Putting desert in its place

Stanford Law Review

Volumn 65, Issue 1, 2013, Pages 77-136

  Slobogin, Christopher ^a   Brinkley Rubinstein, Lauren ^b  

^a VANDERBILT UNIVERSITY   (United States)

^b VANDERBILT UNIVERSITY   (United States)

Author keywords

[No Author keywords available]

Indexed keywords

EID: 84873026693     PISSN: 00389765     EISSN: None     Source Type: Journal    
DOI: None     Document Type: Article

References (195)

1. Michael Tonry, Introduction to WHY PUNISH? HOW MUCH?: A READER ON PUNISHMENT 3, 3 (Michael Tonry ed., 2011).
2. See id. at 7 ("The fundamental contrast between the ideas that punishment is morally justified because people have behaved wrongly and that punishment is morally justified only when it has good consequences has long existed and most likely always will.").
3. Paul H. Robinson, DISTRIBUTIVE PRINCIPLES OF CRIMINAL LAW: WHO SHOULD BE PUNISHED HOW MUCH? 139-140 (2008).
4. These utilitarian goals, which focus on reducing recidivism in individual offenders, are meant to be distinguished from general prevention goals, the best example of which is general deterrence.
5. Richard J. Bonnie et al., CRIMINAL LAW 25-37 (3d ed. 2010) (categorizing specific deterrence, incapacitation, and rehabilitation as "individual prevention" goals, and distinguishing those goals from general deterrence).
6. See infra Part II.A.1.
7. See infra Part II.A.2.
8. See infra Part II.B.
9. See infra Part II.C.
10. See infra Tables 5-7.
11. As noted earlier, see supra note 5, this Article's references to "preventive" purposes of punishment are meant to focus on the "individual" prevention goals of incapacitation, rehabilitation, and specific deterrence, rather than general deterrence. The assumption throughout this Article is that, while some sort of punishment is crucial to deterring crime, no particular punishment is necessary to ensure that goal. As Michael Tonry summarized the research of three National Academy of Sciences panels, "[i]maginable increases in severity of punishments do not yield significant (if any) marginal deterrent effects."
12. Michael Tonry, Purposes and Functions of Sentencing, 34 CRIME & JUST. 1, 28 (2006).
13. This is a standard account of the criteria retributivists use in calculating desert. See, e.g., MODEL PENAL CODE: SENTENCING § 1.02(2)(a)(i) (Tentative Draft No. 1, 2007).
14. On the difficulties of assigning sentences based on general deterrence theory.
15. BONNIE et al., supra note 5, at 21-23.
16. As Herbert Packer put it, "the incapacitative theory is at its strongest for those who, in retributive terms, are the least deserving of punishment."
17. Herbert L. Packer, THE LIMITS OF THE CRIMINAL SANCTION 50-51 (1968).
18. Wayne R. Lafave, CRIMINAL LAW 33 (5th ed. 2010) ("The rehabilitation theory would let the criminal go when (and perhaps only when) he had been reformed." (footnote omitted)).
19. Norval Morris, THE FUTURE OF IMPRISONMENT 59 (1974).
20. Richard S. Frase, Sentencing Principles in Theory and Practice, 22 CRIME & JUST. 363, 363-378 (1997) (describing in detail Morris's theory of limiting retributivism).
21. K.G. Armstrong, The Retributivist Hits Back, in THE PHILOSOPHY OF PUNISHMENT: A COLLECTION OF PAPERS 138, 155 (H.B. Acton ed., 1969) ("For a variety of reasons (amongst them the hope of reforming the criminal) the appropriate authority may choose to punish a man less than it is entitled to, but it is never just to punish a man more than he deserves.").
22. Andrew von Hirsch, Equality, "Anisonomy," and Justice: A Review of Madness and the Criminal Law, 82 MICH. L. REV. 1093, 1105-1106 (1984) (suggesting a range of two to three years for a first offense of armed robbery).
23. Michael Vitiello, Reconsidering Rehabilitation, 65 TUL. L. REV. 1011, 1051, 1053 (1991) ("[W]hen transformation occurs it is difficult to find a continued justification for imposing suffering on that offender.").
24. Andrew von Hirsch, Recent Trends in American Criminal Sentencing Theory, 42 MD. L. REV. 6, 16 (1983).
25. Paul H. Robinson & John M. Darley, JUSTICE, LIABILITY, AND BLAME: COMMUNITY VIEWS AND THE CRIMINAL LAW (1995).
26. Id. at 42-50 (omissions study); id. at 96-105 (complicity study); id. at 14-28 (attempt study).
27. Paul H. Robinson, Competing Conceptions of Modern Desert: Vengeful, Deontological, and Empirical, 67 CAMBRIDGE L.J. 145, 148-150 (2008).
28. See id. at 164 (noting that "moral philosophers simply disagree about just how [moral blameworthiness] translates into specific punishment in a given case").
29. Paul H. Robinson, The Ongoing Revolution in Punishment Theory: Doing Justice as Controlling Crime, 42 ARIZ. ST. L.J. 1089, 1110 (2011) [hereinafter Robinson, Doing Justice as Controlling Crime].
