By Eric Blumenson* and Eva S. Nilsen**

    B. College and Higher Education C. Prison Education

    1. Substantive Due Process and Equal Protection 2. Retroactive Application
    3. Rights Under the Federal Disability Acts
    4. Unconstitutional Punishment

    The United States has been waging a highly publicized war on drugs for a long time. Familiar scenes seem to have been with us for a lifetime: government agents posing with bricks of cocaine piled high, helicopters defoliating coca fields abroad, drug czars promising that our redoubled efforts are finally turning the tide. A systematic accounting of the American casualties of this war is harder to come by, however. Not many Americans are aware that we now imprison two million people, six times as many as we did in 1972, in large part as a result of the drug war that has been waged over those three decades.1 Even fewer know that in the ’90s, educational deprivation became a weapon in the drug war, resulting in the denial of high


* Professor, Suffolk University Law School; J.D. 1972, Harvard Law School.

** Associate Clinical Professor of Law, Boston University Law School; J.D. 1977, University of Virginia Law School; LL.M. 1980, Georgetown University Law Center.

Copyright 2002 by Eric Blumenson and Eva S. Nilsen. We thank many friends and colleagues for their generous and thoughtful counsel, including Karen Blum, Gabriel Chin, Nora Demleitner, Victoria Dodd, Eileen Kaufman, Simone Levine, Eric Lotke, Ken Simons, and Jeremy Travis; and for expert research assistance, Paige Ormond, Samit Patel, Wendy Fritz, Michael Van Dam, and the library staffs at our respective universities.

1. See infra notes 51-56.


62 The Journal of Gender, Race & Justice [6:2002]

school or college opportunities to tens of thousands of students. While Americans are arguably more committed than ever to the ideal of universal education, the drug war has effectively withdrawn this commitment to many children who are most at risk. It has done so in several related ways that we explore in this article.

First, the drug war has combined with public school zero-tolerance policies to remove tens of thousands of adolescents from their public schools.2 Eighty-eight percent of public schools have zero-tolerance policies for drugs,3 and according to one recent study approximately eighty percent of students charged with drug or alcohol infractions are suspended or expelled from school.4 While a small number of these students have undoubtedly committed serious drug offenses, far greater numbers have lost their educational opportunities based on the simple possession of drugsñ sometimes as small as one pill.5

Second, denial of higher education has been adopted as an additional punishment for drug offenders. Under the Drug Free Student Loans Act of 1998, students who have ever been convicted of a drug offense are either temporarily or permanently ineligible for federal college loans and grants.6 This law has led to the withdrawal from school of thousands of college students who have no alternative means of paying for their education.7 As to drug offenders in prison, their access to higher education had already effectively been terminated by a 1994 law that excluded all prisoners from Pell Grants, the federal college aid program that had engendered numerous college programs in prison.8 For thousands of prisoners and college students, these two laws have meant the end of the college dream.

Finally, the war on drugs has targeted massive numbers of drug usersó addicts, serious abusers, and casual users alikeóand siphoned them out of society and into prison. As is often noted, there are fewer young black men

  1. See infra notes 13-34 and accompanying text.

CRIME AND SAFETY 129 tbl. A1 (NCES Pub. No. 2002-113/NCJ-190075, 2001), available at

4. THE NAT’L CTR. ON ADDICTION & SUBSTANCE ABUSE, MALIGNANT NEGLECT: SUBSTANCE ABUSE AND AMERICA’S SCHOOLS 39 fig. 5A (2001) (citing National Center for Education Statistics), available at Twenty seven percent of schools surveyed reported taking 170,000 disciplinary actions on drug and alcohol infractions during the school year 1996-97. Id. at 39. Of these 170,000, 62% were suspensions for 5 or more days, 18% were expulsions, and 20% were transfers to alternative schools. Id.

  1. See infra notes 26-31.
  2. 20 U.S.C. ß 1091(r) (2002).
  3. See infra notes 41-45 and accompanying text.
  4. Violent Crime Control and Law Enforcement Act of 1994, Pub. L. No. 103-322, ß 20411,

108 Stat. 1796 (1994). The act amended 20 U.S.C. ß 1070a(b)(8) (1965), which now states: “(8) No basic grant shall be awarded under this subpart to any individual who is incarcerated in any Federal or State penal institution.” ß 20411.


How to Construct an Underclass 63

in college than are in the correctional system.9 We should also recognize that there are so few young black men in college partly because so many are in prison; and that a great number of them are in prison because of the drug war.10 Whether wittingly or not, Americans have fully embraced that tradeoff for many years now: we are more interested in segregating and punishing drug offenders than educating them.11

The war on drugs has spawned a second frontóa war on education. The casualties of this war are all poor or lower-income people who cannot afford to buy a private education.12 This article details the consequences of this other war, and explores some legislative and litigation strategies for reclaiming educational opportunity for all Americans. First, however, we must set out in more detail the laws and policies that now deploy educational privation as punishment in public schools, colleges, and prisons.

9. In the fall of 1997, 1,532,800 African Americans were enrolled in institutions of higher education, compared to 2,149,900 African Americans who were under correctional supervisionó i.e., in jails, prison, or on parole or probation. These figures are reported respectively in NAT’L CTR. FOR EDUC. STAT., DIGEST OF EDUCATION STATISTICS 2000, at 236 tbl. 207 (2000), available at; U.S. DEP’T OF JUST., BUREAU OF JUST. STAT., CORRECTIONAL POPULATIONS IN THE UNITED STATES, 1997, at 2 tbl. 1.3 (2000), available at

10. See infra notes 51-56. One study concluded that “sentences for drug offenders are the major reason for increases in prison admissions since 1980.” Joan Petersilia, When Prisoners Return to the Community: Political, Economic and Social Consequences, NAT’L INST. JUST., Nov. 2000, available at (citing Alfred Blumstein & Allen J. Beck, Population Growth in U.S. Prisons 1980-1996, in PRISONS, 20-22 (Michael Tonry et al. eds., 1999)) [hereinafter Petersilia].

11. To that end, states in the 1990s have been shifting expenditures from building universities to building prisonsóin 1995 reducing the former by $954 million while increasing prison construction budgets by $926 million. See JUST. POL’Y INST., FROM CLASSROOMS TO CELLBLOCKS: A NATIONAL PERSPECTIVE (1997); JUST. POL’Y INST., TOO LITTLE TOO LATE: PRESIDENT CLINTON’S PRISON LEGACY, at (last visited Apr. 22, 2002) (citing NAT’L ASS’N OF STATE BUDGET OFFICERS, 1995 STATE EXPENDITURES REPORT NASBO 77 tbl. A-6, 98 tbl. A-22 (Apr. 1996)); see also JUDITH GREENE & VINCENT SCHIRALDI, JUST. POL’Y INST., CUTTING CORRECTLY: NEW PRISON POLICIES FOR TIMES OF FISCAL CRISIS 2 (2001) (reporting an eightfold growth of state and local prison expenditures, from $5 billion in 1978 to $40 billion in 2000, the latter comprising seven percent of all state expenditures), available at; The State of Criminal Justice, A.B.A. J., Oct. 2000, at v (reporting that the “rate of increase in aggregate per capita state spending for corrections was almost double that of education from 1990 -1996”); John Greiner & Jack Money, Prison System Criticized, THE SUNDAY OKLAHOMAN, Feb. 10, 2002 (quoting Oklahoma State Senator Dick Wilkerson, Chairman of the Senate Corrections subcommittee, as condemning “the obscenity [of] spending less than $5,000 to educate a child and more than $20,000 on a felon”).

12. See infra notes 76-84 and accompanying text.


64 The Journal of Gender, Race & Justice [6:2002] I. A NEW PUNISHMENT: DENYING OFFENDERS AN EDUCATION

A. Public School Education

Public school zero tolerance rules represent a sea change in American educational policy. After years of campaigns aimed at keeping children at risk in school, the zero tolerance effort seeks instead to identify troublesome students and get them out of school. Longstanding anxiety about student drug abuse was only part of the impetus; more significant were the sporadic episodes of extreme student violence that occurred throughout the 1990s.13 The effect, however, was to cast all students as objects of worry and suspicion, and to establish systems of surveillance and discipline that identified and removed students for any misbehavior, which often involved drugs. School regulations reflect this progression. In 1994, Congress passed a law limiting federal educational aid to only those states that imposed a mandatory one-year expulsion on students bringing weapons to school.14 All

13. For example, between 1990 and 1997, there were shootings at schools in Pearl, Mississippi; West Paducah, Kentucky; Jonesboro, Arkansas; Edinboro, Pennsylvania; and Springfield, Oregon. ELIZABETH DONOHUE ET AL., JUSTICE POL’Y INST., SCHOOL HOUSE HYPE: SCHOOL SHOOTINGS AND THE REAL RISKS KIDS FACE IN AMERICA (1999), available at As profoundly alarming as these events were, “the number of children killed by gun violence in schools is about half the number of Americans killed annually by lightning strikes.” Id.

14. The Gun-Free Schools Act of 1994, 20 U.S.C. ß 8921(b) (1994). The Act conditions federal aid on the state’s adoption of a mandatory one-year expulsion for students who bring certain kinds of weapons to school, although it does permit the school district’s chief administrator to modify this sanction in compelling cases. Id. The law provides in part:

(1) In General. Except as provided in paragraph (3), each State receiving Federal funds under this chapter shall have in effect a State law requiring local educational agencies to expel from school for a period of not less than one year a student who is determined to have brought a weapon to a school under the jurisdiction of local educational agencies in that State, except that such State law shall allow the chief administering officer of such a local educational agency to modify such expulsion requirement for a student on a case-by-case basis. (2) Construction. Nothing in this subchapter shall be construed to prevent a State from allowing a local educational agency that has expelled a student from such a student’s regular school setting from providing educational services to such student in an alternative setting.


Although this section affords the chief administrative officer limited discretion in the prescribed punishment, some schools districts choose not to adopt discretionary policies. Other districts fail to abide by the discretionary policies they have adopted. See, e.g., Lyons v. Penn. Hills Sch. Dist., 723 A.2d 1073 (Pa. 1999) (reversing the expulsion of a seventh grader for carrying a Swiss army knife because the school board violated school department rules by denying the superintendent of schools the right to review expulsion decisions). According to a 1998 Department of Education study of forty-three states, only thirty-four percent of expulsions were for less than one year. See BETH SINCLAIR ET AL., U.S. DEP’T OF EDUC., REPORT ON STATE IMPLEMENTATION OF THE GUN-FREE SCHOOLS ACTóSCHOOL YEAR 1996-1997, at 4 (1998) (providing a compilation of statistics for expulsions in American schools).


How to Construct an Underclass 65

fifty states complied,15 but the great majority did so by instituting far broader zero tolerance sanctions. According to one study, by 1998 eighty- seven percent of public schools had promulgated zero tolerance for alcohol and eighty-eight percent for drugs.16

The disciplinary policies in effect in many schools today apply zero tolerance to public school students in three draconian ways. First, they are blind to the most basic distinctions between types of offenses. In many schools, dangerousness is irrelevant; the penalties are the same for weapons and alcohol, sale and possession, robbery, and disorderly offenses. Offenses that used to be resolved informally with an apology or an after-school detention now lead to formal disciplinary hearings.17 Second, they require a severe sanction, typically suspension or expulsion, for all of these offenses, regardless of the circumstances of the offense or the intent, history and prospects of the offender. Third, these policies generally mandate some degree of information-sharing with law enforcement. This multiplies the consequences of student misconduct in two directions: out-of-school offenses referred to the child’s school may result in suspension or other sanctions,18 and in-school infractions referred to law enforcement agencies may result in juvenile or criminal prosecution.19 Forty-one states and federal law now require schools to report specified kinds of offenses to law enforcement.20

15. See Alicia C. Insley, Suspending and Expelling Children from Educational Opportunity: Time to Reevaluate Zero Tolerance Policies, 50 AM. U. L. REV. 1039, 1047 n.46 (2001) (citing the laws of the fifty states and the District of Columbia).

  1. See KAUFMAN ET AL., supra note 3, at 129 tbl.A1.
  2. George F. Will sees zero tolerance school policies as an example of the ascendancy of

“Bureaucratic Legalism” over “Hidden Law, on which privacy and civilized life generally depend. . . . [C]ivilized life depends on informal rules and measuresósocial winks, . . . preventing such mundane conflicts from becoming legal extravaganzas or occasions for moral exhibitionism. . . . One sound of such a society is the ‘snap’ of handcuffs being placed on a 12-year- old subway snacker.” George F. Will, Zero Tolerance Policies are Getting Out of Hand, BOSTON GLOBE, Dec. 25, 2000, at A23; see also Dirk Johnson, Schools’ New Watchword: Zero Tolerance, N.Y. TIMES, Dec. 1, 1999, at A1 (“[T]he new school-conduct ethos has profoundly changed views about what was once deemed usual, if annoying, behavior by adolescents. No longer is the playground scrap or the kickball tussle deemed a rite of passage best settled by a teacher who orders the combatants to their corners, hears out the two sides and demands apologies and a handshake.”).

18. According to one reporter, Chicago’s school zero tolerance policies are in effect twenty- four hours a day, seven days a week. Johnson, supra note 17, at A1. Massachusetts adopted the so- called Principal’s Bill, which gave principals the power to suspend students who are charged and expel students who are convicted of felonies so long as the principal determines that the student’s presence at school is detrimental to the general welfare of the school. Anthony J. DeMarco, Suspension/Expulsion Punitive Sanctions from the Jail Yard to the School Yard, 34 NEW ENG. L. REV. 565, 568 (2000).

19. Criminal prosecution is increasingly likely for juvenile offenders. From 1992 to 1995, forty states changed their juvenile delinquency laws to make it easier for juveniles to be sentenced as adults. PATRICK GRIFFIN ET AL., U.S. DEP’T OF JUST., TRYING JUVENILES AS ADULTS IN CRIMINAL COURT: AN ANALYSIS OF STATE TRANSFER PROVISIONS (1998).



66 The Journal of Gender, Race & Justice [6:2002]

This blunderbuss approach has worked as intended. In Chicago, expulsions rose from 81 to 1000 during the first three years of zero tolerance.21 Nationally, more than 3.1 million students were suspended and another 87,000 were expelled in the 1998 school year.22 African American students made up a disproportionately large percentage of these expulsions and suspensionsó33%, although they comprised only 17% of students.23 The immediate consequence for these students was a total withdrawal of public education in the many states that do not offer alternative schooling to expelled or suspended students.24 Longer term, studies show that many of

SUSPENDED: THE DEVASTATING CONSEQUENCES OF ZERO TOLERANCE AND SCHOOL DISCIPLINE POLICIES, Executive Summary at 1, 8 (2000) (citing forty one states that require school officials to report students who violate certain disciplinary rules to law enforcement agencies), available at [hereinafter HARVARD REPORT]. Federally, the Gun Free Schools Act requires referral to law enforcement when the student has brought a firearm to school. 20 U.S.C.S. ß 7151(h)(1) (2002) (“No funds shall be made available under any title of this Act to any local educational agency unless such agency has a policy requiring referral to the criminal justice or juvenile delinquency system of any student who brings a firearm or weapon to a school served by such agency.”).


22. OFFICE FOR CIVIL RIGHTS, U.S. DEP’T OF EDUC., FALL 1998 ELEMENTARY AND SECONDARY SCHOOL CIVIL RIGHTS COMPLIANCE REPORT: NATIONAL AND STATE PROJECTIONS (June 2000), cited in HARVARD REPORT, supra note 20, at 1. Suspensions and expulsions continue to rise. See, e.g., Carl Campanile, Suspension DisbeliefóSchool Kids Are Getting Sent Home in Droves, N.Y. POST, Nov. 1, 2001, at O25; Scott S. Greenberger, Expulsion, Suspension Rate Climbs 6 Percent; Study Says Boston Figures Remain Low, BOSTON GLOBE, Dec. 21, 2001, at B7 (reporting “more than 1,400 Massachusetts students were expelled from school or suspended for more than 10 days during the 1999-2000 school year,” a 6% increase over the previous year); Grace Murphy, SAD 6 Policy Causes Leap in Expulsions, PORTLAND PRESS HERALD, Feb.17, 2000, at 1A. It is estimated that approximately 1.5 million students miss a large portion of school annually because of suspensions or expulsions. Philip Daniel & Karen Coriell, Suspension and Expulsion in America’s Public Schools: Has Unfairness Resulted from a Narrowing of Due Process?, 13 HAMLINE J. PUB. L. & POL’Y 1, 15 (1992).

23. HARVARD REPORT, supra note 20, at n. 22 (citing the OFFICE FOR CIVIL RIGHTS, supra note 22). A second study of twelve cities found that “in disproportionate numbers, it is African American and Latino students whose futures are wrecked by zero-tolerance.” APPLIED RES. CTR., supra note 21, at 4; see also infra note 79.

