Volume 87

Fall 1996

Justice Commons

This Criminology is brought to you for free and open access by Northwestern University School of Law Scholarly Commons. It has been accepted for

inclusion in Journal of Criminal Law and Criminology by an authorized editor of Northwestern University School of Law Scholarly Commons.

Recommended Citation

Michael L. Radelet, Ronald L. Akers, Deterrence and the Death Penalty: The Views of the Experts, 87 J. Crim. L. & Criminology 1

(1996-1997)

I. INTRODUCTION

The American public has long been favorably disposed toward

capital punishment for convicted murderers, and that support continues

to grow. In a 1981 Gallup Poll, two-thirds of Americans voiced

general approval for the death penalty. That support rose to 72% in

1985, to 76% in 1991, and to 80% in 1994.1 Although these polls

need to be interpreted with extreme caution, it is clear that there are

few issues on which more Americans agree: in at least some circumstances,

death is seen as a justifiable punishment.

Part of the support for capital punishment comes from the belief

This justification suggests that murderers should be executed for retributive

reasons: murderers should suffer, and the retributive effects

of life imprisonment are insufficient for taking a life. While such

views are worthy of debate, no empirical research can tell us if the

argument is "correct" or "incorrect." Empirical studies can neither

answer the question of what specific criminals (or non-criminals) "deserve,"

nor settle debates over other moral issues surrounding capital punishment.

* We appreciate the helpful comments from Phoebe Ellsworth, William Bailey and

Samuel Gross on early drafts of our questionnaire.

** Professor and Chair, Department of Sociology, University of Florida.

*** Professor of Sociology and Director of the Center for Studies in Criminology and

Law, University of Florida.

1 David W. Moore, Majority Advocates Death Penalty for Teenage Killers, GALLuP PoLL

MONTHLY, Sept. 1994, at 5.

2 Hugo Adam Bedau, Retribution and the Theory of Punishment, 75 J. PHIL. 602 (1978);

On the other hand, much of the support for capital punishment

rests on its presumed value as a general deterrent: we need the death

penalty to encourage potential murderers to avoid engaging in criminal

deterrence rationale in their cries for more and quicker executions

Whether or not the threat or use of the death penalty is, has

been, or could be a deterrent to homicide is an empirical question

that should not-and cannot-be answered on the basis of moral or

political stands. It is an empirical question that scores of researchers,

dating back to a young Edwin Sutherland, writing in the pages of this

Has this long history and sizeable body of research led to any

general conclusions? Can any factual statement be made about the

death penalty's deterrent effects, or are the scholarly studies such that

no conclusions can be reached? At least two valid methods can be

used to answer these questions. One is to examine individual scholarly

opinions, as is done in most published research reports. Here

researchers review the empirical research on deterrence and reach

conclusions based on it and their own research. A second approach is

to gauge the informed opinions of scholars or experts. Indeed, much

research-based public policy rests on known or presumed consensus

of "expert" opinions. It is the aim of this paper to address the question

of the death penalty's ability to deter homicides using this second

approach: by gauging the judgments of a set of America's top

criminologists.

Measuring sentiment on the death penalty is not as easy a task as

it might at first appear. When opinion polls ask respondents whether

they support the death penalty, often no alternative punishments are

given, and respondents are left to themselves to ponder what might

happen if a particular inmate were not executed. Often respondents

erroneously believe that absent execution, offenders will be released

to the community after serving a short prison sentence. 6 Even the

(1925).

6 SeeJames Alan Fox et al., Death Penalty Opinion in the Post-Furman Years, 18 N.Y.U.

most ardent death penalty abolitionists might support capital punishment

if the alternative was to have dangerous murderers quickly released

from prison. When respondents are asked how they feel about

the death penalty given an alternative of life without parole, support

supported the death penalty, but that support would drop to 53% if

life imprisonment without parole were available as an alternative.8

While most deterrence research has found that the death penalty

has virtually the same effect as long-term imprisonment on homicide

rates,9 in the mid-1970's economist Isaac Ehrlich reported that he had

uncovered a significant deterrent effect.10 He estimated that each execution

between 1933 and 1969 had prevented eight homicides."

