Tuesday, August 31, 2010
Karl Smith notes, contra Bryan Caplan, that giving someone life is not like giving them $100 because life can't be freely disposed of. And Adam Ozimek argues, inter alia, that we can't just poll already-existing people to figure out the wishes of possible people.
And various commenters want everybody to stop talking about such obvious stuff.
Thursday, August 19, 2010
Truth is secondary. On its own, it doesn't help us survive or reproduce, and it isn't even necessarily any fun. Truth often does us more harm than good. Human beings have not evolved to actively care about and seek out truth for its own sake. (It's often beneficial to signal that one cares about truth when one does not - this, rather than mere factual inaccuracy, is the genesis of bullshit.)
Truth - and here, I mean factual veracity - is especially vulnerable in the face of legend, the kind of cultural story that invests its believers with feelings of comfort, belonging, and value. Facts that contradict important legends tend to be denied and disbelieved; uncomfortable facts are prevented from reaching consciousness, and if forcibly brought to consciousness, are resoundingly shouted down. (This strong emotional reaction - the shouting-down of facts contradictory to cherished beliefs - is valuable to those who try to serve truth for truth's sake, because the strong emotional reaction is a clue to the existence of an underlying important cultural legend or fixed belief that can, once identified, be examined.)
Legends are not necessarily universal. The pictures is more complicated. Often there are two or more competing, incompatible legends within a large society on the same topic. Subcultures form around countercultural legends, and these new legends are not necessarily any closer to truth as the majority legend.
Such is the case with beliefs about gender. There are some beliefs about gender roles and differences that are so common among human societies as to form a sort of universal legend. In modern times, feminism has provided a countercultural narrative to many of the traditional beliefs about gender (defining the prevailing norms as patriarchy). Feminism was such a popular critique that it became, over a generation or two, arguably the predominant cultural narrative about gender in modern Western society. And a counter-counter-critique which we might call the men's rights movement has challenged the feminist narrative (labeling it gynocracy - only sometimes tongue-in-cheek). From the perspective of history, we can see the back-and-forth of revolution and counter-revolution, but we must make our factual judgments in the present moment. Recognizing the narratives and counter-narratives is a first step in looking for truth - but it is by no means a final step.
Every source has its biases; I myself, as stripped of myth as I consider myself to be, no doubt operate with my own unrecognized legends forming the firmament of my consciousness. But let's try to look beneath at least the truth-threatening legends we can recognize, while remaining open to the possibility that our reasoning may be swayed by unseen narratives.
Legends exist where, for the legend believer, some fact must be true or can't be true. That is, the need to believe the legend is greater than the desire for truth (this is the origin of political correctness - a particular form of the more general class of bullshit). But compare the life work of even someone as purportedly truth-seeking as a scientist; if a scientist becomes identified with a particular theory, it must be mortally terrifying to have this theory threatened. One's life work will be undone. I think we must be circumspect about even the truth-orientation of scientists. But that doesn't make science worthless; only imperfect.
On to the Sex and Violence
I am interested here in violence in sexual relationships, or marital violence, if you prefer. Specifically, I am interested in the question of how violence by men in relationships compares to violence by women in relationships. Murray Straus and others claim that women are just as violent in relationships as men - in terms of frequency and severity of violence. Others dispute this claim. I can't pretend to be neutral, since I had formed the belief prior to writing this article that men are, in fact, more violent than women in relationships; but many people that I respect (including David Benatar) have articulated the opposite belief, and I must ardently promise (okay, signal) that I was, and still am, open for correction on this issue if presented with strong evidence.
At the outset, I ask the reader to imagine himself a sociologist, and to think about what sources we might consult to explore the question of who is more violent in relationships, and what these sources' limitations might be. As a sort of epistemological meta-issue, we would ideally like to find lots of different types of sources using different methodologies; and the more these diverse sources agree, the more confident we may be in our conclusion. Similarly, if a source has a limitation or potential bias, we must examine such limitations and biases. Assuming that a bias exists it is as unscientific as assuming there is no bias - such a claim must be examined, not assumed to be true or false.
