Spouse Murder Defendants in Large Urban Counties

U.S. Department of Justice Office of Justice Programs

Bureau of Justice Statistics Spouse Murder Defendants in Large Urban Counties

In spouse murder cases, wife defendants were less likely to be convicted and to receive severe sentences than husband defendants

11 % Not prosecuted Probation 5 % 46 % Pleaded guilty Of 318 husbands Sentenced to Jail 1 % 41 % Convicted at trial Prison 81 % 2 % Acquitted at trial (Average prison term: 16.5 years)

16 % Not prosecuted Probation 12 % 39 % Pleaded guilty Of 222 wives Sentenced to Jail 1 % 31 % Convicted at trial Prison 57 % 14 % Acquitted at trial (Average prison term: 6 years) U.S. Department of Justice Office of Justice Programs Bureau of Justice Statistics

Spouse Murder Defendants in Large Urban Counties

By Patrick A. Langan, Ph.D. BJS Statistician and John M. Dawson former BJS Statistician

September 1995, NCJ-153256 Contents

U.S. Department of Justice Highlights iii Bureau of Justice Statistics Survey of murder defendants 1 Jan M. Chaiken, Ph.D. Director Characteristics of spouse murder defendants 2

Spouse murder cases disposed by the prosecutor 5

Patrick A. Langan, Ph.D., Statistician, Spouse murder cases disposed Bureau of Justice Statistics, and John by trial verdict 6 M. Dawson, former BJS Statistician, wrote this report. Substantial assis- Spouse murder cases disposed tance was provided by Lawrence A. by guilty plea 11 Greenfeld. Tom Hester edited and Jayne Robinson produced the report. Overview of case disposition of spouse murder defendants 15 Data presented in this report may be obtained from the National Archive of Sentencing of defendants Criminal Justice Data at the University convicted of killing their spouse 17 of Michigan, 1-800-999-0960. The name of the dataset is Murder in Large Processing time in spouse Urban Counties, 1988 (ICPSR 9907). murder cases 20

Less severe dispositions for wife defendants: Summary of evidence 21

Race in spouse murder cases 24

Methodology 25

ii Spouse Murder Defendants in Large Urban Counties Highlights

Number of spouse murder Of the 540 spouse murder defendants, An estimated 12% of the 431 convicted defendants and their demographic 232  or 43%  pleaded guilty to kill- spouse murderers received a sentence characteristics ing their spouse and 238  or 44%  to life imprisonment and 1% received pleaded not guilty and stood trial. The the death penalty. In 1988, the justice system in the remaining 70 persons  or 13%  Nation's 75 largest counties disposed were not prosecuted. Excluding life and death sentences, of an estimated 540 spouse murder the average prison term imposed cases. Husbands charged with killing Outcome for spouse murder was 13 years. their wife outnumbered wives charged defendants who pleaded not with killing their husband. Of the 540, guilty and stood trial Wife defendants less likely to be 318  or 59%  were husband convicted defendants and 222  or 41%  Of the 238 who pleaded not guilty, were wife defendants. 63% were tried by a jury and the Wife defendants had a lower convic- remaining 37% were tried by a judge. tion rate than husband defendants  Blacks comprised 55% of the 540 Together, judges and juries acquitted Of the 222 wife defendants, 70% defendants and whites comprised 16% of the 238 spouse murder defen- were convicted of killing their mate. 43%. Among husband defendants dants and convicted 84%  or 199 By contrast, of the 318 husband 51% were black and 45% were white. persons  of killing their spouse. defendants, 87% were convicted Among wife defendants 61% were Bench trials (trials before a judge) had of spouse murder. black and 39% were white. In 97% of a higher acquittal rate than jury trials: the murders, both spouses were the 26% of bench trials ended in acquittal, Of the 100 wife defendants tried same race. versus 11% of jury trials. by either a judge or jury, 31% were acquitted. But of the 138 husbands Ages of spouse murder defendants Defendants convicted of killing tried, 6% were acquitted. ranged from 18 to 87. The average their spouse age was 39. The average age of hus- Of the 59 wife defendants tried by a band defendants was 41; of wife Of the 540 spouse murder defendants, jury, 27% were acquitted. But of the defendants, 37 years. 431 (or 80%) were ultimately convicted estimated 91 husband defendants tried of killing their spouse. Their conviction by a jury, none was acquitted. Arrest charge was the result of either pleading guilty (232 persons) or being convicted at Convicted wife defendants First-degree murder was the most fre- trial (199 persons). sentenced less severely quent charge at arrest, accounting for 70% of defendants. In descending While most persons arrested (70%) An estimated 156 wives and 275 order of seriousness, charges were for spouse murder were charged with husbands were convicted of killing their distributed this way across the 540 first-degree murder, most persons spouse. Convicted wives were less spouse murder defendants: convicted (52%) of spouse murder had likely than convicted husbands to be 70% first-degree murder negligent or nonnegligent manslaugh- sentenced to prison, and convicted 24% second-degree murder ter as their conviction offense. wives received shorter prison sen- 6% nonnegligent manslaughter tences than their male counterparts  Sentences for defendants convicted 81% of convicted wives but 94% How the justice system disposed of killing their spouse of convicted husbands received a of spouse murder cases prison sentence. Of the 431 defendants convicted of kill- Cases were disposed of in one of three ing their spouse, 89% were sentenced ways: On average, convicted wives to a State prison, 1% were sentenced received prison sentences that were (1) the prosecutor declined to to a county jail, and the remaining 10% about 10 years shorter than what prosecute; or received a sentence of straight proba- husbands received. Excluding life or (2) the defendant pleaded not guilty, tion (no prison or jail confinement). stood trial, and was either acquitted death sentences, the average prison or convicted; or sentence for killing a spouse was 6 (3) the defendant pleaded guilty. years for wives but 16.5 years for husbands. Spouse Murder Defendants in Large Urban Counties iii Among wives sentenced to prison, The average prison sentence for For conviction for nonnegligent 15% received a sentence of 20 years unprovoked wife defendants was 7 manslaughter, the average prison term or more (including life imprisonment years, or 10 years shorter than the (excluding life sentences) was 8 years and the death penalty); among hus- average 17 years for unprovoked in white-victim cases, not significantly bands, it was 43%. husband defendants. different from the average 6 years in black-victim cases. Victim provocation more often Victim's race unrelated to outcomes present in wife defendant cases Defendant's race unrelated to The victim was black in 55% of cases outcomes According to information contained in and white in 43%. The likelihood of a prosecutor files, more wife defendants defendant being convicted of spouse The likelihood of conviction, and of a (44%) than husband defendants (10%) murder was about the same whether prison sentence if convicted, and the had been assaulted by their spouse the murder victim was white or black. length of the prison sentence were (threatened with a weapon or physi- Among spouse murder defendants about the same whether the spouse cally assaulted) at or around the time whose victim was white, 81% were murder defendant was white or of the murder. convicted. Among those whose victim black  was black, 79% were convicted. Self-defense as possible explana- 78% of white defendants were con- tion for wives' lower conviction rate Likewise, the sentence was unrelated victed, not significantly different from to the victim's race. The likelihood of a the 80% of black defendants. In certain circumstances, extreme convicted spouse murderer receiving a victim provocation may justify taking a prison sentence was about the same Among convicted spouse murderers, life in self-defense. Provocation was whether the murder victim was white 93% of white defendants were sen- more often present in wife defendant or black: the convicted spouse mur- tenced to prison, not significantly cases, and wife defendants were less derer was sentenced to prison in 93% different from the 88% of black de- likely than husband defendants to be of cases where the victim was white, fendants. convicted, suggesting that the rela- not significantly different from the 87% tively high rate of victim provocation of cases where the victim was black. Processing time characteristic of wife defendant cases The length of the prison sentence was one of the reasons wife de- imposed on a convicted spouse mur- Three measures of processing time fendants had a lower conviction rate derer was generally unrelated to were taken from the day of the murder than husband defendants. Consistent whether the murder victim was white  to arrest, to indictment, and to final with that, of the provoked wife de- or black  disposition. Most spouse murder defendants, 56% were convicted, signifi- fendants were arrested on the same For conviction for first-degree cantly lower than either the 86% con- day the killing occurred. Average time murder, the average prison term viction rate for unprovoked wife de- to indictment was 4 months. Average (excluding life and death sentences) fendants or the 88% conviction rate time to final disposition was almost ex- was 29 years in white-victim cases, not for unprovoked husbands. actly 1 year. significantly different from the 32 years in black-victim cases. No explanation for why State prison For husbands tried by a jury, 12 ½ sentences were, on average, 10 months was the average elapsed time For conviction for second-degree years shorter for wife defendants from the day of the murder to the con- murder, the average prison term (ex- than husband defendants clusion of the jury trial. For wives tried cluding life sentences) was 19 years by a jury it was significantly longer, in white-victim cases, significantly Wives received shorter prison sen- about 18 ½ months. longer than the 13 years in black-victim tences than husbands (a 10-year dif- cases. However, 23% of convicted ference, on average) even when the second-degree murder defendants in comparison is restricted to defendants black-victim cases received a sentence who were alike in terms of whether or of life imprisonment, compared to 8% not they were provoked  of defendants in white-victim cases.

iv Spouse Murder Defendants in Large Urban Counties Survey of murder defendants

In 1988 State prosecutors, judges, and The 75 largest counties are where a juries across the 75 most populous little over half of all murders in the U.S. counties convicted, acquitted, or Nation occur.1 Consequently, survey otherwise disposed of an estimated results summarized in this report have 9,600 murder defendants. Six percent broad relevance because they are  or an estimated 540 defendants  from the courts where the majority of were persons charged with murdering the Nation's murder trials are held. their spouse. This case processing study of spouse murder defendants is the most geo- Information on the 540 comes from a graphically comprehensive study on representative sample of murder cases the topic. While the survey covers disposed of in 1988. The sample was murder cases processed 7 years ago drawn from State prosecutor files in 33 in 1988, the Bureau of Justice Statis- of the 75 counties. The counties were tics knows from long experience with widely scattered, from Los Angeles surveying courts that changes in case and San Diego, Denver and Dallas, to processing are quite gradual. The re- Philadelphia and Dade County (Miami). port's results are therefore likely to be For each defendant, data collectors applicable today. filled out a lengthy questionnaire and prepared a brief narrative from file in- The same database used in this formation. Prosecutor files include report was previously analyzed by such items as the police arrest report, John M. Dawson and Barbara Boland investigator reports, and information on (Murder in Large Urban Counties, how the case was disposed. Ques- 1988, BJS Special Report, tionnaires and narratives are the NCJ-140614, May 1993) and by John sources of data for this report. M. Dawson and Patrick A. Langan (Murder in Families, BJS Special Report, NCJ-143498, July 1994). 1The Nation consists of 3,109 counties; thus, murders are concentrated in a relatively small number of places.