This research gained widespread attention, in part because Solicitor

General Robert Bork used it to defend the death penalty in the 1970s

when the Supreme Court was considering whether to make permanent

its 1972 ban of the death penalty.12 Although scholars, including

a panel appointed by the National Academy of Sciences, 13 strongly

criticized Ehrlich's work for methodological and conceptual shortcomings,

14 some continue to cite it as proof that the death penalty

does have a deterrent effect.' 5 A student of Ehrlich's, Stephen Layson,

later reported his estimate that each execution deterred approximately

18 homicides.' 6 This research, too, was loudly criticized,' 7 but

nonetheless it continues to be embraced by proponents of the death penalty.' 8

157, 167-71 (1993).

7 See Fox et al., supra note 6, at 514-15; see also Bowers, supra note 6, at 163-64.

8 Alec Gallup & Frank Newport, Death Penalty Support Remains Strong, GALLup POLL

MONTHLY, June 1991, at 40.

9 RAYMOND PATERNOSTER, CAPITAL PUNISHMENT IN AMERICA 217-45 (1991).

10 Isaac Ehrlich, The Deterrent Effect of Capital Punishment: A Question of Life and Death, 65

AM. ECON. REv. 397 (1975).

11 Id. at 398.

12 THE DEATH PENALTY IN AMERICA 95 (Hugo Adam Bedau ed., 3d ed. 1982).

ON CRIME RATES (Alfred Blumstein et al. eds., 1978).

14 See, e.g., Brian Forst, Capital Punishment and Deterrence: ConflictingEvidence?, 74J. CRIM.

Ronald Huff eds., 1981).

15 See, e.g., Thomas Sowell, Death Penalty is Valid Option, ST. Louis POST DISPATCH, Dec.

12, 1994, at lC.

It could very well be that the mere existence of a critique is more

important than the quality of that critique. One researcher finds one

thing, and another claims to refute it. What is left is a net gain of

zero: politicians who never read or understand the original studies

can select either position and cite only those studies that support their position.

Some research has asked the general public whether the death

penalty acts as a deterrent to murder. Such a question is regularly

over 60% of the respondents in Gallup polls said they believed the

death penalty was a deterrent. Furthermore, these polls showed that

the deterrence rationale is an important death penalty justification.

In the 1986 Gallup Poll, respondents were asked if they would support

the death penalty "if new evidence proved that the death penalty does

not act as a deterrent to murder." Given this assumption of no deterrent

Similarly, in the 1991 poll, where 76% of the respondents initially

indicated support for the death penalty, Gallup asked those who favored

the death penalty: "Suppose new evidence showed that the

death penalty does not act as a deterrent to murder, that it does not

lower the murder rate. Would you favor or oppose the death penalty?"

As in the earlier poll, the respondents were less likely (76% vs.

52%) to support capital punishment if it were shown that it is not a

of a deterrent effect is a major factor in public and political endorsement

of the death penalty. If that assumption is undermined, even

those who initially favor the death penalty tend to move away from it.

In another study that sheds light on the public's view of the death

penalty's deterrent abilities, Ellsworth and Ross mailed questionnaires

that 82% of the death penalty proponents, but only 3.1% of the opponents,

agreed with the statement, "We need capital punishment to

show criminals that we really mean business about wiping out crime in

Rights ofthe House Comm. on theJudiciay, 103rd Cong. 228, 253-55 (1993) (statement of Paul

G. Cassell, Associate Professor of Law, University of Utah College of Law).

19 See, e.g., Gallup & Newport, supra note 8, at 40; 7 in 10 FavorDeath Penalty for Murder,

GALLuP REPORT, Jan.-Feb. 1986, at 10; Support for the Death Penalty Highest in Half-Century,

GALLuP REPORT, Jan.-Feb. 1986, at 3.