24. While twenty-six states require school districts to provide alternative schools for these students, eighteen states leave it to the discretion of school districts whether to provide alternative education, and many of these offer no educational program at all. See HARVARD REPORT, supra note 20, at 3. Of the twenty-six states that do require alternative education assignments, the report cites “anecdotal evidence illustrat[ing] that many of these schools fail to provide an adequate education. There is little data revealing the quality of the instruction that occurs in these centers if any is given at all.” Id. The Report’s Appendix III details the availability of alternative education programs in each state. See also KIM BROOKS ET AL., JUST. POL’Y INST., SCHOOL HOUSE HYPE: TWO YEARS LATER 38 (2000); Joan M. Wasser, Zeroing in On Zero Tolerance, 15 J.L. & POL’Y 747, 761-62 (1999) (citing Department of Education statistics estimating that “fifty-six percent of students expelled in 1996-1999, approximately 38,200 students, were not offered any form of alternative education”). The authors note that like other states

Massachusetts does not require schools to provide expelled students with alternative education, either home schooling, or placement in an alternative school. Probably the


How to Construct an Underclass 67

these children will not return to school even when the sanction expires, and those who do return are more likely than other students to fail their courses.25

And for what? Courts and scholars have found that a majority of suspensions are ordered for minor infractions that pose no actual threat to anyone.26 According to numerous media reports, suspensions and expulsions have followed the discovery of such “weapons” as key chains,27 staplers,28 and geometry compasses;29 and such “drugs” as lemon drops, Midol and Advil.30 One student was suspended for “trafficking” in drugs by loaning her asthma inhaler to a student in the midst of an asthma attack.31 Because intent

most disturbing finding in the Massachusetts report is that 37% of youth expelled in 1997-98 did not receive alternative education in another school or special education program. In 75% of those cases, alternative education was not provided because the school district chose not to do so.

BROOKS ET AL., supra, at 38.

25. Pedro Reyes, Factors that Affect the Commitment of Children at Risk to Stay in School, in CHILDREN AT RISK 18, 23 (Joan Lakebrink ed., 1989); see also Russ Skiba & Reese Peterson, The Dark Side of Zero Tolerance: Can Punishment Lead to Safe Schools, 80 PHI DELTA KAPPAN 372, 376 (1999). A significant factor is whether suspended students are given the opportunity to make up schoolwork they missed during the suspension. Some schools allow suspended students to make up their missed work, while others believe not being able to do so should be part of the punishment. See, e.g., Tracy Dell’Angela, Suspended Girl Sues School, CHI. TRIB., June 18, 2001, at 1. Also, if the juvenile offender is incarcerated, he or she is unlikely ever to return to school. ROBERT J. GEMIGNANI, U.S. DEP’T OF JUST., JUVENILE CORRECTIONAL EDUCATION: A TIME FOR CHANGE 3 (NCJ Pub. No. 150 309, 1994).

26. Hawkins v. Coleman, 376 F. Supp. 1330, 1335-37 (N.D. Tex. 1974) (finding sixty percent of offenses punished by suspension or corporal punishment were for minor offenses). One study estimates that about three percent of infractions underlying school removals were major offenses, and another found most were for minor offenses, such as smoking, tardiness, truancy, and dress code violations. Roni Reed, Education and the State Constitutions: Alternatives for Suspended and Expelled Students, 81 CORNELL L. REV. 582, 603 (1996); see Daniel & Coriell, supra note 22, at 15; Gail Paulus Sorenson, The Worst Kind of Discipline, 6 UPDATE ON LAW-RELATED EDUC. 26, 27 (1982).

27. Georgia Pols Want ‘Common Sense’ to Trump ‘Zero Tolerance’ (Fox News television broadcast, Jan. 21, 2002), available at,2933,43666,00.html (“Georgia got its taste of zero tolerance-gone-wild as dozens of children were disciplined for what appeared to be minor infractions,” including a keychain deemed a weapon) [hereinafter Fox News].

28. The student was suspended for several months for holding a stapler as if it were a gun. See James M. Peden, Through a Glass Darkly: Educating with Zero Tolerance, 10 KAN. J.L. & PUB. POL’Y 369, 374 (2001) (citing World News Tonight with Peter Jennings (ABC television broadcast, Feb. 8, 2000)).

29. Jordana Hart, Creeping Violence Worries Suburban Schools; Stricter Rules Enforced to Seize “Weapons,” BOSTON GLOBE, Apr. 18, 1994, at 15 (reporting a broadening of the definition of weapons to include a geometry compass).

30. Cara DeGette, Busted for Lemon Drops, First-Grader Suspended, DENVER POST, Nov. 19, 1997, at A01; Jessica Portner, Suspensions Spur Debate Over Discipline Codes, EDUC. WK., Oct. 23, 1996, at 10 (reporting that a thirteen-year old honor student received a nine-day suspension for possession of Midol tablets in school and a seventh-grader was suspended for a day for bringing Advil to school).

31. Margaret Graham Tebo, Zero Tolerance, Zero Sense, A.B.A. J., Apr. 2000, at 44 (citing “a middle schooler who shared her asthma inhaler on the school bus with a classmate who was


68 The Journal of Gender, Race & Justice [6:2002]

is irrelevant under some zero tolerance policies, students have been suspended for unknowingly possessing a weapon or contrabandña penalty that is proving vulnerable to due process challenges.32 Zero tolerance may also apply against such nebulous infractions as “disrespect” and “defiance of authority,” which are prone to import bias into the decision to discipline.33 Zero tolerance leaves school officials “unable to differentiate between good kids who may be acting imprudently, and the unruly delinquent who can be dangerous.”34

B. College and Higher Education

Denial of a college education is now an additional sanction for drug offenders as the result of two federal statutes: a 1998 law that suspends or forever terminates a drug offender’s eligibility for federal college loans and grants, and (as we shall discuss below) a 1994 law making all inmates ineligible for the Pell Grants that formerly provided the means to obtain a college degree in prison.35

The 1998 Drug Free Student Loans Act denies federal grants, federally- subsidized loans, and work-study funds to college students who have been convicted of any drug offenseñfelony or misdemeanor, sale or possession, heroin or marijuana (but not rape, robbery, or murder).36 However, the

experiencing a wheezing attack was suspended for drug trafficking”).

32. See Seal v. Morgan, 229 F.3d 567 (6th Cir. 2000) (holding the expulsion of a student who unknowingly brought a weapon to school grounds when his friend’s knife was found in his glove compartment was unconstitutional because not rationally related to any legitimate state interest).

33. A.B.A., A.B.A. ZERO TOLERANCE POLICY REPORT 4 (2001), available at

34. Peden, supra note 28, at 373. Among many examples, Peden cites the case of a Virginia middle school student who convinced a friend to give him her knife after she threatened to commit suicide with it. Id. “[H]e put it in his locker. During the course of the day, he went to his locker to change books and another student saw the knife in his locker and reported it to the principal. While his school praised him for his actions, the local school board decided to suspend him because he did not immediately turn the knife over to the school officials.” Id. (citing Leonard Pitts, School Bent on Ignoring Common Sense, MILWAUKEE J. SENTINEL, Dec. 20, 1999, at 10).

35. Violent Crime Control and Law Enforcement Act of 1994, Pub. L. No. 103-322, ß 20411, 108 Stat. 1796 (1994), amending 20 U.S.C. ß 1070a(b).

36. A 1988 law already authorized judges to deny federal college assistance at a drug offender’s sentencing, but because this discretion was rarely exercised, Congress revised the law to make the sanction mandatory. This new law, passed as the Souder Amendment to the 1998 Higher Education Reauthorization Act, 20 U.S.C. ß 1091(r) (2002), was sponsored by Rep. Mark Souder (R-Ind.) and former Rep. Gerald Solomon (R-N.Y.). It states as follows:

Suspension of eligibility for drug-related offenses.

(1) In general. A student who has been convicted of any offense under any Federal or State law involving the possession or sale of a controlled substance shall not be eligible to receive any grant, loan, or work assistance under this subchapter and


How to Construct an Underclass 69

periods of ineligibility for federal financial aid vary depending upon the number of convictions and whether they were for possession or distribution of drugsófrom one year’s ineligibility for a single possession conviction, to permanent ineligibility for a second distribution or third possession conviction.37 Students may shorten the period of ineligibility by successfully completing a certified rehabilitation program that includes two random drug tests.38 A particularly draconian aspect of this law is that ineligibility can be triggered not only by offenses committed during college, but also by those accrued years before.39

The law’s co-sponsor, Congressman Mark Souder, hopes that the law will send “a clear message: Actions have consequences, and using or selling drugs will ruin your future.”40 Whether or not that message is getting through, the Souder Amendment has already succeeded in crippling the future prospects of thousands of students. In its first yearóthe academic year 2000-2001óthe law rendered 9,200 students ineligible to receive federal college assistance due to prior drug convictions.41 These students answered “yes” to the financial aid form inquiry regarding about whether they had a previous drug conviction.42 Significantly, the 279,000 students

part C of subchapter I of chapter 34 of Title 42 during the period beginning on the date of such conviction and ending after the interval specified in the following table:

(2) convicted of an offense involving
óThe possession of a controlled substance: Ineligibility period is: First offense . . . . . .. . . . . .. . .1 year
Second offense . . . . . .. . . . . .2 years
Third offense. . . . . .. . . . . … Indefinite

óThe sale of a controlled substance: Ineligibility period is: First offense . . . . . .. . . . . .. . .2 years
Second offense. . . . . .. . . . . . Indefinite

37. Id. While the statute speaks of “indefinite” ineligibility, this is construed to be a permanent bar. 34 C.F.R. ß 668.40 (b)(1)(iii) cmt. (2001); 34 C.F.R. ß 668.40 (b)(2)(ii) cmt. (2001). Some drug convictions do not disqualify a candidate: convictions that have been dismissed or expunged, and juvenile court delinquency findings. 20 U.S.C. ß 1091(r)(2)(B) (1998).

38. 20 U.S.C. ß 1091(r)(2) provides an exception for rehabilitated offenders as follows: (2) Rehabilitation. A student whose eligibility has been suspended under paragraph (1)

may resume eligibility before the end of the ineligibility period determined under such paragraph ifó
(A) the student satisfactorily completes a drug rehabilitation program that- (i) complies with such criteria as the Secretary shall prescribe in regulations for purposes of this paragraph; and (ii) includes two unannounced drug tests; or

(B) the conviction is reversed, set aside, or otherwise rendered nugatory.

39. See infra notes 158-69 regarding the legal infirmaties with retroactive application of the Souder Amendment.

  1. Mark Souder, Actions Have Consequences, USA TODAY, June 13, 2000, at 16A.
  2. Diana Jean Schemo, Students Find Drug Law Has Big Price: College Aid, N.Y. TIMES,

May 3, 2000, at A12.
42. Free Application for Federal Student AID (FAFSA) for July 1, 2001 – June 30, 2002


70 The Journal of Gender, Race & Justice [6:2002]

who left the question blank last year continued to obtain loans.43 However, the Bush Administration has decided that henceforth candidates who do not answer the question will be deemed ineligible44 ñ a dubious shift coming from a President who refused to answer questions about his own drug history on grounds it was irrelevant to his suitability for the Presidency. The number of students who will be denied college assistance this year is expected to at least quadruple: as of the end of 2001, 43,436 had either been rejected or risked automatic denial for leaving question 35 blank.45 For many of them, this will mean the difference between going to college and dropping out.46 Although in theory a student may regain eligibility by completing a rehabilitation program, this will be an impossible burden for students who either can’t afford or can’t obtain the relatively scarce placements, or who must quit part-time jobs or miss class in order to attend meetings.47

(OMB #1845-0001). Question 35 reads: “Do not leave this question blank. Have you ever been convicted of possessing or selling illegal drugs? If you have, answer yes, complete and submit this application, and we will send you a worksheet in the mail for you to determine if your conviction affects your eligibility for aid.” Id.

43. Schemo, supra note 41, at 12. An unknown number of additional students may have opted out of the federal aid pool altogether because of prior drug involvement that they, perhaps incorrectly, thought would make them ineligible. Id. Others may have falsely answered “no” in order to get federal aid. Id.

44. Carl M. Cannon, A Drug Law’s Long Reach, NAT’L J., July 14, 2001, at 2251. Although Department spokeswoman Lindsey C. Kozberg states that “this is a change in practice, not a policy change” to “better reflect the legislative requirements of the statute,” it is nevertheless a change that will greatly exacerbate the law’s exclusionary impact. Id. at 2252.

45. 43,000 Students with Drug Convictions Face Denial of Aid, N.Y. TIMES, Dec. 29, 2001, at A11 [hereinafter 43,000 Students]. This constitutes a small percentage of the 9.8 million financial aid applications processed, but a devastating blow to each of the individuals affected.

46. One magazine reports the story of “Anne”:

the type of success story that makes college admissions officers smile. She had little money growing up, but with the help of federal loans and a lot of sweat has managed to afford an education at Ohio’s Antioch College. In addition to attending school full-time, she currently works two jobs, one of which involves teaching drama to grade-school children. . . . But to the federal government, Anne is a campus undesirable. In early May, police caught her with an empty one-hit pipe containing residue from the marijuana she occasionally smokes, and now she’s awaiting trial. The charge is only misdemeanor drug possessionóbut if she’s convicted, Anne may lose her federal Pell Grant, which she depends on to pay for school

Jake Ginsky, Smoke a Joint, Lose Your Loan, MOTHER JONES, May 18, 2000, available at

47. America has a paucity of drug rehabilitation programs compared to the demand for them, and the minor offenders most likely to be targets of college loan ineligibility may find an authorized rehabilitation program out of reach. Apart from tuition that may be unaffordable to students who all are dependent on federal aid, many treatment programs give priority to intravenous addicts, a group unlikely to be enrolled in college, and also refuse to accept persons that are not serious substance abusers. See Arianna Huffington, Souder Says Drug-Use Rule Never Meant as Punishment, SO. BEND TRIB., Apr. 29, 2001, at B8 (noting that treatment is typically available only to those who can pay for it or who have serious drug problems, and that three million people are unable to get treatment); Simone Levine, Encountering the 1998 Amendment to the Higher Education Act: A Guide to Litigation (2001) (unpublished manuscript, on file with authors).


How to Construct an Underclass 71

Those who do enroll in rehabilitation programs under the compulsion of this law may not belong there. The law’s assumptionóthat students with a drug conviction have a drug problem that needs rehabilitationóignores statistics on the high frequency of youthful drug experimentation and the sharp decline in drug use that follows in early adulthood.48 Over one third of all Americans have experimented with illicit drugs, but most lose interest without the necessity of rehabilitation or other intervention.49 Opponents of the law object on these grounds and multiple others, including that the law is “racially and economically discriminatory; it ignores the No. 1 substance abuse problem on campus, binge drinking; it might steer students with old convictions . . . into scarce rehab programs that should be reserved for true addicts; it encourages students to lie; [and] it entails a kind of double jeopardy not encountered in cases of more serious felonies.”50

C. Prison Education

For drug offenders in prison, educational privation is the result of two zero tolerance policies operating in tandem. Mandatory minimum drug sentences and other get-tough laws sent them to prison, and zero tolerance clampdowns on prisoners stripped them of access to longstanding prison education programs.

Between 1980 and 2000, the prison population tripled.51 A large share of this increase resulted from escalating drug arrests and prosecutions, and a series of laws that imposed increasingly harsh sentences for virtually all drug

One twenty-three year old student at a state university, who was forced to drop out when he lost his federal aid said, “I tried to get into drug treatment. I was working at a car wash at the time. And I told my boss I needed a month off so I could go help myself and he fired me immediately, said he couldn’t afford to keep my job.” Lost Loans, Lost Future?: Students Lose Aid Because of Past Drug Offenses (ABC News television broadcast, Apr. 28, 2001), available at (quoting Josh Gerstein).

48. Social scientists consider experimentation with a wide range of lawful and unlawful behaviors to be a natural part of the process of maturity. AM. ACAD. OF CHILD & ADOLESCENT PSYCHIATRY, TEENS: ALCOHOL AND OTHER DRUGS NO. 3, at (last visited May 1, 2002).

49. According to the 1998 survey on drug abuse conducted by the Substance Abuse and Mental Health Services Administration (SAMHSA), 35.8% of Americans over 12 have used an illicit drug (not including tobacco or alcohol) at some time in their life, and an estimated 13.6 million Americans were current users of illicit drugs in 1998. SAMHSA, 1998 NATIONAL HOUSEHOLD SURVEY ON DRUG ABUSE, tbl. 3B (1999), available at With approximately 216 million of the U.S. population over 13 in 1998, according to the U.S. Census Bureau, these figures indicate that approximately 82% of those who have violated the drug laws no longer do so.

50. Carl M. Cannon, A Drug Law’s Long Reach, NAT’L J., July 14, 2001, at 2252 (citing the complaints of the organizations Students for Sensible Drug Policy and the Drug Reform Coordination Network).

51. BUREAU OF JUST. STAT., U.S. DEP’T OF JUST., CORRECTION STATISTICS (posted Feb. 3, 2002) (stating that the incarceration rate rose from 139 per 100,000 to 478 per 100,000, an increase of 343.88%).


72 The Journal of Gender, Race & Justice [6:2002]

crimes and mandatory minimum sentences for many of them.52 Of the two million Americans in prison today,53 approximately one quarter are there for drug offenses;54 in federal prisons, the proportion is now 62%.55 Today, 72% of first-time admittees to state prison are non-violent offendersóa figure that suggests a prison population that is less committed to a life of crime and more educable than was formerly the case.56

Yet prisoners as a class are substantially undereducated: almost one- half lack a high school education,57 and one in five are completely illiterate.58 What then could be a better use of five or 10 year’s confinement

52. Arrests: In 1999, there were approximately 1,532,200 non-alcohol drug arrests, 80% of which were possession offenses and 20% of which were sale or distribution offenses. BUREAU OF JUST. STAT., U.S. DEP’T OF JUST., SOURCEBOOK OF CRIMINAL JUSTICE STATISTICS 2000, at 352 tbl. 4.1, 393 tbl. 4.29 (Kathleen Maguire & Ann Pastore eds., 2001) [hereinafter DEP’T OF JUST., SOURCEBOOK].