30. Paul H. Robinson, The Role of Moral Philosophers in the Competition Between Deontological and Empirical Desert, 48 WM. & MARY L. REV. 1831, 1840 (2007) [hereinafter Robinson, The Role of Moral Philosophers] (suggesting that empirically sound research methodologies in social psychology can provide moral philosophers with "more reliable" information about intuitions of justice).
31. Robinson, Doing Justice as Controlling Crime, supra note 28, at 1107.
32. Robinson, The Role of Moral Philosophers, supra note 28, at 1836.
33. For instance, Mary Sigler disputes the relevance of empirical desert to deontological desert, since philosophers are not trying to "measure or describe something as it is," and argues further that empirical desert's positive nature provides no means of reconciling the disagreements that, in a democracy, are inevitable.
34. Mary Sigler, The Methodology of Desert, 42 ARIZ. ST. L.J. 1173, 1181, 1185-1187 (2011).
35. Zachary R. Calo, Empirical Desert and the Moral Economy of Punishment, 42 ARIZ. ST. L.J. 1123, 1131 (2011) ("[E]mpirical desert... has no resources for explaining why an offender deserves punishment other than that doing so achieves [utilitarian] benefits.").
36. Others, pointing to our racist history and similarly undesirable motivations to punish, have noted that societal views may diverge radically from deontologically sound prescriptions.
37. Youngjae Lee, Desert, Deontology, and Vengeance, 42 ARIZ. ST. L.J. 1141, 1149 (2011) ("Punitive passions... can also be excessive and driven by other less desirable, yet no less common, sentiments such as cruelty, sadism, inhumanity, and racial hatred and prejudice."). Robinson recognizes this potential problem and states that empirical desert should not always trump deontological desert (although he does not provide a methodology for deciding when trumping should occur).
38. ROBINSON & DARLEY, supra note 21, at 80, 205.
39. Douglas A. Berman, A Truly (and Peculiarly) American "Revolution in Punishment Theory," 42 ARIZ. ST. L.J. 1113, 1118 (2011).
40. Calo, supra note 33, at 1133-1134.
41. Deborah W. Denno, The Perils of Public Opinion, 28 HOFSTRA L. REV. 741, 744 (2000) ("[P]ublic opinion research may fail to measure accurately the public's fundamental values or 'moral intuitions.'... [I]t may be far more successful in reflecting individuals' erroneous knowledge and distorted attitudes, which can range enormously depending on demographics and personal experiences.").
42. Donald Braman et al., Some Realism About Punishment Naturalism, 77 U. CHI. L. REV. 1531, 1551-1556 (2010) (discussing "empirical evidence contradicting the central claim that evaluations of serious wrongfulness do not vary across social conditions or individuals").
43. Adam J. Kolber, How to Improve Empirical Desert, 75 BROOK. L. REV. 433, 434, 441-448 (2009) (claiming that Robinson has "cherry-pick[ed]... intuitions that are elicited at a particular level of abstraction in ways that hide some of our more passionate intuitions that, for better or worse, motivate many people").
44. Alice Ristroph, Third Wave Legal Moralism, 42 ARIZ. ST. L.J. 1151, 1165, 1167 (2011) (stating that Robinson's scenarios do not simulate real crimes and questioning whether a failure to follow desert causes significant "social disutility").
45. Braman et al., supra note 36, provides some empirical evidence that challenges the consensus hypothesis at the margins, but in the main the attack is conceptual.
46. Christopher Slobogin, Some Hypotheses About Empirical Desert, 41 ARIZ. ST. L.J. 1189 (2011).
47. The ten hypotheses were: (1) that rank ordering of core crimes will vary according to factors other than desert; (2) that significant disagreement exists about appropriate punishments for core crimes; (3) that for crimes outside the core, there is significant variation in both the rank orderings and the appropriate punishment; (4) that "[m]ost departures from empirical desert will not be noticed by the public"; (5) that those departures that are noticed will not occasion negative reaction because they are perceived to achieve other legitimate goals, or (6) will occasion negative reaction only because they depart from goals other than desert; (7) that criminal laws that depart from empirical desert will not cause noncompliance, or (8) will cause noncompliance only in connection with the specific laws at issue; (9) that any noncompliance that does result from departures from empirical desert will be less than the noncompliance resulting from failure to directly address preventive goals; and (10) that preventive dispositions can often satisfy desert. Id. Hypotheses 1-3 are related to the consensus hypothesis described in the text, hypotheses 5-7 are related to the compliance hypothesis, and hypothesis 9 is the converse of the control hypothesis. Hypotheses 4, 8, and 10 are not addressed by the research in this Article, although they are addressed by other research. See infra note 109 (hypothesis 4); infra notes 83-84 (hypothesis 8); infra note 138 (hypothesis 10).
48. Paul H. Robinson & Robert Kurzban, Concordance and Conflict in Intuitions of Justice, 91 MINN. L. REV. 1829, 1848-1865 (2007).
49. Robinson, Doing Justice as Controlling Crime, supra note 28, at 1105 ("When... asked to assign punishment, [laypersons] don't look to the factors that determine dangerousness or deterrence, but rather to the offender's moral blameworthiness.").
50. Robinson, Doing Justice as Controlling Crime, supra note 28, at 1107 ("[T]he crime-control power of the criminal law depends in some significant part upon how well it tracks the community's shared intuitions of justice.").