21 See Gallup & Newport, supra note 8, at 41, 43.

[Vol. 87

assumption of deterrence is one of the most important foundations

for death penalty support in America. Questions from both the Gallup

and the Ellsworth/Ross surveys were used in our own research, so

precise comparisons will be made when our results are discussed below.

One recent survey has been conducted that examines how leading

police officials, who arguably hold more expertise on criminal behavior

than the general public, view the deterrence rationale for

capital punishment. The survey was conducted in 1995 by the Washington,

Telephone surveys were conducted with 386 randomly selected police

chiefs and county sheriffs from throughout the U.S. Little support for

the deterrence argument was found. Among six choices presented as

"primary" ways to reduce violent crime, only one percent of the law

enforcement respondents chose the death penalty. This choice

ranked last among the options. When asked to consider the statement

"The death penalty significantly reduces the number of homicides,"

67% of the chiefs felt the statement was inaccurate, while only

26% said it was accurate. Reacting to the poll, former New York Police

Chief Patrick V. Murphy wrote, "Like the emperor's new clothes,

the flimsy notion that the death penalty is an effective law enforcement

purposes, some of the questions we posed to our sample

(reported below) were taken from this survey.

III. METHODOLOGY

In order to assess what the experts think about the deterrent effect

of the death penalty, we must first define the term "expert." According

to one definition, the law enforcement executives surveyed by

Hart are experts. Another definition would include scholars who have

conducted high-quality scholarly research on the death penalty and

deterrence, such as the panel appointed two decades ago by the National

document the views of these researchers, but such a survey would simply

reflect disagreements that are evident in the scholarly literature,

not evaluate or judge them.

24 RicHARD C. DmER, DEATH PENALliy INFORMATION CENTER, ON THE FRONT LINE: LAW

ENFoRcEMENT VIEws ON THE DEATH PENALTY 2 (1995); PETER D. HART RESEARCH Associ-

ATES, INc., STUDy No. 4286 DEATH PENALTY--PoUCE CHIEFS (1995) (on file with author).

But what about other leading criminologists who are not specialists

in capital punishment research but who have gained more general

visibility and leadership in the field? It is this group of "experts," as

defined by visibility and recognition as leaders among professional

criminologists, that we surveyed for this project. We operationally define

"expert" as one who has been recognized by peers by being

elected to the highest office in scholarly organizations. We contacted

all present and former presidents of the country's top academic criminological

societies. This small and elite group includes many of the

country's most respected and distinguished criminologists. As such,

although few of these scholars have done research on capital punishment

or deterrence, they are generally well versed in central criminological

issues, such as crime causation, crime prevention, and criminal

justice policy issues. The presidents of three associations were surveyed:

• the American Society of Criminology,

• Academy of Criminal Justice Sciences, and

• the Law and Society Association.

The American Society of Criminology (ASC), founded in 1941, is

the country's largest association of professional and academic criminologists,

Justice Sciences (ACJS), founded in 1963, today includes 3,350

the ASC, but its leadership (taken primarily from undergraduate

teaching programs) does not. Only one person in the history of the

Association (LSA), founded in 1964, includes more law professors and

legal scholars among its 1,400 members than either the ASC or

but no one has served as president of LSA and either of the other two.

These three associations are all interdisciplinary and publish what are

among the most respected scholarly journals in criminology and criminal

We began by obtaining names and addresses of current and all

living former presidents of each of the three organizations. A total of

seventy one individuals were identified: twenty nine from the Academy

of Criminal Justice Sciences, twenty seven from the American So-

27 ENCYCLOPEDIA OF ASSOCIATIONS 10803 (SandraJaszczak ed., 31st ed. 1996).

28 Id. at 10742.

29 Each organization elects officers, including a president, by a ballot sent to all members.

To be elected president, one must generally have high visibility in the field, be wellrespected,

and have been active in programmatic and organizational activities.

Association. As noted, one person had served as president of two of

the associations, reducing our sample to seventy. Drafts of the questionnaire

were critiqued by three scholars who have conducted deterrence

research. Numbered questionnaires were mailed to our

respondents, and follow-ups were sent to non-respondents. In the

end, a total of sixty seven responded (95.7%): twenty seven from

ACJS, twenty six from the ASC, and fifteen from LSA.