Prosecutions: Federal drug prosecutions rose sharply from 1980 to 1992 before reaching a plateau. In 1981, when President Reagan took office, there were 8,775 federal drug prosecutions. The Bush and Clinton administrations each averaged just under 26,000 drug prosecutions a year, with the number rising to 28,585 in the election year of 1992. Drug Prosecution Said to Level Off, BOSTON GLOBE, Oct. 20, 1996, at A16 (citing Transactional Records Access Clearinghouse, Syracuse University).

Prisoners: The number of drug offenders in prison increased by 478% between 1985 and 1995, compared to a rise of 119% for all offenses. MARC MAUER, RACE TO INCARCERATE 152 (1999). This reflected an exploding increase in prisoners admitted for drug offenses during that decadeóa 1040% increase between 1986 and 1996. BARRY HOLMAN, NAT’L CTR. ON INST. & ALTERNATIVES, MASKING THE DIVIDE: HOW OFFICIALLY REPORTED PRISON STATISTICS DISTORT THE RACIAL AND ETHNIC REALITIES OF PRISON GROWTH 17 (2001), available at

53. In 2000 there were 1,933,503 prisoners and 4,565,059 others on parole or probation, which amounts to one in every 32 adults who were under the supervision of the criminal justice system. DEP’T OF JUST., SOURCEBOOK, supra note 52, at 488 tbl. 6.1. An inmate population of 2 million was projected to arrive by late 2001. ALLEN J. BECK & JENNIFER C. KARBERG, PRISON AND JAIL INMATES AT MIDYEAR 2000, BUREAU OF JUST. STAT. BULLETIN (2001), available at, at 2. “If recent incarceration rates remain unchanged, an estimated 1 of every 20 persons (5.1%) will serve time in a prison during their lifetime.” BUREAU OF JUST. STAT., U.S. DEP’T OF JUST., CRIMINAL OFFENDER STATISTICS, available at (last revised Feb. 28, 2002).

  1. DEP’T OF JUST., SOURCEBOOK, supra note 52, at 519 tbl.6.39 (compiling 1997 statistics).
  2. Id.
  3. OPEN SOC’Y INST., THE AFTER PRISON INITIATIVE, available at (updated Feb. 1, 2002) (72% figure); WILLIAM J. SABOL & JAMES P. LYNCH, URBAN INST., CRIME POLICY REPORT: DID GETTING TOUGH ON CRIME PAY? 8 (1997). Prison admittees now comprise a more socially integrated groupómore likely to have some education, employment history, and families.

57. A survey of inmates in 1996 showed 46.5% had less than a high school education.” TASK FORCE ON YOUTH IN THE CRIMINAL JUSTICE SYSTEM, A.B.A, YOUTH IN THE CRIMINAL JUSTICE SYSTEM, CORRECTION, 2001, at 5 n.77 (2001). Among those entering state prisons, over 70% have not completed high school, and 16.4% have no high school education at all. BUREAU OF JUST. STAT., U.S. DEP’T OF JUST., SOURCEBOOK OF CRIMINAL JUSTICE STATISTICS 1995, at 567 (Kathleen Maguire & Ann L. Pastore eds., 1996) [hereinafter DEP’T OF JUST., SOURCEBOOK 1995].



How to Construct an Underclass 73

than education? Prior to the 1990s, few questioned the value and importance of prison education programs, at least publicly. Politicians had not yet stricken rehabilitation from their political lexicon, and most prison administrators were content with both the opportunities afforded motivated prisoners and the institutional benefits to be gained from having prisoners who were occupied with homework, classes, discussions and planning for the future.59 Most prisoners could choose from a variety of education opportunities, including grammar and high school courses, vocational training, and classes from community college offerings within the prisons.60 A smaller number of inmates who wanted to study at the college and graduate school level could enroll in diverse offerings made available to prisons by colleges and universities.61 For inmates who did not have the means to pay for the classes, the federal government offered financial assistance in the form of Pell Grants.62 Opportunities for advanced education in prison reached their zenith in 1990, with 1,287 prison higher education programs across the country educating approximately 40,000 students.63

In the midst of these proliferating programs and an ever-expanding, increasingly educable prison population, Congress gutted federal college aid for prisoners, resulting in the near demise of the college programs. In 1994 it

THE PRISON POPULATION FROM THE NATIONAL ADULT LITERACY SURVEY 124 (NCES Publication No. 94-102, 1994). The report finds nineteen percent of adult inmates completely illiterate and forty percent functionally illiterate. Id.

59. In a 1993 Senate Judiciary Committee survey, ninety-three percent of prison wardens strongly supported prisoner education programs. PETER ELIKANN, THE TOUGH-ON-CRIME MYTH: REAL SOLUTIONS TO CUT CRIME 151 (1996). According to one study, “participation in college programs [is a] better predictor of positive institutional behavior than participation in conjugal visitation programs.” Jon Marc Taylor, Deny Pell Grants to Prisoners? That Would Be A Crime, 9 CRIM. JUST. 19, 55 (1994) (citing R. Davis, Education and the Impact of the Family Reunion Program in a Maximum Security Prison, J. OFFENDER COUNSELING, SERV. & REHABILITATION 153 (1988)). For arguments supporting the training of prisoners for the increasingly demanding workplace they will encounter upon their release, see Richard A. Tewksbury & Gennaro F.Vito, Improving the Educational Skills of Jail Inmates: Preliminary Program Findings, 58 FED. PROBATION 55 (1994); Ahmad Tootoonchi, College Education in Prisons: The Inmates’ Perspectives, 57 FEDERAL PROBATION 34 (1993).

60. For many years prisons offered, and sometimes required, basic education classes, first through eighth grade, and later through twelfth grade. See Michael K. Greene, Show Me the Money, 24 CRIM. & CIVIL CONFINEMENT 173, 177 (1998). Even with these requirements, however, many who completed twelfth grade were not ready for college. The community college system filled this gap by offering basic skills courses and vocational education. Students who wished to pursue their education by taking college courses leading to an advanced degree were able to do so once they attained the necessary basic reading and writing skills.

61. See generally R. Tewksbury, et al., Opportunities Lost: The Consequences of Eliminating Pell Grant Eligibility for Correctional Education Students, 31 J. OFFENDER REHAB. 43 (2000).

62. Pub. L No. 102-32, 106 Stat. 481 (1992) amending the Higher Education Act of 1965, made Pell grants available to all financially-needy students including prisoners unless the latter were “serving under sentence of death or any life sentence without eligibility for parole or release.” Id. ß (8)(A).



74 The Journal of Gender, Race & Justice [6:2002]

enacted the Violent Crime Control Law Enforcement Act, which eliminated Pell grants for state and federal prisoners.64 Perhaps this was inevitable in an era when politicians outdid one another denouncing rehabilitation and promoting more punitive approaches to imprisonment, such as chain gangs, three-strikes-and-you’re-out laws and striped uniforms. 65 Congressmen denounced a “taxpayer rip-off” that rewarded prisoners for their crimes with a college education.66 Senator Kay Bailey Hutchinson alleged that providing Pell Grants to “carjackers, armed robbers, rapists, and arsonists” shortchanged 100,000 non-criminal students who were denied Pell Grants because of lack of funds.67 This was extreme hyperbole: at that time prisoners received only .7% percent of Pell Grants issued (300,000 prisoners out of 4.3 million recipients), a similar proportion of Pell dollars ($45 million out of $6.3 billion); and all eligible applicants received aid.68

The elimination of Pell Grants for prisoners marked the end of prison college programs for most prisoners. Without federal financial aid, demand for college programs declined dramatically (the 1995 national student exit rate from such programs was 70%)69 and consequently so did the programs

64. Violent Crime Control and Law Enforcement Act of 1994, Pub. L. No. 103-322, ß 20411, 108 Stat. 1796 (1994). The act amended 20 U.S.C. ß 1070(a)(b), which now states: “(8) No basic grant shall be awarded under this subpart to any individual who is incarcerated in any Federal or State penal institution.” ß 20411. Congress subsequently created a Youthful Offenders Grant program, which provided some post-secondary educational grants to youthful offenders beginning in 1998. Improving America’s Schools Act of 1994 Pub. L. No. 103-382, 108 Stat. 3518 (1994). Compared to Pell Grants, however, this program dispensed a much smaller amount of aid, and only gives to young inmates under twenty-six with less than five years to serve. Peter Schmidt, College Programs for Prisoners, Long Neglected, Win New Support, CHRON. HIGHER EDUC., Feb. 8, 2002, at 26. The Department of Education also awards approximately $15 million in grants to assist states in establishing and operating programs designed to develop the life skills necessary for reentry, pursuant to the National Literacy Act as amended. National Dropout Prevention Act of 1991, Pub. L. No. 102-103, ß 313, 105 Stat. 508 (1991). See Office of Correctional Education website, available at

65. CAL. PENAL CODE ß 667 (1996) (providing for life in prison upon third felony conviction); Sheryl Stolberg, Schools Out For Convicts, L.A. TIMES, Sept. 14, 1995, at A1 (“In Alabama and Arizona, chain gangs are back. In Texas, weightlifting has been banned. . . . In Mississippi, convicts will soon wear striped uniforms. In the long-running debate over whether the purpose of prisons is to rehabilitate or to punish, the pendulum has swung clearly in the direction of punishment.”).

66. COMMUNITY RESOURCES FOR JUSTICE, RETURNING INMATES: CLOSING THE PUBLIC GAP (2001). Massachusetts Governor William Weld told 60 Minutes that inmates “are in prison to be punished, not to receive free education.” 60 Minutes (CBS television broadcast, May 5, 1991); see also Michael K. Greene, Show Me The Money! Should Taxpayer Funds Be Used To Educate Prisoners Under the Guise of Reducing Recidivism?, 24 NEW ENG. J. ON CRIM. & CIV. CONFINEMENT 173 (1998) (arguing against government funding of prison education); Taylor, supra note 59, at 23 (reporting Congressional debate on H.R. 1158).

67. 139 CONG. REC. S15746 (1993) (Congressional debate on Pell Grant eligibility, Nov. 16, 1993).

  1. Taylor, supra note 59, at 23-24.
  2. Richard Tewksbury & Jon Marc Taylor, The Consequences of Eliminating Pell Grant

Eligibility for Students in Post Secondary College Education Programs, 60 FED. PROBATION 60, 61 (1996). The percentage of prison inmates in post secondary education was cut in half over four


How to Construct an Underclass 75

themselves (from an average of 6.1 different post-secondary schools operating in each state’s prisons in 1994-95 to only 3.4 a year later).70 At the same time, non-college prison education programs folded as states decided they could no longer afford to fund them.71 In 2002, “only 9% of prisoners are [enrolled] in full-time job training or education programs.”72


If one deliberately set out to construct an underclass, an especially effective method would be to inflict ignorance, even illiteracy, on an already-disadvantaged population. To impose such a severe disability would insure systematic disadvantages throughout life on those least able to overcome them. The uneducated are primed for unemployment or marginal employment,73 and all that often comes with it: impoverishment, criminal victimization and temptation, poorer health, shorter lives, political powerlessness, and despair. Disproportionate numbers succumb to alcohol or drug abuse. Educational privation is also an excellent way to make a

years, from 7.3% in 1994 to 3.8% in 1998. Schmidt, supra note 64, at 26.

70. Tewksbury & Taylor, supra note 69, at 61; see also Schmidt, supra note 64, at 26 (“Higher education programs for prisoners have becoming increasingly scarce over the past decade, . . .Congress declared prisoners ineligible for Pell Grants in 1994, and many state legislatures have withdrawn tax-dollar support for such programs.”).

71. At least 25 states reduced vocational and technical training programs after Pell Grants were eliminated for prisoners. R. Worth, A Model Prison, THE ATLANTIC MONTHLY, Nov. 1995, at 38; see also Fox Butterfield, Getting Out: A Special Report, N.Y. TIMES, Nov. 29, 2000, at A1 (“[M]any prison rehabilitation programs were eliminated [including] classes, vocational training and halfway houses. . . . The money saved went to building more prisons.”) [hereinafter Butterfield, Getting Out]. More recently, sudden and severe budgetary shortfalls have simultaneously led to the closing of some prison education programs and some prisons, as states seek ways to reduce the share of state expenditures devoted to corrections. Nationally, this share is approximately seven percent of state budgets. Judith Greene & Vincent Schiraldi, Cutting Correctly: New Prison Policies for Times of Fiscal Crisis, JUST. POL’Y INST., Feb. 2001, at 1. Illinois, for example, is saving $5.4 million by eliminating classes for 25,000 inmates and also closing Joliet Correctional Center. Id. at 3; see also Fox Butterfield, Tight Budgets Force States to Reconsider Crime and Penalties, N.Y. TIMES, Jan. 21, 2002, at A1 (reporting efforts to close prisons or otherwise reduce correctional budgets in California, Ohio, Illinois, Michigan, and Washington); Robert Ellis Gordon, My Life as a Prison Teacher, CHRISTIAN SCI. MONITOR, Mar. 12, 2001, at 9 (reporting that in Washington state, a “no- frills” approach to incarceration led to the dismantling of the community-college system, “once a model for the nation. . . . Even high school degrees are no longer offered to those convicts who want them”).

72. Butterfield, Getting Out, supra note 71, at A1 (quoting James Austin, Director of the George Washington University Institute on Crime, Justice and Corrections).

73. Not surprisingly, an A.B.A. study found a positive correlation between high school education and employment, with drop-outs at risk for both unemployment and arrest. TASK FORCE ON YOUTH IN THE CRIM. JUST. SYS., A.B.A. CRIM. JUST. SEC., YOUTH IN THE CRIMINAL JUSTICE SYSTEM: GUIDELINES FOR POLICYMAKERS AND PRACTITIONERS 27 (2001), available at [hereinafter YOUTH IN CRIMINAL JUSTICE SYSTEM]. The study also noted that 45.5% of all jail inmates in 1996 had less than a high school education. Id.


76 The Journal of Gender, Race & Justice [6:2002]

person feel fungible and insignificant. It is a formula for subtracting self- esteem and substituting the disdain of others.74

Of course, the legislators, policy experts, and officials who remove many thousands of students from the classroom each year are not doing so in order to produce these dire consequences. They are surely aiming at other goals, including drug-free schools and colleges. But while drug-free schools remain a fantasy, their policies are contributing to an uneducated underclass that just gets larger, more despairing, and more entrenched. This underclass now includes five million young adults between sixteen and twenty-four who are both out of school and out of work, with few skills and fewer prospects.75 It includes most ex-prisoners, half of whom lack a high school education,76 and most of whom are jobless one year after release.77 And it includes Black Americans and other racial minorities who have never remotely attained the standard of well-being common throughout the developed world.78

To a large degree, these are the groups most burdened by educational privation sanctionsñthe already disadvantaged who are the very people most damaged by the withdrawal of the primary means of advancement, education. These are policies that by definition affect only the needy: those who cannot afford to obtain a private education. Moreover, they are policies that inevitably burden minority students far more than white students. Consider the impact of each of the three policies on African Americans in particular:

74. As the Supreme Court observed in ruling unconstitutional the denial of education to an undocumented alien, illiteracy “will handicap the individual . . . each and every day of his life,” exacting an “inestimable toll . . . on [his] social, economic, intellectual, and psychological well- being. . . .” Plyler v. Doe, 457 U.S. 202, 222 (1982).

75. Bob Herbert, On the Way to Nowhere, N.Y. TIMES, Sept. 3, 2001, at 15. He adds: “[M]ost lack basic job skills as well as solid literacy and numbers proficiencies, and they are neither working nor looking for jobs. They are not in vocational training. They are not in manufacturing. They are not part of the information age. They are not included in the American conversation.” Id.

76. A survey of inmates in 1996 showed 46.5% had less than a high school education. YOUTH IN THE CRIMINAL JUSTICE SYSTEM, supra note 73, at 5.

77. Petersilia, supra note 10. (“One year after release, as many as 60 percent of former inmates are not employed in the legitimate labor market.”).

When the California state legislature investigated the prospects of its own ex-prisoners, it found between seventy and eighty percent were jobless one year after release and fifty percent were illiterate. Butterfield, Getting Out, supra note 71 (citing a report issued a year before by the California State Legislative Analyst’s Office). Butterfield concludes that “[b]ecause states sharply curtailed education, job training and other rehabilitation programs inside prisons, the newly released inmates are far less likely than their counterparts two decades ago to find jobs, maintain stable family lives or stay out of the kind of trouble that leads to more prison.” Id.

78. The United Nations Human Development Index combines longevity, education, and per capita income to formulate a rough scale of well being. According to this index, the United States ranks sixth among the countries of the world, but white Americans alone would rank first and black Americans alone would rank thirty-first, next to Trinidad and Tobago. United Nations Dev. Programme, Human Development Report (1993), at 18 & figs. 1.12-13, cited in Cass R. Sunstein, The Anticaste Principle, 92 MICH. L. REV. 2410 (1994).