51. ROBINSON & DARLEY, supra note 21, at 7.
52. Tom R. Tyler, WHY PEOPLE OBEY THE LAW (1990).
53. Paul H. Robinson & John M. Darley, The Utility of Desert, 91 NW. U. L. REV. 453, 471, 474-475 (1997) (citing TYLER, supra).
54. Paul H. Robinson & John M. Darley, Intuitions of Justice: Implications for Criminal Law and Justice Policy, 81 S. CAL. L. REV. 1, 26 (2007).
55. Janice Nadler, Flouting the Law, 83 TEX. L. REV. 1399, 1417 (2005).
56. Erich J. Greene, Effects of Disagreements Between Legal Codes and Lay Intuitions on Respect for the Law, at iv-v (June 2003) (unpublished Ph.D. dissertation, Princeton University) (on file with authors)).
57. Paul H. Robinson, Geoffrey P. Goodwin & Michael D. Reisig, The Disutility of Injustice, 85 N.Y.U. L. REV. 1940 (2010).
58. See, e.g., id. at 1948 ("The shift to a desert distribution-specifically, empirical desert-will not seriously undermine the criminal justice system's crime-control effectiveness, and indeed may enhance it....").
59. See, e.g., id. at 1947 ("[A] distribution of liability and punishment that tracks lay intuitions of justice would significantly reduce the injustice now present.").
60. See id. at 1948.
61. As revealed below in Appendix D, except for Study 5 (which used law students) and Study 7C (which did not include any African Americans), our participants represented a wide range of races, ages, political affiliations, income categories, and education. Note, however, that in one of our samples (used in Studies 1, 2, 3, and 6), the gender split (63% male, 37% female) was far from even. For a discussion of the difficulties inherent in assessing "ordinary" citizens' views about punishment, see Denno, supra note 35, at 747-58, 762-64 (arguing, in the course of discussing a Robinson and Darley study, that various demographic variables, including race, gender, education, and political views can "substantially impact... results").
62. Robinson & Kurzban, supra note 40, app. A at 1894-98 (describing scenarios).
63. Id. (referring to "intuitions of justice" solely in terms of blameworthiness).
64. John M. Darley, Kevin M. Carlsmith & Paul H. Robinson, Incapacitation and Just Deserts as Motives for Punishment, 24 LAW & HUM. BEHAV. 659, 659 (2000) (reporting a study finding that desert predominates over incapacitative concerns).
65. Robinson, Doing Justice as Controlling Crime, supra note 28, at 1105.
66. Robinson & Kurzban, supra note 40, app. A at 1898 (describing instructions to subjects).
67. Similar comments can be made about the study that Robinson conducted with John Darley and Kevin Carlsmith, which purported to find that desert, not dangerousness, is the primary determinant of people's judgments about punishment. See Darley et al., supra note 56, at 659, 663 (describing the instructions given to survey participants and the conclusions from their study); see also Slobogin, supra note 38, at 1190 & n.6 (criticizing the study).
68. See infra text accompanying notes 154-161.
69. See ROBINSON & DARLEY, supra note 21, app. A at 226 (reporting that participants' punishment choices resulted in standard deviations of over 3.50 in 20% of the presented scenarios).
70. Alfred Blumstein & Jacqueline Cohen, Sentencing of Convicted Offenders: An Analysis of the Public's View, 14 LAW & SOC'Y REV. 223, 223 (1980) (reporting a study that obtained agreement about "relative severity of sentences to be imposed... but disagreement over the absolute magnitude of these sentences").
71. This Study and Studies 2, 3, and 6 made use of the eLab, run by Vanderbilt University's Owen Graduate School of Management. As indicated on eLab's website, "eLab researchers have leveraged use of a diverse international consumer panel to conduct cuttingedge research through robust web-based experiments and surveys." ELAB AT VANDERBILT UNIV., http://elab.vanderbilt.edu (last visited Dec. 19, 2012). The demographics for the samples are found in Appendix D, below. The control and experimental demographics are not reported independently because they are essentially the same here and in the other studies reported in this Article.
72. Michael Wenzel et al., Retributive and Restorative Justice, 32 LAW & HUM. BEHAV. 375, 378 (2008) (distinguishing retribution as backward-looking and remorse as restorative). On the relationship of prior criminal acts to retribution.
73. Andrew Von Hirsch, PAST OR FUTURE CRIMES: DESERVEDNESS AND DANGEROUSNESS IN THE SENTENCING OF CRIMINALS 88-91 (1985) (arguing that repeat offenders deserve somewhat greater punishment).
74. George P. Fletcher, RETHINKING CRIMINAL LAW 460-466 (1978) (questioning whether a prior record should increase an offender's culpability to any degree).
75. Paul H. Robinson et al., Extralegal Punishment Factors: A Study of Forgiveness, Hardship, Good Deeds, Apology, Remorse, and Other Such Discretionary Factors in Assessing Criminal Punishment, 65 VAND. L. REV. 737, 739 (2012) (listing as "extralegal punishment factors" apology, remorse, history of good or bad deeds, public acknowledgement of guilt and various other factors that "go beyond the factors that the criminal law formally recognizes" because they are not "desert-based factors" having to do with the "seriousness of the harm or the evil of the offense and an offender's culpability and mental capacity").
