Death and Deterrence

by Burk Foster

(June 2001)

Appeared in The Angolite, June/July/August 2001, pp. 21-23.

In 17th and 18th century London, the execution of a condemned criminal was a public spectacle, an event that might have been orchestrated by Cecil B. DeMille. After a farewell dinner (and often a farewell party, if the condemned could afford it) the night before, the condemned (one or several, as multiple executions were common) were paraded through the streets to the site of the gallows. Before the huge throng gathered in the square, the condemned person was expected to give a speech and ask for forgiveness. He then dropped a handkerchief to signal that he was ready to be hanged. The point of this ritual, historians say, was to educate the public about the consequences of criminal behavior, and thereby to discourage similar behavior. Today we call this idea deterrence.

In America executions remained public until the post-Civil War era when, first in the Northern states and later in the South, they were moved behind jail and prison walls and made by-invitation-only events. Journalists were always among those allowed to view the executions, to report the details to the larger public no longer allowed to attend in person. Political officials and death penalty proponents continued to speak of the deterrent effect of the death penalty, even as newspaper accounts universally replaced the direct viewing (and sometimes overt participation, for and against) of the earlier era.

Did executions, public and private, deter crime--or more specifically homicide, as the crime most likely to be punished by death? Most research says no. Probably the best known of the studies that says yes is Isaac Ehrlich's 1975 article, "The Deterrent Effect of Capital Punishment: A Question of Life and Death." Looking at executions and homicides between 1933 and 1969, an era when both were in decline, Ehrlich claimed that each execution prevents seven or eight murders. In a 1980 article, sociologist David Phillips, who studied executions and homicides in London between 1858 and 1921, reported that a well-publicized execution resulted in a temporary drop in the homicide rate, a decline which was quickly followed by an increase to an above-normal rate thereafter.

Most of the research has suggested opposite conclusions, finding that executions either had no deterrent consequences or even increased the number of homicides. Recent studies in Texas, California, Oklahoma, and Arizona have found no apparent deterrent effect of executing murderers. William Bowers and Glenn Pierce, writing in "Deterrence or Brutalization: What Is the Effect of Executions?" in 1980, used the term "brutalization effect" to suggest that viewing an execution or reading about one might encourage certain persons to kill. If we believed in deterrence, we would hope that a might-be murderer would identify with the person being executed, and would fear the legal consequences of killing and thus avoid the act itself. But Bowers and Pierce posed an interesting thought: what if the potential killer identifies with the power of the state, used against its victims, to justify his own violence against other victims. If this sounds farfetched, you might consider the case of one Timothy McVeigh, about to become the first federal prisoner executed in four decades.

Many historical observers of executions expressed skepticism that the event was having much of a deterrent effect. Crowds in London in the 1600s and 1700s were often drunk and unruly, making the event more like Mardi Gras than a moral philosophy lecture, and historians have long pointed out that execution crowds were a favorite working place for pickpockets, criminals who knew, even as they plied their trade, that if they were caught it would be their turn on the gallows next. There is the story, told in Robert Johnson's Death Work, of the mother in the crowd, calling out to her son about to be hanged, "Son, I hope you will die courageously, like your father."

Closer to home, in both time and place, Table 1 below was prepared to consider the deterrent effects of executions on the murder rate in Louisiana during the decade of the 1990s.

Does executing criminals deter homicide? No definite relationship can be established from our own recent experience. It is clear, in looking at the FBI's homicide statistics, that the murder rate in Louisiana fell off sharply during the latter half of the 1990s. It is also clear that juries gave out more death sentences from 1995 to 1999 (56) than they did from 1990 through 1994 (29). More death penalties, fewer murders; isn't that what prosecutors have been saying all along?

But when you look at the stats by parish, rather than statewide, a more uncertain relationship emerges. Most of Louisiana's parishes did not give anyone a death sentence during the 1990s. Of those that did, by far the most significant decline in the murder rate occurred in Orleans Parish, where defendants are least likely to get a death sentence. New Orleans, with 11% of the state's population, had 44% of the state's homicides during the 1990s. Another sharp decline in the murder rate took place in Caddo Parish, which also falls lower on the list of parishes ranked by the ratio of homicides to death sentences.