05/16/02 7:57 PM

How to Construct an Underclass

ïAfrican American students are removed from public school at twice the rate of white students,79 a disparity one federal district court has attributed to “institutional racism.”80

ïThe withdrawal of education programs from prisons likewise hits African Americans hardest because their incarceration rate is nine times that of whites.81
ïThe law denying college loans to students with drug convictions badly discriminates because it exactly reproduces the disproportionate percentage of African Americans convicted of these offenses. A statistical study in the mid-’90s found that African Americans comprise 13% of all monthly drug users, but 55% of those convicted of drug possession and 74% of those sentenced to prison for drug possession.82 Similarly, Black juvenile drug offenders are two-and-a-half times more likely than their white classmates to be adjudicated as adults and end up with a drug conviction.83 This discriminatory effect is


CONDITION OF EDUCATION 1997, at 156 (1997) (finding that 19.8 % of all African American students were suspended at least once over a four-year period compared to 9.7 % of their white student counterparts; and that 25% of all African American students nationally are suspended at least once over a four-year period), available at; Herbert, supra note 75, at 15 (citing a 1999 report showing thirteen percent of white sixteen to twenty-four year olds were idle compared to twenty-one of blacks and Hispanics in same age group, excluding the 350,000-360,000 under-twenty four year olds in prison).

80. Hawkins v. Coleman, 376 F. Supp. 1330, 1335-37 (N.D. Tex. 1974) (finding that “institutional racism” resulted in African Americans being suspended more often, and for longer periods, than white students at all grades of the Texas schoolsóand in sixty percent of cases, for minor and non-violent infractions).

81. HOLMAN, supra note 52, at 15 (reporting an African American incarceration rate nine times that of whites, and a Latino incarceration rate almost four times that of whites). For all crimes, blacks and Latinos now comprise seventy percent of the nation’s inmates, although only twenty five percent of the nation’s population. Id. A majority of the Black and Latino inmates sent to prison during those years were sentenced for non-violent drug offenses. Cindy Rodriguez, Latino Prison Count Called Inaccurate, BOSTON GLOBE, June 7, 2001, at A3. A Department of Justice extrapolation from current rates of incarceration for all crimes predicts that “an estimated 28% of black males will enter State or Federal prison during their lifetime, compared to 16% of Hispanic males and 4.4% of white males.” U.S. DEP’T OF JUST., BUREAU OF JUSTICE STATISTICS, CRIMINAL OFFENDER STATISTICS, at (last modified Nov. 14, 2001).

82. MARC MAUER & TRACY HULING, YOUNG BLACK AMERICANS AND THE CRIMINAL JUSTICE SYSTEM: FIVE YEARS LATER 11-12 (1995) (compiling statistics from 1992 and 1993). From 1986 to 1996, “the rate of prison admission for drug offenses increased sixfold for African Americans while the rate of white admissions doubled.” HOLMAN, supra note 52, at 17. The number of black (non-Hispanic) women incarcerated in state prisons for drug offenses multiplied more than eightfold from 1986 to 1991. MAUER & HULING, supra at 20. The net effect is that by 1993 Blacks and Hispanics accounted for almost 90 % of those sentenced to state prisons for drug possession. Id. at 1-2, 13.

83. E. Poe-Yamagata & M. Jones, And Justice for Some: Differential Treatment of Minority Youth in the Justice System, BUILDING BLOCKS FOR YOUTH: WASHINGTON, D.C. (2000). In 1997,

See supra note 23 and accompanying text; see also THOMAS M. SMITH ET AL., THE


78 The Journal of Gender, Race & Justice [6:2002]

compounded by the fact that those who require federal

college aid are disproportionately minorities.84
The bleak consequences of withdrawing educational access sweep well

beyond those directly deprived. They extend to the entire society. A robust economy as well as our democratic survival requires a well-educated population. As the Supreme Court has stated, “Some degree of education is necessary to prepare citizens to participate effectively and intelligently in our open political system if we are to preserve freedom and independence.”85 Another predictable outcome of educational deprivation is an increase in crime. There is a demonstrated correlation between the lack of secondary education and criminal behavior,86 a connection aggravated by expulsions that produce unsupervised free time, bleak future prospects,87 and feelings of unjust treatment.88 One study concludes that “school

0.7% of white juveniles and 1.8% of African American juveniles charged with drug offenses were adjudicated as adults. Id. An Illinois study found African Americans comprise 15.3% of the state’s juvenile population but 88% of the juveniles in adult prisons for drug crimes. Id. (citing JUST. POL’Y INST., DRUGS AND DISPARITY: THE RACIAL IMPACT OF ILLINOIS’ PRACTICE OF TRANSFERRING YOUNG DRUG OFFENDERS TO ADULT COURT (2001), available at The transfers for adult prosecution were pursuant to an Illinois law that provides for automatic transfer of fifteen and sixteen year old drug offenses. 705 ILL. COMP. STAT ß 405/5-130(4)(a) (2002).

84. When selective service non-registrants were barred from receiving post secondary federal assistance, the Supreme Court rebuffed challenges to the law based on Fifth Amendment and Bill of Attainder grounds. Selective Service Sys. v. Minn. Pub. Interest Research Group, 468 U.S. 841 (1984). At that time, a majority of the recipients of most such assistance programs were minorities, even though minorities comprised only 14.3% of college students. Id. at 878 n.20 (1984) (Marshall, J., dissenting).

  1. Wisconsin v. Yoder, 406 U.S. 205, 221 (1972).
  2. Young white men who lack a high school diploma are more than five times as likely to

be incarcerated than those who graduated. Anne Morrison Piehl, Economic Issues in Crime Policy 68 (1994) (unpublished Ph.D. dissertation, Princeton University) (on file with authors). Piehl examined the higher conviction rates of less educated adolescents to determine whether they were committing more crimes or simply more likely to get caught. Id. She found that “more schooling is associated with lower probabilities of committing illegal activities . . . [as well as] lower conviction rates for those people involved in crime. . . . [T]here is potential for affecting crime rates through education programs, either broadbased efforts or targeted to a criminally active population.” Id. at 70, 96; see also YOUTH IN THE CRIMINAL JUSTICE SYSTEM, supra note 73, at 5 (“[I]n 1996, 45.5% of all jail inmates in the United States had less than a high school education.”).

87. The correlation between lack of education and criminal behavior among men “is consistent with an economic model of crime in which education is associated with better legal sector opportunities.” Piehl, supra note 86, at 81.

88. The Harvard Report states that the rigidity and harshness of zero tolerance policies “further alienate students from school and exacerbate the behaviors they seek to remedy.” HARVARD REPORT, supra note 20, at 3, 9. “This damage is particularly acute for children who are already considered ‘at risk’ for school failure and often has the effect of pushing them out of school completely.” Id. Another study shows that “schools that rely heavily on zero tolerance policies continue to be less safe than schools that implement fewer components of zero tolerance. [R]esearch has suggested that misuse of school security measures such as locker or strip searches can create an emotional backlash in students.” R.J. Skiba & R.L. Peterson, School Discipline at the Crossroads, 66 EXCEPTIONAL CHILD. 335, 337 (2000).

It is possible that high school zero tolerance policies for minor infractions replicate the


How to Construct an Underclass 79

personnel may simply be dumping problem students out on the streets, only to find them later causing increased violence and disruption in the community . . . [W]e face serious questions about the long-term negative effects of one of the cornerstones of zero tolerance, school exclusion.”89 As for prisoners, numerous studies show that prison education programs reduce recidivism rates,90 in some cases by a factor of four.91 A Rand study

criminogenic effects that have been found to flow from “invoking the criminal justice system for relatively minor behavior [which may] increase rather than reduce crime through its effect on the life prospects or psychology of the arrested individual.” Philip B. Heymann, The New Policing, 28 FORDHAM URB. L.J. 407, 418 (2000).

89. Skiba & Peterson, supra note 25, at 376.

90. Among the studies reporting this finding are MILES D. HARER, FED. BUREAU OF PRISONS, PRISON EDUCATION PROGRAM PARTICIPATION AND RECIDIVISM: A TEST OF THE NORMALIZATION HYPOTHESIS (1995) (reporting that inmates who actively participate in education programs have significantly lower likelihood of recidivating), available at; DAVID CLARK, N.Y. DEP’T OF CORR. SERV., ANALYSIS OF RETURN RATES OF THE INMATE COLLEGE PROGRAM PARTICIPANTS (Aug. 1991); MILES D. HARER, FED. BUREAU OF PRISONS, RECIDIVISM AMONG FEDERAL PRISON RELEASES IN 1987: A PRELIMINARY REPORT 4 (1994) (reporting that the more educational programs successfully completed for each six months confined, the lower the recidivism rate); Mary Ellen Batiuk, The State of Post-Secondary Education in Ohio, 48 J. CORR. EDUC. 70 (1997) (comparing an overall recidivism rate that was forty percent for Ohio inmates with an eighteen percent rate for those who completed the Associate Degree program); Mary Ellen Batiuk et al., Crime and Rehabilitation: Correctional Education as an Agent of Change-A Research Note, 14 JUST. Q. (1997) (concluding that college education increases the likelihood of post-release employment, which in turn reduces the risk of recidivism); Piehl, supra note 86, at 83. Cf. M. S. BRUNNER, OFFICE OF JUVENILE JUST. & DELINQUENCY PREVENTION, REDUCED RECIDIVISM AND INCREASED EMPLOYMENT OPPORTUNITY THROUGH RESEARCH-BASED READING INSTRUCTION 1, 6 (NCJ Publication No. 141324, 1993) (reporting twenty percent reduction in juvenile recidivism through reading instruction programs). These studies demonstrate that prison education programs are far more efficacious than researchers thought three decades ago, when a seminal survey of mid-century studies found little impact. See Robert Martinson, What Works? Questions and Answers About Prison Reform, PUB. INT., Spring 1974, at 22, 24.

91. A City University of New York research study found that 7.7% of women inmates at New York State’s Bedford Hills Correctional Facility who took prison education courses returned to jail, compared to 29.9% of those who did not. Michelle Fine et al., The Impact of College in a Maximum Security Prison (2001), available at Another study of Texas inmates found that prison college program graduates had less than a 15% re- arrest rate, compared to a general Texas re-arrest rate of 60%. C. Tracy & C. Johnson, Review of Various Outcome Studies Relating Prison Education to Reduced Recidivism, Windham School System: Huntsville, TX 7 (1994). A similar finding nationally is reported in Alexandra Marks, One Inmate’s Push to Restore Education Funds for Prisoners, CHRISTIAN SCI. MONITOR, Mar. 20, 1997 at 3, available at 1997 WL 2799990; see also J. Chase & R. Dickover, University Education at Folsom Prison: An Evaluation, 34 J. CORR. EDUC. 3, 92-96 (1983) (reporting recidivism rates within three years of release of 55% for the state’s general prison population and 0% for those who had completed a baccalaureate degree in prison); Dennis J. Stevens & Charles S. Ward, College Education and Recidivism: Educating Criminals Meritorious, 48 J. CORR. EDUC. No. 3 106-111 (1997) (comparing recidivism rates of inmates generally with inmates who received prison college degrees in Alabama, 35% and 1% respectively; Maryland, 46% and 0%; New York, 45% and 26%; and Texas, 36% and 10%); Jon M. Taylor, Post Secondary Correctional Education: An Evaluation of Effectiveness and Efficiency, 43 J. CORR. EDUC. 132 (1992) (reporting one study of a Canadian prison college program that produced a recidivism rate of 14% compared to 52% of the matched group of non-student prisoners, and another study showing that inmates at the New Mexico State Penitentiary who took college courses had an average recidivism rate of 15.5% compared to 68% of the entire inmate population).


80 The Journal of Gender, Race & Justice [6:2002]

concluded that education is the most cost-effective crime prevention program available, and other studies confirm that investment in prisoner education more than pays for itself .92 The exploding incarceration rate and the termination of prison education programs were intended to “get tough” on criminals. But given the consequencesña multiplying recidivism rate93 at a time when an unprecedented 600,000 prisoners are returning to society per year94ñthese policies are proving particularly tough on America.95

James Gilligan, a psychiatrist who directed the Massachusetts Prison Mental Health Service, reports that:

the most successful of all [anti-recidivism programs], and the only one that had been 100 percent effective in preventing recidivism, was the program that allowed inmates to receive a college degree while in prison. Several hundred prisoners in Massachusetts had completed at least a bachelor’s degree while in prison over a 25-year period, and not one of them had been returned to prison for a new crime.

James Gilligan, Reflections from a Life Behind Bars: Build Colleges, Not Prisons, CHRON. HIGHER EDUC., Oct. 16, 1998, at B7.

92. PETER W. GREENWOOD ET AL., DIVERTING CHILDREN FROM A LIFE OF CRIME: MEASURING COSTS AND BENEFITS 37-41 (1996) (citing the Rand study). Another study conducted by the Correctional Education Association for the U.S. Department of Education Office of Correctional Education studied more than 3,600 inmates who had been released for at least three years in Maryland, Minnesota, and Ohio. It found that “simply attending school behind bars reduces the likelihood of re-incarceration by twenty-three percent. Translated into savings, every dollar spent on education returns more than two dollars to the citizens in reduced prison costs.” STEPHEN STEURER ET AL., THE THREE STATE RECIDIVISM STUDY (2001), available at A third study examined the return on public investment in vocational and educational programs in Florida prisons, concluding that those who completed certain educational programs produced a return of $3.53 per $1.00 invested. FLA. DEP’T CORR., RETURN ON INVESTMENT FOR CORRECTIONAL EDUCATION IN FLORIDA (1999), available at

93. See Butterfield, Getting Out, supra note 71, at A1 (“In 1977, only 788 inmates who had been released on parole were returned to prison in California, compared with 90,000 in 1999.”); Outside the Gates, TIME, Jan. 13, 2002, at 57, 58 (“The more people you lock up, the more you must one day let out. For 40% of those now in state prisons, that day arrives in the next 12 months. . . . The result is that each day this year, an average of 1,726 men and womenñmostly menñwill walk out of penal institutions having spent more time behind bars, with less preparation for their return to society and slimmer chances of success there, than those who came before them.”); Petersilia, supra note 10 at 1 (“Fully two-thirds of all parolees are rearrested within 3 years. In 1980, they constituted 17 percent of all admissions, but they now make up 35 percent.”).

  1. Petersilia, supra note 10, at 1.
  2. Jeremy Travis, an expert on the subject of prisoner-reentry, believes that our almost

exclusive focus on incarceration and punishment not only produces prisoner recidivism but also threatens community dissolution in a variety of ways:

[T]he fact that we have increased fourfold the rate of removal, imprisonment, and reentry over the past generation has significantly weakened the capacity of communities to do the work that communities should do: raise children, provide a healthy environment for families, provide jobs for young and old, and sustain a vibrant civic life. . . . [I]s it possible that a weakening of community capacity through a policy of mass incarceration will actually result in higher crime rates? Two researchers at the City University of New York, Dina Rose and Todd Clear, are testing this proposition and have found, in one study, that crime rates went up after a certain tipping point was reached.


How to Construct an Underclass 81

In other ways too numerous and too obvious to detail here, the creation and continuation of an alienated, uneducated underclass inflicts deep wounds on our national interests. It is sufficient to restate the Supreme Court’s query in its opinion striking down the exclusion of undocumented aliens from public school: “It is difficult to understand precisely what the state hopes to achieve by promoting the creation and perpetuation of a subclass of illiterates within our boundaries, surely adding to the problems and costs of unemployment, welfare and crime.”96

Unhappily, the educational privation laws we have surveyed are but one element in a comprehensive legal onslaught against ex-offenders. Federal and state laws hobble rehabilitation and reintegration in countless other ways as well.97 For example, if you were once convicted of selling a vial of crack, long after you complete your sentence you will continue to bear what the law calls “civil disabilities” (and what prisoners call “the mark of Cain”)98 that may put many of the material and psychological essentials of life beyond reach.99 You will be ineligible for food stamps, federal health care benefits, and other federal assistance. 100 In some states, you will be

Jeremy Travis, Prisoner Reentry Seen Through a Community Lens, Address at the Neighborhood Reinvestment Corporation Training Institute 2 (Aug. 23, 2001), available at [hereinafter Prisoner Re-entry].

  1. Plyler v. Doe, 457 U.S. 202, 230 (1982).

CONVICTED FELONS: A STATE-BY-STATE SURVEY app. B (1996) (listing each state’s civil disability laws).

  1. Webb Hubbell, The Mark of Cain, S.F. CHRON., June 10, 2001, at D1.
  2. For descriptions, analyses, and critiques of the burgeoning “civil disabilities” imposed on

offenders even after completion of their sentence, see JEREMY TRAVIS, COLLATERAL DAMAGE: THE SOCIAL COST OF MASS INCARCERATION (Mayeda Chesney-Lind & Marc Mauer eds., 2001) [hereinafter TRAVIS]; Gabriel Chin, Effective Assistance of Counsel and the Consequences of Guilty Pleas, 87 CORNELL L. REV. 697 (2002); Nora Demleitner, Preventing Internal Exile: The Need for Restrictions on Collateral Sentencing Consequences, 11 STAN. L. & POL’Y REV. 153 (1999); Kathleen M. Olivares et al., The Collateral Consequences of a Felony Conviction: A National Study of State Legal Codes 10 Years Later, 60 FED. PROBATION 10 (1996).

100. Federal law imposes a lifetime ban on food stamps and federally funded public assistance for drug felons unless a state elects otherwise. 21 U.S.C. ß 862a (2002). A recent study reports that 22 states now impose the ban in full and 20 others in part, placing an estimated 92,000 women and 135,000 children at risk in the affected states. PATRICIA ALLARD, THE SENTENCING PROJECT, LIFE SENTENCES: DENYING WELFARE BENEFITS TO WOMEN CONVICTED OF DRUG OFFENSES (2002), available at Federal law also imposes a mandatory ineligibility for federal health care benefits for those convicted of distribution offenses. 42 U.S.C. ß 1320a-27a (1999). Grants, licenses, contracts, and some other federal benefits are restricted as to drug offenders under 21 U.S.C. ß 862. Section (a) provides that, in the discretion of the court, individuals convicted of a first federal or state drug trafficking offense may be rendered ineligible for all federal benefits for up to five years after conviction, and second offenders for up to ten years; third offenders are permanently ineligible as a mandatory sanction. Id. Under section (b), in the discretion of the court, individuals convicted of a first federal or state drug possession offense may be rendered ineligible for all federal benefits for up to one year, and second offenders for up to five years; and third offenders are mandatorily ineligible permanently. Id. Section (b) sanctions may be waived if a person declares himself to be an addict and undergoes treatment or is declared rehabilitated. Id.