76. Dean J. Spader, Megatrends in Criminal Justice Theory, 13 AM. J. CRIM. L. 157, 168 (1985) ("Factors dealing with the present, such as the defendant's social or economic status, repentance and desire to make restitution, or factors dealing with the future, such as the defendant's predicted dangerousness, rehabilitation, and recidivism, are not truly compatible with pure desert sentencing.").
77. See infra text following note 68.
78. The left-hand column of Table 1 ranks the crimes in the same order as Robinson and Kurzban's subjects did. However, contrary to Robinson and Kurzban's findings, see Robinson & Kurzban, supra note 40, at 1868-69 & tbl.1, in our study the average sentence for the microwave theft slightly exceeded the average sentences for both the assault at the store and the assault at the football game. The microwave theft scenario was also one of the two unmanipulated scenarios (the other involving assault at a store) that produced a different average sentence (slightly lower) in the experimental group compared to the control group. 67.
79. Jeremy A. Blumenthal, Perceptions of Crime: A Multidimensional Analysis with Implications for Law and Psychology, 38 MCGEORGE L. REV. 629, 649 (2007) (reporting a study finding that "people may agree about how serious a crime may be, but not why it is so serious").
80. Dena M. Gromet & John M. Darley, Restoration and Retribution: How Including Retributive Components Affects the Acceptability of Restorative Justice Procedures, 19 SOC. JUST. RES. 395, 411, 423 (2006) (finding that subjects assigned reduced sentences to offenders who successfully completed restorative procedures but also finding that subjects insisted on some punitive prison for serious offenses).
81. One review of public opinion surveys circa 2000 that looked at data about public attitudes toward both relative and absolute punishment concluded that "the central tendency in public opinion is to be punitive and progressive.... In short, do justice, protect society, and reform offenders."
82. Francis T. Cullen et al., Public Opinion About Punishment and Corrections, 27 CRIME & JUST. 1, 60 (2000) (capitalization altered).
83. Robinson's own research verifies this point. See Robinson et al., supra note 64, at 825 (summarizing studies finding that "there is broad support for the use of some [extralegal punishment factors] in determining punishment"). It is not clear how he reconciles these results with his statement that people base their punishment decisions on desert.
84. Like Robinson and Kurzban, we calculated the Kendall coefficient, which ranges from 0, indicating no agreement, to 1, indicating complete agreement or concordance. We did not run statistical significance tests on the differences between our results and theirs because we did not have access to their raw data.
85. The punishment options were: not liable, liable but no punishment, one day, two weeks, two months, six months, one year through fifteen years in one-year increments, twenty through fifty years in five-year increments, life sentence, and death penalty.
86. David L. Faigman et al., MODERN SCIENTIFIC EVIDENCE: THE LAW AND SCIENCE OF EXPERT TESTIMONY, app. B at 252 (2005) ("The [standard deviation] indicates how far a typical element deviates from the average.").
87. See id. at 204 n.4 ("[W]hen the distribution follows the normal curve, about 68% of the data will be within one standard deviation of the mean, and about 95% will be within two standard deviations of the mean."). Our distributions were normal.
88. See Kolber, supra note 36, at 450 ("[I]t seems unlikely that a person will care very much if the penalty for grand theft is a little higher or lower than the penalty for a minor battery if penalties for all offenses are an order of magnitude higher or lower than where he thinks they should be.").
89. Robinson & Darley, supra note 45, at 33-34.
90. Life Without Parole, DEATH PENALTY INFO. CTR., http://www.deathpenaltyinfo.org/life-without-parole (last visited Dec. 19, 2012) (indicating 33 states with the death penalty, 16 states with a maximum sentence of life without parole, and Alaska with neither).
91. Andrew Von Hirsch, DOING JUSTICE 139 (1976) (arguing for a five-year maximum sentence for all but the worst murders); Cullen et al., supra note 67, at 10, 15, 17-18, 20 (discussing widespread disagreement among the public over the death penalty and three strikes laws).
92. Robinson implicitly recognizes this point in the studies he conducted on the compliance hypothesis, which tested people's reaction to specific sentences, not discrepancies in ordinal rankings. See infra notes 97-107 and accompanying text.
93. Robinson and Kurzban acknowledge, based on other studies they conducted, that "true disagreements do exist for intuitions about wrongdoing outside the core of physical aggression, unconsented-to takings, and deception or deceit in exchanges." Robinson & Kurzban, supra note 40, at 1880; see also id. at 1880-90 (reporting significant disagreement on ranking of marijuana possession, cocaine possession, prostitution, a third theft, abortion, and rape, among other crimes).
94. Robinson et al., supra note 46, at 1948.
95. TYLER, supra note 44, at 45-46, 64.
96. Robinson & Darley, supra note 45, at 25 (alteration in original) (quoting TYLER, supra note 44, at 64).
97. TYLER, supra note 44, at 64.
98. Several of these studies are reported in Robinson et al., supra note 46, at 2011-16.
99. Herbert Jacob, Deterrent Effects of Formal and Informal Sanctions, 2 LAW & POL'Y Q. 61, 70 (1980) (finding that people who disagreed with speeding and marijuana possession laws are much more likely to violate those laws).