Spouse Murder Defendants in Large Urban Counties 1 Characteristics of spouse murder defendants

Sex of defendant white husbands killing black wives, Nonnegligent manslaughter is willful white wives killing black husbands, or killing, without premeditation, in the Husbands outnumbered wives as black wives killing white husbands  heat of passion. spouse murder defendants. Of the accounted for the remaining 2%.3 540 defendants, 318  or 59%  Less serious classes of homicide  were husbands charged with killing Ages of spouse murder defendants namely, negligent manslaughter and their wife, and 222  or 41%  were ranged from 18 to 87. On average, justifiable homicide  were not wives charged with killing their hus- spouse murder defendants (mean band (table 1). By definition, the 540 age 39) were considerably older than Table 1. Spouse murder defendants included no divorced couples but did murder defendants overall (mean in 75 largest counties, 1988 include both common-law and sepa- age 28). The average age of husband 2 Murder Total Percent rated spouses. defendants was 41 years; of wife defendant number of total defendants, 37 years. All 540* 100% Year and location of murder Arrest charge Husband 318 59 About eighty percent of the alleged Wife 222 41 * Excludes divorced persons but does include murders that were disposed of in 1988 The sample consisted exclusively of common-law and separated spouses. occured in 1988 or 1987 (table 2). The persons charged with (in descending rest were earlier, dating to 1975 in one order of seriousness) first-degree case. Nearly 90% took place in the murder, second- (or third-) degree Table 2. Spouse murder defendants: home. murder, or nonnegligent (or voluntary) Year and location of murder, and time manslaughter  to arrest, 1988 Time to arrest Percent of spouse First-degree murder is premeditated murder defendants murder (for example, murder by In about 80% of cases, defendants poisoning, murder while lying in wait) Year of murder All 100% were arrested either on the day of the or felony murder (for example, rape- murder or on the following day. 1988 30 murder, robbery-murder). 1987 48 1986 14 Murder weapon 1985 3 Second- (or third-) degree murder is 1984 4 all other murder. Pre-1984 1 More of the wives (95%) than the Murder location 3 husbands (69%) used a gun or knife Historical data for the United States indicate that All 100% to inflict death (table 3). For a smaller the risk of spouse murder is over 7 times greater Victim's home 86 person to kill a larger person (such as for spouses in interracial than intraracial mar- Elsewhere 14 riages. See James A. Mercy and Linda E. Saltz- for a wife to kill her husband), a lethal man, "Fatal Violence among Spouses in the How long after murder defendant was arrested weapon is generally required. United States, 1976-85," American Journal of All 100% Public Health, May 1989, Vol. 79, No 5. Same day 62 Race and age of defendants Next day 17 Later 21 Of the 540 defendants, 55% were black and 43% were white. Among Table 3. Spouse murder defendants: Weapons used, 1988 husbands charged with spouse murder, 51% were black and 45% were Percent of spouse murder defendants Firearms white (table 4). Among wives charged, Murder Total All 61% were black and 39% were white. defendant number Total types Handgun Shotgun Rifle Knife Other* Ninety-seven percent of the murders were intraracial. One percent were All 540 100% 53% 45% 4% 4% 26% 21% black husbands killing white wives. Husband 318 100 50 41 6 3 19 31 Other interracial murders  either Wife 222 100 58 51 2 5 37 5

2 Case narratives indicate that 8% were separated, *Hands, feet etc. but the actual figure was probably greater than that because separation status was probably not always recorded. 2 Spouse Murder Defendants in Large Urban Counties sampled. However, some defendants Percentage with history of mental Differences between defendants initially charged with one of the more illness serious homicides were ultimately 11% of husband defendants Certain characteristics did vary signifi- convicted of negligent homicide. Like- and 15% of wife defendants had a cantly between husband defendants wise, some initially charged with crimi-  history of mental illness. and wife defendants nal homicide were later exonerated on More husbands (20%) than wives grounds they acted in self-defense. Percentage that involved a firearm (10%) had killed in a fit of jealousy over the mate's real or imagined First-degree murder was the predomi- 50% of husbands and 58% of wives infidelity. nant initial charge. Charges were had used a firearm. distributed in this way  first-degree murder: 70% Table 4. Spouse murder defendants: Race and age, 1988 second-degree murder: 24% Percent of spouse murder defendants nonnegligent manslaughter: 6% Race Age (in years) (table 5) Murder Total Mean defendant number Total White Black Other 18-19 20-29 30-59 60 or older age Similarities between defendants All 540 100% 43% 55% 2% 1% 22% 66% 11% 39 yrs

Certain characteristics did not vary Husband 318 100 45 51 4 0 14 74 12 41 significantly between husband de- Wife 222 100 39 61 0 2 33 55 10 37 fendants and wife defendants (table 6)  Number of victims Table 5. Spouse murder defendants: Arrest offense, 1988 4% of husbands and 1% of wives Percent of arrest offenses of spouse murder defendants had been charged with killing more First- Second- than one person.4 Murder Total degree degree Nonnegligent defendant number Total murder murder manslaughter

Percentage that were contract killings All 540 100% 70% 24% 6% 3% of husbands and 6% of wives Husband 318 100 69 28 3 had been charged with a contract Wife 222 100 71 18 11 killing.

4Counting as multiple murder all those committed against pregnant women, the percentage of men Table 6. Spouse murder defendants: Defendant use of firearm, history with multiple victims then becomes 5%, a conserva- of mental illness, charged with multiple or contract murder, motivation, 1988 tive estimate since pregnancy was not recorded in every case. Percent of spouse murder defendants History Murder Murder Total of mental Multiple Contract motivated defendant number Firearm use illness murder murder by jealousy

All 540 53% 13% 3% 4% 16%

Husband 318 50 11 4 3 20 Wife 222 58 15 1 6 10

Spouse Murder Defendants in Large Urban Counties 3 More husbands (31%) than wives Table 7. Spouse murder defendants: (9%) had a history of drug abuse Substance abuse by defendants, (table 7). 1988 Percent of spouse More husbands (22%) than wives murder defendants (3%) had been using drugs at the time Drug use of the murder. Alcohol use Murder At time at time of defendant of murder In past murder More husbands (66%) than wives (37%) had been drinking alcohol at the All 14% 22% 55% 5 time of the murder. Husband 22 31 66 9 5Altogether, 55% of defendants were drinking at Wife 3 37 the time of the murder. Drinking, either by the de- Note: Of the 540 cases, drug use at the time fendant or the victim, was involved in 66% of of murder was known in 255; in past, 540; alcohol use, 335. cases.

4 Spouse Murder Defendants in Large Urban Counties Spouse murder cases disposed by the prosecutor

Following the defendant's arrest, Spouse murder defendants not prosecuted prosecutors review the case, deciding whether to screen it out or to pro- Notes taken on prosecutors' records* secute.6 Dallas case #79 The wife (the victim) is Columbus (OH) case #2 The wife, age 89 and has been married 65 years. A 75, has mental problems. One morning, Of the 540 defendants, 70  or 13%  recent stroke leaves her in terrible pain. while her husband (the victim), a were not prosecuted. Prosecutors She pleads with the doctor to kill her. 75-year-old retired welder, is still asleep, did not screen out significantly more The doctor refuses. The 87-year-old she stabs him to death. She says she of the wives than the husbands  husband goes to the hospital and shoots heard a voice tell her to kill him. 16% of wives and 11% of husbands her. He is immediately arrested. Orange County (CA) case #64 When were not prosecuted (table 8). New Orleans case #54 For years, the police arrive at the scene, they find the 43-year-old husband (the victim), a dry body of the wife (the victim) with 15 Most of the wives not prosecuted were cleaner operator, has beaten his stab wounds and the husband's body screened out because prosecutors 35-year-old wife. At the time of the mur- with 5. Police believe there was an ar- concluded from the evidence that der the two are fighting, and the hus- gument, the husband stabbed the wife in these wives had killed their husband in band stabs her in the back. She grabs a struggle, and then sliced his own neck. self-defense. By contrast, self-defense the knife and stabs him, causing him to Both have been dead for 1 or 2 days. was rarely the reason prosecutors bleed to death. She is arrested the gave for not prosecuting husbands. same day. She claims self-defense, and San Diego case #89 Three weeks be- For husbands, reasons were more the victim's family voices no objection. fore the murder the 32-year-old husband diverse: he committed suicide after beats his 35-year-old wife (the victim) New Orleans case #95 The 28-year- severely. She tells others that next time killing his wife; he took her life in a old husband (the victim) has a long his- he is going to kill her. She wants a di- mercy killing or in a gun accident. tory of assaulting his 25-year-old wife. vorce, but he wants a reconciliation. On 6As used throughout the report, cases "disposed At the time of the murder a witness sees the day of the murder, the two are at a by the prosecutor" include cases screened out, the husband in the kitchen chasing the bar, and he is drinking. Later a witness rejected, declined for prosecution, cases classi- wife with a machete in his hand. The sees her outside a car in the middle of fied as nolle prosequi, and cases in which the fight ends when the wife stabs the hus- the street. The wife screams, then judge dismissed charges. band once. She is arrested the same slumps to the ground, dying from multi- day and claims self-defense. ple stab wounds in the chest. The husbands flees the scene. Later he is found Table 8. Spouse murder defendants: Los Angeles case #21 During an argu- dead from a drug overdose. Percent prosecuted, 1988 ment the 50-year-old housewife (the vic- Percent of spouse tim) pulls out a gun and threatens to kill Orlando (FL) case #7 The husband murder defendants her 39-year-old unemployed husband. comes home drunk and demands Not Murder Total Prose- pros- The two struggle. He flees the house money from his wife. She refuses, and defendant number Total cuted ecuted and gets in his car. She moves in front he attacks her with a metal pipe. She of the car, raises the gun and takes aim. gets a butcher knife from the kitchen All 540 100% 87% 13% He runs over her. He is arrested a day and stabs him once through the heart. later. Husband 318 100 89 11 Wife 222 100 84 16 *Case numbers are those assigned in the dataset. Notes are not necessarily complete in having all relevant details.