The presidents were dearly asked in both the cover letter and on

did not ask for their personal opinions on the death penalty-information

on this might be interesting, but it is irrelevant to the goal of

the present study. Eleven questions, all relating to deterrence issues,

were included on the questionnaire; the responses to all eleven are

reported below.

A. GENERAL QUESTIONS ON DETERRENCE

The first question explored concerns how the presidents generally

view the deterrence question. Table I begins by replicating the

question asked in the Gallup polls, "Do you feel that the death penalty

acts as a deterrent to the commitment of murder-that it lowers the

murder rate, or what?" It can be seen that the criminologists are more

than twice as likely as the general population to believe that the death

penalty does not lower the murder rate-41 % of the population held

this belief in 1991, the last year that Gallup published responses to this

question, compared to 83.6% of our experts. Among the sixty four

presidents who voiced opinions on this question, fifty six (87.5%) believe

the death penalty does not have deterrent effects.

Table 1 also compares responses to deterrence questions between

our respondents and the members of the general public in northern

California surveyed by Ellsworth and Ross. Here 86.5% of the criminologists

and 46% of the general public say they are "sure" or "think"

that "abolishing the death penalty (in a particular state) would not

have any significant effects on the murder rate (in that state)." As

would be expected, substantially more members of the general public

than the criminologists (32.6% vs. 10.4%) say they have no idea

whether this statement is true or false.

Similarly, as shown in the third part of Table 1, the criminologists

31 One of these former presidents is a co-author of this paper (RIA).

are much less likely than members of the general public to agree that

"Over the years, states which have had the death penalty have had

lower murder rates than neighboring states which did not have a

death penalty." Nearly 80% of the criminologists said that they were

sure or they thought this was not true, compared to 37% of the general

public. Interestingly, more criminologists stated that they had no

idea whether this statement was true or false than did members of the

general public (14.9% vs. 6.0%).

The results of Table 1 clearly show that approximately 80% of the

experts in criminology believe, on the basis of the literature and research

in criminology, that the death penalty does not have significant

deterrent effects. In addition, no matter how measured, it is clear that

the criminologists are much more likely than the general public to

dismiss the deterrence argument.

Table 2 compares the beliefs of our experts to those of top criminal

justice administrators, specifically to the beliefs of the police chiefs

surveyed by Peter D. Hart Research Associates in 1995 (discussed

above) .32 Overall there is widespread agreement between the criminologists

and the police chiefs on the deterrent value of the death

penalty (or lack thereof), with the criminologists even less likely than

the chiefs to see any deterrent value. As seen in Table 2, all of the

criminologists, and 85% of the police chiefs, believe it is totally or

largely accurate that "politicians support the death penalty as a symbolic

way to show they are tough on crime." Almost 87% of the criminologists

and 57% of the chiefs find it totally or largely accurate to say

that "debates about the death penalty distract Congress and state legislatures

from focussing on real solutions to crime problems." None of

the criminologists, and only about a quarter of the chiefs, believe

there is any accuracy in the statement, "the death penalty significantly

reduces the number of homicides." These statements indicate that

both academic criminologists and police chiefs view the death penalty

as more effective in political rhetoric than as a criminal justice tool.

Table 3 asks general questions about deterrence in two different

ways. We developed the wording for these questions ourselves, so no

comparisons with other opinion polls are possible. However, we believe

these questions word the issue more precisely than the questions

taken from other surveys. Given the widespread availability of "life

displayed in Table 3 is perhaps the clearest statement of the de-

32 DirER, supra note 24, at 10 fig.4, 14-15.

[Vol. 87

terrence issue as actually faced by researchers and policy makers

today. It focuses on the unique deterrent effect of the death penalty

above and beyond available alternatives of long imprisonment. Only

three of our respondents (4.5%) agreed, and none strongly agreed,

with the statement, "overall, over the last twenty years, the threat or

use of the death penalty in the United States has been a stronger deterrent

to homicide than the threat or use of long (or life) prison

sentences." Those disagreeing or strongly disagreeing included

92.6% of the respondents, and 96% of those with an opinion. Responses

to the next question indicate that only three respondents felt

that the empirical support for the deterrent effects of the death penalty

had moderate support; none believed it had strong support. Instead,

94% of the criminologists felt the argument had weak or no

support.

