Why parents kill their children

The daily reports of murders are bad enough but I think that people are struck with a particular horror when they read of parents killing their own children, especially when the children are young, before killing themselves. It has the effect of infuriating me and making me wonder why, if they think that their lives are so bad, such people not simply kill themselves without taking innocent children with them. Of course, people who commit such horrendous acts are clearly in a highly disturbed state of mind and are not thinking rationally in the first place so we should not expect them to think like others would. But still, there must be some reason that drives them to do something so awful.

A 2005 study looked at the reasons why such things happen. You can read the journal article here.

Several recent cases of filicide, child murder by parents, have drawn national attention to this archetypal tragedy. Specific motives for filicide were initially described by Resnick, classified as (1) altruistic, (2) acutely psychotic, (3) accidental filicide (fatal maltreatment), (4) unwanted child, and (5) spouse revenge filicide. Altruistic filicide is murder committed out of love to relieve the real or imagined suffering of the child. Altruistic filicide may be associated with suicide. For example, a mother who is suicidal may not be willing to leave her child motherless in a “cruel world.” Distinct from this, acutely psychotic filicide occurs when a parent in the throes of acute psychosis (e.g., experiencing command hallucinations) kills his or her child with no comprehensible motive. Fatal maltreatment filicide may occur as a result of child abuse, neglect, or Munchausen syndrome by proxy. Parents committing spouse revenge filicides kill children in a specific attempt to make the spouse suffer. Furthermore, filicide may occur within the context of familicide, the extermination of the entire family.

Resnick reported a “relief of tension” after altruistic and acutely psychotic filicides. The expulsion of energy after the child’s death explains why some parents who had intended filicide-suicide did not complete the act of suicide. Conversely, other parents, “upon realization of the gravity of their act…may attempt suicide even if it was not planned”

The authors also hypothesized that motives for filicide-suicide would most likely be altruistic or acutely psychotic, because parents who had fatally abused a child, had an unwanted child, or sought revenge on a spouse would be less likely to kill themselves after having killed their children. In all three of these scenarios, the offending parents would be less likely to merge their own fate with that of their children. The parent may fail to commit suicide because, in the case of fatal maltreatment, the child usually has been neglected or abused but unintentionally killed; in the case of an unwanted child, there is a lack of identification with the child; and in the case of spouse revenge, there may be a desire to remain alive to take pleasure in the vengeance.

The sample of 30 had twice as many fathers (20) as mothers (10). There were other gender differences.

The fathers (mean age ± SD, 38.2 ± 8.8 years), were significantly older than the mothers (mean age, 31.8 ± 5.7 years, t = −2.369, p < .026). … Sixty-five percent of the fathers attempted to kill their wives as well as their children, whereas no mothers attempted to kill their husbands. In all, 55 percent of the fathers, but none of the mothers, attempted familicide, that is, annihilation of the entire family.

If I had to guess, I would have said that acutely psychotic filicide was the most common type followed by spouse revenge filicide. But I would have been wrong.

The majority (70%) of the motives for filicide-suicide were identified as altruistic, that is, the parents (90% of the mothers and 60% of the fathers) were motivated by the desire to alleviate real or imagined suffering in their children. The altruistic category was subdivided into psychotic altruistic and non-psychotic altruistic. For example, a psychotic altruistic motive would include taking a child’s life because of the delusional belief that the child was in acute danger of a worse fate. Alternately, a non-psychotic altruistic case would be taking the life because of a belief that a severely medically ill child would be better off.

That altruism was the most common motive for killing one’s children surprised me. Of course the total sample size is small because filicide-suicide is relatively rare.


  1. says

    There were some studies back in the day about human males being more likely to kill their mates’ children by another male, by a pretty large amount. However, like all social science studies, I’m inclined to dismiss it.

    For example, the definition of “psychotic” in the last statistic -- that expands out to: “in cases where a psychiatrist diagnosed the killer as psychotic” which is problematic since psychosis is just a label for a set of behaviors: there is no objective test for psychosis. In other words, there’s no actual “disorder” there are “symptoms that indicate a disorder.” In the case of a murderer it’s quite plausible that someone might be arguing they are psychotic in order to get different sentencing guidelines, or to get sympathy from a jury. At present I’d say that social scientists don’t even know what “psychosis” is* and there’s no way to tell if a given doctor in a given case actually believes any given killer is psychotic or not.