Spouse Murder Defendants in Large Urban Counties 5 Spouse murder cases disposed by trial verdict

Overall, 44% of spouse murder de- Insanity acquittals, however, were Those who went to trial and those fendants pleaded not guilty and stood equally likely  ultimately convicted at trial did not trial. Percentages standing trial were differ in terms of arrest offense  Among spouses who stood trial, 3% nearly identical for husband de- of wives and husbands were found not Most of those who went to trial (69%) fendants and wife defendants  guilty by reason of insanity.7 (table 11) and most of those who were 45% of wives and 43% of husbands convicted at trial (70%) had first- 7 stood trial (table 9). The number who pleaded not guilty by reason of degree murder as their arrest offense insanity but who were convicted is unknown. (table 12). About 2% had a nonnegli- Convictions and acquittals gent manslaughter arrest.

Among spouse murder defendants Table 10. Spouse murder defendants who stood trial: who stood trial, 16% were acquitted Percent acquitted, 1988 and 84% were convicted. Judges and juries acquitted more wives than Percent of spouse murder husbands  defendants who stood trial Acquitted 31% of wives who stood trial were Murder Number Insanity acquitted, compared to just 6% of defendant standing trial Total Total acquittal Convicted husbands (table 10). All 238 100% 16% 3% 84%

Husband 138 100 6 3 94 Wife 100 100 31 3 69

Table 9. Spouse murder defendants: Table 11. Spouse murder defendants who stood trial: Percent tried, 1988 Arrest offense, 1988

Percent of spouse Percent of spouse murder defendants who murder defendants stood trial and whose arrest offense was  Murder Total Not Murder Number First-degree Second-degree Nonnegligent defendant number Total Stood trial tried defendant standing trial Total murder murder manslaughter

All 540 100% 44% 56% All 238 100% 69% 29% 2%

Husband 318 100 43 57 Husband 138 100 68 30 2 Wife 222 100 45 55 Wife 100 100 70 27 3

Table 12. Defendants convicted at trial of killing their spouse: Arrest and conviction offense, 1988

Percent of tried and convicted spouse murder defendants Murder Number tried First-degree Second-degree Nonnegligent Negligent Weapon defendant and convicted Total murder murder manslaughter manslaughter offense

Arrest offense

All 199 100% 70% 27% 3% 0% 0%

Husband 130 100 68 30 2 00 Wife 69 100 76 21 3 00

Conviction offense

All 199 100% 35% 31% 24% 9% 1%

Husband 130 100 37 30 22 10 1 Wife 69 100 32 33 29 60

6 Spouse Murder Defendants in Large Urban Counties However, trial conviction offenses were ever the reason, where there was a Despite a sizable difference in the less serious than arrest offenses  conviction, it was to a lesser charge bench trial acquittal rate between in about half of cases, whether the de- wives and husbands, the difference Among those convicted at trial, 70% fendant was the wife or the husband  was not statistically significant, possi- had first-degree murder as their arrest bly because estimates were based on offense but 35% had first-degree mur- 49% of wives and 47% of husbands too few sample cases  der as their conviction offense. A convicted at trial had a conviction of- quarter of those convicted at trial had fense less serious than their original Judges acquitted 37% of wives and nonnegligent manslaughter as their arrest offense (table 13). 17% of husbands (table 16). conviction offense. Bench trials (trials before a judge) Table 15. Spouse murder The reduction from first-degree murder had a higher acquittal rate than jury defendants who stood trial: to a lesser charge did not occur at in- trials  Type of trial, 1988 dictment: Of all trial defendants, 95% Percent of spouse 26% of bench trials ended in acquit- murder defendants were indicted at the same level as their who stood trial tal, versus 11% of jury trials (table 14). arrest offense. Nor did the small num- Murder Number defendant standing trial Total Jury Bench ber of first-degree murder convictions Overall, 63% of defendants who stood result from a higher rate of acquittal: trial were tried by a jury, and the All 238 100% 63% 37% Acquittal rates did not vary significantly remaining 37% were tried by a judge. 150 66 by charge level. Husband 100 34 Wives were not significantly more Wife 88 100 59 41 likely than husbands to choose a Rather, the reason might have been bench trial  that the evidence did not support conviction to the highest charge, or per- 41% of wife defendants and 34% of Table 16. Spouse murder defendants haps juries could seldom agree on husband defendants chose a bench who stood trial: Percent acquitted, conviction to the highest level. What- trial rather than a jury trial (table 15). by trial type, 1988 Percent of spouse Wives were more likely than husbands murder defendants Table 13. Defendants convicted at to be acquitted by a jury  who stood trial Type of trial Number trial of killing their spouse: and murder stand- Ac- Con- Convicted of arrest offense or lesser Juries acquitted 27% of wives but defendant ing trial Total quitted victed 8 offense, 1988 none of the husbands (table 16).

8 Jury trial Percent of spouse The unweighted sample size was 34 husbands murder defendants tried before a jury. Not one of the 34 was acquit- All 150 100% 11% 89% tried and convicted  ted. Extrapolated to the 75 largest counties, the of 34 represent an estimated 91 cases. Husband 91 100 0 100 Number Wife 59 100 27 73 Murder tried and Arrest Lesser defendant convicted Total offense offense Table 14. Spouse murder defendants Bench trial who stood trial: Percent acquitted, All 199 100% 52% 48% All 88 100% 26% 74% by trial type, 1988 Husband 130 100 53 47 Husband 47 100 17 83 Wife 69 100 51 49 Percent of spouse murder Wife 41 100 37 63 defendants who stood trial Type Number of standing trial trial Total Acquitted Convicted

All 238 100% 16% 84%

Jury 150 100 11 89 Bench 88 100 26 74

Spouse Murder Defendants in Large Urban Counties 7 Sentencing of those convicted at trial Table 17. Defendants convicted at trial of killing their spouse: Type of sentence, 1988 While wives were not significantly less likely than husbands to receive a Percent of spouse murder defendants  prison sentence, the length of their convicted at trial and sentenced to Murder Number tried Mean prison prison sentence was shorter than defendant and convicted Total Prison Jail Probation term* what husbands received  An estimated 82% of wives and All 199 100% 90% 3% 7% 17 years 95% of husbands convicted at trial Husband 130 100 95 2 3 21 of killing their spouse received a Wife 69 100 82 3 15 10 prison sentence (table 17).9 *Excludes life and death sentences. Wives convicted at trial received an average sentence to prison of 10 years, 11 years shorter than what husbands received (an average 21-year term in prison).

9"Convicted of killing their spouse" is not pre- cisely correct, since one of the defendants, a husband, was convicted of only a weapon offense. All the rest were convicted of killing or assisting in the killing of a spouse.

8 Spouse Murder Defendants in Large Urban Counties Spouse murder defendants who stood trial: Trial acquittals Notes taken on prosecutors' records

