{[ promptMessage ]}

Bookmark it

{[ promptMessage ]}

DalyWilson.1988.Homocide.8-1

DalyWilson.1988.Homocide.8-1 - The Logic of Same-Sex...

Info icon This preview shows pages 1–13. Sign up to view the full content.

View Full Document Right Arrow Icon
Image of page 1

Info icon This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
Image of page 2
Image of page 3

Info icon This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
Image of page 4
Image of page 5

Info icon This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
Image of page 6
Image of page 7

Info icon This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
Image of page 8
Image of page 9

Info icon This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
Image of page 10
Image of page 11

Info icon This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
Image of page 12
Image of page 13
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: The Logic of Same-Sex Conflict ——8——- ’" Men become embroiled in dangerous competitive interactions far more ”T -- often than do women. This is a sex difference that we share with other ’1 ‘ effectively polygynous mammals, and there is no mystery about why 1 , this should be so: As we saw in Chapter 7, the sex difference in fitness ,. , variance means that mammalian males typically compete for bigger prizes than do females. Bigger prizes warrant bigger gambles. As the variance in individual fitness associated with competitive H’ ~ success or failure increases—as the outcomes for winners and losers ’9 become more disparate—the temptation to resort to dangerous compet- itive tactics is also likely to increase. A simulation model may clarify this point for some readers, and so we present such a model in the following box. The simulation, with its arbitrary parameters, is not a realistic ' description of a competitive situation. Rather, it is intended to demon- - strate in principle that the competitive strategy that is favored by natural '1 selection can vary according to the exact relationship between compet- itive success and fitness. In particular, selection may favor behavior that is increasingly dangerous, both to one’s opponent and to one’s self, as » fitness differentials increase. - t Throughout human history, the winners have got the women, and L: the biggest winners—Moulay Ismael the Bloodthirsty is an extreme ase—have got the most women. But when we consider the psychology of dangerous risk-taking, what is possibly even more crucial than the chance of winning big is the high probability of losing totally. Any creature that is recognizably on track toward complete reproductive failure must somehow expend effort, often at risk of death, to try to _ mprove its present life trajectory. If, for example, the energetic return ' from foraging in a predator-free area is sufficient to maintain oneself but nsufficient to breed, then the only foragers to leave descendants will be those willing to forage outside the predator-free area, even though they must therefore incur some predation risk. If one’s present status guarantees lifelong exclusion from mating opportunities, then one must strive to raise that status. Wilson and Herrnstein (1985) have suggested / 163 164 Box 8.1 Fitness Payoffs of Risky Competition: A Simulation This simulation illustrates the increasing utility of dangerous ' competitive tactics as the prize for victory increases. The parame- ters have no real-life significance, but were chosen for expository convenience. In the simulation, each individual possesses one of three possi-‘ ble competitive strategies: high-risk, medium-risk, low-risk. Compet- 2 itive interactions take the form of fights between two individuals. In every fight, there is one winner. The other contestant may.- simply lose, or may be killed. , When two low-risk individuals fight, each has a 0.5 probability ,- of winning and there is no risk of death. A riskier strategy raises its practitioner's chance of winning, but also raises both contestants’ ‘, chances of being killed. Thus, for example, a medium-risk individ- ual has a 0.6 probability of winning a tight with a low-risk”: individual, but each contestant also has a 0.05 probability of death. ‘ The complete matrix of outcome probabilities is this: —continued— " ‘1 Logic of Same—Sex Conflict ‘ I . Risky Competition A’s outcome B’s outcome Win Lose Die Win Lose .50 .10 .40 .50 .10 .60 .10 .30 .40 .30 .70 .10 .20 .30 .50 Med. .40 .30 .30 .60 .10 Med. .50 .40 .10 .50 .40 Med. Low .60 .35 .05 .40 .55 Low High .30 .50 .20 .70 . 10 Low Med. '. 40 .55 .05 .60 .35 Low Low .50 .50 .00 .50 .50 _—___________.—.