    (* other than “psychosis is manifesting behaviors that we define as being associated with psychosis”)

  2. says

    We had a case very recently in Alberta where a man killed his two sons and then himself while they were with him (the mother had divorced him and re-married). It came out that she had told the courts he was abusive.


    Of course to people outside of the relationship, he seemed like a great father…


    Anyway, what I’m not seeing in the study is any talk about how people feel a sense of ownership over their child. “Those are my children, I can do what I want with them!” Any attempts to consider children’s rights are met with huge push back screaming about parental rights.

  3. jrkrideau says

    However, like all social science studies, I’m inclined to dismiss it.

    A bit like me, after the cold fusion & water memory findings and the CERN faster than light report I tend to discount most modern physics reports.

  4. says

    Henry Gale@#4:
    Why isn’t this type of behavior subject to evolutionary pressures?

    It could be. Or it could be a social behavior. It’s very very hard to isolate which.

  5. says

    Are you a physicist? I’m curious as to whether what you’re expressing is: “I know enough about the field that I am very unimpressed with how it’s reported” or something else.

    I have a BA in psychology, and came to be unimpressed with the field in general, but only after it was too late to change majors. Fortunately for me, it didn’t matter because I’d been programming computers for a long time in high school and didn’t have to try to make a career out of peddling psychology BS.

  6. flex says

    Actually, miss-guided altruism has been a suspicion of mine for years in the case of workplace shootings.

    My thought is that their workplace was likely an abusive place to work, and the employee was already contemplating suicide. Then the employee decides that if they were ready to commit suicide, they might as well kill the person/people they blame for driving them to that stage, as a way to help their co-workers escape from the abusive boss.

    Clearly people who think this way are not likely thinking particularly well. But there is a rationality, of a sort.

  7. Heidi Nemeth says

    I am not a lawyer, and I am not a doctor. But my husband committed suicide, as did my daughter. So I have looked into the matter a bit.

    You are psychotic if you are a danger to yourself or others. So, really, anyone who kills himself or others is psychotic. Trying to rationalize why the killer killed is enough to convince almost anyone that the killer was crazy/irrational. I spent two years speculating why my husband had killed himself. I knew I was done questioning when I felt satisfied with “It was God’s will.” I was and am an atheist (and reasonably sane)! Now I would just say he was so mentally ill he could no longer live.

    When he killed himself by jumping off a bridge, my husband was living with our teenage son. The cops told me first. Then they went with me to tell my son. For an hour my sleeping son did not respond to phone calls, knocks on the door or the ringing door bell. The cops were quietly restless, pestering me with questions about my son’s sleeping habits and responses to phone calls and door bells. Eventually a key was obtained. I went into the apartment and bedroom and awoke my son. A day later I learned the cops had been terribly worried that my husband had killed my son before he killed himself.

    One in ten people who kills himself has killed someone else first.

    I had a protection order at the time of my husband’s death. I knew I was in danger. I hadn’t thought much about our son being in danger. In any case, I felt I had no way of protecting my son. Everything I did just made my husband madder.

    Nowadays in Ohio, the spouse with a protection order automatically gets custody of all minor children. The protection order covers the children, too. In other words, the threatening spouse is not allowed any contact with the minor children.

    I would guess that filicide-suicide is way more common than is generally perceived. I know most suicides do not make the news. I would not be at all surprised if most filicide-suicides do not make the news, either. As a survivor, I did not want my husband’s or daughter’s death by suicide publicized. I felt enormous guilt when my husband committed suicide, and a large amount of guilt when my daughter did 9 years later, even though I knew neither death was my fault. That guilt made even talking about their deaths hard for years.

  8. jaxkayaker says

    For selection against such behaviors to evolve, it would be necessary for such behaviors to have a strong genetic basis, i.e. be heritable, and have a strong enough effect on differential reproductive output to reduce fitness in individuals with the genetic composition that predicts such behavior. Even then, selection is frequently not capable of completely purging such genes from a population, especially if the alleles in question are recessive in character or if the trait is polygenic.

    Minor point, but it’s odd that the t-test statistic was reported with a negative value, since the formula for the t statistic has an absolute value in the numerator.

  9. KG says

    I have a BA in psychology -- Marcus Ranum@6

    Ah, so absolutely qualified to dismiss the whole of social science.

Leave a Reply

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