 |
|||||
|
|
|
|
|
||
|
|
|||||
|
|
|
|
|
||
A. The Harm Done by Habitual Criminals.
At long last, America is making progress against crime. As the solid line in Figure 1 shows, the rate of FBI Index Crime has fallen 28% from the 1980 peak of 5,950.0 index crimes per 100,000 population to 4,266.8 in 1999. See Federal Bureau of Investigation, Crime in the United States 1999, p. 64 (2000) (1980-1999 data). While this is progress, we have much further to go, as a comparison with the 1960 rates shows. The rates of forcible rape and robbery each remain about 3.4 times the respective 1960 rates. Compare ibid., with U. S. Dept. of Justice, Bureau of Justice Statistics, Sourcebook of Criminal Justice Statistics 1999, Table 3.120, p. 266 (2000) (1960-1998 data). That means about 246,000 people were robbed and 63,000 were raped who would not have been if crime rates were at 1960 levels.
Crime is a complex problem with multiple causes and hence no one solution. The criminal justice system is responsible for one vital component of a multi-front attack on crime: punishment. The importance of this component must not be underestimated. The dotted line in Figure 1 shows the strength of sentencing policy as measured by the number of criminals in custody relative to the crime rate. Sourcebook, supra, Table 6.43, at 516 (data for graph). Lenient sentencing in the 1960s and 1970s was followed by a skyrocketing crime rate, and tougher sentencing from the 1980s forward was followed by a dropping crime rate. This is not a coincidence. See Scheidegger & Rushford, The Social Benefits of Confining Habitual Criminals, 11 Stan. L. & Pol'y Rev. 59, 60 (1999).
There are three primary purposes of punishment: retribution, deterrence, and incapacitation. Retribution is the subject of endless philosophical debates, see, e.g., 1 W. LaFave & A. Scott, Substantive Criminal Law § 1.5(a)(6), p. 36, n. 44 (1986), while deterrence is the subject of empirical debates, id., at 34, n. 27. Incapacitation, however, is not debatable. So long as five-time felon Earthy Daniels is confined in prison, he will not be breaking into the homes of law-abiding citizens, and he will not be robbing them, although he may commit crimes on other prisoners. Nor will anyone "take his place" committing robbery and burglary, because there is no fixed number of places to take. See Scheidegger & Rushford, 11 Stan. L. & Pol'y Rev., at 60.
If society is going to take criminals off the street, it can either do so indiscriminately by raising penalties across the board, or it can impose longer sentences on those it believes most likely to commit further crimes if released. The latter, selective approach is both less expensive to society and less severe in its treatment of those given the lesser sentences. "[P]unishing first offenders with twenty-five year sentences does not deter crime as much as it ruins lives." Deal v. United States, 508 U. S. 129, 146, n. 10 (1993) (Stevens, J., dissenting) (quoting United States v. Jones, 965 F. 2d 1507, 1521 (CA8 1992)). Both justice and efficacy require focusing on career criminals.
Research has long established that a fairly small percentage of criminals commit a disproportionate share of the crimes. H. R. Rep. 99-849, p. 2 (1986) (report on Career Criminal Amendments Act of 1986, summarizing research). One study noted that "incarcerating one robber who is above the 90th percentile for one year would prevent more robberies than incarcerating 18 offenders who are below the median for the same period of time." P. Greenwood, Selective Incapacitation 45-47 (1982).
Identification of the "high-rate" offenders is no small task, however. While they freely "self-report" in confidential interviews with researchers, id., at 14-15, they are obviously not going to speak freely during sentencing. Furthermore, there are moral and constitutional restraints on the factors that can be used in sentencing. Even if research were to find a correlation between growing up in a broken home and recidivism, for example, it would be obviously unjust to base a sentence on a factor beyond one's control.
Criminal record is the single most important factor in identifying the career criminal. Common sense, backed up by research, indicates that felons with multiple convictions are more likely to commit new crimes upon release.
"It is a psychological truism that the best predictor of future behavior is past behavior. Not surprisingly, one of the best predictors of future criminal conduct is past criminal conduct, and the parole-prediction literature amply supports this fact. From the earliest studies [citations] to the latest [citations], indices of prior criminal conduct consistently are found to be among the most powerful predictors of parole violations, arrest for the commitment of new offenses, and conviction and reincarceration for these." Gottfredson & Gottfredson, Accuracy of Prediction Models, in 2 Criminal Careers and "Career Criminals" 239-240 (A. Blumstein ed. 1986).