In brief, some of the sources we might consult - and their expected limitations - are as follows: (please suggest other sources in the comments section)
- General population surveys, utilizing various methodologies. (Limitations: depends on both truthfulness and accuracy of subjects)
- Targeted surveys using various methodologies. (Limitations: findings may not reflect general population; depends on truthfulness and accuracy of subjects
- Emergency room records. (Limitations: only catches violence that produces injury, and for which treatment is sought. Lesser and unreported violence will not be counted. Cause of injury may be partly determined by report of subjects, so some of the same limitations as survey data.)
- Data on people seeking help with domestic violence (shelters, hotlines, etc.). (Limitations: only catches reported violence for which help is sought. Depends on truthfulness and accuracy of responders.)
- Police assault records. (Limitations: Only catches violence that is reported to police. Conclusions may be partially based on reports by subjects, so truthfulness and accuracy of subjects is still an issue.)
- Homicide records. (Limitations: only catches the most severe violence, which may or may not reflect general levels of violence.)
The error in these sources may be random (noise), or it may be systematic - effectively biasing the results in a particular direction. The latter type of error is more dangerous, from our perspective.
One type of systematic bias we must consider is that men may be less likely to report violence than women. It is shameful, we might hypothesize, for a man to let a woman smack him around; he might feel a bit silly going to the hospital or to police, or responding to a survey, even, especially given that human males are on average about 15% more massive than human females. This is related to the fact that, at the outset, we probably assign the greater probability to males being more violent in relationships; the idea that women are just as violent in relationships as men is counterintuitive, but that might just be because there is a huge, unrecognized epidemic of battered husbands too humiliated to speak out. It might be that the belief that men are more violent than women, that battered wives are more prevalent than battered husbands, is just a cultural legend that is ripe for toppling.
What the Case for Sexual Symmetry of Marital Violence is Based Upon
1. Conflict Tactics Scale
Beginning in the 1970s and continuing until the present day, Murray Straus began claiming that women are as violent as men in relationships. His claim in the initial studies was based on a single source of data: a survey of married or cohabiting couples, using the Conflict Tactics Scale (CTS) as its methodology. Other researchers have since joined Straus; generally, they use the exact same methodology as the studies in the 1970s, that is, the Conflict Tactics Scale. This reliance on a single source of data for confirmation is endemic to those who claim sexual symmetry in marital violence; one 2000 meta-study found that out of 82 empirical studies that found marital violence to be equal between genders, 76 of them used CTS measures exclusively.
The CTS is nearly the only methodology that produces the counterintuitive result of gender symmetry in relationship violence. Where a hypothesis is supported by only a single source of data, and contradicted by all other sources, this does not necessarily mean that the single source is wrong. It does, however, necessitate careful review of the source producing the aberrant result. I will examine the CTS shortly. But there is one other source which produces similar results, and that is:
2. American Homicide Data
In the United States, women are nearly as likely to kill their husbands as men are to kill their wives. The Sex Ratio of Killing (SROK) - the ratio of women who kill husbands (legal or de facto) to men who kill wives - is about 75 in the United States, meaning about 75 women kill their husbands or live-in boyfriends for every 100 men who kill their wives or live-in girlfriends.
And that's pretty much . . . it.
What the Case for Asymmetry is Based Upon
Evidence from criminal and divorce courts, police, women's shelters, and emergency rooms all support the hypothesis that males commit considerably more violence than females in relationships, as do survey data using methodologies other than CTS.
American SROK is Not Reflected by Homicides in Other Countries
Again, in the United States there are nearly equal homicide rates between husbands and wives (75 husband-killings for every 100 wife-killings). In every country other than the United States, however, the Sex Ratio of Killing (SROK) is quite low - in no case higher than 40, and generally much lower than that. This peculiarity of SROK data to the United States pretty much destroys its probative value for sexual symmetry of violence. As Dobash et al. note,
. . . U.S. homicide data and CTS data from several countries have been invoked as complimentary pieces of evidence for women's and men's equivalent uses of violence (citations omitted). One cannot have it both ways. If the lack of sex differences in CTS results is considered proof of sexually symmetrical violence, then homicide data must somehow be dismissed as irrelevant, since homicides generally fail to exhibit this supposedly more basic symmetry. Conversely, if U.S. homicide counts constitute relevant evidence, the large sex differences found elsewhere surely indicate that violence is peculiarly symmetrical only in the United States, and the fact that the CTS fails to detect sex differences in other countries must then be taken to mean that CTS is insensitive to genuine differences.