100. Karyl A. Kinsey, Deterrence and Alienation Effects of IRS Enforcement: An Analysis of Survey Data, in WHY PEOPLE PAY TAXES: TAX COMPLIANCE AND ENFORCEMENT 259, 276, 282 (Joel Slemrod ed., 1992) (finding that news about acquaintances being treated unfairly by the IRS increased willingness to evade taxes).
101. Matthew Silberman, Toward a Theory of Criminal Deterrence, 41 AM. SOC. REV. 442, 457 (1976) ("When public sentiment in general disapproves [of] a given offense, it is relatively unlikely to occur.").
102. Tom R. Tyler, WHY PEOPLE OBEY THE LAW 273 (2006) (new afterword to 2006 reprinting).
103. Jaime L. Napier & Tom R. Tyler, Does Moral Conviction Really Override Concerns About Procedural Justice? A Reexamination of the Value Protection Model, 21 SOC. JUST. RES. 509, 509 (2008) (contesting the notion that procedural justice concerns are trumped by "moral mandates").
104. The third study reported by Robinson, Goodwin, and Reisig in The Disutility of Injustice, supra note 46, at 2016-23, also may have focused on the effect of attitudes about procedural, rather than substantive, justice. This study involved an analysis of data from a 2000 survey entitled Public Opinion on the Courts in the United States. The measure Robinson, Goodwin, and Reisig used to determine satisfaction with the criminal law was the participants' answer to the question "how well [do] you think the courts in your community handle" violent, drug, and juvenile cases?
105. David B. Rottman et al., PUBLIC OPINION ON THE COURTS IN THE UNITED STATES, 2000, app. A at 1-2, available at http://www.icpsr.umich.edu/cgi-bin/file?comp=none&study=3864&ds=1&file_id=892958 (subscription required).
106. That type of question could trigger reactions based on procedural fairness as well as, or instead of, reactions based on case outcomes. Even if respondents chose to answer this question with only substantive outcomes in mind, however, the survey does not reveal the grounds on which substantive outcomes were judged-desert, utilitarian factors, or both.
107. TYLER, supra note 44, at 64.
108. Nadler, supra note 45, at 1399.
109. Id. at 1413-14. In addition to the subjects who read the newspaper stories, Nadler also administered the questionnaire to a control group that did not read any of this material. Id. at 1411.
110. Id. at 1421 ("The theory of moral credentialing holds that people feel licensed to act on questionable motives when they have previously established their credentials as a person of pure motives."). In this study, male subjects who read the fair stories may have felt that they established their pure motives by agreeing with stories' content. See id. at 1421-22.
111. Robinson et al., supra note 46, at 1999-2000, app. C at 2031-32.
112. For the content of the Moral Credibility Scale, see id. at 1999 tbl.5.
113. Id. at 2004-05. In a between-subjects design (to be distinguished from a "within subjects" design), the manipulation is hidden from the subjects; its effect is studied by using two or more samples, ideally matched in all relevant respects, with each sample receiving a different independent variable (here, a group that reads scenarios depicting "just" sentences and a second group that reads scenarios depicting "unjust" sentences).
114. Id. at 2007 & tbl.7. It should also be noted, however, that while the authors found a statistically significant difference in scores between the control and experimental situations, that difference was not particularly large. The "baseline condition" (in response to "just" scenarios) for the eight Moral Credibility Scale answers averaged 6.22 on a 9-point Likert scale (with 9 indicating complete agreement with statements that grant maximum moral credibility to the law). Id. The two groups exposed to the seven real-world injustice scenarios averaged slightly above 5.2 and 5, respectively, which on the Scale's 9-point Likert spectrum was midway between "strongly agree" and "strongly disagree." Id. at 2003, 2007.
115. See id. at 1999.
116. See id. at 1999.
117. The subjects could not return to the prompts for Studies 1 and 2 after they accessed the prompts for Study 3.
118. Slobogin, supra note 38, at 1193-94 (hypothesizing that "[m]ost departures from empirical desert will not be noticed by the public," in part because, as Robinson himself has noted in his criticism of general deterrence theory, "'what we know from studies is that even criminals commonly have no idea what the [criminal law's] rule is'" (emphasis in first quotation omitted) (alteration of second quotation in original) (quoting Robinson, Doing Justice as Controlling Crime, supra note 28, at 1093)). Without reporting results relevant to this matter in detail, we can indicate that our results show that very few people have accurate information about sentencing laws in their jurisdictions.
119. See Nadler, supra note 45, at 1413-14. Nadler's questionnaire included eight crimes: drunk driving, parking in a no-parking zone, failing to pay required taxes, making illegal copies of software, eating a small item without paying in the grocery store, exceeding the posted speed limit, drinking alcohol under age twenty-one, and taking home office supplies for personal use. Id. Simply out of curiosity, we added burglary and unjustified homicide to the list. With respect to both of these crimes, willingness to comply was at or near 100% regardless of the conditions to which the subjects were exposed.
120. P-value ≤ 0.118. The correlation test used was a Pearson's R, which like most correlation measures produces a range of 0.00 (indicating no relationship between the measured variables) to 1.0 or -1.0 (indicating a very strong positive or negative relationship, respectively).
121. FAIGMAN et al., supra note 71, app. B at 245 (defining correlation coefficients).
122. P-value ≤ 0.001.
123. FAIGMAN et al., supra note 71, at 223-224.
124. Also worth noting is that the reliability of the correlation coefficient increases with its absolute value, so that a difference of 0.1 between two correlations is much less significant if the two coefficients are 0.15 and 0.25 than if they are 0.8 and 0.9.
