1 There are many dimensions to the “there's really no disparity” position. One argument is simply that the proper statistical controls are not introduced in the usual disparity exercises. With these in place—e.g., controlling for prior record, amount of violence, weapon use and son on—disparity tends to disappear.Google Scholar

A second argument that advocates of this position make is that the disparity that remains once proper controls for standard sentencing variables are introduced can be justified by differences in other attributes of the defendants themselves. So, for example, they argue that age, or employment status, or drug usage ought to influence sentencing, and that drawing distinctions on the basis of these considerations leads to justifiably disparate sentences, and not “pejorative” disparity.Google Scholar

We need not enter the debate here about which variables justify disparate treatment, for this would take us far afield and lead us into complex matters (e.g., how should age or drug usage affect sentence?—Are these mitigators or aggravators?) For our purposes, disparity can simply be defined as the differential treatment of identically situated defendants. If defendants commit the same crime and have the same prior background but receive different sentences, this disparity is considered unjustified and fits our “pejorative” categorization.Google Scholar

Finally, two very interesting arguments can be made in defense of even this pure and more easily assailable kind of disparity. The first focuses on geographical differences and argues that it is appropriate for judges sitting in different parts of a state (in state systems) or in different regions (in the federal system) to sentence similarly situated defendants differently because of local considerations. The second argument is based on the reality of scarce resources and assumptions about deterrence. In a world in which prison space is limited, and with the assumption that incarceration of some defendants can deter other defendants in the same line of work, this argument suggests that selectively incarcerating similarly situated defendants will have a deterrent effect. The system may not have the resources to be tough on every crack dealer, but if it cracks down on every fifth one, the argument goes, some deterrence, albeit at the price of disparate treatment of essentially identical defendants, will be achieved. In this paper we will set these arguments aside as well.Google Scholar

2 See, e.g., Pierce O'Donnell, Michael Churgin, & Dennis Curtis, Toward a lust and Effective Sentencing System (New York: Praeger, 1977). For a careful and excellent analysis of the complexity of assessing disparity, see Alfred Blumstein et al., eds., 1 Research on Sentencing: The Search for Reform (Washington, D.C.: National Academy Press, 1983). And for an interesting study which shows that when controls for offense seriousness, prior record, and victim/defendant relationship are introduced there may be less disparity than first appears to be the case, see Vera Institute of Justice, Felony Arrests (rev. ed. New York: Longman, 1981).Google Scholar

In the Yale Law School Guggenheim Sentencing Seminar, which ran from 1980 until 1988, specially selected judges from a number of states participated in various sentencing exercises, including sentencing the same case from a paper record. The program was run by Professor Daniel Freed of the Law School. It was evident when we examined the judges' sentences that there was a substantial disparity. Interestingly, it was not uncommon for the judges to look at these same data and conclude that they were “not very far apart.” 1 think this kind of “love feast” perception left many of the judges feeling better than would have a recognition of the substantial differences which separated them.Google Scholar

3 A prior question, which I set aside in this paper, is whether judges are even consistent with themselves. For a discussion of an instance of self-inconsistency see Milton Heumann, “Interviewing Trial Judges,”in Symposium Article: Charles Johnson, ed., “Strategies for Judicial Research: Soaking and Poking in the Judiciary,” 73 Judicature 200–202 (198990).Google Scholar

4 For an excellent discussion of the standard criminal justice goals, and of the evidence marshalled in support of each, see Timothy Bynum et al., “Sentencing and Social Research: A Review of the Literature on Deterrence, Incapacitation and Rehabilitation,” in Crime and Punishment in New York City: An Inquiry into Sentencing and the Criminal Justice System-Appendix, Report to Governor Hugh Carey, Prepared by the Executive Advisory Committee on Sentencing (March 1979).Google Scholar

Too often philosophies about criminal sentencing become wedded to one purpose or another of incarceration and insist that this single purpose be determinative in sentencing all defendants. For example, in the past two decades we abandoned a long-held commitment to rehabilitation and have substituted a more just deserts approach. The problem in this shift is not with one or the other goal per se; it is that we seem to go too far in one direction or the other. As the respondents in Sitting in Judgment note, an important task is to calculate the weights of different goals, and not simply to work with a single goal in kneejerk fashion. What is needed is a kind of jurisprudence of purposes, a body of thinking about the conditions under which one goal or another ought to loom larger in a case.Google Scholar

5 For a discussion of the drawing of an explicit in/out line (in Minnesota) see Dale Parent, Structuring Criminal Sentences 77–98 (Stoneham, Ma.: Butterworth Legal Publishers, 1988). Minnesota developed an explicit grid with a line running within it; all sentences below the line were definite “in” sentences. It is reported that prosecutors in Minnesota commonly refer to the line as the AMF line, since for all defendants below the line it is “Adios MF….” Larry Millott, “Crime and Punishment: New Sentencing Guidelines Have Revolutionary Impact,”St. Paul Sunday Pioneer Press, Feb. 21, 1982.Google Scholar

6 Newman, Jon, “Foreword” to “Parole Decisionmaking and the Sentencing Process,” 84 Yale LJ. 812–13 (1975).CrossRefGoogle Scholar

7 442 U.S. 178 (1979).Google Scholar

8 Id. at 181.Google Scholar

9 “At the time sentence was imposed, this Court expected that petitioner would receive a meaningful parole hearing—that is, a determination based on his institutional record and the likelihood of recidivism—upon the completion of one-third (1/3) of his sentence. The Court anticipated—assuming an appropriate institutional adjustment and good behavior while confined—that petitioner would be actually confined for a period of approximately three and one-half to four years of the ten-year sentence, in view of the fact that he was a first-offender and that there appeared to be little probability of recidivism, given the circum stances of the case and his personal and social history. This sentencing expectation was based on the Court's understanding—which was consistent with generally held notion—of the operation of the parole system in 1970.” These remarks by Addonizio's sentencing judge are quoted in id. at 181 n.3.Google Scholar