Enhancement for prior convictions is based on an objective, previously adjudicated fact, and it is based on the defendant's voluntary, antisocial choice to commit the prior crime. The predictive power of a criminal record to forecast future recidivism is obviously far from perfect. See, e.g., Greenwood, supra, at 66. However, there is no other factor for differentiating among people who have committed similar crimes which satisfies as well the concurrent requirements of objectivity, moral basis, and prediction of future recidivism. Consequently, every jurisdiction in the United States has enacted some kind of habitual criminal statute. Parke v. Raley, 506 U. S. 20, 26-27 (1992) (citing Department of Justice, Statutes Requiring the Use of Criminal History Record Information 17-41 (1991) (NJC-129846)). This Court has long recognized the states' strong and legitimate interest in punishing repeat offenders. Oyler v. Boles, 368 U. S. 448, 451 (1962); Moore v. Missouri, 159 U. S. 673, 677 (1895); Raley, 506 U. S., at 27.
For a system based on prior crimes to be just and evenhanded, it must reliably identify those prior crimes. A limitation to prior convictions, rather than unadjudicated prior conduct, is for the defendant's benefit even though the Constitution does not require this limitation. See Nichols v. United States, 511 U. S. 738, 747-748 (1994). But if the priors are thrown out on fortuitous circumstances having little or no connection with the defendant's guilt of the prior crime, then the evenhandedness of the system is compromised. For example, in Parke v. Raley, the Sixth Circuit had created a rule which, in effect, would throw out virtually all prior convictions where the transcript of the Boykin(3) colloquy had been lost, destroyed, or never made. This Court wisely rejected that rule. 506 U. S., at 30.
Congress has provided, in 28 U. S. C. §2255, a mechanism for challenging criminal judgments on grounds which cannot be raised on appeal because they require facts outside the appellate record. See United States v. Hayman, 342 U. S. 205, 212 (1952) (habeas for "facts dehors the record"); id., at 219 (scope of §2255 intended to be the same as habeas). Claims which could have been raised on appeal, but were not, generally cannot be raised in a §2255 proceeding. See Reed v. Farley, 512 U. S. 339, 354 (1994). Claims raised on appeal and resolved against the appellant will generally not be reconsidered as well. See Kaufman v. United States, 394 U. S. 217, 227, n. 8 (1969); Withrow v. Williams, 507 U. S. 680, 720-721 (1993) (Scalia, J., concurring in part and dissenting in part) (collecting cases). Yet petitioner now proposes to convert this proceeding into an open-ended attack on prior judgments on grounds that were or could have been raised and decided long ago. In the typical case, the defendant has had or waived his appeal on the prior. He has filed or foregone his collateral attack, if he was incarcerated. He has been back on the street and committed a new felony, and now he wants to reopen the long-closed book on the prior crime, placing a further burden on a system already groaning under a surfeit of litigable issues. Such a step would require an exceptionally compelling justification.
B. Striking Priors and Unjust Results.
"Striking" a prior conviction, so as to disregard it for sentencing purposes, is fundamentally different from reversing a judgment on appeal or even from granting habeas corpus relief. On appeal or habeas, the grant of relief typically does not, by itself, result in the defendant's release. The usual result is a retrial. A conviction may be more difficult to obtain after a long lapse of time, see Vasquez v. Hillery, 474 U. S. 254, 280 (1986) (Powell, J., dissenting), but it is quite often possible, particularly where the original case was strong.
Striking priors is different. The remedy typically sought is to resentence the defendant without considering the prior offense. See, e.g., J. A. 51-52 (relief sought in District Court in the present case). From a finding of procedural irregularity alone, the defendant completely escapes the sentence-enhancing effect of his prior crime. He would be sentenced as a first offender regardless of how clear the evidence of guilt of the prior may be. This is a truly drastic remedy, and it can have disastrous consequences.
For example, in California the mid-term penalty for a first-offense rapist is 6 years, Cal. Penal Code §264(a), with up to 15 percent worktime credit. See id., §2933.1(a). A second offense of forcible rape draws 25-to-life. See id., §667.61(a), (c)(1), (d)(1). Striking the prior means that a serial rapist who should have been locked up for life instead gets released in 5 years to brutalize more women.
The criminal career of Andre Brigham Young, presently before this Court in Seling v. Young, No. 99-1185, illustrates the dire consequences of failure to adequately punish repeat offenders. See In re Young, 857 P. 2d 989, 994-995 (Wash. 1993). Despite a prior conviction on four counts of rape, Young served only three years for the fifth. The attack on victim number six should never have happened. The cause of this travesty was lax sentencing laws, but the same consequence would follow from a needlessly stricken prior conviction.
In Parke v. Raley, supra, 506 U. S., at 32, this Court noted the state's "legitimate interest in differentially punishing repeat offenders." Although true, this is a dry understatement. The "interest" here goes to the very lives of the people victimized by habitual criminals. Innocent people will be needlessly robbed, raped, and murdered if criminals who are, in fact, recidivists are wrongly sentenced as first offenders. That is what is at stake in this case.
|
|
||||
|
|
|
|
||
3. Boykin v. Alabama, 395 U. S. 238 (1969).