125. What Are Basic Statistics?: How to Determine Whether Two Correlation Coefficients Are Significant, STATSOFT, http://www.statsoft.com/textbook/basicstatistics/#Correlationsc (last visited Dec. 19, 2012).
126. For Studies 4A, 4B, 7A, and 7B, declining funding compelled us to use the online survey service Amazon Mechanical Turk instead of the Owen School eLab, see supra note 62. The Amazon service enables researchers to programmatically access an on-demand, diverse workforce that is paid per task. Amazon's service is less expensive and may be somewhat more likely to solicit inattentive participants, but on the whole has proven trustworthy.
127. Jennifer Jacquet, The Pros & Cons of Amazon Mechanical Turk for Scientific Surveys, SCI. AM. (July 7, 2011), http://blogs.scientificamerican.com/guilty-planet/2011/07/07/thepros-cons-of-amazon-mechanical-turk-for-scientific-surveys.
128. See supra Table 2.
129. See Robinson et al., supra note 46, at 1999. The Moral Credibility Scale uses a 9-point Likert scale (which we modified to 5 points) to gauge the extent to which subjects would: (1) believe that a person the criminal justice system sentences to life must deserve the punishment he received; (2) consider whether a law prohibiting posting false comments online criminalizes a condemnable act; (3) conclude that a sentence of twenty-eight to thirty two years for a financial maneuver on taxes is morally condemnable; (4) report a person who has illegally taken an arrowhead from an important historical site; (5) take a handgun found in an alley to the police; (6) report a neighbor who has six dogs, three more than is legally allowed; (7) return to pay for a tank of gas they inadvertently did not pay for even though they are unlikely to get caught; and (8) return to pay a restaurant bill they inadvertently did not pay even though it is unlikely they would get caught.
130. See infra Appendix B for details on the manipulations.
131. Particularly so since Study 4B was conducted two days after the jury returned a verdict of acquittal on the most serious charges in the Casey Anthony case, a decision that occasioned major outcry about the criminal justice system.
132. Amy Pavuk & Bianca Prieto, Casey Anthony Not Guilty of Murder, ORLANDO SENTINEL (July 5, 2011), http://articles.orlandosentinel.com/2011-07-05/news/os-casey-anthony-verdict-20110704.
133. Under empirical desert theory, this event should have created a greater tendency toward noncompliance in our sample.
134. FAIGMAN et al., supra note 71, at 212 (stating that the "most common" threshold for significance is 0.05, roughly meaning a 95% probability that the difference is not due to chance). However, higher p-values do not necessarily mean differences are nonexistent.
135. D.H. Kaye, Is Proof of Statistical Significance Relevant?, 61 WASH. L. REV. 1333, 1354 (1986).
136. The correlation was 0.28, p < 0.023. In no other group was this correlation significant.
137. At first glance, comparison of the composite cooperation scores in Studies 4A and 4B may seem to confirm this tendency because the participants in Study 4B, who received more "unjust" scenarios than the participants in Study 4A, had appreciably higher cooperation scores. However, since these two studies involved different samples, direct comparisons cannot be made. See supra note 50.
138. Nadler's excessive lenience scenario involved an individual who witnessed his friend molesting a young girl and did not report it, yet was not prosecuted or even arrested. Nadler, supra note 45, at 1417. Her excessive harshness scenarios were described in three newspaper stories reporting legislation that unfairly imposed taxes, permitted government confiscation of property, and infringed civil liberties. Id. at 1413. For a description of Robinson et al.'s scenarios, see supra text accompanying notes 97, 103-107.
139. Nadler, supra note 45, at 1421.
140. This sampling method was necessary because of the two-stage nature of this Study, something that cannot be easily carried out using online survey programs in which participants are anonymous.
141. Note that composite compliance scores reported in Table 4 are higher (by about fifteen points) than the composite compliance scores reported in Table 3, even though the same measure (Nadler's Likelihood of Criminal Behavior Scale) was used in both studies. There are at least three explanations for this difference. First, of course, the prompts in Studies 4A and 4B were different from those used in Study 5. See supra text accompanying notes 122-124. Second, the subjects in the first two studies, obtained through an online survey service, were approximately eight years older, on average, than the law students used in Study 5. See infra Appendix D. Social scientists have long known that age and risk averseness are related.
142. Christopher Slobogin & Mark R. Fondacaro, JUVENILES AT RISK: A PLEA FOR PREVENTIVE JUSTICE 19-35 (2011) (detailing developmental and neurological research indicating that juveniles are much more impulsive and risk-seeking than adults).
143. A third difference may have to do with assumptions about anonymity. Although the online samples were guaranteed anonymity, some of the participants in Studies 4A and 4B may still have been reluctant to report willingness to commit crimes. The students in Study 5, in contrast, filled out their surveys in rooms where they could "see" that anonymity was guaranteed.
144. Robinson, Doing Justice as Controlling Crime, supra note 28, at 1106 ("[A] criminal justice system that regularly does injustice and/or fails to do justice is one that risks prompting resistance and subversion...." (emphasis added)).
