.
  • Supporting non-abusive parents and challenging abusive parents.

It is hardly a secret that the Women’s Aid interpretation of the “power and control dynamics of domestic abuse” is simply that it is men who are the violent abusers and  women who are their victims. Once one recalls this, the third of the above Key Themes no longer seems so benign since it will be interpreted in practice as meaning “support for the mother and  scepticism about the father”. But more importantly, based upon this ideological view of domestic violence, Key Themes 1 and 2 become incompatible. A blinkered view of who might pose a threat to children is inconsistent with recognising the potential for harm to children.

The key fact, which will be demonstrated below, is that more mothers are responsible for the deaths of children than are fathers and other male partners combined. But this key fact is precisely what Women’s Aid will never encourage exposing to public view.

Moreover, the Child First campaign focuses on children killed by their father whilst on contact visits. A further key fact which Women’s Aid fails to advertise is that there are also cases where residence orders have been granted to the father and the child is subsequently killed by the mother whilst on contact visit to her.

The Women’s Aid campaign makes various Recommendations, addressed to the Government, the family court judiciary and CAFCASS. The chief of these is the call for an independent, national oversight into the implementation of Practice Direction 12J – Child Arrangement and Contact Orders: Domestic Violence and Harm. The direction which any change in practice would take, if Women’s Aid had their way, is clear and reinforces the true purpose of their Campaign: frustration of fathers’ contact with their children.

The cases reported in Nineteen Child Homicides were identified using the Serious Case Reviews (SCRs). Here we conduct a more comprehensive review based on this same source of information.

The message of the present review is simple: children should be protected from harm whether the harm is likely to come from the mother or the father or another person. The Child First campaign’s exclusive concentration on the risk posed by certain fathers is misplaced since the empirical evidence is that mothers pose the greater risk.

2. Selection of Cases in Nineteen Child Homicides

Nineteen Child Homicides makes the following statement,

In this study Women’s Aid aimed to identify those cases where a child had been killed by a perpetrator of domestic abuse in circumstances relating to child contact (formally or informally arranged).

The selection of cases was done on the following criteria,

  • a child had been killed
  • the perpetrator was the child’s parent and had perpetrated domestic abuse against the other parent
  • the parents were separated and child contact had been arranged informally or formally

The report continues,

After applying these criteria, 12 reports were found to be relevant. We did not apply any exclusion criteria regarding the gender of the perpetrator of domestic abuse. However, in all of the relevant cases (12 families) it was the father who was the perpetrator.

This is disingenuous bordering on mendacity.

The impression is given that only fathers were perpetrators of abuse and only fathers were responsible for killing a child whilst on contact visits. But notice the filter on case selection provided by criterion (ii), above: the perpetrator had perpetrated domestic abuse against the other parent.

It is very unlikely that fathers were ever asked if they had been subjected to domestic abuse by the mothers. The reverse is, of course, standard practice. Consequently, filter (ii) acts as a virtual guarantee that only fathers will be selected.

The fraudulence of this methodology is exposed most clearly by the identification in Section 9, below, of cases where the mother killed whilst the child was on a contact visit with her, the father having residency.

This observation alone discredits the claim of the campaign to be “child first”. Rather it is “mother first” and anti-father.

3. “Child First” Cases where contact was informal

A further problem with the selection of cases in Nineteen Child Homicides is the reference to contact possibly being “informal”. The thrust of the report is a criticism of the courts’ practice on contact arrangements, and a recommendation that they be reviewed. But this can have no effect upon contact which has been informally arranged, i.e., outwith the courts. So it is illogical to criticise the courts for contact arrangements for which the courts were not responsible, yet this forms part of the Women’s Aid case. About half the cases in Nineteen Child Homicides involved only informally arranged contact. Since there was no court ruling on contact in these cases they are irrelevant to the Women’s Aid objective of tightening the courts’ practice on fathers’ contact, though spuriously used to justify it. (This is not to deny that there are some cases of fathers killing children whilst on court-ordered contact visits).

4. Review of the Serious Case Reviews

The basis of the review reported here is the database of Serious Case Reviews (SCRs).

The SCRs relating to children may be found in the NSPCC’s on-line library. All the SCRs for the seven years 2009 to 2015, inclusive, have been reviewed, confining attention to only those cases which involved child deaths. In many cases the Executive Summaries of the SCRs provide sufficient information to identify the probable perpetrator(s). If not, the full SCR report (where available on-line) was examined to see if this shed more light on the perpetrator.