Advocates of the sexual symmetry of violence hypothesis point to the reporting bias - mentioned above - to justify the poor match between CTS data and all the other data (except international homicide data, which, as they cannot be dismissed as products of reporting bias, go unexplained). Men, it is posited, are less likely to seek a protective order, to call the police, to go to the emergency room, to seek help from a hotline or shelter, to press charges against an offender, and to report being victimized to the National Crime Victims Survey (though not, for some reason, on CTS). Is this factually accurate? One study found that male victims of domestic violence are actually significantly more likely than women to report domestic violence to the police - twice as likely, in fact - and less likely to drop the charges. Without data, we should not assume that a particular bias exists, any more than we should assume a particular data set is perfect and without bias. It is also a bit silly to assume that all violence against women is reported; women have strong reasons not to report marital violence, just as men do.
Why the CTS is Kind Of Retarded
1. Zero Interobserver Reliability
If CTS surveys were factually accurate, and were in fact detecting real phenomena, we would expect the reports of husbands and their wives to match. The validity of the CTS depends on people telling the truth; a necessary (thought not sufficient) condition for factual accuracy is that witnesses of the same incident agree with each other. In the case of marital violence and the CTS, they don't. A given husband and wife interviewed separately using CTS methodology are no more likely than chance to agree in their reports of violence in the relationship. From Dobash et al.:
Szinovacz (1983) found that 103 couples' accounts of the violence in their interactions matched to a degree little greater than chance. On several CTS items, mainly the most severe ones, agreement was actually below chance . . . . In a similar study, Jouriles and O'Leary (1985) administered the CTS to 65 couples attending a marital therapy clinic, and 37 couples from the local community. For many of the acts, the frequency and percentage data reported are impossible to reconcile [that is, data reported are mathematically impossible. -ed.]; for others, Jouriles and O'Leary reported a concordance statistic (Cohen's Kappa) as equalling [sic] zero when the correct values were negative. Straus (1990b) cites this study as conferring validity on the CTS, but in fact, its results replicated Szinovacz's (1983): husband/wife agreement scarcely exceeded chance expectation and actually fell below chance on some items. [Bolded emphasis mine.]
2. Conflation of Serious with Minor Violence
The Conflict Tactics scale asks responders to relate the violence they have perpetrated or experienced in their relationships according to acts: whether one has "pushed," "slapped," "kicked," etc. the other. Certain acts are grouped together as mild, moderate, or severe, regardless of the specifics of the situation or the degree of injury, if any. Throwing an object qualifies as a "severe" assault, regardless of the nature of the object, the context of the "assault," and whether the blow even landed. When case histories are examined closely, the retardedness of this classification becomes obvious. Again quoting Dobash et al.:
In a study of 103 couples, Margolin (1987) found that wives surpassed husbands in their use of "severe violence" according to the CTS, but unlike others who have obtained this result, Margolin troubled to check its meaningfulness with more intensive interviews. She concluded:While CTS items appear behaviorally specific, their meanings still are open to interpretation. In one couple who endorsed the item "kicking," for example, we discovered that the kicking took place in bed in a more kidding, than serious, fashion. Although this behavior meets the criterion for severe abuse on the CTS, neither spouse viewed it as aggressive, let alone violent. In another couple, the wife scored on severe physical aggression while the husband scored on low-level aggression only. The inquiry revealed that, after years of passively accepting the husband's repeated abuse, this wife finally decided, on one occasion, to retaliate by hitting him over the head with a wine decanter.
CTS pretends to measure, but does not actually measure, the severity of violence.
Violence Outside Relationships
Men are, across the board, more violent than women. Gender (male) is the single greatest predictor of criminal violence. In non-marital contexts, this is not at all controversial. One problem with the sexual symmetry hypothesis is that its proponents fail to provide any theory or explanation for why violence would be asymmetrical in nearly all contexts, but symmetrical in this one limited context. In fact, female violence this context is particularly in need of explanation, since it is particularly bizarre for an actor to choose a victim much larger than she, which husbands generally are. Which brings us to . . .