82 The Journal of Gender, Race & Justice [6:2002]

prohibited from driving101 or voting;102 in some, you will be unable to watch your child’s soccer game because school grounds are off-limits to ex- felons.103 Finding a job, never easy for ex-offenders, will be that much harder thanks to a welter of state and federal laws that foreclose numerous occupations104 as well as military service105 and public office.106 Under some circumstances you will be effectively ineligible for public housing. 107 If you are an alien, you are likely to be deported.108 In short, you will be

101. See 23 U.S.C. ß 159 (1998) (denying ten percent of federal highway funds to states which do not suspend offender’s licenses of drug felons).

102. Thirty-two states bar voting by offenders on parole or probation, and at least twelve states bar an ex-felon from voting for life. Lifetime voting bans in Alabama, Florida, Iowa, Mississippi, New Mexico, Virginia and Wyoming have resulted in the permanent disenfranchisement of one quarter of their black male citizens. Prisoner Re-entry, supra note 95, at 5. Currently four million Americans are barred from voting, of whom approximately one third are African Americans (thirteen percent of the adult black male population). Id.

103. Hubbell, supra note 98, at D1 (“[S]ome states, including California, have passed laws banning ex-felons from school grounds.”).

104. See, e.g., 18 U.S.C. ßß 3563(b)(6), 3583(d), 5F1.5(a) (authorizing the sentencing court to place occupational restrictions as a condition of probation); 21 U.S.C. ß 862(d), ß 5F1.6 (limitations on federal licenses to drug offenders); 29 U.S.C. ßß 504, 1111 (ineligibility from listed positions in labor unions or employee benefit plans). In California, parolees are barred from working in law, real estate, medicine, nursing, physical therapy, and education; in Colorado, from working as dentists, engineers, nurses, doctors, pharmacists, or real estate agents. Petersilia, supra note 10. Most jobs requiring federal or state licenses are off limits. Former Deputy Attorney General and former prisoner Webb Hubbell offers the following as a partial list of jobs unavailable to him as an ex-felon:

It is highly unlikely that I could ever obtain a custom broker’s license, an export license, a merchant marine license, or even a locomotive engineer’s license. . . . I cannot become a director, officer, employee or controlling stockholder of any federally insured institution such as a bank or a savings and loan. I cannot be an adviser, officer or director of a labor organization for 13 years after my conviction. . . . The Securities and Exchange Commission prohibits me from becoming an investment adviser for 10 years. The secretary of Health and Human Services can bar me from working in any aspect of health care if federal, state or local dollars are involved. I couldn’t even sweep the floors of a nursing home. . . . Any job that requires a state license is probably out. . . . Other occupations most likely closed to me include: certified public accountant, physician, dentist, insurance agent, nurse, real estate broker, pharmacist, landscape architect, law enforcement officer, teacher, day-care worker, veterinarian, bartender, dietitian, engineer, barber, cosmetologist, mortician, speech pathologist, social worker. . . .

Hubbell, supra note 98, at D1.

  1. 10 U.S.C. ß 504 (2001).
  2. Twenty-five states restrict felons from public office. Olivares et al., supra note 99, at


107. 42 U.S.C. ß 13661(a) (1998) provides that if you were previously evicted from federally-assisted housing by reason of drug related criminal activity, you are ineligible for admission to any federally-assisted housing for three years. Under a separate provision you may be evicted from public housing for the drug violation of a guest of family member, even if unknown to you. Dep’t of Housing & Urban Dev. v. Rucker, 122 S. Ct. 1230 (2002) (interpreting 42 U.S.C. ß 1437d(l)(5) (1994), redesignated in 1998 as d(l)(6)).

108. ß 237(a)(2)(B); 8 U.S.C. ß 1227(a)(2)(B) (1999). Deportation of criminal offenders rose from 7,338 in 1989 to 56,011 in 1998. TRAVIS, supra note 99, at 11 (citing Peter H. Schuck & John


How to Construct an Underclass 83

branded as a kind of untouchableóa caste status that any layman (but not the law) would recognize as an extra layer of punishment. This is the humiliating and overwhelming plight that confronts the 13 million Americans with felony convictions, and to a lesser degree the 47 million Americans with some kind of criminal history on file with state or federal criminal justice agencies.109 Ex-offenders re-entering society with $200, the shirt on their back, and virtually no socially recognized identity, scarcely have a chance without some help in obtaining housing, a job, educational opportunities, medical care, and counseling.110 What is needed, for their sake and ours, are reintegration programs, not exclusionary policies that promote desperation and failure.


Each year that educational privation sanctions remain in place does irreparable damage to our most vulnerable citizens and to ourselves as a nation. We believe that there are reasonable prospects for reform, however, particularly as the relatively sorry state of American education is now near the top of the public agenda. Given the dire consequences of consigning many thousands of children and adult offenders to an uneducated underclass, we have reason to hope that politicians, educators, policy experts and courts will sooner or later devise more effective and humane sanctions than educational privation. We first survey the prospects for political reform and then turn to some possible court challenges.

A. Political Prospects

Reversing high school zero tolerance exclusions is a formidable undertaking, not least because school administrators will be reluctant to abandon a policy that efficiently rids the classroom of troublemakers, and also removes students who are most likely to score poorly on the standardized tests increasingly used to evaluate their schools.111

Williams, Removing Criminal Aliens: The Pitfalls and Promises of Federalism, 22 HARV. J. L. & PUB. POL’Y 367, 384-85 (1999).

  1. TRAVIS, supra note 99, at 5
  2. The N.Y. Times reports the story of one man that no doubt reflects the experience of

hundreds of thousands of other emerging prisoners. Butterfield, Getting Out, supra note 71, at A1. Steve Butler, forty-four, was “released from prison after serving a one-year sentence for possession of cocaine. Records show he was given his $200 and a bus ticket back to Los Angeles, where he had been arrested. But Mr. Butler was homeless at the time of his arrest, with no family here, so the first night after getting off the bus, he said, he went back to sleeping on the same skid-row street just east of downtown where he had lived before. With no education, job skills or hope, he said, he used some of his money to buy dope to make himself feel better.” Id.

111. The American Federation of Teachers, the nation’s largest teachers’ union, has been one of the most enthusiastic supporters of mandatory expulsions of students who bring weapons or drugs to school. See Wasser, supra note 24, at 749-50. For more on the national debate among school administrators and critics on the merits of zero tolerance, see Johnson, supra note 17 at A20; Paul


84 The Journal of Gender, Race & Justice [6:2002]

Nevertheless, there are increasing signs that the zero tolerance approach is vulnerable. In Georgia, after dozens of students were suspended for possessing innocuous items deemed “weapons” (such as a Tweety Bird wallet with a long keychain attached, and a broken axe found during a random search of students’ cars), Republican and Democratic leaders in the legislature introduced a bill to make suspensions and expulsions discretionary. A January 2002 newspaper poll showed that ninety-six percent of Georgians agree.112 The American Bar Association (ABA) has adopted a similar position, voting in February 2001 to oppose school zero tolerance policies that fail to take into account the circumstances or nature of the offense or the accused’s age and history.113 Moreover, the first comprehensive national study of public school zero tolerance policies strongly recommended that they be discontinued. The Harvard University study found that zero tolerance produces “devastating consequences,” including educational privation, teacher-student confrontation and distrust, and unjustifiable criminalization of children. “Policymakers, educators and parents should be very concerned with the long-term implications of denying educational opportunities to millions of children, particularly when the effectiveness of these policies in ensuring school safety is highly suspect.”114

The denial of college loans to drug offenders has also generated opposition, particularly from students and educators on college campuses. An organization attempting to repeal this law, Students for a Sensible Drug Policy, has established 130 chapters on college campuses, and more at high schools throughout the country.115 Sixty-six student governments, the

H.B. Shin, Suspensions Go Through the Roof; Schools Tossed Out Record 26,820 Kids, N.Y. DAILY NEWS Nov. 5, 2000, at 6.

112. Fox News, supra note 27. Only four percent wanted zero tolerance sanctions to remain mandatory or become stricter. The report states that co-sponsor Sen. Richard Marable expects Senate Bill 335 to “go through the process as fast . . . as any can. . . .” Id.

113. In its report to the A.B.A.’s House of Delegates, the Criminal Justice Section strongly opposed zero tolerance measures applied to schools. “It has redefined students as criminals, with unfortunate consequences.” A.B.A., FINAL REPORT, BI-PARTISAN WORKING GROUP ON YOUTH VIOLENCE 1 (2000), available at

114. HARVARD REPORT, supra note 20, at 1, 3 (2000), available at zt.report2.html; see also Skiba & Peterson, supra note 25, at 376. Skiba finds that without providing services to suspended or expelled students

school personnel may simply be dumping problem students out on the streets, only to find them later causing increased violence and disruption in the community. In sum, we lack solid evidence to support the effectiveness of harsh policies in improving school safety, and we face serious questions about the long-term negative effects of one of the cornerstones of zero tolerance, school exclusion.


115. The figure is as of January 28, 2002. The chapters are listed on the SSDP website, which is available at /chapters. Another opposition group is the Coalition for HEA Reform, the web site of which is


How to Construct an Underclass 85

American Council on Education, and the National Association of Student Financial Aid Administrators have urged repeal,116 along with such civil rights organizations such as the NAACP, ACLU and NOW,117 and even DEA Administrator Asa Hutchinson.118 Some colleges and organizations are providing substitute loans to students affected by the law. 119

In Congress, Rep. Barney Frank, D-Mass., has introduced HR 786, a bill to repeal the law that has attracted fifty co-sponsors to date.120 Separately, a subcommittee of the House Education and Work Force

116. STUDENTS FOR A SENSIBLE DRUG POLICY, NATIONAL AGENDA (2001), available at (discussing support from student organizations); 43,000 Students, supra note 45, at A11 (reporting support from the American Council of Education). The president of Hampshire College, Gregory S. Prince, Jr., felt compelled to state the obvious when he asked, “[W]hy would you want to exclude people from the education stream when trying to keep them in the stream is the most important thing to do?” Ariana Huffington, Souder Says Drug-Use Rule Never Meant as Punishment, SO. BEND TRIB., Apr. 29, 2001, at B8.

117. Additionally, the A.B.A. may soon draw attention to the defects that the Souder amendment shares with other collateral consequences of criminal convictions. Its Task Force on Collateral Consequences is about to complete a proposed revision of the A.B.A. standards on the imposition of collateral consequences. CIVIL DISABILITIES OF CONVICTED PERSONS, 4 A.B.A. STANDARDS FOR CRIMINAL JUSTICE Chap. 23 (1986). It appears the new standards will “conceive collateral consequences as essentially a form of discrimination against convicted persons. . . .” Margaret Colgate Love, Deconstructing the New Infamy, 16 CRIM. JUST. 30, 31 (2001). In both its previously published standards and in those that are forthcoming, the A.B.A. strongly recommends that to ensure truth in sentencing any consequences to criminal convictions be codified and considered by the sentencing judge. Id. The standards oppose the mandatory approach to sanctions and oppose any retroactive imposition. Id.

118. After his nomination to be DEA Administrator, Hutchinson said he favored

allowing convicted drug offenders to remain eligible for federal student loans. Hutchinson said many offenders who left prison years ago are now finding that they cannot get financial aid “even though they’ve turned their lives around. And what you’re doing is punishing those people,” Hutchinson said. “There is some unfairness in that.” Such financial aid, he said, is an important component in letting drug offenders “get back to leading useful, productive lives.”

Josh Meyer, New DEA Chief Suggests “Compassionate” Policy, L.A. TIMES, Aug. 2, 2001, available at

119. Hampshire, Swarthmore, Yale, and Western Washington Universities have established funds to lend money to students whose financial aid has been cut off due to this law. Yilo Zhao, Yale Takes a Stand with Policy on Drugs and Financial Aid, N.Y. TIMES, Apr. 13, 2002, at A27. Additionally, in March 2002, activists launched the John W. Perry Fund, an organization that will raise money to provide scholarships to students denied aid under the Souder provision. See The John W. Perry Fund, at (last visited Apr. 22, 2002).

120. Last year, Congressman Frank’s bill (then HR 1053) was co-sponsored by fifty members of Congress and had the support of such groups as the ACLU, NAACP, NOW, the American Public Health Organization, and the Coalition for HEA Reform. Cannon, supra note 44, at 2253. Frank argued that:

these low-income students are, in effect, being thrown out of school for doing what George Bush and Al Gore have done. Now, people might not be enamored with either Bush or Gore, but I don’t think anybody would say that America was disserved by them completing their college education.

Huffington, supra note 47, at B8.


86 The Journal of Gender, Race & Justice [6:2002]

Committee surveyed post-secondary educators to create an agenda for regulatory relief, and as a result now includes possible repeal of the Souder amendment as issue 49 on its financial aid agenda.121 Even Congressman Souder, the primary sponsor of the law, is seeking to reduce its scope. He says he never intended the law to be used to punish people for convictions incurred prior to the student’s receipt of federal funds, and last year offered an amendment to limit the law’s application to drug offenses that occur while the offender is receiving federal college assistance. After winning a House vote, Souder’s revision died in the Senate.122 With all sides pledging to continue reform efforts, some degree of change in the law is likely.

Advocates for prison education press for the restoration of prison educational programs, but so far with little success. In 2000 Rep. Bobby Scott, D-Va., attempted to reinstate Pell Grants for prisoners, but his bill was defeated amidst concerns about the rising cost of college education, which is increasing the Pell Grant budget for non-prisoners. 123 At the state level, concern over the release of unprecedented numbers of unprepared prisoners has led some legislators to favor rehabilitation and reentry programs, including the reinstatement of some prison education programs.124 Meanwhile, prison education advocates are researching and publicizing alternative sources of funding that might permit some inmates to obtain secondary and post-secondary degrees while in prison.125

121. COMM. ON EDUC. & THE WORKFORCE, FED UP INITIATIVE: REGULATORY RELIEF PROPOSALS (2001), available at The agenda describes the rationale for considering repeal, as suggested by the survey, as follows: “There are regulations and statutory requirements that have no bearing on a student’s propensity for educational success. These regulations add complexity to the student aid application and process without providing any significant social benefit.” Id.

  1. Cannon, supra note 44, at 2253.
  2. The bill was introduced in the House of Representatives of the 106th Congress on Nov.

14, 2000, as H.R. 5632, 106th Cong. (1999).

124. See Peter Schmidt, College Programs for Prisoners, Long Neglected, Win New Support, CHRON. HIGHER EDUC., Feb. 8, 2002, at A26. Schmidt reports that Vermont is likely to institute new prison education programs to facilitate prisoner re-entry, and notes that prison educational opportunities could expand through distance education utilizing technologies such as the internet and interactive television. Id. “Even as many such programs teeter on the brink of oblivion, sentiment seems to be growing among criminal-justice experts and some state lawmakers that they provide benefits that make them worth keeping.” Id.

125. Jon Marc Taylor, a Missouri prisoner, writer, and long-time campaigner for prison education, provides a painstaking roadmap to alternative funding in a three part series, Jon Marc Taylor, Piecing Together a College Education Behind Bars Part I, 2 CELL DOOR Issue 4 (2000), available at; Jon Marc Taylor, Piecing Together a College Education Behind Bars Part II, 3 CELL DOOR Issue 1 (2001), available at; Jon Marc Taylor, Piecing Together a College Education Behind Bars Part III, 3 CELL DOOR Issue 2 (2001), available at For additional information of prisoner education options, see EUR. PRISON EDUC. ASS’N, available at (last visited May 7, 2002); OFFICE CORR. EDUC., available at (last visited May 7, 2002); CORR. EDUC. ASS’N, available at (last visited May 7, 2002); Books Through Bars, (last visited May 7, 2002) (an organization that


How to Construct an Underclass 87 B. Court Challenges

Opponents of educational privation laws should also challenge them in court. For public school students who have been removed from school, there are a number of legal grounds that might be invoked against the denial of their right to a public educationóa right that is guaranteed by all state constitutions and one which could yet be recognized, to at least to a limited degree, in the federal constitution.126 These potential legal claims transcend the drug war that is the subject of this symposiumóthey seek to reclaim a public education for all children, including those expelled for non-drug offensesóand we shall devote a subsequent article solely to these constitutional theories.

Aspiring college students should also consider litigation, but they will have less to work with than adolescents expelled from high school: whatever constitutional right to education exists is unlikely to apply to advanced, post- secondary education.127 Moreover, suits seeking to restore Pell Grants and other educational opportunities to prisoners have been unsuccessful to date.128 However, we believe the Souder Amendment’s denial of financial

supplies books to prisoners).

126. The Supreme Court has repeatedly and explicitly left unresolved the question whether the Constitution affords a limited right to a “minimally adequate education.” See San Antonio Indep. Sch. Dist. v. Rodriguez, 411 U.S. 1, 36-37 (1973); Papasan v. Allain, 478 U.S. 265, 284-86 (1986).