For ease of exposition we shall use the term Serious Case Review (SCR), which is strictly the term appropriate in England, to refer to the corresponding reviews for all four nations of the UK. Hence this term can be understood to include the Child Practice Reviews which replaced SCRs in Wales from 2013, the Significant Case Reviews applicable in Scotland, and the Case Management Reviews applicable in Northern Ireland.

Cases of child suicide were distinguished from child killings by others. In some cases this may conceal the consequences of abuse which drove the child to such an extreme, but the underlying culpability in these cases is too difficult to establish in this brief review.

The SCRs are reviewed by year, but note that this refers to the year when the SCR was published, not the year of the death(s). The deaths were often the previous year, and sometimes several years earlier.

5. Do the SCRs cover all child homicides?

No.

The number of child homicides based on ONS data exceeds the number of SCRs relating to non-suicide child deaths in every year.

Moreover, there is evidence that the ONS data under-estimates the actual number of child homicides, possibly because the ONS count only cases where homicide has been ruled officially as the cause of death. The annual report from Ofsted for 2007/8 used the SCRs to identify 210 child killings, including suicides, over the 17 month period April 2007 to August 2008 (see Table 1). After subtracting the suicides, and scaling down to a 12 month period, this implies a rate of child killings of about 128 per year. This is far greater than identified by ONS. The reason is probably that some of the causes of death listed in Table 1 are not counted as homicide by ONS.

Even the estimated 128 child killings in 2007/8 is likely to be an under-estimate because, despite the particularly large number of SCRs that year, Ofsted noted, with some suspicion, that many Local Safeguarding Children Boards (LSCBs) had not submitted any SCRs. To quote the 2007/8 Ofsted report,

“The geographical distribution of reported serious incidents across Government Office regions varies significantly. A survey conducted by Ofsted showed that around a quarter of LSCBs had not carried out any serious case reviews, while 5% had completed five or more. Differences in the numbers of serious incidents and deaths of children and young people between local authority areas do not fully account for these variations; they are also due to varying interpretations of the guidance in “Working together to safeguard children” by LSCBs.”

The fact that the current review was conducted based on the SCRs alone may imply a skew in the findings if there are systematic features which tend to preclude child deaths from being subject to SCR. It is beyond the scope of this review to investigate this possibility.

Table 1: Taken from Robert Whiston and Nigel Hawkes (click to enlarge)

child-deaths-2007_8

6. Identification of Perpetrators

For the purposes of this exercise, cases have been categorised according to the following perpetrator classes,

  • Mother as perpetrator;
  • Father as perpetrator;
  • Mother’s partner as perpetrator (where the partner is not the biological father);
  • Joint perpetration by mother and father;
  • Joint perpetration by mother and partner;
  • Other perpetrators;
  • No culpability or perpetration could not be determined.

The determination of the perpetrator category based on the SCRs is not always simple. It is not merely a matter of counting because the SCRs themselves do not identify perpetrators in many cases. The approach adopted here is to use the facts of the case, as described by the SCR, to come to a reasonable judgment regarding the perpetrator(s). In most cases this will be uncontentious, for example if arrests have been made. Being charged with an offence is not the same as being guilty, but the present exercise is based on the balance of probability. Where no criminal charges have been brought, there may nevertheless be indications in the case history suggesting likely culpability. Again, balance of probability is the guide. However, there will be room for differences of opinion in some cases.

The methodology adopted is described in more detail in the Appendix.

A brief résumé of the cases where the mother has been identified as the perpetrator, or as a co-perpetrator, is provided here. This file is organised with cases listed in the following order: (i) single mother as sole perpetrator; (ii) cohabiting mother as sole perpetrator; (iii) mother and father joint perpetrators; (iv) mother and non-paternal partner as joint perpetrators.

In many cases a perpetrator could not be assigned, either due to lack of information or because it was not clear where culpability lay. A résumé of these cases is also provided here. For further discussion of the difficulties in deciding on perpetrator category see the Appendix.

7. Single mothers

To the majority of people, who live relatively orderly lives, it is a clear cut matter whether someone is cohabiting or not living with another adult. But the families to which almost all the SCRs relate are anything but orderly. In many cases the mother has a partner – perhaps several partners – whose presence in the same living accommodation may be rather indeterminate. In such cases the partner may drift in and out. For our purposes, a mother is defined to be living with no other adult (i.e., a single mother) if no adult male had been living with the mother at around the date of the child’s death.