Miami case #84 The couple has an on- Philadelphia case #47 The 35-year old St. Louis case #9 The wife, now age 42, again, off-again relationship for 20 years. husband (the victim) comes home drunk has grown increasingly depressed over Several weeks prior to the murder, she after work and begins fighting with his the years. On the day of the murder she sees her common-law husband (the vic- 31-year-old common-law wife over money is so distraught she shoots her 51-year- tim) leave a motel with another woman. he is missing. The husband is throwing old husband (the victim) several times in Subsequently, the couple has several vio- things at the wife and her children until the head while he is napping. Outcome: lent confrontations until one day she she gets a knife and stabs the husband The jury found her not guilty by reason of shoots him. Outcome: The jury acquitted once in the chest. Outcome: At a bench insanity. her of all charges. trial the judge acquitted her. Detroit case #98 The two are out drink- Chicago case #15 The couple is Cambridge (MA) case #16 The ing. They return home and begin argu- arguing when the 25-year-old wife finds a 38-year-old husband has been hospital- ing. According to the wife, age 30, her love letter to her 25-year-old husband ized several times for mental illness. He 25-year-old husband (the victim) attacked (the victim, a fast-food restaurant em- feels everyone is out to get him. He her. She grabs a knife and stabs him. ployee) from a co-worker. The wife gets comes home one evening and sees his She claims he has beaten her before. No a kitchen knife and stabs him. She 30-year-old wife (the victim) talking with a witnesses are present at the time of the claims she was a victim of battered wife friend. He immediately thinks she is talk- murder. Outcome: She was acquitted of syndrome and was only defending her- ing about him, and he later strangles her. second-degree murder at a bench trial. self. Outcome: At a bench trial the judge Outcome: At a bench trial the judge acquitted her. found him not guilty by reason Manhattan case #72 She, a 29-year-old of insanity. artist, has lived with her 47-year-old Chicago case #28 The couple is common-law husband (the victim), a mu- arguing when the 64-year-old husband Dallas case #65 The unemployed sician, for a few years. He frequently (the victim) swings a pipe at his 34-year- 58-year-old husband is released from a beats, starves, and tortures her. He re- old wife. She gets a knife and stabs him mental hospital two weeks before the peatedly tells her he is going to kill her. to death. Outcome: The jury acquitted murder. He fears his 56-year-old wife For some time leading up to the murder, her. (the victim), a secretary, will leave him for he has not let her eat or shower. On the another man and have him put back in day of the murder they are going to a bar Chicago case #52 The 49-year-old the mental hospital. He stabs her to when she asks him to let her go home husband (the victim) is drunk and gets death. Outcome: At a bench trial the because she is tired. He says "no" and into an argument with his 50-year-old judge found him not guilty by reason of begins hitting her and calling her names. wife. According to her, at some point he insanity. The beating ends when she stabs him. throws a fan at her. She gets a knife and He has numerous past arrests and con- lunges at him, cutting his abdomen. Out- victions. Outcome: The jury acquitted come: The jury acquitted her. her.

Spouse Murder Defendants in Large Urban Counties 9 Spouse murder defendants who stood trial: Trial convictions Notes taken on prosecutors' records

Oklahoma City case #24 The 59-year testify that he has beaten her for at least Seattle case #32 Two weeks before the old husband (the victim) comes home 15 years. She has been to hospitals murder, the 43-year-old husband, an air- intoxicated and continues drinking. His throughout the city for cuts and bruises craft worker, is jailed for assaulting his 52-year old wife claims he became violent inflicted by him. He has a long history of 31-year-old wife (the victim). She and the and began to beat her. She says she arrests and convictions. Outcome: The children then move out of the house. On shot him in self-defense. The prosecution jury convicted him of negligent man- the day of the murder the children are be- counters that there are no bruises or slaughter. He was sentenced to 3 years ing watched by a babysitter in their new other signs of a beating. The prosecution in prison. lodgings. The wife returns home and the contends she shot him while he lay in bed husband appears out of hiding. He or- sleeping. Outcome: The jury convicted Oklahoma City case #1 The 35-year-old ders the babysitter to lie down and tells her of first-degree murder. She was sen- husband has threatened his 29-year-old the wife that she is going to watch the tenced to life imprisonment. wife (the victim) in the past. The threats babysitter die. He begins stabbing the cause her to leave him. On the day of babysitter in the back, but the babysitter Rochester (NY) case #26 Throughout the murder she is drinking and returns to breaks free. The husband turns his at- their turbulent 4-year marriage, the hus- his apartment with her mother to get tack on the wife, who dies from multiple band (the victim) verbally and physically clothes and food. He is there and also stab wounds. Outcome: The jury con- abuses his wife. She never once leaves has been drinking. He stabs to death victed him of first-degree murder. He him, though. Prior to the night of the mur- both her and her mother. He then flees. was sentenced to 45 years in prison. der, the husband buys a gun and puts it He has numerous prior arrests and con- in his dresser. The night of the murder, victions. Outcome: The jury found him Brooklyn case #54 The 52-year-old hus- the husband is slapping the wife around, guilty of two counts of first-degree mur- band suspects his 45-year-old wife (the telling her to get out of the house. He der. He was sentenced to the death victim) is having an affair. While at home calls the police to have her removed from penalty. the two argue and she, a factory worker, the premises. While he is on the phone, admits to the affair. He bludgeons her telling police "she's a whore," the wife San Diego case #15 He, age 41, has with a hammer. Outcome: The jury grabs his gun and shoots him four times. been sadistically abusing his 41-year-old found him guilty of using a dangerous Outcome: The jury convicted her of wife (the victim) for about 8 years. He weapon (other than a firearm). He was second-degree murder. She was sen- frequently binds, gags, and blindfolds her sentenced to straight probation (no con- tenced to 15 years in prison. and forces her to perform sex acts on finement in prison or jail). him. He sometimes cuts and burns her Bakersfield (CA) case #34 The husband while she is bound. A year prior to the Dallas case #178 The wife (the victim), (the victim) and wife jointly own a bar. murder she calls the police following a a 24-year-old postal clerk, is having an af- She claims he showed up at the bar an- beating. He is arrested and convicted fair with another man. The 26-year-old gry at her about something and slapped and then thrown out of the Navy, termi- husband arrives at the other man's house her. She says she became scared, saw a nating a 22-year career. Soon after, they and finds his wife in bed with the man. gun, and shot him. The prosecution dis- separate. On the night of the murder, he The husband shoots and kills them both. putes that, contending instead she mur- has drunk about 20 beers and calls his Outcome: The jury convicted him of two dered him so that she would gain sole estranged wife, ordering her to take off counts of nonnegligent manslaughter. He ownership of the bar. Outcome: The jury her clothes and await his arrival. When was sentenced to 50 years in prison. found her guilty of nonnegligent man- he enters her apartment, she is nude and slaughter. She was sentenced to 8 years holds out her wrists to him. He binds, Riverside (CA) case #1 The husband in prison. gags and blindfolds her, cuts her with a (the victim) beats and abuses his wife knife, and wraps tape and rope around for many years. One night, after being Philadelphia case #169 They have been her neck, strangling her. Hours later he is abused, she gets up, grabs a rifle and married 20 years. Both are heavy drink- arrested. He admits killing her and in- shoots him while he is sleeping. She ers. During an argument he, a longshore- tending to do so for some time. He then takes her children and flees. man, hits his 53-year-old wife (the victim) blames her for ruining his career, saying Outcome: The jury found her guilty on the head, and several days later she he was going to kill her sooner or later. of nonnegligent manslaughter. She was dies. Evidence is uncovered that she was Outcome: The jury convicted him of first- sentenced to 8 years in prison. a battered wife. The couple's children degree murder. He was sentenced to 26 years in prison.

10 Spouse Murder Defendants in Large Urban Counties Spouse murder cases disposed by guilty plea

Of the 540 defendants, 43% pleaded Table 19. Defendants who pleaded guilty to killing their spouse: guilty to killing their spouse. Wives Arrest and conviction offense, 1988 and husbands were about equally likely to plead guilty  Percent of convicted spouse murder defendants who pleaded guilty 39% of wives and 46% of husbands Number First- Second- Murder pleading degree degree Nonnegligent Negligent pleaded guilty to killing their spouse defendant guilty Total murder murder manslaughter manslaughter (table 18). Arrest offense The vast majority of pleas were to a reduced charge  All 232 100% 67% 23% 10% 0% Most of those who pleaded guilty Husband 145 100 64 31 5 0 Wife 87 100 70 11 19 0 (67%) had first-degree murder as their arrest offense (table 19). Ten percent Conviction offense had nonnegligent manslaughter. All 232 100% 10% 20% 58% 12%

The guilty plea was to first-degree Husband 145 100 12 28 50 10 murder in just 10% of cases and to Wife 87 100 6772 15 nonnegligent manslaughter in 58% of the spouse murder cases (table 19). husbands pleaded guilty, the plea was Plea bargains have advantages for to manslaughter in 60% (table 19). both sides. To the prosecutor ham- Comparing husbands and wives who pered by an evidence problem, for in- pleaded guilty to killing a spouse, wife Nevertheless, a plea to reduced stance, the guilty plea at least assures defendants were more likely than charges was no more likely among conviction, even if not to the most husband defendants to have been wives than husbands  serious charge. To the guilty de- arrested for the least serious degree fendant, a plea bargain will usually When wives pleaded guilty, the plea of murder, nonnegligent manslaughter result in a reduced sentence from what was to a reduced charge in 71% of (19% of wives and 5% of husbands). might otherwise have been imposed. Having less serious arrest offenses cases; and when husbands pleaded in the first place, the offenses they guilty, the plea was to a reduced pleaded guilty to therefore tended also charge about as often, in 76% Table 20. Defendants who pleaded to be less serious than those of of cases (table 20). guilty to killing their spouse: husbands  Conviction offense compared to Sentencing of those who pleaded guilty arrest offense, 1988 When wives pleaded guilty, the plea Percent of convicted was to manslaughter (negligent or non- Probably the majority of guilty pleas spouse murder defen- negligent) in 87% of cases; but when dants who pleaded were the product of negotiation be- guilty to  tween the prosecution and the de- Number Table 18. Spouse murder defendants: fense. In such negotiations a "bargain" Murder pleading Arrest Lesser defendant guilty Total offense offense Percent who pleaded guilty to killing is struck. In exchange for the prosecu- their spouse, 1988 Percent of spouse tor reducing the seriousness or num- All 232 100% 26% 74% murder defendants ber of charges (a charge reduction), Husband 145 100 24 76 Disposed the defendant agrees to plead guilty. Murder Total Pleaded some Wife 87 100 29 71 defendant number Total guilty other way Or, in exchange for the prosecutor's All 540 100% 43% 57% recommending a less severe sentence (a sentence reduction), the defendant Husband 318 100 46 54 agrees to plead guilty as charged. Wife 222 100 39 61