127. But see Selective Serv. Sys. v. Minn. Pub. Interest Research Group, 468 U.S. 841, 879 n.22 (1984) (Marshall, J., dissenting).

[W]here our prior cases have focused particularly on the extraordinary importance to the individual of elementary and secondary education, our concern that burdening access to education creates permanent class distinctions and political disadvantage is equally relevant here. Postsecondary education is the necessary prerequisite to pursuit of countless vocations, both professional and technical.


128. In Tremblay v. Riley, 917 F. Supp. 195 (W.D. N.Y. 1996), the court found the law denying Pell grants to prisoners constitutional, rejecting claims that it violated (1) the ex post facto prohibition; (2) procedural due process, because any property rights in the Pell grants were extinguished by the law; (3) the Eighth Amendment bar on cruel and unusual punishment, finding that the law does not impose punishment; and (4) equal protection and substantive due process finding the law rationally related to the government’s interests in allocating scarce resources, increasing funding for law-abiding students, and eliminating fraud in the administration of Pell grants to prisoners. See also Johnson v. Galli, 596 F. Supp. 135, 139 (D. Nev. 1984) (declaring prisoners do not have a constitutional right to rehabilitation or vocational programs); Nicholas v. Riley, 874 F. Supp. 10, 11-12 (D.C. Cir. 1995) (“Plaintiff’s arguments regarding potential correlations between the provision of educational resources and reduced recidivism, while not without arguable merit, are beyond the ken of the Court’s inquiry.”); Garza v. Miller, 688 F.2d 480, 486 (7th Cir. 1982) (“[T]here is no constitutional mandate to provide educational, rehabilitative, or vocational programs . . . [to prisoners].”); Russell v. Oliver, 392 F. Supp. 470, 474 (W.D. Va. 1975) (stating that there is “[n]o federal constitutional right to vocational training . . . for inmates in a correctional system”). Courts have also rejected prisoners’ suits claiming a right to rehabilitation. Marshall v. United States, 414 U.S. 417, 421 (1974) (drug rehabilitation is not a fundamental right); French v. Heyne, 547 F.2d 994, 1002 (7th Cir. 1976) (“[P]risoners possess no abstract right to rehabilitation and . . . the failure to provide rehabilitative programs does not constitute cruel and unusual punishment. . . .”); Landman v. Royster, 333 F. Supp. 621, 644 (E.D. Va. 1971) (“Even now


88 The Journal of Gender, Race & Justice [6:2002] aid to drug offenders remains untested and may well be vulnerable to the

following constitutional and statutory approaches.
1. Substantive Due Process and Equal Protection Violations

With neither a fundamental constitutional right nor a suspect classification implicated by the laws denying college aid to offenders,129 equal protection or substantive due process challenges to these laws will confront the least favorable standard of judicial review: “rational basis” review. The rational basis standard affords the law a “strong presumption of validity,”130 a presumption that can be overcome only by a showing either that (1) the law does not have a legitimate governmental purpose, or (2) there is no rational relation between that purpose and the means the law utilizes to achieve it.131 This test affords great deference to the legislature (as is evident when compared to the more exacting review applied to laws that do affect a fundamental right or suspect classó”strict scrutiny,” requiring that the governmental purpose be “compelling” and the means “narrowly tailored” to achieve it without unnecessary burdens on the suspect class or the constitutional right.)132 Yet even the deferential rational basis test is not toothless,133 and perhaps is discerning enough to recognize the blindness and futility involved in denying educational access to people at risk who are trying to turn their lives around.

What, then, is the purpose of the Souder amendment? The only purpose that the law seems to achieve is the imposition of further punishment on

no court has required that states adapt their penal system to the goal of rehabilitation.”).

129. A statute will be subject to “strict scrutiny” only if it (1) impinges on a fundamental right or (2) discriminates against a “suspect class”óa class defined on the basis of race, national origin, alienage, or will trigger heightened but only “mid-level” scrutiny, if the class is defined by gender or illegitimacy. See, e.g., Lyng v. Int’l Union, 485 U.S. 360, 370 (1988); Plyler v. Doe, 457 U.S. 202, 216-17 (1982); Rodriguez, 411 U.S. at 16-17. Although denying college loans to offenders affects only financially needy students, the Supreme Court has decided that classifications either based on wealth, or having different effects on the wealthy and the poor, are not “suspect classifications” triggering strict scrutiny. Kadrmas v. Dickenson Pub. Sch., 487 U.S. 450, 458 (1988); Harris v. McRae, 448 U.S. 297, 322-23 (1980); Maher v. Roe, 432 U.S. 464, 470-71 (1977).

  1. FCC v. Beach Communications Inc., 508 U.S. 307, 313, 315 (1993).
  2. A rational basis challenge must “‘negative every conceivable basis which might support

it,’ . . . whether or not the basis has a foundation in the record.” Heller v. Doe, 509 U.S. 312, 320-21 (1993). But the relationship of the classification to its goal cannot be so “attenuated as to render the distinction arbitrary or irrational.” City of Cleburne v. Cleburne Living Ctr., 473 U.S. 432, 447 (1985); see also Vacco v. Quill, 521 U.S. 793, 807 (1997); Romer v. Evans, 517 U.S. 620 (1996); Hodel v. Indiana, 452 U.S. 314, 331-32 (1981); Holt Civic Club v. Tuscaloosa, 439 U.S. 60, 71 (1978); McGowan v. Maryland, 366 U.S. 420 (1961).

  1. Wygant v. Jackson Bd. of Educ., 476 U.S. 267, 280 n.6 (1986).
  2. Mathews v. Lucas, 427 U.S. 495, 510 (1976); see also Romer, 517 U.S. at 632 (noting

that “even in the ordinary equal protection case calling for the most deferential of standards, we insist on knowing the relation between the classification adopted and the object to be attained. The search for the link between classification and objective gives substance to the Equal Protection Clause”).


How to Construct an Underclass 89

drug offenders, yet this is the one rationale that cannot be asserted in defense of the law. If the denial of loans is conceded to be punishment, then it is an unconstitutional ex post facto punishment as applied against anyone whose offense occurred prior to its passage, and arguably multiple punishment in violation of the double jeopardy bar as well.134 As punishment, ineligibility would have to be issued by a sentencing judge,135 and also noted as part of all guilty plea colloquys.136 Only by maintaining the fiction that the denial of loans is a non-punitive “collateral consequence” or “civil disability” can the Souder amendment avoid invalidation on these grounds.

In legislative debate and commentary, one finds an array of other rationales for the Souder amendment, all so poorly served by this law that they are hard to take seriously. Congressman Goodling argued for the law as a way to “ensure safety on our nation’s campuses [by keeping] them drug- free.”137 There are, however, three problems with this rationale. First, if campus safety is the goal, is there any rational basis for believing that a person once convicted of marijuana possession poses more risks than convicted murderers, arsonists, rapists, and other violent criminals untouched by this law? Or than those previously convicted of driving while intoxicated, who continue to be eligible because the law specifically exempts alcohol-related convictions138? Second, if the goal is to remove either drugs or drug offenders from campus, the Souder amendment by its terms does neither. Drug offenders can go to college so long as they can afford the

134. U.S. CONST. art. I, ß 9, cl. 3 (ex post facto prohibition); U.S. CONST. amend. V (double jeopardy); Hudson v. United States, 522 U.S. 93, 99 (1997) (“The Clause protects only against the imposition of multiple criminal punishments for the same offense… when such occurs in successive proceedings.”); Kansas v. Hendricks, 521 U.S. 346 (1997) (ex post facto bar only applies to laws imposing criminal punishments).

135. Punishment issued by Congress rather than the judiciary constitutes a bill of attainder in violation of U.S. CONST. art. I, ß 9, cl. 3. Nixon v. Adm’r of General Serv., 433 U.S. 425, 468, 472 (1977); United States v. Brown, 381 U.S. 437, 442 (1965) (finding that a bill of attainder bar is designed to safeguard “against legislative exercise of the judicial function, or more simplyótrial by legislature”). In Selective Serv. Sys. v. Minn. Pub. Interest Research Group, 468 U.S. 841, 847 n.3 (1984), the Supreme Court found that denial of federal aid for higher education to those who had not registered for military service was not an unconstitutional bill of attainder, in part because it found the denial not to be punishment.

136. Durant v. United States, 410 F.2d 689, 692 (1st Cir. 1969) (collateral consequences are civil and, therefore, need not be part of the plea colloquy).

137. 144 CONG. REC. H2510, 2516 (daily ed. Apr. 29, 1998); see also Josh Gerstein, Lost Loans, Lost Futures?, Students Lose Aid Because of Past Drug Offenses, ABC NEWS, Apr. 28, 2001, available at

138. Illegal drugs are controlled substances as defined by the Controlled Substances Act, 21 U.S.C. 801(6), ß102(6)(2002), and do not include alcohol or tobacco. See also Dep’t of Educ. Regs., 34 C.F.R. ß 668.4 ß 3 (2002). According to a report issued in February 2002 by the National Center on Addiction and Substance Abuse at Columbia University, eighty-one percent of high school students have used alcohol, compared to forty-seven percent who have used marijuana, and almost thirty-three percent of high school students say they binge drink at least once a month. According to Susan Foster, the center’s vice president and director of policy research and analysis, “Alcohol is far and away the top drug of abuse for American kids.” Underage Drinking in U.S. Is on the Rise, N.Y. TIMES, Feb. 26, 2002.


90 The Journal of Gender, Race & Justice [6:2002]

tuition. Finally, as to those less affluent ex-drug offenders who are excluded, this rationale conclusively presumes that they all present a safety risk. It is worth remembering that of all drug arrests, roughly one-third are for nothing more than possession of marijuana,139 and eighty percent are for drug possession rather than distribution.140 Yet the law imposes a period of ineligibility on every person ever convicted of a drug misdemeanor or felony who has not completed a formal rehabilitation program, without regard to the nature of the offense, her current sobriety, or her conduct and achievements since conviction.

This same extreme overbreadth undermines another rationale Congressman Souder offered for the law, that tax dollars should not be wasted on students who are in college to use drugs rather than to learn.141 It is hardly rational to presume without inquiry that everyone who once used drugs continues to do so, even after apprehension and punishment, and even after a college has found them suitable for admission. Nor can the Souder amendment be justified as a measure to deter drug crimes, insofar as many of those currently sanctioned by the law committed their offenses prior to its enactment.142 No deterrent effect is even possible in these cases, and even when applied prospectively it is hard to envision a significant group of drug offenders whose conduct will change as a result of this additional sanction.

139. The FBI’s 1995 figures were: 1,476,100 arrests for drug abuse violations. Thirty-four point one percent of these arrests were for marijuana possessionómore than the combined total number of arrests for murder, manslaughter, robbery, arson, vagrancy, rape, and all sex offenses including prostitution. FED. BUREAU OF INVESTIGATION, UNIFORM CRIME REPORTS FOR THE UNITED STATES 1995, at 207-08 (2002).

  1. See DEP’T OF JUST., SOURCEBOOK, supra note 52, at 352 tbl. 4.1, 393 tbl. 4.29.
  2. Souder’s remarks on the subject of drug use in college vividly reveal the overheated and

unfounded stereotypes that fueled the enactment of his amendment: “A college degree, if you’re taking drugs, isn’t worth a whole lot. You’re likely to be messed up for the rest of your life.” Cannon, supra note 50, at 2252. Such severely debilitating ongoing drug addiction undoubtedly afflicts a miniscule percentage of the offenders affected by this law.

Congressman Souder also argued that “[t]axpayers have a right to know that students who have a drug-abuse problem aren’t using tax dollars to go through school.” Phil Zabriskie, The New Anti-War Protesters, ROLLING STONE, Oct. 26, 2000, at 1. To the degree that the remark implies a further rationaleñthe conservation of federal loan money for more suitable studentsóit is worth noting that the loans to ex-drug offenders are so small a percentage of the total as to have almost no effect on the grants available to others. In Nyquist v. Mauclet, 432 U.S. 1 (1977), the Supreme Court rejected a similar rationale offered to justify the denial of scholarships to aliens. The Court found that although funding education for non-aliens who can vote is a compelling state interest, the ineligibility of aliens had too insubstantial an impact to serve this interest. Id. Although this decision utilized strict scrutiny because the law burdened a suspect class (aliens), the Court’s objection to the mismatch between means and ends would seem to fail the rational basis test as well.

142. Congress commissioned no study and considered no evidence of the law’s deterrent value, but some supporters and opponents debated the Souder amendment’s deterrent value. One sponsor, Rep. Gerald Solomon, R-N.Y., described college aid ineligibility as “the best way to combat illegal drug use at the sourceóby sending a message to college students and high school seniors applying for college that illegal drug use is intolerable.” Cannon, supra note 50, at 2252; see also 144 CONG. REC. H2510, 2516 (Apr. 29, 1998) (statement by Rep. Barney Frank, D-Mass., “[a]nyone who thinks this law will deter drug use isn’t paying attention”).


How to Construct an Underclass 91

Few enough drug offenders consider the consequences of apprehension; this group would not only calculate the risks, but do so according to an implausible psychology that is deterred by Higher Education Act (HEA) ineligibility but not by the long prison sentences and steep fines already on the books.

One problem these arguments must address is that the rational basis test allows for a substantial degree of under- and over-inclusiveness.143 Unlike strict scrutiny, rational basis review does not demand that a law utilize the most precise means to its end. At one point the Supreme Court did apply an “irrebutable presumption” doctrine requiring that when criteria presume a fact that is not universally true (for example, physical incompetence due to pregnancy), affected individuals must have an opportunity to show that presumption untrue in their case.144 But the Court has virtually gutted that doctrine, so that such overbroad laws must either be struck down as unconstitutional or upheld as reasonable in light of the effectiveness of the classification and the costs of precision:

[T]he question raised is not whether a statutory provision precisely filters out those, and only those, who are in the factual position which generated the congressional concern reflected in the statute. . . . The question is whether Congress, its concern having been reasonably aroused by the possibility of an abuse which it legitimately desired to avoid, could rationally have concluded both that a particular limitation or qualification would protect against its occurrence, and that the expense and other difficulties of individual determinations justified the inherent imprecision of a prophylactic rule.145

This rational basis criterion grants deference, but does not confer immunity on imprecision no matter how extreme. The constitutionality of an under- or over-inclusive statute is necessarily a matter of degree. As the mismatch becomes increasingly pronounced, the legal instrument becomes increasingly unfair and futileóand by what indicator are we to judge instrumental rationality if not by the effectiveness of the means chosen? Thus, numerous cases have struck down the blanket exclusion of felons from

143. The Court has said that “[i]n the area of economics and social welfare, a State does not violate the Equal Protection Clause merely because the classifications made by its laws are imperfect.” Weinberger v. Salfi, 422 U.S. 749, 769 (1975) (citing Dandridge v. Williams, 397 U.S. 471 (1970); Lindsley v. Natural Carbonic Gas Co., 220 U.S. 61, 78 (1911)).

144. Cleveland Bd. of Educ. v. LaFleur, 414 U.S. 632 (1974) (striking down a compelled leave of absence for pregnant public school teachers as a violation of substantive due process); see also Vlandis v. Kline, 412 U.S. 441 (1973); Stanley v. Illinois, 405 U.S. 645 (1972).

145. Weinberger v. Salfi, 422 U.S. 749, 777 (1975). The Court stated that there is “no basis for our requiring individualized determinations when Congress can rationally conclude not only that generalized rules are appropriate to its purposes and concerns, but also that the difficulties of individual determinations outweigh the marginal increments in the precise effectuation of congressional concern which they might be expected to produce.” Id. at 785.


92 The Journal of Gender, Race & Justice [6:2002]

particular occupations as not rationally related to any legitimate state goal. For example, an Alabama exclusion from towing contracts of persons convicted of a crime of force, violence, or moral turpitude was struck down on equal protection grounds as “totally irrational.”146 Several federal courts struck have down blanket exclusion of ex-offenders from government positions as too unrelated to legitimate state interests, one noting “the punitive effects across-the-board ‘felon bans’ can have on individuals seeking to rehabilitate themselves.”147 When the New York City Transit Authority dismissed an employee convicted of manslaughter pursuant to a policy requiring the dismissal of anyone convicted of a felony, a federal court found the policy unconstitutionally overbroad: “Before excluding ex- felons as a class from employment, a municipal employer must demonstrate some relationship between the commission of a particular felony and the inability to adequately perform a particular job.”148 On the other hand, courts have repeatedly upheld exclusions of ex-offenders from specified occupations when the kind of offense poses particular risks in that field: arsonists may be excluded from fire departments, for example.149 All of

146. Lewis v. Alabama Dep’t of Pub. Safety, 831 F. Supp. 824, 825-28 (M.D. Ala. 1993) (finding no rational basis because the exclusion did not allow consideration of the nature, circumstances, age, or seriousness of the crime in relation to the job sought, nor the degree of the offenders rehabilitation).