B. THE QUESTION OF REFORM

Proponents of the death penalty might concur with the critics of

the deterrence argument, but say that the lack of a clear deterrent

effect is a result of the fact that only a small proportion of those on

death row are executed each year, or that the wait on death row between

condemnation and execution is too long. Increasing the frequency

and celerity of the death penalty could produce a deterrent

effect. The experts responding to our survey, however, disagree with

such a position. Almost 80% disagree or strongly disagree with the

statement, "if the frequency of executions were to increase significantly,

more homicides would be deterred than if the current frequency

of executions remained relatively stable." As seen in the

second portion of Table 4, nearly three quarters (73.2%) of the experts

disagreed or strongly disagreed with the position that decreasing

the time on death row would deter more homicides. Much of the

research that informs these experts' opinions was done with data from

the 1930's, 1940's, and 1950's, when the frequency of executions was

higher and the average time spent on death row was shorter than it is

today. Hence, criminologists do have some research at their disposal

that would enable accurate predictions of what would happen if these

proposed death penalty reforms were actually enacted.

C. SUPPORT FOR THE BRUTALIZATION HYPOTHESIS

In a final question, the experts were asked how they felt about the

so-called "brutalization hypothesis." This argument, supported by

34 William C. Bailey, Deterrence and the Death Pnaltyfor Murder in Oregon, 16 WILIAMETTE

human life and sends a message that tells citizens that killing people

under some circumstances is appropriate. However, as shown in Table

5, this hypothesis does not have widespread support among the

experts. Two-thirds (67.1%) of the respondents either disagree or

strongly disagree with the statement, "overall, the presence of the

death penalty tends to increase a state's murder rate rather than to

decrease it."

The responses to this item help us address some possible reservations

about our overall findings: Is there anti-capital punishment bias

among the respondents? Were the responses made based on an understanding

of the research or are our respondents merely liberal academics

who object to the death penalty on moral grounds and would

report opinions that might undermine it, even if the empirical evidence

showed otherwise? The responses to the question on brutalization

suggest that the answers to these questions are negative. If the

respondents simply responded to any question in a way that buttresses

the abolitionist position, there should be strong agreement with the

notion that the death penalty actually increases the homicide rate,

since this is an anti-capital punishment argument. It appears, instead,

their appraisal of existing research. The brutalization hypothesis, in

fact has not been tested very well and the research supporting it remains

more suggestive than definitive. As on the other questions, the

respondents appear to have reacted to the state of knowledge on this

question (as they were instructed), not to personal preferences.

V. CONCLUSIONS

The results of this project show that there is a wide consensus

among America's top criminologists that scholarly research has

demonstrated that the death penalty does, and can do, little to reduce

rates of criminal violence. Hence, these leading criminologists do not

concur with one of the most important public justifications for the

death penalty in modern society.

Do politicians and policy makers pay any attention to expert

opinions among members or leaders of scholarly societies? There is

some evidence in the recent ASC task force panel reports to the Attor-

107, 110-30 (1994).

[Vol. 87

studying a dozen crime control policy options, did not examine the

issue of capital punishment. The advice we would offer, reflecting the

opinions of the presidents of the major criminological organizations,

is to shift public debates about how to reduce criminal violence in

America away from the death penalty.

Capital punishment will continue to generate much public debate

in the early decades of the next century and various bodies of

opinion will be consulted. One important body of opinion has been

revealed by this study. The results show that the question of whether

or not the death penalty can reduce criminal violence is - at least for

the presidents of the major scholarly societies in criminology - a settled

issue. Hopefully this study will provide policy makers with information

that might help move political debate beyond "gut" feelings

and simplistic demands for the death penalty as a way of "getting

tough" on crime. Careful consideration of alternatives can build a

public consensus around more effective policies that really hold promise

in reducing America's high rates of criminal violence.