That Whole Sexual Dimorphism in Body Size Thing
Proponents of the sexual symmetry hypothesis generally concede that women are more likely to be injured by marital violence than men, yet still maintain that "women initiate and carry out physical assaults on their partners as often as men do." Focusing on the frequency of violence, rather than the severity and consequences, is silly and dishonest in the context of a physically mismatched pair such as a typical heterosexual couple. Imagine reading a study that found that children were as likely to hit their parents as vice versa, or elderly dependent adults were as likely to hit their caretakers as vice versa. We might say - interesting, but so what? Violence by the physically stronger party is simply more dangerous - more of a "social problem" - than violence by a weaker party. (This is aside from the issue that violence initiated by the stronger party against a weaker party is vastly more common than the reverse in other contexts, casting even more doubt on the veracity of the symmetry claim.)
However, violence against men is not nonexistent; in my experience as a domestic violence attorney, men with physical disabilities were particularly vulnerable to abuse by their female partners.
The idea of sexual asymmetry in marital violence is not just a feminist idea; it is one grounded in evolutionary psychology. Men have much more to gain from violence against their wives than women do from violence against their husbands. Violence is an effective means of achieving men's evolutionary aims (maintaining exclusive access to a woman's reproductive capacity); women are unlikely to advance their evolutionary aims by physical violence against husbands. Women must use other means of getting what they want, which brings me to my . . .
Closing Words: Toward an Evolutionary Biology of the Attack Heifer
Over the past several years in my personal life, I have been baffled by a phenomenon affecting several of my male friends. Four of my male friends are or have been married to extremely mean, unpleasant, downright emotionally abusive women. The abuse was (or is) so severe that all of my friends developed a perceptible depression; one actually began to wet his pants at work (he's still married, God save him). From an evolutionary perspective, I could understand my friends tolerating this abuse if the females in question were attractive (had high reproductive value); however, in each case, the female is significantly less attractive than the male. Two of the four women are clinically obese as well as ugly; the other two are merely ugly. Based on the baffling (to me) combination of emotional abusiveness and ugliness, I have termed this surprisingly common beast the attack heifer.
My male friends in question are characterized by innocence, lack of experience, and early age at marriage. They are also characterized by IQs within the top tenth of a percent of American adults (easily). Why did (or do) they stay?
My thinking is that this phenomenon - an objectively less "valuable" mate being crappy to a more "valuable" mate - is analogous to the fact that a man is more likely to kill a young wife than an older wife, and that age difference is a strong predictor of interspousal homicide. A young wife has a higher reproductive value than an older wife; the husband of a young wife has "more to lose," and is more likely to use violent tactics to prevent or punish his wife's infidelity or attempts to leave the marriage.
Similarly, attack heifers perceive that they have a great deal to lose, and escalate their tactics to maintain control within the relationship - not through violent means, which wouldn't be effective anyway, but using skills at which women surpass men: emotional manipulation. Physical violence is not necessary to cause extreme suffering. Women do not need to be as violent as men to be as evil as men.
1. I dislike the fuzziness of the concept of "romantic relationships," and marriage is too limiting for what we are discussing, which includes de facto marriages as well as legal unions. I tend to reduce this kind of relationship to its sexual component, which, to me, is what separates romantic relationships from friendships. The special value of sex compared to other, more nebulous concepts involved in romantic relationships, and the questionable value of monogamous relationships, are certainly something of legends for me, albeit examined ones.
2. Straus, Murray A. "Measuring intrafamily conflict and violence: The Conflict Tactics (CT) Scales." Journal of Marriage and the Family 51:75-88 (1979).
3. Archer, John. "Sex Differences in Aggression Between Heterosexual Partners: A Meta-Analytic Review." Psychological Bulletin 126.4:651-680 (2000).
4. Kimmel, Michael. "'Gender Symmetry' in Domestic Violence: A Substantive and Methodological Research Review." Violence Against Women 8.11:1332-1363 (2002).
5. Wilson, M. I., and Daly, M. "Who kills whom in spouse killings? On the exceptional sex ratio of spousal homicides in the United States. Criminology 30:189-215 (1992). No, it's not because of guns, and it's not because American women are more violent in general. Read the paper - it's fascinating.