——-———————- The simulation begins with a first generation of 60 individuals: 20 of each risk-type. Everyone starts with a score of 0. Pairs of ,surviving individuals are randomly selected to fight, and the fight’s result is generated randomly according to the probabilities in the above matrix. The winner gets one point added to his score, and the loser gets one subtracted. This process continues until either 90 fights have been completed or only 10 survivors remain. The survivors are then the parents of the next generation of 60 individuals, according to a rule relating competitive success to fitness. This rule is what is varied below. —continued-— 165 166 Offspring inherit the parental risk-type (as in asexual reproduc- tion, or a trait carried on a Y chromosome). Any risk-type may thus gain or lose in numerical representation from one generation to the next. The entire process is repeated, generation after generation, until the population consists entirely of a single risk-type (”fixa- tion”) and the other two are extinct. Rule 1. Survivors: All surviving individuals have equal fitness. Twenty simulations were run with this rule and the low-risk strategy went to fixa- tion every time (requiring anywhere from 18 to 57 gen- erations to do so). This result should not be surprising. The only objective under this rule is survival. A loss is as good as a win. There is therefore no utility in taking risks in order to win. Rule 2. Winning survivors: All survivors with a positive score (i.e., wins exceed losses) have equal fitness. This rule is perhaps analo- gous to a territorial system, in which some individuals hold a resource that permits mo- nogamous breeding status, and others do not. There is no premium on running up a large positive score, but one must win at least one more contest than one loses. Twenty simulations were run with this rule and the medi- um-risk strategy went to fixa- tion 18 times, with each of the others going to fixation once. No. of Individuals No. ot Individuals Logic of Same-Sex Conflict ' V 60 50 40 30 20 10 13 5 7 9 1113151719212325 Generations 60 50 Medium risk 40 30 20 10 13 5 7 9 ”13151719212325 Generations —continued— Risky Competition 167 Rule 3. Linear: Survivors with a positive score have fitness proportional to that score. 50 This rule is more like a domi- nance hierarchy with greater fit- ness the reward for a higher aggressive ranking. Twenty sim- ulations were run with this rule. The high-risk strategy went to fixation 9 times, and the medi- um-risk strategy 11 times. 5 O Medium risk No. of Individuals 13 5 7 91113151719212325 Generations Rule 4. Squared: Survivors with a positive score have fitness proportional to the square of that score. . . . 60 This rule raises the premium on victory still higher, producing the greatest fitness variance yet. Twenty simulations were run with this rule, and the high-risk strategy went to fixation 16 times, the medium-risk strategy 4 times. 50 High risk 40 30 20 No. of individuals Medium 13 5 7 9 1113151719212325 Generations What this simulation shows is that selection can favor relatively risky competitive strategies despite substantial mortality costs. Under all four rules, high-risk individuals experienced the highest mortality during the fighting phase, and low-risk individuals _ experienced the lowest mortality. But surviving the fighting phase is only the first step to fitness, and the payoffs for high positive scores can offset the mortality risks. The same principle applies to competition in the real world. 168 Logic of Same-Sex Conflict that men who engage in predatory violence and other risky criminal activity have different ”time—horizons” than law-abiding men, weighing the near future relatively heavily against the long term. Several lines of evidence support this idea. What Wilson and Herrnstein do not empha— size is that such adjustment of one’s personal time-horizons is probably an adaptive response to predictive information about one’s prospects for longevity and eventual success. Natural selection will especially tend to favor risk-proneness in circumstances where one's anticipated life trajectory, in the absence of I risk, is so poor that one has little or no expected fitness to lose. As a particular example of this general proposition, dangerous competitive tactics are predicted to be especially prevalent within those demographic categories in which the probability of reproductive failure is high. The Demography of Homicide Figure 8.1 presents the risk of becoming involved in a homicide in Canada, either as killer or victim, as a function of age and sex. The sex difference in homicidal violence requires no further discussion, but what about age? Killing is concentrated among young men, and so, to a lesser extent, is the risk of being killed. Men appear to be most conflictual in late adolescence and young adulthood, a life stage at which competitive f striving to achieve status, resources, and marriageability has surely been essential in historical societies, and probably still is. It may be argued, then, that a willingness to resort to violence has been most strongly favored in the age—sex class that has historically experienced the most intense sexual selection. Conversely, however, one might argue that where two men find themselves similarly disenfranchised—unem- ployed, broke, isolated from relatives—that it is the older, not the younger, who has less to lose and should therefore be readier to resort to such dangerous tactics as armed robbery or violent confrontation. . An empirically based choice between these alternative Darwinian models is not so easily achieved as one might initially suppose. It is clear that young men are more often violent than their elders, and many writers have implied that this difference reflects a maturational change in inclinations and attitudes. But of course circumstances change over the life span, too. If celibacy and childlessness, for example, are circumstan- tial factors contributing to risk-proneness, then the proportion of men who are risk—prone will decline with age simply because the proportions unwed and childless decline. Whether there is a significant aging effect upon risk-proneness and violence, over and above the effects of Risky Competition 60 50 Male vlctlms 40 30 20 10 0 100 90 Male ottenders 80 70 60 50 40 30 20 Homicides per million persons per annum 10 Age (years) Figure 8.1. Age- and sex-specific homicide rates in Canada, 1974—1983. 