Local officials in New Orleans and Shreveport, when asked to explain why their murder rates have declined, typically mention such influences as a decline in violence associated with crack cocaine distribution, police tactics aimed at keeping guns off the streets, the decline of gang violence, expansion of alternatives for victims of domestic violence, and improved employment conditions in urban areas. The role of the death penalty, as a public policy influencing criminal violence, is so negligible that it is scarcely mentioned.

For the decade as a whole, two parishes, East Baton Rouge and Jefferson, gave out the most death sentences, a total of 30 between them. These two parishes, with 20% of the state's population and 17% of its homicides, meted out 35% of its death sentences. Criminals: are you listening to the tough-talking DA's in East Baton Rouge and Jefferson? Maybe in Jefferson, where the murder rate did decline substantially in the latter half of the decade.

But the experience of East Baton Rouge lends support to the old argument that the death penalty is much more about politics than it is either justice or crime control. East Baton Rouge, which gave more death sentences than any other parish in the 90s, was the only urban parish in which the murder rate went up in the second half of the decade. The more death sentences, the more murders. So if East Baton Rouge wanted to follow the homicide trend in Caddo and Orleans, it should givefewer death sentences, and make its streets safer.

Statewide only about one in 84 homicides results in a death sentence, mostly because of the very low rate of death sentences in Orleans Parish. This is slightly below the recent national average of one death sentence for every 60 homicides. But even if we gave ten times the number of death sentences we do now, only about one in every eight or nine murders would result in a death sentence, which would still likely not be enough to significantly increase the deterrent effect of death.

We might keep in mind the two main reasons why England and America moved away from the death penalty and other corporal punishments at the end of the 1700s, and toward the use of the penitentiary in the 1800s: (1) the physical punishments were widely viewed as inhumane and uncivilized in a modern society; and (2) because the legal system did not often impose these punishments, they were viewed as being largely ineffective in deterring crime. The penitentiary replaced death as a punishment (for all but the most unlucky or heinous murderers), precisely because it was viewed as a more certain punishment. So it remains today, especially in a state like Louisiana which imposes true life sentences to incapacitate criminals who were not deterred.























Table 1

Homicides and Death Sentences in Louisiana: 1990-1999*

Homicides by Parish (Population in 1,000s)

Homicides Rate (per Death Sentences (484) (457) (396) (247) (177) (173) (167) (146) (126) (100)

Louisiana 100,000) Louisiana Orlns Jeff EBR Cad Lafy Calc St.T Ouac Rapid Terre

1990 724 17.2 1 304 63 57 68 9 15 (7)³ 6 (11)⁴ 9

1991 720 16.9 7 345 62 65 53 15 (13)² (5)³ 11 (9)⁴ 3

1992 747 17.4 6 279 50 72 55 20 (18)² (6)³ 26 (12)⁴ 14

1993 874 20.3 8 395 65 86 83 20 15 10 23 25 6

1994 856 19.8 7 424 54 79 69 11 19 6 15 11 7

1995 740 17.0 12 363 38 80 63 8 12 7 9 11 4

1996 762 17.5 8 351 36 86 54 10 21 12 9 11 (2)¹

1997 682 15.7 13 267 39 70 42 20 18 7 16 8 (4)¹

1998 560 12.8 12 230 33 73 37 12 9 8 12 10 6

1999 468 10.7 11 158 33 64 35 12 5 6 14 10 3

Total: 7,133 Total: 85 3,116 473 732 559 137 145 74 141 118 58

Total homicides in urban parishes: 5,573 (78.1% of state total)

Total homicides in rural parishes: 1,560 (21.9% of state total)

¹Houma P.D. did not report statistics.

²Calcasieu S.O. did not report statistics.

³Slidell P.D. did not report statistics.

⁴Rapides S.O. did not report statistics.

1990s Death Sentences by Parish Ratio of Homicides to Death Sentences

E. Baton Rouge 16 46/1

Jefferson 14 38/1

Orleans 12 260/1

Caddo 8 70/1

Rapides 4 30/1

Calcasieu 4 36/1

Ouachita 3 47/1

Terrebonne 2 29/1

St. Tammany 1 74/1

Lafayette 0 None

Others 21 74/1

*Homicide statistics from FBI Uniform Crime Reports Crime in the United States.