Spouse Murder Defendants in Large Urban Counties 11 The advantage of pleading guilty to These longer sentences for defendants Although wives were not significantly homicide is not always immediately convicted at trial were to some extent less likely than husbands to receive obvious to the defendant deciding in line with the lower likelihood of their a prison sentence, the length of their whether to plead guilty. A trial offers receiving a charge reduction  prison sentence was shorter than that at least the chance of acquittal,  When defendants pleaded guilty, the of husbands whereas pleading guilty guarantees plea was to a reduced charge in 74% Of those who pleaded guilty, 81% conviction and makes prison a near of cases (table 20); but when they of wives and 93% of husbands were certainty. However, the possible were convicted at trial, the conviction sentenced to State prison (table 21). advantage of a guilty plea becomes was to a reduced charge in 48% of clear when sentences are compared cases (table 13). 4 years was the average prison between spouse murder defendants sentence length for wives who pleaded convicted at trial and those convicted Partly reflecting their less serious con- guilty, 10 years shorter than the through a guilty plea  viction offenses, wives were given less 14-year average for husbands who Though guilty plea and trial convic- severe sentences than husbands. entered a guilty plea (table 21). tion cases were about equally likely to receive a prison sentence (88% and 90%, respectively), a life sentence was given to 1% of those who pleaded Table 21. Defendants who pleaded guilty to killing their spouse: guilty but 25% of those who were con- Type of sentence, 1988 victed at trial (prison term sentences Percent of spouse murder defendants in tables 21 and 17; life sentences for pleaded guilty and sentenced to  pleas or convictions not shown in a Murder Number plead- Mean prison table). defendant ing guilty Total Prison Jail Probation term*

Among those sentenced to prison All 232 100% 88% 0% 12% 10.5 years but not to a life term, the average Husband 145 100 93 0 7 14 prison sentence was 10 ½ years for Wife 87 100 81 0 19 4 defendants who pleaded guilty but 17 *Excludes life and death sentences. years for defendants convicted at trial.

12 Spouse Murder Defendants in Large Urban Counties Spouse murder defendants who pleaded gulity Notes taken on prosecutors' records

Philadelphia case #69 Both are retired. have been repeatedly physically abused Ft. Lauderdale case #62 Late one night They have been married 47 years. The by the victim. Hospital examination at the the husband (the victim) and the wife, a 65-year-old wife (the victim) has recently time of the murder finds multiple bruises mail carrier, are arguing. Police arrive suffered a stroke. Her health and state of on the child and the defendant in various and convince the husband to leave for mind deteriorate rapidly. According to stages of healing. Outcome: She the night. The wife locks the door, arms her 68-year-old husband, she had be- pleaded guilty to nonnegligent man- herself with a handgun and sleeps in the come "mentally disturbed." He feels he slaughter and was sentenced to 10 hall. The husband returns the next morn- can no longer meet her health demands months in prison. ing, enters the apartment, and finds the and decides to "put her out of her mis- wife locked in the bathroom. When the ery." He shoots her with a rifle. Out- New Haven case #2 She, a 28-year old husband tries to get into the bathroom, come: He pleaded guilty to nonnegligent secretary, suffers years of physical and the wife fires a shot through the door, manslaughter and was sentenced to sexual abuse at the hands of her striking the husband in the leg. straight probation (no confinement in 30-year-old husband (the victim). Sev- The action moves into the bedroom prison or jail). eral times he tries to kill her. She stays where the husband is shot in the head. with him at first because she thinks he Outcome: She pleaded guilty to non- Queens (NY) case #5 According to will stop; then because she fears he will negligent manslaughter and was sen- neighbors, the husband and his common- find her wherever she goes; and then be- tenced to straight probation (no law wife (the victim) regularly get drunk cause she fears losing her kids. At some confinement in prison or jail). and argue. On the day of the murder, point she buys a gun to defend herself. they are arguing and the common-law On the night of the murder she thinks he Austin (TX) case #4 The husband (the husband places his 69-year-old wife in a is possibly going to kill her. In the middle victim) and his common-law wife, a sofa bed and closes it. He later says he of a beating she grabs the gun from un- housekeeper, frequently drink and fight. tried but was unable to open it. She dies der the mattress and shoots him. Out- On the day of the murder, both are intoxi- in the sofa bed. Outcome: He pleaded come: She pleaded guilty to negligent cated and begin fighting. The wife gets a guilty to negligent manslaughter and was manslaughter and was sentenced to gun and shoots her husband. She then sentenced to straight probation (no con- straight probation (no confinement in calls the police. The wife has injuries she finement in prison or jail). prison or jail). says she received from being beaten by her common-law husband. Outcome: San Diego case #1 The husband is a Pittsburgh case #15 The husband (the She pleaded guilty to negligent man- construction worker and the wife is on victim) has a history of beating his wife. slaughter and was sentenced to 10 years welfare. The 23-year-old husband (the On the night of the murder, the husband in prison. victim) beats his 21-year-old wife and comes home and begins ordering her their 18-month-old daughter the night around, as he frequently does. The wife Dayton (OH) case #3 The husband (the prior to the murder. On the night of the leaves the room. When she returns she victim) and his common-law wife are ar- murder, the wife and her husband are notices him looking through the closet for guing about a variety of things. Through- asleep in their bedroom and the daughter his gun. The wife earlier hid it under the out the argument the husband beats her. is in another room. The child begins to bed. While he is searching, the wife re- When the common-law husband comes cry, waking the husband. He orders the trieves the gun and shoots him repeat- after her in the kitchen, she grabs a knife. wife to stop the child's crying. The wife edly. She claims she was tired of the She stabs him in the back as he is walk- leaves the bedroom, returns with a gun abuse. Outcome: She pleaded guilty to ing away. Outcome: She pleaded guilty and shoots the husband. Evidence nonnegligent manslaughter and was sen- to negligent manslaughter and was sen- emerges that the wife and the daughter tenced to straight probation (no confine- tenced to 2 years in prison. ment in prison or jail).

Spouse Murder Defendants in Large Urban Counties 13 Spouse murder defendants who pleaded guilty (continued) Albuquerque case #9 Both are unem- New Haven (CT) case #7 The couple Cambridge (MA) case #2 Both the ployed. One week before the murder, the have a history of physical abuse. The 58-year-old husband, a mail clerk, and his 22-year-old husband is released from 56-year-old husband has a stab wound 44-year-old wife (the victim) have been prison and moves in with his 22-year-old inflicted by his 38-year old wife (the vic- hospitalized for psychiatric disorders. On wife (the victim), who is living at his tim) in an earlier altercation. On the day the day of the murder the wife returns grandmother's house. At the time of the of the murder the husband claims his wife from a 6-month stay at a mental hospital. murder the wife tells the husband that she jumped on him and said she was going She refuses to take her medication and, wishes to leave him to be with a man she to kill him. He stabs her to death. He has according to the defendant, goes crazy. was seeing while the husband was in a long history of arrests and convictions. She threatens the husband with a broken prison. She also informs him that she Outcome: He pleaded guilty to nonnegli- bottle, begins hallucinating, and beats the might be pregnant by the other man. The gent manslaughter and was sentenced to husband. For hours he tries to calm her husband strangles her, tells the grand- 20 years in prison. until, at one point, he applies too much mother he killed his wife, and calls police. pressure to her neck and she dies. Out- Family members and friends say he has a Denver case #18 He is a 34-year-old come: The husband pleaded guilty to history of assaulting his wife and threaten- steel mill worker, and she is a 24-year-old nonnegligent manslaughter and was sen- ing her life. He also has a long criminal waitress. An argument develops between tenced to 20 years in prison. record. Outcome: He pleaded guilty to the two, partly over whether the pregnant nonnegligent manslaughter and was sen- wife (the victim) should have an abortion. Oklahoma City case #30 The 29-year- tenced to 16 years in prison. The husband tells her to leave. She re- old wife (the victim) accuses the 32-year- fuses. She threatens him with a knife. old husband of being interested in another Prince George's County (MD) case #13 He gains control of the knife and stabs woman. An argument turns into a fight. On the night of the murder, the husband her. Outcome: He pleaded guilty to The wife is armed with a kitchen knife. and a friend pick up the wife (the victim) second-degree murder and was sen- The husband has scissors. Both sustain and are driving down the road when the tenced to 24 years in prison. stab wounds, but her wounds are fatal. husband stops to "repair the car." While Outcome: The husband pleaded guilty to the husband is looking under the hood, Memphis case #2 The 39-year-old negligent manslaughter and was sen- the friend begins slashing the wife with a husband claims that, during an argument, tenced to 5 years in prison. razor. The husband states he earlier his 34-year-old wife (the victim) beat him joked with the friend about killing her, but with a phone. The wife is much larger Philadelphia case #96 Both are he claims he did not really mean it. See- than the husband. The husband states retired. The 73-year-old husband has ing her suffering, he says, he shoots her that the wife went for a handgun kept in a Alzheimer's disease, causing him to be to put her out of her misery. Outcome: confused at times. On the day of the closet and the two struggled over it. In The husband pleaded guilty to first- murder he decides to kill his 73-year-old the struggle the gun fires once, killing the degree murder and was sentenced to life wife (the victim). He beats her to death wife. The husband then calls police. in prison. with a crowbar and then calls police. Out- Outcome: The husband pleaded guilty come: The husband pleaded guilty to to negligent manslaughter and was sen- nonnegligent manslaughter and was sentenced to 2 years in prison. tenced to straight probation (no confinement in prison or jail).

14 Spouse Murder Defendants in Large Urban Counties Overview of case disposition of spouse murder defendents

All sampled defendants were charged Although wives were not significantly Conversely, 70% of wives were with murdering a spouse, and all were less likely than husbands to be prose- convicted of killing their spouse, disposed of in 1988. How they were cuted (16% versus 11%), or to stand versus 87% of husbands (table 23). disposed of varied  trial (45% versus 43%), or to plead guilty (39% versus 46%) (table 22), Wives were about half as likely as Not prosecuted: 13% they were less likely than husbands to husbands to be convicted of first- Pleaded guilty: 43% be convicted of killing their spouse  degree or second-degree murder  Convicted at trial: 37% Acquitted at trial: 7% 30% of wives but 13% of husbands 25% of wives but 46% of husbands were either not prosecuted or were were convicted of first-degree or prosecuted but acquitted at trial. second-degree murder.