169 170 Logic of Same—Sex Conflict circumstantial variables associated with age, is a question that is wide open for future research and well worth pursuing. It would be extremely interesting to knowto what extent a man’s risk—proneness declines when he becomes a father, if at all. Unfortu— nately, information on the incidence of fatherhood—either in the population-at-large or in such criterion groups as homicide offenders— is nonexistent. Some other demographic variables that may be related to present and expected fitness can be assessed, however, namely employ- ment status and marital status. Our study of homicide in Detroit in 1972 provides the best available information on the relationships of various demographic factors to the chance of becoming involved in a homicide. The age- and sex-specific rates of such involvement are shown in Figure 8.2. The distributions are roughly similar in shape to those for Canada (Figure 8.1), but note that the rates are substantially higher. Note too that the Detroit male victimization rate is more conspicuously age-dependent than the Cana- dian male victimization rate. Indeed, Detroit victim and offender distri- butions look very similar to one another, a fact that reflects the prevalence of ”trivial altercations" (see Chapter 6) in this sample. In Detroit in 1972, 43% of adult male homicide victims and 41% of adult male offenders were unemployed, compared to 11.2% of all adult men in the city (Figure 8.3). Sixty-nine percent of adult male victims and 73% of adult male offenders were unmarried, compared to 43% of all adult men in the city (Figure 8.4). Thus, it is indeed the case that circumstan- tial variables are related to the probability of becoming involved in lethal violence, even in comparisons within an age—sex class. The victim and offender populations are remarkably alike, and this similarity extends to other variables too. Among both male victims and male offenders, 36% had previous criminal records (excluding convic- tions for motor vehicle violations, drunkenness, and narcotics offenses). The high proportion of altercations among these homicides is largely responsible for this similarity between killers and victims: In this most prevalent type of conflict leading to homicide in Detroit, the hostilities are reciprocal, and it is often an open question which party will end up dead. But of course not all homicides involve such reciprocity, and further analysis of the factors associated with becoming a ”participant” in homicide will require some differentiation of motive categories. The Problem of “Motive" “Altercation of relatively trivial origin; insult, curse, jostling, etc." That was Wolfgang’s (1958) characterization of the predominant ”mo- ”Motive” 171 2000 Male victims. 1 000 2000 1000 Homicides per million persons per annum Female otlenders vmvmvmvm mvmvm I'v-v-NNMM mmwcgg’gg omolul’olul,llllllll|llN omomomom Fa—annvvmmowrepg Age(years) Figure 8.2. Age- and sex-specific rates of homicide in Detroit, 1972. (From Wilson 8: Daly, 1985.) . tive" in American homicide, and all subsequent students of the subject have been influenced by his categories. The description of altercations and the recognition of their prevalence were important insights, but it is debatable whether they have truly illuminated the question of motive, or have instead deflected attention away from that question. The essence of the “altercation” category is the spontaneity of the dispute, its interactive face-to-face character, and its uninterrupted development to a dénouement. These elements are crucial to police and prosecutors, because of the paramount issues of “premeditation” and ”provocation” in determining the charge under which a homicide can be successfully prosecuted. If a verbal insult provokes a direct violent 172 Logic of Same-Sex Conflict Population-at-Iarge m Ottenders so [:1 Vlctlms ' Percent unemployed 16—19 20—24 25-44 45-54 55-64 265 Age In years Figure 8.3. Unemployment rates among male homicide offenders, male victims, and the male population-at-large in Detroit, 1972. (From Wilson 8: Daly, 1985.) response, for example, then the killer is guilty of manslaughter at most, and the motive will be recorded as an ”argument" or ”altercation.” But should the identical insult inspire its recipient to brood for a week and then to ambush his antagonist, the prosecutor can bring a charge of murder, and the case will be assigned to some other motive category such as ”revenge.” Thus, the category of ”altercations" cuts