8. Review of the SCRs: Statistics on Perpetrators

The data in Table 2 refer to the number of cases. In brackets are the number of children killed where different. The two differ only when one or more cases involve more than one child death. Where only one figure is given, only one child was killed in all the relevant cases. (For consistency with this blog format Table 2 has been split into two parts, Tables 2a and 2b).

Table 3 breaks down the data differently, in particular it identifies single mothers as perpetrators.

Table 2a

Summary of Statistics Obtained from Serious Case Reviews of Child Deaths

The data in the Table refer to the number of cases. The data in brackets are the number of children killed where different.

Item 2009 2010 2011 2012
Total child homicides# 90 56 47 67
Number  of child suicides or overdoses 18 23 11 8
Number of SCRs involving child deaths excluding suicides/overdoses 79 54 39 28
Perpetrator was mother 17(18) 17(19) 11(12) 9
Perpetrator was father 16(19) 5(6) 9(10) 7(8)
Perpetrator was partner 3 4 2 4
Perp mother plus father 11 3 3(4) 1
Perp mother plus partner 4 4 1 0
Perpetrator was someone else 9 1 6 2
Suicide* of mother 1 1 1 3
Suicide* of father 0 0 1 2
Perpetrator not determined 19 20 7 5

*Including attempts      #Approximate only, relates to April-to-March (ONS data)

Table 2b

Summary of Statistics Obtained from Serious Case Reviews of Child Deaths

The data in the Table refer to the number of cases. The data in brackets are the number of children killed where different.

Item 2013 2014 2015 TOTAL

2009-2015

Total child homicides#
Number  of child suicides or overdoses 11 5 10 85
Number of SCRs involving child deaths excluding suicides/overdoses 38 34 35 307
Perpetrator was mother 7(8) 10(14) 15(17) 86(97)
Perpetrator was father 7(9) 8 3 55(63)
Perpetrator was partner 2 2 3 20
Perp mother plus father 9 3(8) 5 35(41)
Perp mother plus partner 3 2 2 16
Perpetrator was someone else 3 2 1 24
Suicide* of mother 0 2 2 9
Suicide* of father 3 2 1 9
Perpetrator not determined 7 7 6 71

*Including attempts      #Approximate only, relates to April-to-March (ONS data)

Table 3: Mothers as Perpetrators: Single Mothers versus Cohabiting Mothers and Percentage of Cases Involving Mothers as Perpetrators@

year Single Mother Cohabiting Mother Mother with Father or Partner Father or Partner Other Person Mother involved in % of cases@
2015 12 3(5) 7 6 1 76%
2014 7(10) 3(4) 5(10) 10 2 58%
2013 6(7) 1 12 9(11) 3 59%
2012 6 3 1 11(12) 2 43%
2011 7(8) 4 4(5) 11(12) 6 47%
2010 15(16) 2(3) 7 9(10) 1 71%
2009 9(10) 8 15 19(22) 9 53%
TOTAL 62(69) 24(28) 51(57) 75(83) 24 58%

@Percentage of cases for which a perpetrator has been assigned which involve the mother as the perpetrator or one of the perpetrators

If the SCRs are indicative of all child killings, then the findings of our review suggest that,

  • One or both parents are culpable in the killing of at least 65% of children – and probably a much greater percentage if culpability could be assigned in every case.
  • Mothers’ male partners, where not the father, are culpable for ~11% of child killings. This might be a somewhat larger percentage if culpability could be assigned in every case.
  • Children are killed by someone other than their parents or the mother’s male partner in only ~7% of cases.
  • Where culpability is established, the mother is the lone perpetrator in 36% of cases and either a lone or a co-perpetrator in over half of cases (58%).
  • Mothers are more likely to be responsible for a child death than fathers and male partners combined.
  • Single mothers are the demographic most likely to be responsible for the deaths of children (62 cases versus 55 cases of fathers as sole perpetrators).

No culpability was assigned for 23% of the SCRs involving child deaths (excluding suicides/overdoses). In some cases this was due to lack of sufficient information. However, large numbers of these SCRs might have been classed as neglect or negligence but the circumstances were such that I erred on the side of caution and did not assign culpability. Some of these cases might otherwise have implied culpability by both parents or carers. But, had a harsher line been adopted, virtually all such cases would have implied the mother’s culpability (since neglect implies that all carers must be neglectful). Consequently the approach adopted will, if anything, tend to minimise the statistics of perpetration by the mother. A résumé of the cases where perpetrators were not assigned is provided here.

9. Cases of mothers killing while on contact

As a by-product of our review of the SCRs, two cases were found relating to a child killed by the mother whilst on contact visit to the mother, the father being the resident parent, i.e., the gender-reverse of the cases considered by Nineteen Child Homicides. These cases relate to 2014 and 2015 and are described briefly below.