Nevertheless, the percentage of Table 22. Spouse murder defendants: Type of disposition, 1988 husbands and wives convicted of a lesser offense (an offense less serious Percent of spouse murder defendants than the one they were arrested for) Tried did not differ significantly  Murder Total Not Pleaded defendant number Total prosecuted Total Acquitted Convicted guilty 43% of wives and 54% of husbands were convicted of a lesser offense 540 All 100% 13% 44% 7% 37% 43% (table 24). Husband 318 100 11 43 2 41 46 Wife 222 100 16 45 14 31 39

Table 23. Spouse murder defendants: Percent convicted of killing their spouse, by conviction offense, 1988

Percent of spouse murder defendants Convicted of  Not convicted: First- Second- Murder Total disposed some degree degree Nonnegligent Negligent Weapon defendant number Total other way Total murder murder manslaughter manslaughter offense

All 540 100% 20% 80% 17% 20% 34% 9% --

Husband 318 100 13 87 21 25 32 9 -- Wife 222 100 30 70 12 13 37 8 0

--Less than 1%.

Table 24. Spouse murder defendants: Percent convicted of killing their spouse, by whether conviction was to arrest offense or lesser offense, 1988

Percent of spouse murder defendants Convicted of  Not convicted: Murder Total disposed some Arrest Lesser defendant number Total other way Total offense offense

All 540 100% 20% 80% 30% 50%

Husband 318 100 13 87 33 54 Wife 222 100 30 70 27 43

Spouse Murder Defendants in Large Urban Counties 15 The likelihood of conviction did not The likelihood of conviction to reduced Wives were less likely than husbands vary significantly by arrest offense  charges was associated with the to be convicted and sentenced to seriousness of the arrest offense. The prison  78% of defendants arrested for first- more serious the arrest offense, the degree murder were convicted, com- 57% of wives were convicted and more likely the conviction was to a pared to 84% of defendants arrested sentenced to prison for killing their reduced charge  for second-degree murder or nonnegli- spouse, compared to 81% of husbands gent manslaughter (table 25). 53% of defendants arrested for first- (table 27). degree murder were convicted of a lesser offense, versus 29% of defendants arrested for nonnegligent manslaughter (table 26).

Table 25. Spouse murder defendants: Percent convicted of killing their spouse, by arrest and conviction offense, 1988

Percent of spouse murder defendants Convicted of  Not convicted: First- Second- Total disposed some All degree degree Nonnegligent Negligent Weapon Arrest offense number Total other way offenses murder murder manslaughter manslaughter offense

All 540 100% 20% 80% 17% 20% 34% 9% --

First-degree murder 377 100 22 78 25 15 31 7 -- Second-degree murder 129 100 16 84 0 40 36 8 0 Nonnegligent manslaughter 34 100 16 84 0 0 55 29 0 --Less than 1%.

Table 26. Spouse murder defendants: Percent convicted, by arrest offense and by whether conviction was to arrest offense or lesser offense, 1988

Percent of spouse murder defendants Convicted of  Not convicted: Total disposed some All Arrest Lesser Arrest offense number Total other way offenses offense offense

All 540 100% 20% 80% 30% 50%

First-degree murder 377 100 22 78 25 53 Second-degree murder 129 100 16 84 40 44 Nonnegligent manslaughter 34 100 16 84 55 29

Table 27. Spouse murder defendants: Percent convicted, by type of sentence, 1988

Percent of spouse murder defendants Convicted and sentenced to  Not convicted: Prison Murder Total disposed some Total Total Straight defendant number Total other way sentences prison Life Death Jail probation

All 540 100% 20% 80% 71% 10% 1% 1% 8%

Husband 318 100 13 87 81 13 2 1 5 Wife 222 100 30 70 57 6 0 1 12

16 Spouse Murder Defendants in Large Urban Counties Sentencing of defendants convicted of killing their spouse

Among convicted defendants, virtually murder than nonnegligent manslaugh- Among convicted defendants whose all were convicted of either murder or ter  arrest was for nonnegligent man- manslaughter (table 28). Because of slaughter, 34% had a conviction Among convicted defendants whose charge reduction, conviction offenses offense less serious than their arrest arrest was for first-degree murder, were generally less serious than arrest offense. 68% had a conviction offense less seri- offenses. Charge reduction was more ous than their arrest offense (table 29). common in arrests for first-degree Among those convicted, the conviction offense differed markedly from the Table 28. Convicted spouse murder defendants: arrest offense  Conviction offense, by arrest offense, 1988 The predominant arrest offense of Percent of spouse murder convicted defendants was the most  defendants convicted of serious form of homicide, first-degree First- Second- Non- Negligent Wea-- murder (68%) (table 30). Number degree degree negligent man- pon Arrest offense convicted Total murder murder manslaughter slaughter offense But the predominant conviction All 431 100% 22% 25% 42% 10% 1% offense was the least serious form, negligent or nonnegligent First-degree murder 294 100 32 19 40 8 1 Second-degree murder 108 100 0 48 42 10 0 manslaughter. Nonnegligent manslaughter 28 100 0 0 66 34 0 Convicted wives were more likely than convicted husbands to have Table 29. Convicted spouse murder defendants: Conviction to arrest manslaughter as their conviction offense or lesser offense, by arrest offense, 1988 offense 

Percent of spouse murder 64% of convicted wives had defendants convicted of  manslaughter (negligent or nonnegli- Number Arrest Lesser gent) as their conviction offense, Arrest offense convicted Total offense offense compared to 46% of convicted All 431 100% 38% 62% husbands.

First-degree murder 294 100 32 68 Second-degree murder 108 100 48 52 Nonnegligent manslaughter 28 100 66 34

Table 30. Spouse murder defendants convicted of killing their spouse: Arrest and conviction offense, 1988

Percent of convicted spouse murder defendants Convicted Second- First-degree degree Nonnegligent Negligent Weapon Defendant Total murder murder manslaughter manslaughter offense Arrest offense

All 100% 68% 25% 7% 0% 0%

Husband 100 66 31 3 00 Wife 100 73 15 12 00

Conviction offense

All 100% 22% 25% 42% 10% 1%

Husband 100 24 29 36 10 1 Wife 100 17 19 53 11 0

Spouse Murder Defendants in Large Urban Counties 17 Still, charge reduction was not more A smaller percentage of convicted imprisonment and the death penalty). common among convicted wives than wives (81%) than convicted husbands Among wives sentenced to prison, convicted husbands  (94%) received a prison sentence 15% received a sentence of 20 years (table 32). or more; among husbands, 43%. 61% of convicted wives had a lesser offense (an offense less serious than their arrest offense) as their conviction However, convicted wives (8%) were Excluding life or death sentences, offense, compared to 62% of convicted not significantly less likely than con- the average prison sentence was 6 husbands (table 31). victed husbands (15%) to receive a life years for wives convicted of killing their sentence. husband, about 10 years shorter than the average 16 ½ years husbands Wives convicted of killing their hus- received for killing their wife. band were generally sentenced less There was a significant difference severely than husbands convicted between the percentages of wives and of killing their wife  husbands receiving a prison sentence of 20 or more years (including life

Table 31. Convicted spouse murder Table 32. Defendants convicted of killing their spouse: defendants: Conviction to arrest Type of sentence imposed, by conviction offense, 1988 offense or lesser offense, 1988 Percent of spouse murder defendants Percent of spouse convicted and sentenced to  murder defendants Prison convicted of  Murder Number Arrest Lesser Murder Number Total Total Straight Mean prison defendant convicted Total offense offense defendant convicted sentences prison Life Death Jail probation term*

All 431 100% 38% 62% All 431 100% 89% 12% 1% 1% 10% 13 yrs

Husband 275 100 38 62 Husband 275 100 94 15 2 1 5 16.5 Wife 156 100 39 61 Wife 156 100 81 8 0 1 18 6 *Excludes life and death sentences.

18 Spouse Murder Defendants in Large Urban Counties Penalties imposed on convicted spouse murder de- murder, with first-degree murder punished most severely fendants varied according to the seriousness of the and negligent manslaughter least severely (table 33) 

For conviction for first-degree murder For conviction for nonnegligent manslaughter least severe penalty  10 years in prison least severe penalty  straight probation most severe penalty  death most severe penalty  50 years in prison life in prison  36% life in prison  0% average non-life prison sentence  31 years. average non-life prison sentence  7 years.

For conviction for second-degree murder For conviction for negligent manslaughter least severe penalty  straight probation least severe penalty  straight probation most severe penalty  life in prison most severe penalty  10 years in prison life in prison  17% life in prison  0% average non-life prison sentence  16 years. average non-life prison sentence  5 years.

Table 33. Convicted spouse murder defendants: Type of sentence imposed, by conviction offense, 1988

Percent of spouse murder defendants convicted and sentenced to  Prison Mean Number Total Total Straight prison Conviction offense convicted sentences prison Life Death Jail probation termb

All 431a 100% 89% 12% 1% 1% 10% 13 yrs

First-degree murder 93 100 100 36 6 0 0 31 Second-degree murder 108 100 94 17 0 0 6 16 Nonnegligent manslaughter 183 100 89 0 0 0 11 7 Negligent manslaughter 45 100 54 0 0 11 35 5 aIncludes convictions for weapon offense bExcludes life and death sentences.