147. Butts v. Nichols, 381 F. Supp. 573, 580 (S.D. Iowa 1974) (holding an exclusion of ex felons from most civil service jobs unconstitutional on equal protection grounds); see also Kindem v. City of Almeda, 502 F. Supp. 1108, 1111-13 (D.C. Cal. 1980) (holding unconstitutional a ban on hiring ex-felons for city jobs; “it has not been demonstrated that the sole fact of a single prior felony conviction renders an individual unfit for public employment, regardless of the type of crime committed or the type of job sought.”); Davis v. Bucher, 451 F. Supp. 791 (D.C. Pa. 1978) (“[A] regulation which bars former users and addicts from city employment, without any consideration of the merits of each individual case, [is] overbroad and irrational in violation of the Equal Protection Clause.”); Hyland v. Fukuda, 402 F. Supp. 84, 93-94 (D. Haw. 1975) (state violates equal protection and due process by excluding a person convicted of armed robbery from work as a corrections officer without allowing him to demonstrate his fitness for employment). Cf. Harvey Prager, 422 Mass. 86, 91 (1996) (regarding ex-offender ineligibility for bar membership, “no offense is so grave as to preclude a showing of present moral fitness”); Smith v. Fussenich, 440 F. Supp. 1077, 1080-81 (D. Conn. 1977) (striking down a state law excluding ex-felons from private detective or security guard work as lacking rational basis because the “across-the-board disqualification fails to consider probable and realistic circumstances in a felon’s life, including the likelihood of rehabilitation, age at the time of conviction, and other mitigating circumstances related to the nature of the crime and degree of participation.”).

148. Furst v. N.Y. City Transit Auth., 631 F. Supp. 1331, 1338 (E.D. N.Y. 1986). A Massachusetts superior court held likewise in a recent suit brought by two plaintiffs whose respective manslaughter and armed robbery convictions barred them for life from Health and Human Services jobs involving “potential unsupervised contact with program clients.” Cronin et al. v. O’Leary, No. 00-1713-F (Mass. Sup. Ct. Aug. 9, 2001) (Memorandum of Decision and Order on Plaintiff’s Motion for Partial Summary Judgment).

149. Carlyle v. Sitterson, 438 F. Supp. 956, 963 (D.N.C. 1975). For other examples of the kind of occupational exclusions that have been upheld, see FW/PBS, Inc. v. City of Dallas, 837 F.2d 1298, 1304-05 n.22 (5th Cir. 1988) (upholding five year exclusion from sexually oriented businesses for those convicted of sex or obscenity crimes); Darks v. City of Cincinnati, 745 F.2d 1040, 1043-44 (6th Cir. 1984) (denying felons a dance hall license); Schanuel v. Anderson, 708 F.2d 316, 319 (7th Cir. 1983) (finding ex-felon may be disqualified from obtaining detective license for ten years


How to Construct an Underclass 93

these cases are essentially applications of a 1957 Supreme Court ruling that the due process and equal protection clauses prohibit a state from imposing occupational qualifications unless they have a rational connection with the applicant’s fitness or capacity for that occupation.150 Their holdings have real bite when applied to the Souder amendmentóa much broader law that excludes misdemeanants as well as felons, and excludes them from learning opportunities rather than from sensitive occupations like law on which others depend.

The rational basis test has also been sufficient to strike down other “civil disabilities” that, like this one, seem primarily designed to burden ex- offenders who have already paid for their crimes. In one such case, the Supreme Court struck down a Kansas recoupment law on equal protection grounds, finding no rational basis for denying indigent criminal defendants the exemptions from wage garnishment afforded to all other debtors.151 The Court said:

It is in the interest of society and the State that. . .a defendant, upon

because the public trust might be undermined by assigning the guarding of persons and property to ex-offenders); Upshaw v. McNamara, 435 F.2d 1188, 1190 (1st Cir. 1970) (upholding exclusion of felons from police work even if pardoned).

150. Schware v. Bd. of Bar Examiners, 353 U.S. 232 (1957) (arrest record, use of aliases, and former Communist party membership insufficient to exclude applicant from bar membership on character grounds). In Schware, the Court said that a state

cannot exclude a person from the practice of law or from any other occupation in a manner or for reasons that contravene the Due Process or Equal ProtectionClause of the Fourteenth Amendment. . . . A State can require high standards of qualification, such as good moral character or proficiency in its law, before it admits an applicant to the bar, but any qualification must have a rational connection with the applicant’s fitness or capacity to practice law.

Id. at 238-39. For a recent example of an occupational qualification found insufficiently rational, see Craigmiles v. Giles, 110 F. Supp. 2d 658 (2000), striking down on equal protection and substantive due process grounds a requirement that casket-sellers hold a Tennessee funeral director’s license.

151. James v. Strange, 407 U.S. 128 (1972). The Kansas law exempted many categories of property from recoupment against civil judgment debtors (including limits on wages subject to garnishment), but did not afford many of these exemptions to indigent criminal debtors, whose wages could be entirely garnished to reimburse the state for their legal expenses. Id. at 135. Although the case concerned reimbursement for court-appointed counsel, the Court explicitly decided the case on equal protection-rational basis review, not right to counsel grounds. Id. at 134

In its opinion, the Court cited its decision six years earlier in Rinaldi v. Yeager, 384 U.S. 305 (1966), striking down a statute requiring only indigent defendants who were sentenced to imprisonment to reimburse the State the costs of a transcript on appeal.

In Rinaldi, as here, a broad ground of decision was urged, namely, that the statute unduly burdened an indigent’s right to appeal. The Court found, however, a different basis for decision, holding that “to fasten a financial burden only upon those unsuccessful appellants who are confined in state institutions . . . is to make an invidious discrimination” in violation of the Equal Protection Clause. Rinaldi affirmed that the Equal Protection Clause “imposes a requirement of some rationality in the nature of the class singled out.”

Id. (internal citations omitted).



The Journal of Gender, Race & Justice [6:2002]

satisfaction of the criminal penalties imposed, be afforded a reasonable opportunity of employment, rehabilitation and return to useful citizenship. . . . [T]he Equal Protection Clause ‘imposes a requirement of some rationality in the nature of the class singled out.’ This requirement is lacking where, as in the instant case, the State has subjected indigent defendants to such discriminatory conditions of repayment. . . . State recoupment laws, notwithstanding the state interests they may serve, need not blight in such discriminatory fashion the hopes of indigents for self- sufficiency and self-respect. The statute before us embodies elements of punitiveness and discrimination which violate the rights of citizens to equal treatment under the law. 152

Statutes burdening ex-offenders extend well beyond the areas of college assistance, occupational ineligibility, and garnishment laws, as we have noted. We can expect many years of litigation regarding the constitutionality of the multitude of “civil disabilities” that are now imposed, each of which will inform the disposition of others yet to be resolved.153

  1. Id. at 139-42.
  2. At this point, examples of this developing case law include, inter alia: (1) Drug felon’s

ineligibility for food stamps: upheld as a rational means of deterring drug felonies and reducing food stamp fraud, which the legislative record showed often takes the form of trading food stamps for drugs. Turner v. Glickman, 207 F.3d 419, 425 (7th Cir. 2000). (2) Automatic registration of sex offenders. Cf. People v. Adams, 581 N.E.2d 637 (Ill. 1991) (finding sex offender registration rationally furthers the state’s interest in safety), with Doe v. Attorney Gen., 715 N.E.2d 37 (Mass. 1999) (Doe No. 5) (finding sex offender registration requirement unconstitutionally overbroad under the Massachusetts Constitution, for conclusively presuming that particular offenses posed a risk to public safety). In Doe No. 5, the Massachusetts Supreme Judicial Court found that even persons convicted of child rape might no longer present a risk of reoffense, and therefore “the general legislative category does not adequately specify offenders by risk so as to warrant automatic registration of every person convicted under that statute.” Id. at 44; see also Doe v. Attorney Gen., 686 N.E.2d 1007 (Mass. 1997) (Doe No. 3) (holding unconstitutional a sex-offender registration requirement placed on anyone previously convicted of indecent assault and battery, because the conviction does not establish that the offender “is a threat to those persons for whose protection the Legislature adopted the sex offender act”). In both cases, the Court required an individualized hearing at which the ex-offender could demonstrate he no longer posed a threat, and perhaps for that reason framed its decisions as a matter of procedural due process under the Massachusetts Constitution. See Doe, 715 N.E.2d at 45 and Doe, 686 N.E.2d 1007. Although cases invalidating irrefutable presumptions test the borderline between procedural and substantive due process, they are best understood as reflecting substantive due process concerns. In the Doe cases, the court’s objection was to the over breadth in a regulation that conclusively presumed these ex-offenders to pose a continuing riskóand thus to the substantive irrationality of the criteria, not the procedure by which the criteria was found to apply to the plaintiff. (3) Prohibition on firearm possession by felon. United States v. Harris, 537 F.2d 563 (1st Cir. 1976) (found rationally related to government’s interest in assuring public safety). Cf. U.S. v. Jester, 1389 F.3d. 1168 (7th Cir. 1998). (4) State disenfranchisement of felons: found constitutional because the Fourteenth Amendment, Section 2, specifically excludes disenfranchisement of criminals from its guarantee. (5) Driving license suspension. Cf. People v. Lindner, 535 N.E. 2d 829 (Ill. 1989) (finding revocation for convicted sex offenders unconstitutional because there is no rational relationship between sex offenses and safe driving), with Rushworth v. Registrar of Motor Vehicles, 596 N.E. 2d 340 (Mass. 1992) (holding five year suspension for drug traffickers upheld as rationally related to state’s interest in deterring drug crimes and promoting safe driving).


How to Construct an Underclass 95

Developing case law will also guide litigants toward the most promising specific doctrinal claims. Here we have challenged the rational basis of the Souder amendment in light of its purported goals, but have not addressed the variety of substantive due process and equal protection frameworks in which a rational basis challenge might be pursued, nor some more specific objections that would flow from each.154 An equal protection claim, for example, could identify the burdened and unaffected classes in a number of ways. To contrast ex-drug offenders with non-offenders questions the rationality of the sanction at the most general level, as does a substantive due process argument. However, one might also argue that the sanction irrationally distinguishes among classes of ex-offenders: why are drug offenders singled out for ineligibility, while arsonists, rapists and drunk driving offenders remain untouched? Does imposing this additional sanction on top of the court’s sentence serve some permissible governmental interest in drug cases that it would not in alcohol-related crimes? This kind of comparison may help establish that this classification was unconstitutionally “drawn for the purpose of disadvantaging the group burdened by the law.”155 According to the Supreme Court, the Equal Protection Clause “must at the very least mean that a bare . . . desire to harm a politically unpopular group cannot constitute a legitimate governmental interest.”156 A third equal protection claim juxtaposes economic classes: affluent ex-drug offenders are unaffected, but needy ex-offenders lose access to higher education. The Souder Amendment thereby “lays an unequal hand on those who have committed intrinsically the same quality of offense,”157 and the government

We also note that a very recent Supreme Court case may inform civil disabilities doctrine (although not itself a civil disability case because the sanction was triggered by criminal conduct rather than conviction). In Dep’t of Hous. & Urban Dev. v. Rucker, 122 S. Ct. 1230 (2002), the Supreme Court upheld “one-strike-and-you’re-out” federal housing regulations, which require the eviction of families from federally-assisted housing if anyone, including a guest, uses drugs on the premises, whether or not known to the tenant.

154. The Fifth Amendment due process clause has been construed to include an equal protection component. Richardson v. Belcher, 404 U.S. 78, 81 (1971).

  1. Romer v. Evans, 517 U.S. 620 (1996).
  2. United States Dep’t of Agric. v. Moreno, 413 U.S 528, 534 (1973) (“Of course Congress

may not invidiously discriminate among such claimants on the basis of a ‘bare congressional desire to harm a politically unpopular group.'”). On the other hand, when confronted with a New York Transit Authority regulation that prohibited the employment of methodone users in drug treatment programs, but disciplined employees found drinking on the job on an individual basis that might allow them to retain their jobs, the Court found no equal protection violation. “[T]he fact that the TA has the resources to expend on one class of problem employees does not by itself establish a constitutional duty on its part to come up with resources to spend on all classes of problem employees.” N.Y. City Transit Auth. v. Beazer, 440 U.S. 568, 591 n.37 (1979).

157. Skinner v. Oklahoma, 316 U.S. 535, 541 (1942). Skinner concerned Oklahoma’s Habitual Criminal Sterilization Act, a law that authorized courts to order sterilization of any defendant who had been convicted of three felonies involving moral turpitude. Id. at 536. While larceny was such a crime, the Act excluded embezzlers from its application. Id. at 537. The Supreme Court held Oklahoma’s Habitual Criminal Sterilization Act unconstitutional as a violation of the Equal Protection Clause:


96 The Journal of Gender, Race & Justice [6:2002] should be required to demonstrate what legitimate governmental interest is

furthered by targeting only the needy for this additional deprivation. 2. Challenge to Retroactive Application

The Department of Education applies the Souder amendment to every applicant who has ever been convicted of a drug offense. This means that an offender who was convicted before the statute’s enactment may now suffer an additional, retroactive sanction she could not have anticipated. If, for example, she waived trial and pled guilty to a third marijuana possession misdemeanor a decade ago, she did so without knowing that it would permanently foreclose federal college aid. By enforcing the Souder Amendment retroactively, the Department of Education has injected an additional element of unfairness as well as greatly expanded its scope.

Were the ineligibility sanction recognized as criminal punishment (as we argue it should be in the next section), the Constitution’s prohibition on ex post facto laws would prevent its application to convictions occurring before its enactment.158 If the courts deem this sanction to be a non-punitive “civil disability,” however, the legality of its retroactive enforcement requires a more complicated analysis. In the past decade, the Court has treated the retroactivity of civil burdens largely as a matter of statutory interpretation, holding that while Congress has the constitutional power to legislate retroactively, it will be presumed not to have done so “absent a clear indication. . .that it intended such a result.”159 The Court has explained

When the law lays an unequal hand on those who have committed intrinsically the same quality of offense and sterilizes one and not the other, it has made as an invidious a discrimination as if it had selected a particular race or nationality for oppressive treatment. . . . Sterilization of those who have thrice committed grand larceny with immunity for those who are embezzlers is a clear, pointed, unmistakable discrimination.

Id. at 541.

Skinner, however, involved an acknowledged fundamental right, procreation, and, therefore, the Court applied “strict scrutiny” review. Id. at 541. A Court hearing a challenge to the Souder amendment would have to decide whether the less demanding “rational basis” standard is also unsatisfied by a dual system of sanctions that discriminates between rich and poor defendants whose offenses are identical. Unfortunately, the Court has already found that denying federal aid to higher education for those who fail to register for the Selective Service rationally promoted the government’s legitimate goals of draft registration and the fair allocation of resources. Selective Serv. Sys. v. Minn. Pub. Interest Research Group, 468 U.S. 841, 859 n.17 (1984). The Court noted, however, that ineligibility was not based on any criminal offense but on a continuing failure to register, so that a substantial increase in registrations could be expected to result from the threat of ineligibility. Id. at 855.

158. U.S. CONST. art. I, ß 9, cl. 3 (ex post facto prohibition); Kansas v. Hendricks, 521 U.S. 346 (1997) (ex post facto clause bars retroactive criminal punishments); Lindsey v. Washington, 301 U.S. 397, 401-02 (1937) (retroactive removal of possibility of early release without parole violates ex post facto clause). If recognized as punishment, the ineligibility sanction would also abridge a variety of other constitutional guarantees. See supra notes 134-36.

159. INS v. St. Cyr, 533 U.S. 289 (2001); Landgraf v. USI Film Products, 511 U.S. 244 (1994). In Landgraf, the Court noted that the Constitution contains no blanket ban on retroactive


05/16/02 7:57 PM

How to Construct an Underclass

this presumption as a matter of fairness:

Elementary considerations of fairness dictate that individuals should have an opportunity to know what the law is and to conform their conduct accordingly; settled expectations should not be lightly disrupted. For that reason, the “principle that the legal effect of conduct should ordinarily be assessed under the law that existed when the conduct took place has timeless and universal appeal.” . . . Requiring clear intent assures that Congress itself has affirmatively considered the potential unfairness of retroactive application and determined that it is an acceptable price to pay for the countervailing benefits.160


Additionally, the legislature’s “responsivity to political pressures poses a risk that it may be tempted to use retroactive legislation as a means of retribution against unpopular groups or individuals.”161 As others have noted, both of these reasons suggest a more robust restriction on retroactive civil burdens than the mere requirement that Congress clearly indicate its intent to inflict unfairness or retribution.162 But even the Court’s interpretive default rule should be fully sufficient to severely limit the reach of the

application of civil laws as it does criminal punishments, Landgraf, 511 U.S. at 268, but “the presumption against retroactivity applies far beyond the confines of the criminal law.” Id. at 272.

160. See Landgraf, 511 U.S. at 265, 272-73. The Court also noted that “it will frequently be true . . . that retroactive application of a new statute would vindicate its purpose more fully. That consideration, however, is not sufficient to rebut the presumption against retroactivity.” Id. at 268.

161. Id. at 266. When the retroactive legislation imposes a criminal sanction, Landgraf observes that the ex post facto clause “restricts governmental power by restraining arbitrary and potentially vindictive legislation.” Id. at 267 (citing Weaver v. Graham, 450 U.S. 24, 28-29 (1981)); see also James v. United States, 366 U.S. 213, 247, n.3 (1961) (retroactive punitive measures may reflect “a purpose not to prevent dangerous conduct generally but to impose by legislation a penalty against specific persons or classes of persons.”); Charles B. Hochman, The Supreme Court and the Constitutionality of Retroactive Legislation, 73 HARV. L. REV. 692, 693 (1960) (a retroactive statute “may be passed with an exact knowledge of who will benefit from it.”).

162. Daniel Kanstroom asks,

Can it be seriously maintained that a constitutional court really addresses such problems by requiring the legislature to be clear about it? Indeed, a legislature that is truly aiming at an unpopular group may be, if anything, more likely to be meticulously clear than one which retroactively deprives a more empowered group (such as the Landgraf defendants) of something arguably of value.