9.1 April 2015 – Lancashire LSCB – Child N

Death of a 4-year-old boy and his mother as the result of a house fire in Liverpool. Parents had separated acrimoniously some 4 years earlier. Due to concerns about his safety and on-going contact disputes, Child N’s care was subject to a number of court proceedings. The court’s decision in the second case resulted in the father, who lived in Lancashire, being granted a residence order and the mother a contact order. During the fourth and final set of proceedings, whilst Child N was on a contact visit, the mother made allegations of child sexual abuse which resulted in Child N staying with her in Liverpool. Police evidence indicated that the mother had purchased petrol in a container which had been used to start the fire. The Coroner’s report states “… it is found that the fire was started deliberately … with the intention of causing death or being reckless”. The Coroner’s verdict was that Child N had been unlawfully killed and mother had taken her own life after deliberately starting the fire. [This case has been reported in the press, Child N being revealed as Jai Joshi and the father as Paresh Patel].

This may be classed as a spite killing

9.2 August 2014 – Oxfordshire LSCB – Child N

Death of a 1-year-old girl in May 2013 whilst in the care of her mother, recorded as an unexplained death but treated as suspicious. Parents grew up in Africa. Child N had been the subject of contested proceedings for residence and contact in the family court between her mother and the father. Child N is believed to have been in her mother’s care during the last days of her life and her mother is known to have left the UK in the hours following the child’s death. A Residence Order had been granted in favour of the father due to fear of the mother leaving the country with Child N. The day after the order was granted the mother left the country and Child N was found dead in her flat. At the date the SCR was written there had been no inquest. The initial post mortem examination of Child N was unable to ascertain the cause of her death; however it noted that Child N did not have the commonly observed symptoms of unexplained sudden infant death. The death of Child N is therefore being treated as suspicious and is the subject of a continuing criminal investigation. There is no evidence or reason to suspect that Child N was ever caused any harm by her father.

This appears to have been a spite killing.

Neither of these cases were discussed in Nineteen Child Homicides on the preposterous basis that there was no record of the father being subject to partner violence from the mother.

Given the relative rarity of residence being granted to fathers, and hence the rarity of mothers being the parent involved in contact orders, it is noteworthy that two cases like those above have been found in 7 years of data – compared to just 12 cases of fathers killing children whilst ‘on contact’ in 10 years of data.

Nineteen Child Homicides selected cases by requiring that the perpetrator had a history of partner abuse against the other parent. The cases identified above illustrate how this criterion works to spuriously select only fathers as perpetrators, contrary to the facts. This ideological approach elides the danger that mothers can also present to their children.

10 Other Spite Killings or Murder-Suicides by Mothers

10.1 The Samira Lupidi Case

The Samira Lupidi case does not involve killing whilst ‘on contact’, but it does involve a mother killing her children during a break-up from the children’s father. Lupidi stabbed to death her daughters, 17-month-old Jasmine and three-year-old Evelyn. Incredibly the murders took place whilst Lupidi was resident in a women’s refuge in Bradford, West Yorkshire. She was found guilty of murder and sentenced to 24 years. The claim was made that Lupidi was suffering from “a complete misinterpretation of reality”, having falsely accused her 31-year-old partner of domestic violence. One wonders if this is the same sort of complete misinterpretation of reality from which false accusers generally suffer.

However she was interpreting reality, the judge noted that, even a week later, Lupidi was telling the prison medical staff that the most important thing was that Carl Weaver  (her ex) was suffering. This is a case of ‘spite killing’.

I cannot pretend to have any understanding of the psychology behind acts like this. However, the evidence seems to suggest that, whatever the psychology, it is not gender specific. The cases of fathers killing their children identified by the Child First report are distressingly true, and Women’s Aid suggest the motivation was revenge. Be that as it may, one supposes that Lupidi’s motivation was probably much the same. And the same goes for the mothers identified in Section 9, above, and probably also those in Sections 10.2, 10.3 and 11.1 below.

It is understandable that Women’s Aid would wish to sweep the Lupidi case under the carpet, and not only because it undermines the ‘argument’ of the Child First campaign. It is also because Lupidi and her history refute the feminist line on domestic violence. Lupidi would appear to be a text book case in which the tendency to domestic violence runs in families. Lupidi was herself subject to domestic abuse as a child by her own mother. No patriarchal power and control in sight, is there?