145. Slobogin, supra note 38, at 1198-99.
146. Robinson, The Role of Moral Philosophers, supra note 28, at 1836.
147. Slobogin, supra note 38, at 1201-1202.
148. Scott W. Henggeler et al., MULTISYSTEMIC TREATMENT OF ANTISOCIAL BEHAVIOR IN CHILDREN AND ADOLESCENTS 252-254 (1998)).
149. ROBINSON & DARLEY, supra note 21, at 141 (reporting a study indicating that the average penalty assigned by lay participants to fourteen-year-olds who commit murder was over six years).
150. Elizabeth S. Scott et al., Public Attitudes About the Culpability and Punishment of Young Offenders, 24 BEHAV. SCI. & L. 815, 823 (2006) (finding that participants assigned the same level of culpability to an offender whether they believed the offender was a fifteen-year-old or a twenty-year-old).
151. Research indicates that the public is not averse to community-based dispositions for juveniles when allowed to consider "effectiveness" rather than desert.
152. Elizabeth S. Scott & Laurence Steinberg, RETHINKING JUVENILE JUSTICE 281 (2008) ("Adult punishment and long incarceration are approved, for the most part, only as a means to protect the public from violent young criminals;... if other more lenient sanctions are effective, they are favored over incarceration.").
153. Paul H. Robinson & John M. Darley, Does Criminal Law Deter? A Behavioural Science Investigation, 24 OXFORD J. LEGAL STUD. 173, 174, 204-205 (2004) (cataloguing reasons why differences in criminal dispositions have little general deterrent effect).
154. Robinson et al., supra note 46, at 1974.
155. See infra Appendix A for a description of the scenarios.
156. Robinson et al., supra note 46, at 1972 tbl.4.
157. Martin H. Pritikin, Is Prison Increasing Crime?, 2008 WIS. L. REV. 1049, 1108.
158. Robinson has argued that, for at least some crimes, alternatives to prison can have as much "punitive 'bite'" as prison.
159. Paul H. Robinson & John M. Darley, The Role of Deterrence in the Formulation of Criminal Law Rules: At Its Worst When Doing Its Best, 91 GEO. L.J. 949, 996 (2003).
160. But the duration of these alternatives would have to be extreme in order to equate with the prison terms described in the text.
161. Robert E. Harlow, John M. Darley & Paul H. Robinson, The Severity of Intermediate Penal Sanctions: A Psychophysical Scaling Approach to Obtaining Community Perceptions, 11 J. QUANTITATIVE CRIMINOLOGY 71, 86 (1995) (finding, for example, that survey participants view a six-month prison term as equivalent to eighteen months of intermediate sanctions, and that "[n]o intermediate sanctions were seen as equivalent to prison terms of 2 years or more").
162. The subjects were unable to return to the prompts for Studies 1, 2, or 3 after they accessed the prompts for this study.
163. Some research indicates that putting theft offenders on probation upon condition of undergoing some type of treatment reduces recidivism rates.
164. Ted Bartell & L. Thomas Winfree Jr., Recidivist Impacts of Differential Sentencing Practices for Burglary Offenders, 15 CRIMINOLOGY 387, 394 (1977).
165. Some substance abuse treatment programs appear to reduce recidivism rates.
166. William D. Bales et al., Substance Abuse Treatment in Prison and Community Reentry: Breaking the Cycle of Drugs, Crime, Incarceration, and Recidivism?, 3 GEO. J. ON POVERTY L. & POL'Y 383, 389 (2006) (stating, with several qualifications and notable exceptions, that "studies indicate that offenders who participate in prison-based substance abuse treatment programs experience more successful community reentry, as compared to offenders who do not participate in the programs while incarcerated").
167. Blueprints for Violence Prevention Model Programs: Multisystemic Therapy (MST), CTR. FOR STUDY & PREVENTION OF VIOLENCE, http://www.colorado.edu/cspv/blueprints/modelprograms/MST.html (last visited Dec. 19, 2012) (stating that MST reduces long-term recidivism rates by 25% to 70% and reduces outof-home placements by 47% to 64%, with the result that MST has been found to be "the most cost-effective of a wide range of intervention programs aimed at serious juvenile offenders").
168. Peggy Fulton Hora et al., Therapeutic Jurisprudence and the Drug Treatment Court Movement: Revolutionizing the Criminal Justice System's Response to Drug Abuse and Crime in America, 74 NOTRE DAME L. REV. 439, 502 (1999) ("[Drug courts] across the country have recorded substantial success in retaining participants in treatment programs, reducing recidivism rates, and saving criminal justice system resources.").
169. Robinson et al., supra note 64, at 774-775.
170. The subjects could not return to the prompts for Study 4A after they accessed the prompts for this Study.
171. Robinson et al., supra note 64, at 781 & tbl.4.
172. The subjects could not return to the prompts for Study 4B after they accessed the prompts for this Study.
173. As indicated in Appendix D, this sample did not include any African Americans. If this lack of diversity biases the results in any direction, it is probably toward less leniency.
174. Peter H. Rossi & Richard A. Berk, JUST PUNISHMENTS: FEDERAL GUIDELINES AND PUBLIC VIEWS COMPARED 205 (1997) (concluding that African American ethnicity is correlated with somewhat more lenient sentence choices, although educational attainment is the strongest demographic correlate of sentencing attitudes).