Spouse Murder Defendants in Large Urban Counties 19 Processing time in spouse murder cases

Time to arrest: All cases Most rejected by the prosecutor (or dis- Jury versus bench trials Cases dis- spouse murder defendants were missed in court), acquitted or con- posed by bench trial (elapsed time arrested on the same day the killing victed at trial, or concluded through of about 13 months) were not disposed occurred. Consequently, the median a guilty plea. of significantly faster than those dis- number of days from the day of the posed by jury trial (elapsed time of murder to the day the defendant was All cases Median elapsed time from about 15 months). arrested was 0 days. The mean was the day of the murder to the day the 28 days (table 34). case was finally disposed was 9 Jury trials For husbands tried by a months. The mean was almost jury, average elapsed time from the Time to indictment: All cases The 1 year. day of the murder to the conclusion of median period of time from the day the trial was about 12 ½ months; for of the murder to the day of indictment Trials versus pleas Median disposi- wives it was significantly longer, about (for those who were indicted) was just tion time from the day of the murder to 18½ months. under 2 months and the mean was 4 final disposition was 8½ months for months. plea cases and 1 year for trial cases. Mean disposition time was 1 year for Time to final disposition Final dis- plea cases and 1 year and 2 months position is the date the case was either for trial cases.

Table 34. Spouse murder defendants: Median and mean number of days from the murder to arrest, indictment, and final disposition, by case type, 1988

Number of days from date of murder to  Case type, Total Arrest Indictment Final disposition defendant number Median Mean Median Mean Median Mean

All casesa All 540 0 days 28 days 54 days 125 days 267 days 378 days Husband 318 0 30 46 98 232 353 Wife 222 0 26 58 163 303 414

Trial cases All 238 0 42 45 106 357 430 Husband 138 0 43 45 105 292 385 Wife 100 0 40 46 107 471 490

Bench cases All 88 0 2 46 81 367 392 Husbandb 47 0 2 71 106 446 393 Wifeb 41 0 1 32 56 357 392

Jury cases All 150 0 65 45 121 331 451 Husband 91 0 63 41 105 217 382 Wifeb 59 0 69 62 143 508 559

Plea cases All 232 0 21 59 155 255 373 Husband 145 0 24 46 97 263 339 Wifeb 87 0 16 88 243 243 429 aIncludes cases disposed by prosecutor. bMean and median are based on fewer than 30 sample cases.

20 Spouse Murder Defendants in Large Urban Counties Less severe dispositions for wife defendants: Summary of evidence

Though wives were about as likely as Self-defense as a possible explanation likely than husband defendants to have husbands to be charged with first- for wives' lower conviction rates been severely provoked by their degree murder (table 5) and about as spouse and therefore more likely to likely to be prosecuted (table 8) or In some cases the wife or the husband have been absolved on grounds of indicted, in other respects the justice killed the spouse only after being self-defense. system treated wives less severely physically attacked or threatened with than husbands. Most notably  a weapon by the spouse. In such Grounds for claiming self-defense Wives had a lower conviction rate cases the defendant may claim self- more often present in wife defendant than husbands  defense. Justice system officials then cases than in husband defendant decide whether and to what extent the cases 70% of wives but 87% of husbands victim's own conduct contributed to were convicted of killing their mate his/her death. Additionally, officials Possible evidence to support a self- (table 23). consider whether the defendant could defense claim  evidence that, have fled the scene rather than stay somewhere along the chain of events At trial, judges and juries acquitted and defend him/herself and whether leading to the murder, the victim 31% of wives but 6% of husbands the defendant could have used less provoked the defendant  was more (table 10). than lethal force. In certain cases of often present when wives killed their extreme victim provocation, where the husband than when husbands killed Among cases tried before a jury, homicide is ultimately determined to their wife  27% of wives were acquitted but none have been justifiable, the prosecutor More wives (44%) than husbands of the husbands (table 16). declines to prosecute, the grand jury (10%) had been assaulted by their votes not to indict, or the judge or jury spouse (threatened with a weapon or Convicted wives were less likely than acquits the defendant. In less extreme physically assaulted) at or around the convicted husbands to be sentenced cases the charge or conviction offense time of the murder (table 35.)10 to prison, and convicted wives is lowered. received shorter prison sentences 10In some cases the survey questionnaire did not than their male counterparts  The possibility investigated below is indicate that the defendant had been provoked, but the case narrative clearly revealed provoca- 81% of convicted wives but 94% of that wives were less likely to have been convicted than husbands tion. In such instances the defendant was classi- convicted husbands received a prison fied as provoked. sentence (table 32). because wife defendants were more

Excluding life or death sentences, Table 35. Spouse murder defendants: Victim provocation, 1988 the average prison sentence for killing  a spouse was 6 years for wives but Percent of All Husband Wife 16½ years for husbands. Evidence of victim provocation defendants defendants defendants

Among wives sentenced to prison, Victim assaulted defendant At time of murder 24% 10% 44% 15% received a sentence of 20 years At time of murder or in past 29 10 58 or more; among husbands, 43%. Defendant had prior arrest The likelihood of a spouse murder or conviction for any crime 52 70 27 defendant ultimately being convicted Victim had prior arrest and sent to prison was lower for wives or conviction 34 12 51 than husbands  Victim was the only one drinking at time of murder 12 4 25 Of the 222 wife defendants, 57% Victim was armed ended up in prison. Of the 318 at time of murder 15 11 21 husband defendants, 81% were sen- Note: Of the 540 cases, information on provocation was known in 460; tenced to prison. on defendant criminal history, 409; on armed victim, 507; on victim use of alcohol. 265. Table percentages are based on known cases.

Spouse Murder Defendants in Large Urban Counties 21 More wives (58%) than husbands Victim provocation reduces likelihood Victim provocation indirectly mitigates (10%) had been assaulted by their of conviction the sentence through reduced charges spouse at the time of the murder or in the past. Consistent with that, more Victim provocation (meaning that the To illustrate, the prosecutor takes wives (51%) than husbands (12%) had defendant was apparently assaulted or provocation into account by agreeing killed a spouse who had at least one threatened with a weapon by the victim to a plea to a reduced charge. The prior arrest or conviction (not neces- at or around the time of the homicide) sentence the provoked wife receives is sarily an arrest or conviction for a appeared to reduce the likelihood less severe than what she would have crime against the defendant, though). of conviction  received had the prosecutor not made allowance for provocation. Similarly, Of the provoked wives, 56% were The possibility that more wife de- the judge or jury takes provocation into convicted.11 That is lower than either fendants than husband defendants account by convicting the wife de- the 86% for unprovoked wives or the killed in self-defense is further sup- fendant of a lower charge. Again, the 88% of unprovoked husbands. ported by two victim differences, result is a less severe sentence than though neither difference is statistically what she otherwise would have No difference between provoked and significant  received. Such indirect mitigation is unprovoked wives in terms of likelihood not necessarily observable when Situations in which the victim is of receiving a prison sentence or sentences are compared between armed are probably more characteris- length of prison sentence provoked and unprovoked defendants. tic of provoked than unprovoked murders. If more wife defendant than Risk of conviction is lower for pro- husband defendant cases involved an voked than unprovoked defendants. Despite provoked wives' testimony that armed victim, that would tend to cor- But if convicted, the provoked they had been assaulted, their testi- roborate other evidence suggesting defendant gets no obvious break at mony in some cases was insufficient to that more wife than husband de- sentencing  convince prosecutors, judges, or juries. fendants were provoked. Of rele- Perhaps their claim was uncorrobo- 84% of convicted wife defendants vance, then, is the fact that the victim rated. Unsupported defendant testi- were sent to prison, whether or not was armed in 21% of wife defendant mony is generally less persuasive than they were provoked. cases but 11% of husband defendant eyewitness testimony, physical signs cases. of abuse on the defendant, or docu- The average prison sentence was mentary evidence from hospital or 5 years for provoked wife de- legal records of past abuse by the Previous research has shown that, fendants and 7 years for unprovoked victim. Another possibility is that compared to unprovoked murders, wife defendants, but the difference prosecutors, judges, or juries believed provoked spouse murders tend more was not statistically significant. often to involve alcohol use by the vic- provocation had occurred but also tim. One additional check of whether believed that the defendant could have These results do not necessarily more wife defendants than husband fled the scene or, even if that was not mean, however, that the justice system defendants were provoked is to com- possible, did not have to resort to disregards victim provocation at sen- pare the two in terms of victim alcohol lethal force. tencing. Following are three alternate, use. They should differ, with victim al- but not mutually exclusive, interpreta- cohol use being more prevalent in wife tions of why the average sentence The small sample size is a reason the defendant than husband defendant lengths do not differ between wives 5-year sentence for provoked wives is cases. Relevant to the question of who killed their husband upon provoca- not significantly shorter than the 7-year provocation, then, is the fact that the tion and those who were not provoked. sentence for unprovoked wives. victim had been the only one drinking With a larger sample, the 2-year differ- in 25% of cases where the wife killed 11No conviction rate is given for provoked hus- ence would be statistically significant. the husband. That compares to just bands because there were too few cases for a re- 4% of husband defendant cases, liable estimate. Also, cases used to investigate the effect of provocation on sentencing excluded where the wife had been the only one cases where provocation was unknown. Conse- drinking. quently, average sentences in this section may differ from averages shown elsewhere.