The fact that Women’s Aid can publish Nineteen Child Homicides just two months after this mother murdered her two children in one of their own refuges – motivated by the sort of revenge which they are implying is the sole province of fathers – tells you all you need to know about their standards of honesty.

10.2 Birmingham 2009 case No.10

The Abstract of this Serious Case Review reads,

Case no. 10 involves a Bangladeshi family in Birmingham. The family home was violent and volatile. The mother self-harmed twice, once when she was pregnant, but a lack of interpreter and reliance on her husband meant that the risks were not properly assessed by professionals. The mother and two children were found deceased in their home. The children were unlawfully killed. Recommendations include acknowledging that women from BME communities can be perpetrators as well as victims of domestic violence.

The Keywords refer to “abusive mothers” and “suicide”, so, despite not finding the full report on-line this seems to be a case of the mother being responsible for the deaths of her two children followed by her suicide. This case was omitted from the original post “330 Child Homicides” and accounts for this revised post being titled “332 Child Homicides”. 

This case has been classed as a spite killing following reading the report “Learning from CAFCASS Individual Management Reviews (IMRs) – Case Dynamics: Executive Summary”, November 2013.  This report reviews a small number of SCRs between 2009 and 2013. It identifies a number of spite killings, stating,

One of the six spite killings subject to a Cafcass IMR was perpetrated by a woman. We have recently been notified of another SCR into a case in which the mother is thought to have killed her child shortly after the Cafcass FCA recommended change of residence to the father (not included in the sample).

The first of the spite killings alluded to in the CAFCASS report may be the case addressed here, although this has not be ascertained with certainty. 

The timing and circumstances suggest that the second of the spite killings by a mother alluded to in this 2013 CAFCASS report could either be that discussed in Section 9.2 above or that discussed in 10.3 below. (Note that because neither names nor other unique identifiers are used in the SCRs or the CAFCASS report it is not possible to make a certain identification). However, the 2013 CAFCASS report clearly failed to pick up all such spite killings by mothers in the time period claimed. 

10.3 Gwynedd & Anglesey 2011, Child G and Child H

The SCR Abstract reads, “Review into the death of a 5 year old boy (Child G) and his 2 year old halfbrother (Child H) in December 2010. Both children were suffocated by their pregnant mother who committed suicide. Family had contact with several agencies because of concerns over the impact of custody and contact disputes. Mother was involved in 2 separate sets of proceedings, one involving her three older children and the other in relation to Child H.”

Press reports have now revealed that the mother was Melanie Stevens and the murdered children were Philip (PIp) and Izaak. Izaak’s father, Nick Smith, has contested the findings of the SCR, namely that “the only way to have saved the children would have been to remove them from their mother, but there were no grounds to do so“. He claims there were ample grounds to remove the children from their mother, and that he had spent 18 months telling professionals that this could happen. Ms Stevens, who was pregnant at the time, killed the children and herself on the day Mr Smith was due to collect Izaak for his first overnight stay with him.

This appears to be both a spite killing and an instance of murder-suicide.

Incidentally, the burden of the press report into this case, namely that child deaths due to neglect tend to be given a rather muted treatment in the SCR process, is borne out by my review. Other family members, often the father, are typically unable to influence the conclusions of the SCR, and deaths through neglect are frequently not prosecuted and hence do not appear in the official homicide statistics – see the compilation of summaries

10.4 Croydon, April 2015 – Josh

SCR Abstarct: Death of a 3-year-old boy in March 2013. Mother carried Josh into the path of an oncoming train, killing them both. This is a case of murder-suicide.

10.5 Suffolk, January 2014, The Anderson Family

SCR Abstract:  Death of three children and their mother. Children were aged 3-years, 2-years and 13- months at the time of their deaths and mother was 7 months pregnant. Evidence suggests mother killed the children prior to committing suicide. All three children were subject to child protection plans under the category of neglect. Unknown whether there was an element of spite, but clearly a case of murder-suicide. 

10.6 Bury, 2012, Child D (Arlene aged 11)

This is a case of mudrer-then-attempted-suicide. SCR Abstract: Death of a 4-year-old girl from multiple stab wounds. Suicide notes indicated that Child D was killed by her mother who then attempted to kill herself. Mother had recently been dismissed from her job and was due to attend court in relation to offences in contravention of the Data Protection Act. Following her dismissal, mother maintained the pretence that she was employed to family, friends and Child D’s nursery. Identifies additional stress factors that likely contributed to mother’s actions, including her failure to gain employment because of unsatisfactory references and the breakup of her relationship.