TABLE 1

COMPARISON OF -RESPONSES OF CRIMINOLOGISTS AND GENERAL PUBLIC

A. Do you feel that the death penalty acts as a deterrent to the commitment

of murder - that it lowers the murder rate, or what?

Yes:

No:

No Opinion:

N

Criminology Presidents

11.9

83.6

4.5

67

Gallup 198536

62

31

7

1,523

Gallup 199137

51

41

990

B. Abolishing the death penalty (in a particular state) would not have any

significant effects on the murder rate (in that state).

I'm sure it is true

I think it's true

I have no idea

whether it is

true or false

I think it's false

I'm sure it's false

N

Criminology Presidents Ellsworth and Ross, 198338

32.8 10.2

53.7 35.8

32.6

18.0

3.4

500

C. Over the years, states which have had the death penalty have had lower

murder rates than neighboring states which did not have a death penalty.

I'm sure it is true

I think it's true

I have no idea

I think it's false

I'm sure it's false

N

Criminology Presidents

6.0

14.9

40.3

38.8

67

Ellsworth and Ross39

4.6

22.4

6.0

32.0

5.0

37 Gallup & Newport, supra note 8, at 41.

38 Ellsworth & Ross, supra note 22, at 141. The Ellsworth and Ross question read,

"Studies have not found that abolishing the death penalty has any significant effect on the

murder rate in a state."

39 Id.

[Vol. 87

CHIEFS40 TO SAME QUESTIONS (N=386) (IN PERCENTS)

A. Politicians support the death penalty as a symbolic way to show they are

tough on crime.

Totally accurate

Largely accurate

Largely inaccurate

Totally inaccurate

Not sure

Presidents

38.8

0

Police Chiefs

33

52

B. Debates about the death penalty distract Congress and state legislatures

from focussing on real solutions to crime problems.

Totally accurate

Largely accurate

Largely inaccurate

Totally inaccurate

Not sure

Presidents

49.3

37.3

Police Chiefs

11

46

30

11

2

C. The death penalty significantly reduces the number of homicides.

Totally accurate

Largely accurate

Largely inaccurate

Totally inaccurate

Not sure

Presidents

00

41.8

52.2

6.0

Police Chiefs

4

22

45

22

7

TABLE 3

RESPONSES OF CRIMINOLOGISTS TO GENERAL QUESTIONS ON

A. Overall, over the last twenty years, the threat or use of the death penalty in

the United States has been a stronger deterrent to homicide than the threat

or use of long (or life) prison sentences.

Strongly agree 0

Agree 4.5

Disagree 43.3

Strongly disagree 49.3

Missing 3.0

B. Overall, how would you evaluate the empirical support for the deterrent

effects of the death penalty?

Strong support 0

Moderate support 4.5

Weak support 44.8

No support 49.3

Missing 1.5

[Vol. 87

TABLE 4

RESPONSES OF CRIMINoLoGISTS TO BELIEF THAT REFORMS CouLD

PRODUCE A DETERRENT EFFECT (N=67) (IN PERCENTS)

A. If the frequency of executions were to increase significantly, more

homicides would be deterred than if the current frequency of executions

remained relatively stable.

Strongly agree 3.0

Agree 14.9

Disagree 44.8

Strongly disagree 34.3

Missing 3.0

B. The average time on death row between sentence and execution is now

between eight and ten years. If that period was reduced significantly, there is

reason to expect that the death penalty would deter more homicides than it

does today.

Strongly agree 4.5

Agree 22.4

Disagree 44.8

Strongly disagree 28.4

16 RADELET & AKERS [Vol. 87

CRIMINOLOGISTS' RESPONSES TO THE BRUTALIZATION HYPOTHESIS

Overall, the presence of the death penalty tends to increase a state's murder

rate rather than to decrease it.

Strongly agree 4.5

Agree 23.9

Disagree 52.2

Strongly disagree 14.9

Missing 4.5