across more substantive issues. The motive categories that Wolfgang invented for his Philadelphia study are listed in Table 8.1. His list has certain virtues: The categories are broad enough to encompass most cases, and they have a certain face validity, corresponding roughly to the recurring ”types” of cases that experienced homicide detectives recognize as familiar. But Wolfgang’s list is a conceptual hodgepodge. His two leading categories do not refer to the substantive issues at all, and ”domestic quarrel" is the only ”motive” on the list that is defined by the particular relationship between killer and victim. Even the more substantive issues are problematic because of the way they cross-cut other motives. Sexual rivalry and/or infidelity are apparently at issue in all 68 ”jealousy” cases, for example, but this same substantive issue could also be the point in a case ”Motive” 173 E Populatlon-at-Iarge m Offenders l: Vlctlms 100 80 60 40 Percent not married 20 14-19 20-24 25—34 35-44 45-54 55—64 Agetnyears Figure 8.4. Proportions unmarried among male homicide offenders, male victims, and the male 0 ulation-at-l ' ' Wilson 45: Daly, 1985.) p p arge in Detr0it, 1972. (From categorized as a ”domestic quarrel" or an ”altercation of relatively trivial origin. Wolfgang’s unsatisfactory taxonomy of motives remains pretty much the state of the art. Most subsequent students of homicide have adopted his categories, have modified them slightly, or have ignored the question altogether. The upshot is that the prevailing criminological conception of motives in homicide is a woolly amalgam of several potentially independent dimensions: spontaneity versus premeditation the Victim—offender relationship, and only a relatively small dose of those substantive issues that murder mystery writers and ordina speakers of English mean when they speak of ”the motive ” Thri: unsatisfactory state of affairs exists partly because the popular concep- tion of a motive is primarily appropriate to premeditated murders, and not to the sort of reactive, unplanned assaults that produce most homicides. In a murder mystery, the question of motive is “Why did Killer want Victim dead?” In real life, Killer often did not especially want Victim dead at all, or at least did not approach the conflict in those terms. Nevertheless, simply because a homicide occurs spontaneously 174 Logic of Same-Sex Conflict Insult and Redress 175 Table 8.2. Five Hundred Twelve Closed Homicide Cases in the City Table 8.1. Motive Categories and the Number of Cases (Victims) of Detroit, 1972' Classified by the Type of Case and by the Within Each, for 588 Criminal Homicides in the City of 1 h 1948—1952 ViGilli-Offender Relationship“ Philade p ia’ a _—______—________——————————-——— “Motive" Number of cases Percentage of total lype of case // i - SOCial At V y 206 35 0 l CI me triVial origin; insult, - tl' , etc. Genealogical relatives 1 31 0 curse,‘ ’05 mg! 83 14.1 Spouses 1 79 0 Domestic quarre 11 6 _ . ‘ Jealousy 68 10-5 Marital relatives (affine, step) 0 15 0 Altercation over money 62 68 Friends or acquaintances 47 193 3 Robbery :3 5:3 Strangers 119 19 0 [Reveéignetal 23 3.9 Unknown __0 _.Z .2. Seclcfideense 8 1-4 Total 168 339 5 Halting of felon 7 1%) “From Wilson and Daly (1985), Table 1. Escaping arrest 6 1‘ 0 Concealing birth 6 3' 4 , Other 20 4' 8 7‘ ‘ tion has been drawn between “social conflict” cases and those incidental 28 - " , to the commission of another crime (which usually means robbery). Just such a classification—”social conflict" versus "crime-specific”— was employed by sociologist Marie Wilt (1974) in her study of the 690 1,; homicides that occurred in Detroit in 1972, a study that preceded our "' own research on the same cases. Out of 512 solved cases, about one-third were adjudged to have been crime-specific, and about two- thirds to have been social conflicts (Table 8.2). Wilt originally classified the Detroit social conflict cases into four categories, which she had derived not from any particular theory but .7 simply from a preliminary reading of 100 cases. These four types she ‘ referred to as ”jealousy conflicts,” ”business conflicts," ”family con- ' flicts,” and “arguments between friends, acquaintances, or neighbors." - Like Wolfgang’s categories, then, Wilt’s were an uneasy mixture: partly ; defined by the substance of the dispute, and partly by the relationship between the parties. For that reason, we retained her substantive V categories of ”jealousy” and ”business,” but then tried to classify the 3 remaining cases under similarly substantive issues. The resulting clas- sification is shown in Table 8.3. More than half of the cases summarized in Table 8.3 (124 of 212) fall into one of the first two categories, both of which may be considered matters of status competition and the maintenance of face. These are homicide motives that we discussed at some length in Chapter 6. Most Unknown ”From Wolfgang (1958). - and with questionable intent to kill, it does not follow that “altericatiop or ”domestic quarrel" is an adequate characterization of the. ispug. Violence arises from conflicts about something, difficult thou...
View Full Document

{[ snackBarMessage ]}