10.7 Warwickshire 2012, Death of child G.A.

Murder-suicide case. SCR Abstract: Death of a 6-year-old girl in the summer of 2011. It is believed G.A was killed by her mother who then took her own life. The post-mortems recorded the causes of death as follows: Mother: hanging; Child: “interference with the normal mechanisms of respiration”. Father lived with mother and G.A for most of her life, interrupted by periods of parental separation; the most recent period of separation began several weeks before the incident. 

10.8 Leicester 2012, Baby L

This is a case of mudrer-then-attempted-suicide. SCR Abstract: Death of Baby L (aged 7 months) in 2011 after being smothered by her mother who then attempted suicide. Mother pleaded guilty to infanticide and was ordered to undergo secure psychiatric treatment for severe postnatal depression. Review finds that Baby L’s death could not be predicted but there were missed opportunities.

11 Mothers killing children after the father was subject to a non-molestation order or a criminal accusation

A number of cases have been identified in which mothers were responsible for the deaths of children after the father had been removed by a non-molestation order, or by virtue of a criminal accusation. In these cases one wonders if the right person had been separated from the child. They are as follows,

11.1 November 2014 – Torbay – C42

Death of 2 siblings and their mother. Child A and his mother died on the 12 July 2013 following a fall. The body of Child B was found later that same day at their home address. The inquest into the deaths found that the mother took her own life and that Child A was unlawfully killed. An open verdict was recorded in respect of Child B. Children were living with their mother following a short period in foster care whilst their father was charged with the assault of their mother.

This may be classed as a spite killing or ‘family annihilation’.

11.2 April 2014Devon – CN08

Death of a 2½-year-old boy. Mother admitted to killing the child; she was convicted of manslaughter and received a hospital order. Mother had been detained under the Mental Health Act for a period in 2011. A history of domestic abuse reported against the father had led to a Restraining Order against the father being in place at the time of the incident.

11.3 November 2013Bradford – Hamzah Khan

Death of a 4-year-old boy in December 2009, as a result of chronic neglect. Hamzah’s body was discovered by police during a search of the family home. (The child had been dead for two years and this fact had been covered up).  Mother was convicted of manslaughter and child cruelty in October 2013. Maternal history of chronic alcohol dependency; depression; social isolation; domestic abuse; and reluctance to engage with services, including registering children for health and education services. Father was made the subject of a non- molestation order in 2008 following an arrest for assault against mother.

11.4 October 2014 Bedford – Child A1301

Death of a 19-month-old child in April 2013, as the result of a non-accidental head injury. Child A’s mother and her partner were arrested but it was not possible to establish who caused the injury. Child A’s parents’ separated almost immediately after Child A’s birth and mother entered into a new relationship. Parents separation was acrimonious; allegations were made against father and against mother. Key issues: preconception of father as controlling leading to his concerns over mother’s parenting being minimised; insufficient challenge to information provided by mother, which was later found to be untrue; and drink-driving allegations made against mother not being shared with the police agency responsible for assessing potential harm to children.

12. Cases of single fathers killing children

Attention was paid to whether, in any of the SCRs reviewed, the killed child had been living full-time with the father or other male carer (as distinct from being temporarily looked after by a male carer, e.g., whilst the usual carer attended an appointment or was on holiday or the child was on a contact visit). There were no such cases in the 306 SCRs involving child deaths reviewed.

This is in contrast to the 61 identified single mothers being responsible for 67 child deaths.

Of course, there are far fewer children living with single fathers than with single mothers. Estimates are that single fathers account for between 5% and 10% of single parents. Hence 3 to 7 cases of single male carers being responsible for child deaths might have been expected from the SCRs reviewed on a purely random basis. The fact that none have been identified may be of some significance.

13. Informal Observations on the SCRs

The author is not a professional in this field and has had no previous exposure to SCRs of any sort. As such he can bring an outsider’s perspective to bear.

After reading a few hundred SCR Executive Summaries it becomes clear that, in almost every case, we are dealing with an extreme end of society which is entirely uncharacteristic of society as a whole. Consequently, basing recommendations for changes to court practices on these highly unrepresentative, extreme cases is obviously very questionable, even before one looks more closely at the details.

The SCRs almost always make reference to a history of domestic abuse suffered by the mother, whether from her current or former partner, of from her birth family. The virtual ubiquity of the reports of domestic abuse against the mother begins to seem rather obsessional after reading hundreds of cases. As for the father and/or male partner, they are mentioned primarily in the context of their record of violence. Of course, this may all be a perfectly faithful rendition of the facts. But often these factors figure large in the reports even when they play no part in the child’s death, e.g., when the death has been attributed to SIDs, say, or accidental drowning whilst unsupervised in the bath.