22 Spouse Murder Defendants in Large Urban Counties No explanation for why State prison Further reading sentences were, on average, 10 years shorter for wife defendants than Key findings reported here correspond husband defendants closely to what Professor Marvin Wolf- gang found four decades ago in his The average prison sentence for classic study of the justice system's convicted wife defendants was 6 handling of homicides in Philadelphia years, or about 10 years shorter than from 1948 to 1952. Comparing out- the average 16.5 years for husband comes between spouse murder defen- defendants (table 32). dants, Wolfgang reported these apparent disparities13  More spouse killings by wives than a higher proportion of husbands husbands were mitigated by victim (64%) than wives (55%) were found provocation, and wives received guilty; shorter sentences than husbands, but attributing wives' shorter sentences to a higher proportion of wives (34%) this one particular mitigating factor is than husbands (4%) were acquitted; not necessarily justified. Wife and husband defendants differed on factors husbands were convicted of more besides provocation that might explain serious degrees of homicide than were the 10-year difference. Moreover, the wives. 10-year disparity persists even when the comparison is restricted to de- No firm explanation for the disparities fendants who were alike in terms of  was offered, but Wolfgang did suggest whether they were provoked that they might largely be explained by The average prison sentence for differences between defendants in unprovoked wife defendants was 7 terms of victim provocation. Sixty years, or 10 years shorter than the percent of wife defendants were average 17 years for unprovoked strongly provoked by their mate, com- husband defendants.12 pared to 10% of husband defendants.

12It would also be desirable to compare sentence 13 See Patterns in Criminal Homicide (Philadel- lengths between provoked husbands and pro- phia: University of Pennsylvania Press, 1958) voked wives. That comparison was not made p. 217. because too few husband defendants were provoked to form the basis for a reliable comparison.

Spouse Murder Defendants in Large Urban Counties 23 Race in spouse murder cases

The race of the murder victim was Sentence unrelated to race of victim received a sentence of life imprison- known in 529 of the 540 cases. Of the ment, compared to 8% of defendants 529 total, the victim was black in 292 The likelihood of a convicted spouse in white-victim cases. cases  or 55%  and white in 225 murderer receiving a prison sentence  or 43%. Other races (Asian, was about the same whether the mur- For conviction for nonnegligent man- American Indian, Pacific Islander, or der victim was white or black  slaughter, the average prison term Alaska Native) made up the balance (excluding life sentences) was 8 years The convicted spouse murderer was of the 529. in white-victim cases, not significantly sentenced to prison in 93% of cases different from the average 6 years in where the victim was white, not signifi- black-victim cases. Conviction rate unrelated to race cantly different from the 87% of cases of victim where the victim was black (table 37). Race of defendant also unrelated The likelihood of a defendant being The length of the prison sentence to outcomes convicted of spouse murder was about imposed on a convicted spouse the same whether the murder victim Virtually all  97%  of spouse  murderer was generally unrelated to was white or black whether the murder victim was white murders were intraracial. When a white was murdered by a spouse, the The defendant was convicted in 81% or black 14 likelihood was about 97% that the al- of cases where the spouse murder vic- For conviction for first-degree leged murderer was white. When the tim was white, not significantly different murder, the average prison term spouse murder victim was black, the from the 79% of cases where the vic- (excluding life and death sentences) likelihood was 97% that the alleged tim was black (table 36). was 29 years in white-victim cases, not murderer was also black. Given the significantly different from the 32 years highly intraracial nature of spouse Table 36. Spouse murder in black-victim cases. murder, statistics relating the victim's defendants: Percent convicted, race to specific outcomes are inter- by race of victim, 1988 For conviction for second-degree changeable with those relating the murder, the average prison term Percent of spouse defendant's race to outcome. murder defendants (excluding life sentences) was 19 Total Con- Not years in white-victim cases, signifi- The overall conclusion that the victim's Victim race number Total victed convicted cantly longer than the 13 years in race is unrelated to outcome applies All 540* 100% 80% 20% black-victim cases. However, 23% equally well to the defendant's race. of convicted second-degree murder The likelihood of conviction, and of a 225 White 100 81 19 defendants in black-victim cases prison sentence if convicted, and the Black 292 100 79 21 length of the prison sentence were 14 *Includes victims of unknown race and other Sentences for negligent manslaughter are not about the same whether the spouse races not shown separately. Victim race was compared because they are based on too few known in 529 of the 540 cases. cases to be considered statistically reliable. murder defendant was white or black  Table 37. Defendants convicted of killing their spouse: Type of sentence, by race of victim, 1988 78% percent of white defendants were convicted, not significantly differ- Percent of convicted spouse murder defendants ent from the 80% of black defendants. Prison Number Total Straight Among convicted spouse murderers, Victim race convicted Total prison Life Death Jail probation 93% of white defendants were sentenced to prison, not significantly All 431* 100% 89% 12% 1% 1% 10% different from the 88% of black de- White 181 100 93 12 2 2 6 fendants. Black 230 100 87 10 1 1 12

*Includes victims of unknown race and other races not shown separately. Victim race was known in 529 of the 540 cases.

24 Spouse Murder Defendants in Large Urban Counties Methodology

Sample selection A total of 2,539 murder cases were Nonavailability of cases sampled. These cases were a sample The 33 counties in the sample were of about half of all murder-charge The survey goal to track murder cases selected to be representative of the cases disposed of in the sampled 33 across justice system stages was not Nation's 75 largest counties. The counties in 1988. Not eligible for sam- met in nine counties. In one of the ranking of counties in which the 75 ple selection were nonmurder de- nine, legal restrictions barred access largest were identified was based on a fendants or any whose most serious to cases rejected by the prosecutor. combination of crime data (1980 and charge was attempted murder, negli- In the remaining eight counties, some 1984 Uniform Crime Report Part I gent or involuntary manslaughter, or of the sampled cases could not be arrests) and population data (1980 vehicular homicide. In counties with located. population from the Census Bureau's 200 or fewer disposed murder cases City County Data Book). Rankings in 1988, all were selected for inclusion Computation of estimates from sample reflected the size of the prosecutors' in the sample. In counties with more data offices. The original sample plan iden- than 200, a systematic sample of 200 tified 34 counties, 1 of which declined was chosen. Only 6 of the 33 counties Case weights were applied to statistics to participate. had more than 200 murder cases. on the sampled cases to form estimates for the universe of the 75 The following are the 33 counties Virtually all cases meeting the 1988- largest counties, the key assumption whose prosecutors' offices participated disposition criterion were disposed for being that cases not sampled were in the study reported here  all defendants in the case. Of the similar to the cases sampled. A case 3,119 defendants on whom data were weight was the inverse of the probabil- Arizona New Mexico obtained, only 13 had not yet had their ity that a case would be in the survey. Pima Bernalillo cases adjudicated at the time the sur- That probability was the product of California New York vey was carried out in 1990. Another both the probability that a given county Los Angeles Kings 25 defendants had died of suicide or would be chosen and the probability of Orange Monroe other causes, either at about the time selection of that case in that county. Kern New York of the murder or later, before final Case weights were adjusted to com- San Diego Queens disposition. pensate for the loss of one nonpartici- Riverside Ohio pating county. Colorado Franklin Sample of spouse murder defendants Denver Montgomery Statistically weighted, the 3,119 de- Arapahoe Oklahoma Of the 3,119 sampled murder defendants in the sample cases repre- Connecticut Oklahoma fendants, 187 were spouse murder sented 9,576 murder defendants in the New Haven Pennsylvania defendants. The 187 consist of 111 Nation's 75 largest counties. Florida Philadelphia husband defendants and 76 wife de- Dade Allegheny fendants. Extrapolating to the 75 larg- Orange Tennessee est counties, the 187 represent an Broward Shelby estimated 540 spouse murder defen- Illinois Texas dants. The figure 540, then, is based Cook Dallas on a sample size approximately one- Louisiana Tarrant third of 540. Readers wishing to deter- Orleans Travis mine for themselves the approximate Maryland Washington sample size on which any of the Prince King report's statistics are based need only George's to multiply the table estimate by Massachusetts one-third. Middlesex Michigan Wayne Missouri St. Louis

Spouse Murder Defendants in Large Urban Counties 25 Response rates difference in which there is at least 90% confidence. Except as noted below, this report focused exclusively on characteristics Except where explicitly indicated other- that were successfully obtained in a wise, all differences discussed in this high percentage of sample cases report had a confidence level at or ("response rate"). above 90% (virtually all were above 95%). Standard error estimation was The case records identified age, race, based on software that takes into sex, and ethnicity for nearly all de- account the survey's sample design fendants (approximately 98%). features.

Also obtained in nearly all cases were Limited data on provocation the relationships between victims and defendants; the circumstances The report reaches no firm conclusion preceding the homicide; the arrest about the possible impact of victim or indictment charge; and whether the provocation on case outcomes, in part defendant was convicted, and if so, the because thorough analysis requires conviction offense. In incarceration or more cases and more details than are probation cases, the length of the term available from this study's spouse of sentence was usually known. murder database. For instance, the survey does not show  Defendant criminal history was available in three-quarters of the cases, but which defendants actually claimed victim criminal history was obtained in self-defense; only a third of the cases. The juvenile portion of the criminal history informa- which of the spouses in each case tion was probably less complete than was the first to strike or threaten the the adult portion. other;

Sampling error which defendants received a charge or sentence reduction because prose- Data collected in this murder study cutors, judges, or juries decided victim were from a probability sample rather provocation was present;15 than a complete enumeration. Be- cause counties and cases were sam- which claims of self-defense were pled, a sampling error (standard error) supported by strong evidence; is associated with each number in the report. In general, if the difference be- which defendants claiming self- tween two numbers is greater than defense had the option of fleeing twice the standard error for that differ- rather than using deadly force. ence, there is at least 95% confidence 15 that the two numbers are in fact differ- However, the survey did document instances where the prosecutor screened out the case ex- ent; that is, the apparent difference is pressly because of victim provocation. not simply the result of surveying a sample rather than the entire population. Similarly, if the difference between two numbers is greater than 1.6 standard errors, we are at least 90% confident that the two numbers are different. In this report the term "statistically significant" was used to denote a

Spouse Murder Defendants in Large Urban Counties 26