6. Dobash, R.P., Dobash, R.E., Wilson, M., Daly, M. "The myth of sexual symmetry in marital violence." Social Problems 39:71-91 (1992).
7. Felson, Richard. "Big People Hit Little People: Sex Differences in Physical Power and Interpersonal Violence." Criminology 34.3:433-452 (1996).
8. Henning, Kris, and Lynette Felder. "A Comparison of Men and Women Arrested for Domestic Violence: Who Presents the Greater Threat?" J. Family Violence 19.2:69-80 (2004).
9. There are lots more, of course, but I
10. Kinkaid, Pat. The Omitted Reality: Husband-Wife Violence in Ontario and Policy Implications for Education. Maple, Ontario: Learners Press, 1982. Cited in Dobash et al. (1992), supra.
11. Szinovacz, Maximiliane. "Using couple data as a methodological tool: The case of marital violence." Journal of Marriage and the Family 45:633-644 (1983).
12. Jouriles, Ernest, and Daniel O'Leary. "Interspousal reliability of reports of marital violence." Journal of Consulting and Clinical Psychology 53:419-421 (1985).
13. Straus, Murray. "The Conflict Tactics Scales and its critics: An evaluation and new data on validity and reliability." In Physical Violence in American Families, ed. Murray A. Straus and Richard J. Gelles, 49-73. New Brunswick, N.J.: Transaction Publishers (1990).
14. Margolin, Gayla. "The multiple forms of aggressiveness between marital partners: How do we identify them?" J. Marital and Family Therapy 13:77-84 (1987).
15. Straus, Murray. "Physical Assaults by Wives: A Major Social Problem." In Current Controversies on Family Violence, Richard Gelles and Donileen Loseke, eds. Newbury Park, California: Sage Publication (1993).
16. Daly, Martin, and Margo Wilson. Homicide. New York: Aldine de Gruyter (1988); see especially pp. 205-207.
17. Wilson, Margo, and Martin Daly. "Lethal and nonlethal violence against wives and the evolutionary psychology of male sexual proprietariness." Pp. 199-230, in Violence Against Women: International and Cross-disciplinary Perspectives, Dobash & Dobash, eds. Thousand Oaks, California: Sage Publications (1998).
18. "A man will rip off your arm and throw it into a river, but he will leave you as a human being intact. He won't mess with who you are. Women are non-violent but they will shit inside of your heart." - Louis CK
Thursday, August 12, 2010
Carroll's reader laments:
I . . . find it frustrating and upsetting when people make me feel I’m wrong or crazy when I, very diplomatically, describe a more rational, objective, or philosophical explanation, when other people follow irrational paths.
Carroll correctly notes that rationality, the practice of critical reasoning, is a very unnatural mode for humans - it's not what we evolved to do. Valuing truth above one's own interests is hardly evolutionarily beneficial behavior. And people don't tend to admit that you've changed their minds.
But, Carroll says, rational argument has several major purposes, even if it doesn't seem to change anyone's mind: first, argument benefits us directly by promoting our own truth-seeking function:
It is pleasurable to seek out the best evidence available and construct the best argument possible. It is pleasurable to explore a strong argument that goes against what you believe. Either you find weaknesses and fallacies in the argument (strengthening the confidence in your conviction) or you realize the error of your ways. Either way, you benefit. Examining arguments, especially arguments that seem counterintuitive, is the only way we can arrive at the most reasonable beliefs possible.
Mostly, though, argument serves the purpose of (a) potentially changing an observer's mind (especially important for web arguments), (b) changing an interlocutor's mind later, when face-saving is no longer an issue; and (c) figuring out whether we ourselves might be wrong. Carroll says:
The dynamics of changing minds are complex, but I hope for two things by confronting the errors of others in a public forum: I hope they will later reconsider their views in light of the evidence and arguments I present, and I hope others who are not directly in the fray, but who are interested in the subject and interested in getting it as right as possible, will read the discussion and see that I have the better evidence and arguments. I also remain open to the possibility that I might be wrong and that some observer will provide me with the evidence and argument to show me the error of my ways.
Take heart, fellow antinatalists and other thought criminals.