Another factor which rapidly becomes clear is the focus by professionals on the mother’s needs and welfare, often to the exclusion of the needs of the children. The SCRs themselves frequently make this very point. There appears to be a cultural bias towards addressing the needs of the mother whilst the potential danger that the mother may pose to her children tends to go unrecognised. The potential danger of the father or male partner is clearly uppermost in the protection agencies’ minds, but the potential for danger to the child arising from the mother is less often recognised. At this point we echo the words of Polly Neate, but with a very different intended meaning,

While it is impossible to prevent every killing of a child, when the risks are known no other consideration should be more important – yet there is evidence here that other considerations were rated more highly.

14. Conclusions

If the SCRs are indicative of all child killings, then the findings of our review suggest that,

  • One or both parents are culpable in the killing of at least 65% of children – and probably a much greater percentage if culpability could be assigned in every case.
  • Mothers’ male partners, where not the father, are culpable for ~11% of child killings. This might be a somewhat larger percentage if culpability could be assigned in every case.
  • Where culpability was established, children were killed by someone other than their parents or the mother’s male partner in only ~7% of cases.
  • Where culpability was established, the mother was the lone perpetrator in 36% of cases and either a lone or a co-perpetrator in over half of cases (58%).
  • Mothers are more likely to be responsible for a child death than fathers and male partners combined.
  • Single mothers are the demographic most likely to be responsible for the deaths of children (62 cases versus 55 cases of fathers as sole perpetrators).
  • No cases of single fathers killing children were identified in the 307 SCRs reviewed.
  • Six cases of mothers executing ‘spite killings’ or ‘family annihilations’ have been identified (involving 10 children’s deaths) in the period 2009 – 2015, of which two occurred whilst the mother was “on contact” (see Sections 9, 10 and 11.1). 
  • Nine cases, involving 15 children’s deaths, have been identified as either spite killings by mothers or murder-suicides by mothers, in the seven year period 2009 – 2015.
  • Four cases have been identified of mothers killing children whilst the father was subject to a restraining order or had been arrested for assault of the mother.

Armed with these observations let us return to the objectives of the Child First campaign. Firstly, to…

stop avoidable child deaths and make sure children are put first in the family courts.

Contrary to the claims of the Child First campaign this objective will not be met by further restrictions on child contact by fathers – for the simple reason that it is mothers who are responsible for most child deaths – and single mothers in particular. When all the available data is used – as opposed to the specially selected 12 cases presented in Nineteen Child Homicides – the Child First campaign’s recommendations are seen to be the opposite of putting the safety of children first.

And secondly, in respect of the plea that…

no parent should have to be told that their child has been harmed in an act of revenge or rage

we have seen in Section 9 that mothers are just as capable of such destructive behaviour as fathers. The 12 cases presented in Nineteen Child Homicides were specially selected to give a false impression – a misinterpretation of reality.

The Child First campaign by Women’s Aid is profoundly dishonest.

Appendix: Categorisation Methodology

The determination of culpable parties (perpetrators) of child deaths based on the SCRs is not simple. It is not a mere matter of counting because the SCRs themselves do not identify perpetrators in many cases. The methodology used is described below.

Where a criminal prosecution has led to a conviction on a charge which identifies responsibility for death (e.g., manslaughter or ‘causing the death of a child’, etc.) then this defines the perpetrator. If two or more people were so convicted, then they are all perpetrators. However, if one person was convicted on a charge which identifies responsibility for death (e.g., manslaughter or ‘causing the death of a child’) but another person was convicted only of some lesser crime such as child cruelty, then in general only the first person has been defined as the perpetrator. This especially applies if the conviction for child cruelty refers to a sibling who was not killed. This ruling tends to penalise men because, in most cases where the mother and her partner have been abusive to children, it tends to be the man who faces the more serious charge, though not always.

So far, so easy. More problematical are cases where criminal charges have been brought but the outcome of the trial was unknown when the SCR was written. In general I have assumed the accused to be guilty as charged as a convenient means of assigning culpability. Essentially this relies upon the police’s judgment and their greater knowledge of the details of the case. This also will tend to disadvantage men (it being well known that the police more readily charge men, especially in domestic abuse cases).

In some cases the police comment that a criminal case could be raised but that prosecution “would not be in the public interest”. These are treated in the same manner, i.e., as if the charge had been brought.