175. Christopher Slobogin, Is Justice Just Us? Using Social Science to Inform Substantive Criminal Law, 87 J. CRIM. L. & CRIMINOLOGY 315, 325 (1996) (stating that Robinson and Darley's research in Justice, Liability, and Blame: Community Views and the Criminal Law, supra note 21, "certainly could, and probably will, make theorists, especially adherents of the [Model Penal Code], rethink their positions").
176. Berman, supra note 34, at 1118 ("[I]f modern American democracy prompts voters and their representatives to embrace laws that produce and perpetuate obvious injustices, then it would seem our foundational political system has a fundamental flaw that runs much deeper than any problems that could flow from poorly designed criminal laws.").
177. Nadler, supra note 45, at 1416 & n.75 (stating "we cannot definitively predict behavior from [self-reports]" and further noting the questionable ethics of research that motivates illegal behavior (emphasis omitted)).
178. Research confirms that support for prison terms for all but the most serious crimes decreases markedly when individuals are provided contextual information about the costs and effectiveness of intermediate sanctions. Cullen et al., supra note 67, at 43-44.
179. Richard S. Frase, Theories of Proportionality and Desert, in THE OXFORD HANDBOOK OF SENTENCING AND CORRECTIONS 131, 135 (Joan Petersilia & Kevin Reitz eds., 2012) ("[S]pending scarce resources to bring additional charges or impose additional punishments solely because they are deserved cannot be justified given other pressing needs for these resources within and outside of the criminal justice system.").
180. Robinson et al., supra note 46, at 1983-1994.
181. See id. at 1979-80 (suggesting, in the course of discussing why criminal justice provisions "are in conflict with the community's shared intuitions of justice," that "while there may be some latent crime-control concerns present in the public consciousness,... in fact, the public may have been merely 'riding the wave' of concern actuated by a politician's previous comments").
182. Paul H. Robinson et al., Realism, Punishment, and Reform, 77 U. CHI. L. REV. 1611, 1630 (2010) ("[O]ur program challenges the dominant theory of crime control in the United States for the past several decades, one based upon intentionally and regularly doing injustice in the name of general deterrence and incapacitation by its reliance upon doctrines like three strikes, high penalties for drug offenses, adult prosecution of juveniles, abolition or narrowing of the insanity defense, the felony murder rule, and the use of strict liability." (footnote omitted)).
183. William J. Stuntz, The Pathological Politics of Criminal Law, 100 MICH. L. REV. 505 (2001) (focusing on the pathologies of legislatures and prosecutors).
184. James Q. Whitman, A Plea Against Retributivism, 7 BUFF. CRIM. L. REV. 85, 88 (2003) (asking why "the renaissance of neo-retributivism [has] also been the age of epochally harsh punishment").
185. See supra text accompanying notes 16-20.
186. Joan Petersilia, California's Correctional Paradox of Excess and Deprivation, 37 CRIME & JUST. 207, 252-253 (2008) ("[Under California's determinate sentencing regime, a] large percentage of Californians who are nonviolent criminals are accumulating very extensive criminal records... [and yet] may not be any more dangerous than offenders in other states who are left 'on the street' and successfully handled through an array of community based intermediate sanctions. [At the same time], California's sentencing system also releases violent offenders who amass lengthy criminal records-individuals who, in a system more carefully tailored to protect public safety, probably should not have been released in the first place.").
187. Richard Rosenfeld et al., The Contribution of Ex-Prisoners to Crime Rates, in PRISONER REENTRY AND CRIME IN AMERICA 80, 102-103 (Jeremy Travis & Christy Visher eds., 2005) (concluding that "expanded use of discretionary parole supervision" would better protect the public from the safety threat posed by released prisoners).
188. Joanna Shepherd, Blakely's Silver Lining: Sentencing Guidelines, Judicial Discretion, and Crime, 58 HASTINGS L.J. 533, 574 (2007) ("The more mandatory are the guidelines, the larger is the increase in crime.").
189. Yan Zhang et al., Indeterminate and Determinate Sentencing Models: A State-Specific Analysis of Their Effects on Recidivism, CRIME & DELINQ. 1, 18 (Dec. 8, 2009), http://cad.sagepub.com/content/early/2009/12/08/0011128709354047.full.pdf+html (finding, in a multistate, pre/post study, that determinate sentencing was worse than discretionary parole release at reducing recidivism in New York and North Carolina, better than discretionary release at reducing re-offending in Maryland and Virginia, no better or worse as a recidivism-reducing mechanism in Texas and Oregon, and that much depends upon the specific back-end programs in place).
190. Christopher Slobogin, Prevention as the Primary Goal of Sentencing: The Modern Case for Indeterminate Dispositions in Criminal Cases, 48 SAN DIEGO L. REV. 1127 (2011).
191. See id. at 1154-1168.
192. Frase, supra note 153, at 144-145 (theorizing that the persistence of indeterminate sentencing in part "may be due to the widespread support for a hybrid approach").
193. See supra note 70 for a description of the options. The same dispositional options were used for all of the Study 6 scenarios.
194. In the actual survey, subjects who chose to change the sentence were given several lines on which to provide their answer.
195. Subjects who chose to change the sentence were given several lines on which to provide their answer. The same is true for the final question on the following survey.