10 To be sure, with the abolition of parole in some systems, the federal among them, there will be greater “truth in sentencing.” But even in these systems, some ambiguity will remain about “real time” (e.g., as a result of pardon practices, special release programs be cause of jail crowding, home confinement alternatives, etc.). On sentencing under the new federal guidelines, see United States Sentencing Commission, Guidelines Manual (Washington, D.C., October 1987); and United States Sentencing Commission, Supplementary Report on the Initial Sentencing Guidelines and Policy Statements (Washington, D.C., June 18, 1987) (“Guidelines Manual”). For an example of postsentencing adjustments in time caused by jail overcrowding, see “Michigan Law Aims to Ease Prison Crunch,” 12 Crim. L. Newsletter, Jan. 19. 1981, at 3. And for an informative guide to the various clemency practices in the states, practices which in the years ahead probably will increase in importance because of our current love affair with long sentences, see National Governors' Association Center for Policy Research, Guide to Executive Clemency Among the American States (Washington, D.C., March 1988).Google Scholar

11 It is easier, for example, to clarify what the “real time” of a sentence is than it would be to achieve a consensus on “how long is long.”For some intriguing ideas about the latter, see David Rothman, “Doing Time,” 19 Int'l J. Comp Sociology 121–38 (1977).Google Scholar

12 Guidelines Manual.Google Scholar

13 The quantitative data will be presented in a forthcoming volume. See Sitting in Judgment at ix. Some of their data, though, are already available. See Wheeler, Stanton, Mann, Kenneth, & Sarat, Austin, “Sentencing the White-collar Offender,” 17 Am. Crim L. Rev. 479–93 (1980); and Wheeler, Stanton, Weisburg, David, & Bode, Nancy, “Sentencing the White-collar Offender: Rhetoric and Reality,” 47 Am. Sociological Rev. 641 (1982).Google Scholar

14 See generally Alfred Blumstein, Jacqueline Cohen, & Daniel Nagin, eds., Deterrence and Incapacitation: Estimating the Effects of Criminal Sanctions on Crime Rates (Washington D.C.: National Academy of Science, 1978).Google Scholar

15 Minnesota was so concerned about the correlation between employment status and race that it barred the use of employment status as a sentencing factor. See Parent, Structuring Criminal Sentences at 67, 127 (cited in note 5). More generally it is plain that we need to think more systematically about how to weigh these extralegal variables. For example, in common crimes as well as white-collar crimes, incarceration may mean that the defendant's family will be placed in financial jeopardy. Should this be a sentencing consideration? Does it matter if the defendant is supporting a number of young children in addition to a spouse, or if he or she is responsible for the day-today care of the children? In the case of a female offender, should pregnancy be a sentencing consideration? If all things are equal, should this keep a defendant out? As with the judge's weighing the consequences of an incarcerative decision on the defendant's employees in the white-collar area, these other sorts of “dependency” variables implicate difficult and important policy issues about which we presently know very little.Google Scholar

16 Though when there is a “fit” in values even in common crimes, there is some evidence of greater judicial empathy. See Levin's argument that Pittsburgh judges were more steeped in a traditional political culture than were Minnesota judges, and hence were more likely to be sensitive to the particular problems defendants faced. Martin Lvin, Urban Politics and the Criminal Courts (Chicago: University of Chicago Press, 1977).Google Scholar

17 For example, the Sentencing Project headquartered in Washington, D.C., has been assisting defense organizations around the nation in their advocacy of alternative sentences. See Sentencing Project, “Report to the Advisory Committee of the Board” (Washington, D.C., February 28, 1989).Google Scholar

18 Some judges dissented from this view and felt that the memos were prepared mostly to placate clients. Sitting in Judgment at 49.Google Scholar

19 See, e.g., Marvin Frankel, Criminal Sentences (New York: Hill & Wang, 1972); and Pope, Peter, “How Unreliable Factfinding Can Undermine Sentencing Guidelines,” 95 Yale LJ. 1258–82 (1986).CrossRefGoogle Scholar

20 Of course, with the new federal guidelines some of the judge's discretion to consider all of the dimensions of the “Big Three” (even when substantial information about them is available in the presentence report) seemingly has been removed. Wheeler et al. note that this is particularly true with respect to the blameworthiness prong. Sitting in Judgment at 187 n.26. And it is still too early to discern much about the direction appellate courts will take in their review of departures from the guidelines. For a review of the cases to date, see the monthly issues (beginning in June of 1988) of Daniel Freed & Marc MIller, eds., Federal Sentencing Reporter (New York: Vera Institute of Justice).Google Scholar

21 See generally Blumstein et al., eds., Deterrence and Incapacitation.Google Scholar

22 See, e.g., Robinson, Paul, “A Sentencing System for the 21st Century?” 66 Tex. L. Rev. 53–58 (1987).Google Scholar

23 See note 10 supra and Mistretta v. United States, 102 L.M. 2d 714 (1989), upholding the guidelines against a constitutional challenge.Google Scholar

24 For a thoughtful elaboration on this theme, see D. A. Thomas, “Equity in Sentencing,” Sixth Annual Pinkerton Lecture, Albany University, School of Criminal Justice (April 1977); and Sitting in Judgment at 185–87.Google Scholar