The most problematic are those cases where no criminal charges have been brought – and this is a very large proportion of cases. As an example of balance of probability, even in the absence of a prosecution, if there is a statement, “Child X had been in the sole care of Mr Y when the serious head injuries were sustained” then I take that as indicative of probable guilt for the purposes of this exercise, unless there are other factors which negate this impression.

Most difficult are cases of neglect or of unexplained death – and these account for a large proportion of all cases. In cases of neglect the history of the family as depicted in the SCR full report is key. In almost all cases where neglect is suspected there is a history of some sort, perhaps of child protection orders being in place, either for the child in question or a sibling, or of third parties expressing concerns to the authorities, or of hospital visits with unexplained injuries, and so on. By this means a judgment can be made as to culpability. The judgment will be fallible but has been made based on best endeavours.

In cases where neglect may be suspected but there is no ‘guilty history’ then in general no perpetrator will have been assigned and the case is placed in the “don’t know” category.

Next we have the unexplained deaths, which includes SIDS (sudden infant death syndrome) and the pathologists’ assignment as “cause of death unascertained”. Whilst the bulk of SIDS in the general population may well be of natural causes, those cases subject to SCR are not typical. Recall that there are around 200 cases of unexplained infant death in England and Wales per year, but only perhaps half a dozen are subject to SCR. These are the cases which the authorities have selected as being suspicious or having complicating factors. It should not surprise, then, to find that virtually all such cases in the SCR record involve families with histories of one sort or another. Generally they were known to the protection agencies beforehand. So it is possible to adopt the same approach to unexplained deaths as to the suspected cases of neglect, namely assignment of likely culpability based on history.

As an example, I counted the following as a mother being the perpetrator even though no charges were brought,

March 2015 – Coventry – Child T
Death of a 3-week-old girl; coroner classified cause of death as ‘unascertained’. Background: Following Child T’s death, a home visit found that the family were living in dirty and unhygienic conditions. When the family home was visited immediately following Child T’s death, it was found to be squalid with dirty nappies, rubbish and food all over the floor. It was very hot and there were flies everywhere. In the police officer’s opinion it was not fit for human habitation. The police investigation concluded that whilst a case for criminal neglect had been established, it was not in the public interest to prosecute.

On the other hand, in many cases the death might have been genuinely without culpability. In other cases, whilst there might be signs of mistreatment, there is no clear knowledge of the perpetrator. These cases have necessarily been classed as “perpetrator unknown”.

If anything I have been rather over-generous in categorising cases of apparent neglect as “perpetrator unknown”. An typical example is below,

November 2014 – Sunderland – Baby A and Child C
Death of a 1-week-old in April 2013. Coroner’s Inquest concluded no explanation for cause of death. Baby A was admitted to hospital following a cardiac arrest and later died, following the withdrawal of life support. Mother had a history of abusive relationships and had received support for a long-standing heroin addiction for a number of years. Mother was well known to children’s social care. Baby A was the subject of a Child Protection Plan at the time of the incident and had been discharged from hospital, post-birth, into the care of maternal grandparents. Mother’s eldest child and Child C were also living with maternal grandparents at this time and mother’s second eldest child was living with their father. Child C was subject to Child in Need or Child Protection Plans for most of their life prior to the incident.

A précis of those cases which have been put in the “no perpetrator” bucket has been provided separately here.

7 thoughts on “332 Child Homicides

  1. Zac

    William, I just sent this page to my sister after she shared a news article with me about calls for an inquiry into abusive parents’ access to children. I was most disturbed when she responded that her ‘high’ parental control settings in her BT wifi have blocked it. Were you aware this was happening to your site? I wonder if it relates to the dark subject matter of this post or if their motivations are political. Keep up the good work. Zac

    Reply
    1. William Collins Post author

      I have been told previously that some services or search engines block access, yes. There are a number of articles with rape in the title, which might do it. But I suspect it might be political really. I may update the 332 Child Homicides report in response to the recent Proudman thing as there are now three further full years of data to go at. It is a wicked deceit to use these terrible, tragic cases as a stick to beat the average man with.

      Reply
  2. Pingback: Women Trump Children? | The Illustrated Empathy Gap

  3. Pingback: CAFCASS: Learning from Serious Case Reviews | The Illustrated Empathy Gap

  4. Pingback: My Presentation at ICMI18 | The Illustrated Empathy Gap

  5. Pingback: All Hale the Top Dog | The Illustrated Empathy Gap

  6. Pingback: UK PV Perpetrator Programmes – Part 1 | The Illustrated Empathy Gap

Leave a Reply

Your email address will not be published. Required fields are marked *