Daniel Kanstroom, St. Cyr or Insincere: The Strange Quality of Supreme Court Victory, GEO. IMMIGR. L. J. (forthcoming 2002). He urges that the due process clause be construed as including some of the norms against retroactivity that, on the criminal side, are found in the ex post facto clausesóan approach utilized by Judge Weinstein in Mojica v. Reno, 970 F. Supp. 130 (E.D.N.Y. 1997). Id. The Supreme Court has indeed noted the due process implications of retroactive civil laws, stating that “[t]he Due Process Clause also protects the interests in fair notice and repose that may be compromised by retroactive legislation; a justification sufficient to validate a statute’s prospective application under the Clause ‘may not suffice’ to warrant its retroactive application.” Landgraf, 511 U.S. at 266 (citing Usery v. Turner Elkhorn Mining Co., 428 U.S. 1, 17 (1976)); see also Stephen R. Munzer, A Theory of Retroactive Legislation, 61 TEX. L. REV. 425, 471 (1982) (rule of law entitles “persons to have their behavior governed by rules publicly fixed in advance.”).


98 The Journal of Gender, Race & Justice [6:2002]

Souter amendment to offenses occurring after its enactment in 1998. There is no reason to doubt that the presumption would apply here, because retroactive enforcement of HEA ineligibility to drug offenders produces precisely the unfairness the presumption is designed to prevent.163 The Court continues to apply the presumption zealously, most recently last year in INS v. St. Cyr,164 a case striking down the retroactive elimination of discretionary deportation relief for offenders. Moreover, the Court’s criterion for rebutting the presumption is a demanding one: as described in St. Cyr, laws “will not be construed to have retroactive effect unless their language requires this result. . . . Cases where this Court has found truly ‘retroactive’ effect adequately authorized by statute have involved statutory language that was so clear that it could sustain only one interpretation.”165

The Souder amendment lacks any such explicit statement regarding its application to convictions accrued prior to enactment. On the contrary, its language creates ambiguity by limiting the act to “a student who has been convicted of any offense,” not “a person.”166 This might reasonably be construed to exempt those whose convictions occurred prior to their

163. When applied retroactively, the Souder amendment seems a good illustration of both the unfairness and the retribution decried in St. Cyr; it attaches unanticipated and severe consequences to the guilty pleas of a group of offenders distinguished not by their dangerousness or malevolence but by their political unpopularity. The targeted group is politically disempowered as well because many of them have lost the vote as a result of their convictions. St. Cyr, 533 U.S. at 289.

We also note that according to the Supreme Court’s definition, the Department of Education is clearly applying the Souder Amendment retroactively. In INS v. St. Cyr, the Court stated that a law has retroactive effect when it “attaches a new disability, in respect to transactions or considerations already past.” St. Cyr, 533 U.S. at 321 (citing Landgraf, 511 U.S. at 269). In St. Cyr, as in many cases involving HEA ineligibility, the defendant had pled guilty prior to the enactment of a law imposing an additional disability to those convicted of the offense. Id. at 323-24. The Court said,

Plea agreements involve a quid pro quo between a criminal defendant and the government. . . . Now that prosecutors have received the benefit of these plea agreements, agreements that were likely facilitated by the aliens’ belief in their continued eligibility for ß 212(c) relief, it would surely be contrary to “familiar considerations of fair notice, reasonable reliance, and settled expectations” . . . to hold that IIRIRA’s subsequent restrictions deprive them of any possibility of such relief.

Id; see also Miller v. Florida, 482 U.S. 423 (1987) (holding a law retroactive if it “changes the legal consequences of acts completed before its effective date.”).

164. In St. Cyr., 533 U.S. at 289, an alien had pled guilty to a crime with the expectation that he could apply for discretionary relief from deportation. Prior to St. Cyr’s deportation proceeding, Congress passed legislation eliminating such relief. Id. at 293. Although the government claimed that repeal was purely prospective because it would affect only subsequent deportation proceedings, the Court found that the repeal had been applied retroactively to St. Cyr and ruled in his favor. Id. at 321-26. The Court held that the law could not be applied retroactively to reach St. Cyr’s pre- enactment conviction absent a clear indication of such intent from Congress, which was lacking. Id.

165. St. Cyr, 533 U.S. at 315-17 (citing Bowen v. Georgetown Univ. Hospital, 488 U.S. 204, 208 (1988) and Lindh v. Murphy, 521 U.S. 320, 328 n.4 (1997), for first and second points respectively).

166. 20 U.S.C. ß 1091(r) (2001) (italics added).


How to Construct an Underclass 99

enrollment at college, especially since under HEA regulations a “regular student” is defined as someone “enrolled or accepted for enrollment at an institution for the purpose of obtaining a degree . . . or other recognized educational credential. . . .”167 In fact, as noted above this is largely what Congressman Souder intended by the language he crafted. He says he never expected his law to be applied retroactively, but only to students who violate the drug laws while in college:

The Clinton administration interpreted [the law] either through malicious intent or incompetence to mean that if you ever committed three drug crimes, when you were 15 years old and you applied to college when you’re 28, you can’t get a loan. Congress would never have passed that . . . and its ridiculous . . . I’m an evangelical Christian. I believe people change.168

In 2001 Souder tried to persuade the Department of Education to apply the law in this more restrictive way, but he was unsuccessful and denounced the administration for a “simply shocking” enforcement policy.169 It is difficult to believe any Court would find that the act unambiguously mandates retrospective application when even its author says that the Department of Education is misconstruing that issue.

3. Rights Under the Federal Disability Acts

Drug addiction is among the disabilities protected against discrimination by both the Americans with Disabilities Act (A.D.A.) and the Rehabilitation Act of 1973,170 and recently the Ninth Circuit found that

167. 34 C.F.R. ß 600.2 (2002); see also 34 C.F.R. ß 668.2 (2002) (section concerning drug offenders, which incorporates by reference the definition of “regular student” in ß 600.2).

168. The O’Reilly Factor: Interview with Rep. Mark Souder (Fox television broadcast, Jan. 16, 2001); see also Clarence Page, College Loans are Casualties in Drug War, CHI. TRIB., Apr. 29, 2001, at C19 (Rep. Souder “intended to penalize students for drug violations committed while they are students, not for their prior offenses.”).

  1. 43,000 Students, supra note 45, at A11.
  2. Regarding drug addiction as a disability under the two laws, see generally Thompson v.

Davis, 282 F.3d 780 (9th Cir. 2002) (holding that prior drug addiction is a disability protected by the A.D.A., and that this protection extends to parole decisions), citing 28 C.F.R. ß 35.104 (2000) (“The phrase physical or mental impairment includes . . . drug addiction. . . .”); Buckley v. Consol. Edison, 127 F.3d 270, 272-73 (2d Cir. 1997) (recovering drug addict is a “qualified individual with a disability” under the A.D.A.), vacated, 155 F.3d 150 (2d Cir. 1998) on other grounds; Teahan v. Metro-North Commuter R.R. Co., 951 F.2d 511, 517 (2d Cir. 1991) (“[S]ubstance abuse is a ‘handicap’ for purposes of the Rehabilitation Act.”); Davis v. Bucher, 451 F. Supp. 791, 796 (D. Penn. 1978) (finding drug addicts substantially impaired in performing major life activities and therefore handicapped under the Rehabilitation Act).

Title II of the A.D.A. provides that “no qualified individual with a disability shall, by reason of such disability, be excluded from participation in or be denied the benefits of the services, programs, or activities of a public entity, or be subjected to discrimination by any such entity.” 42 U.S.C. ß 12132 (2001). The Rehabilitation Act of 1973, amended in 1978, states that “[n]o otherwise qualified individual with a disability in the United States, as defined in section 705(20) of this title, shall,


100 The Journal of Gender, Race & Justice [6:2002]

denial of parole based on a history of drug addiction violates the A.D.A.171 Arguably, the Souder provision similarly violates the two statutes by denying college aid to individuals on the basis of this disability.

Of course, under the Souder provision what triggers HEA ineligibility is a drug conviction, not drug addiction per se. A plaintiff will have to show that the former serves as proxy for the latter, so that the law effectively denies a public benefit “by reason of the disability.”172 But it is not difficult to establish that (1) the Souder provision on its face presumes every drug offender is a drug addict, and (2) this is enough to come within the ambit of both the A.D.A. and the Rehabilitation Act.

The Souter provision equates drug offenders with drug addicts because it requires even minor misdemeanants to undergo a certified drug rehabilitation program in order to restore eligibility.173 Explaining this requirement, Congressman Souder has said that if you take drugs, “you’re likely to be messed up for the rest of your life [s]o the best thing we can do for education is to get somebody clean. . . .”174 The law’s assumption that drug offenders are addicts, even if invalid, confers “disabled” status upon those students denied loans. As noted, a person whose drug addiction substantially limits a major life activity qualifies as disabled under both the A.D.A. and the Rehabilitation Act (although separate provisions exclude current users from the Acts’ protections)175óbut so does a person who is mistakenly identified as such. Both acts define “disability” as:

(A) a physical or mental impairment that substantially limits one or more of the major life activities of such individual;

(B) a record of such an impairment; or

solely by reason of her or his disability, be excluded from the participation in, be denied the benefits of, or be subjected to discrimination under any program or activity receiving Federal financial assistance or under any program or activity conducted by any Executive agency or by the United States Postal Service. . . .” 29 U.S.C. ß 794 (2001).

  1. Thompson, 282 F.3d at 780.
  2. See supra note 170. Under either law, to prove that HEA ineligibility constitutes

disability discrimination, the plaintiff must allege that he or she: (1) is an individual with a disability; (2) is otherwise qualified to receive the benefit of some public entity’s program, but (3) was denied the benefit (4) by reason of the disability. See Thompson, 282 F.3d at 783-84.

  1. 20 U.S.C. ß 1091(r)(2) (2001).
  2. Cannon, supra note 50, at 2252.
  3. 42 U.S.C. ß 12210(a) (1994) restricts the A.D.A.’s coverage by providing that “the term

‘individual with a disability’ does not include an individual who is currently engaging in the illegal use of drugs, when the covered entity acts on the basis of such use,” but subsection (b) specifically enumerates as “disabled” any individual who “(1) has successfully completed a supervised drug rehabilitation program and is no longer engaging in the illegal use of drugs, or has otherwise been rehabilitated successfully and is no longer engaging in such use; (2) is participating in a supervised rehabilitation program and is no longer engaging in such use; or (3) is erroneously regarded as engaging in such use, but is not engaging in such use. . . .” In 1978 the Rehabilitation Act was similarly amended to exclude current drug users from its protection, with almost identical language. 29 U.S.C. ß 705(20)(C) (2001).


How to Construct an Underclass 101 (C) being regarded as having such an impairment.176

Because the Souder provision considers every past drug offender to be an addict in need of rehabilitation, affected students should argue that they are “regarded as having such an impairment.” (Those with a record of drug addiction will also come within protected class “B”.) Once it is established that the Souder law regards all offenders as addicts, and delays or denies federal funds unless they complete a rehabilitation program, it follows that this law violates the A.D.A. and Rehabilitation Act by imposing a “denial of benefits . . . by reason of the plaintiff’s disability.”

4. Unconstitutional Punishment

We noted earlier that if denying college aid to drug offenders were deemed criminal punishment, it would be unconstitutional on a number of grounds: as legislative punishment it would constitute a bill of attainder; as multiple punishment it would abridge the double jeopardy clause; and if applied retroactively, it would be an ex post facto law.177 Construed as punishment, educational deprivation might also implicate the Eighth Amendment’s ban on cruel and unusual punishment. These claims should be considered as part of any legal challenge, even though the doctrinal terrain is quite steep: recent Supreme Court cases have often, but not always, allowed Congress to escape all of these constitutional strictures simply by characterizing its sanctions as “civil disabilities” rather than punishment.178

176. 42 U.S.C. ß 12102(2) (1994) (A.D.A. definition). The Rehabilitation Act defines disability in substantially identical language. 29 U.S.C. ß 705(20)(B) (2001); see also 42 U.S.C. ß 12210(b)(3) (1994) (including within A.D.A. protection any individual who is “erroneously regarded as engaging in such use, but is not engaging in such use”).

  1. See supra notes 134-36.
  2. The courts have occasionally policed the civil/criminal borders, but most often have

been willing to defer to whatever description Congress has applied to its statute. Some cases have come close to making a Congressional decision to dispense with procedural protections self justifying: for example, a recent Seventh Circuit case found that when Congress authorizes an agency rather than court to impose a sanction, that sanction “is presumptively civil . . . because agency enforcement mechanisms do not contain the same procedural safeguards that criminal proceedings do.” Turner v Glickman, 207 F.3d 419, 429 (7th Cir. 2000) (citing Helvering v. Mitchell, 303 U.S. 391, 402 (1938)). For examples of restraints and sanctions the Supreme Court has found non-punitive and unencumbered by criminal procedural protections, see Hudson v. United States, 522 U.S. 93 (1997) (holding Double Jeopardy Clause does not bar subsequent criminal prosecution of bankers who suffered monetary penalties and occupational debarment because the latter were civil sanctions); Kansas v. Hendricks, 521 U.S. 346, 361 (1997) (rejecting double jeopardy and ex post facto challenges to “civil commitment” of a convicted sexual predator directly following completion of his sentence, on grounds the commitment was not punishment for his past offense but protection against future dangerousness); United States v. Ursery, 518 U.S. 267 (1996) (holding forfeiture is remedial, not punitive, for purposes of double jeopardy clause); United States v. Salerno, 481 U.S. 739 (1987) (pretrial detention of dangerous defendants is non-punitive, remedial restraint designed to protect the community); United States v. Ward, 448 U.S. 242 (1980) ($5,000 civil penalty for each water pollution violation not punitive and thus reporting requirement does not violation the privilege against self-incrimination).


102 The Journal of Gender, Race & Justice [6:2002]

This is an ominous trend, because all of the fundamental criminal procedural guarantees in the Bill of Rights will be at risk if a semantic fig leaf is all its takes to place a sanction beyond their reach. At this point, the vexing and vague borders separating civil disabilities from criminal punishment vary according to which constitutional guarantee is at issue,179 and have been the subject of numerous scholarly examinations.180 Here we can sketch only some of the issues that pertain to the Souder amendment in particular.

Recent Supreme Court cases have placed numerous “collateral consequences” of conviction outside the Bill of Rights’ criminal procedural guarantees, yet two important features of the Souder Amendment might distinguish it from these precedents. First, it is significant that Congress initially enacted college loan ineligibility in 1988 as a punishment for drug offenders, to be issued by the sentencing court in its discretion.181 Evidently

Occasionally, however, the Court has looked past the legislature’s designation to find a purportedly civil statute “punitive” and therefore requiring at least some of the criminal procedural protections. See United States v. Bajakajian, 524 U.S. 321 (1998) (holding amount of forfeiture violated Eighth Amendment bar on excessive fines); Dep’t of Revenue of Montana v. Kurth Ranch, 511 U.S. 767 (1994) (state marijuana tax is punishment to which double jeopardy clause applies); Austin v. United States, 509 U.S. 602 (1993) (holding that the Eighth Amendment’s Excessive Fines Clause applies to in rem civil forfeitures of drug-offense-related property because it is at least partly punitive); United States v. Halper, 490 U.S. 435 (1989) (finding violation of double jeopardy after previously convicted defendant was subjected to civil fine for Medicaid fraud that was so large as to constitute punishment); see also Kansas v. Crane, 122 S. Ct. 867 (2002) (holding that confinement of a sexual predator is civil rather than criminal only if accompanied by a finding that the individual lacks some capacity to control himself).

179. Kurth Ranch, 511 U.S. at 767 (finding the Double Jeopardy Clause but not other criminal procedural guarantees applicable to a punitive marijuana tax); Austin, 509 U.S. at 607-10 (holding that civil forfeiture is non-criminal punishment, and therefore Eighth Amendment restrictions on excessive punishment apply but criminal procedural protections do not). Professor Carol Steiker discerns a cacophony of doctrines associated with different constitutional provisions and observes that “[o]n the rare occasion when the Court has attempted to define punishment more globally, it has resorted to a list of “factors . . . for which it has been unable to offer an underlying rationale.” Carol S. Steiker, Punishment and Procedure: Punishment Theory and the Criminal-Civil Procedural Divide, 85 GEO. L.J. 775, 781 (1997).

180. See, e.g., Mary M. Cheh, Constitutional Limits on Using Civil Remedies to Achieve Criminal Law Objectives: Understanding and Transcending the Criminal-Civil Law Distinction, 42 HASTINGS L.J. 1325 (1991); Susan R. Klein, Redrawing the Civil-Criminal Boundary, 2 BUFF. CRIM. L. REV. 679 (1999); Steiker, supra note 179.

181. The “McCollum Amendment” authorizes both federal and state courts to deny certain federal benefits, including HEA loans and grants, to individuals convicted of drug offenses. 21 U.S.C. ß 862 (2002) (formerly 853(a)). This ineligibility sanction remains one of the sentences that may be issued by a federal judge under the United States Sentencing Guidelines. See U.S.S.G. ß 5F1.6 (“The court, pursuant to 21 U.S.C. ß 862, may deny the eligibility for certain Federal benefits of any individual convicted of distribution or possession of a controlled substance.”). These features undoubtedly mark the sanction as punishment for reasons Justice Thomas offered in his opinion for the Court holding a forfeiture to be punishment:

We have little trouble concluding that the forfeiture of currency ordered by ß 982(a)(1) constitutes punishment. The statute directs a court to order forfeiture as an additional sanction when ‘imposing sentence on a person convicted of’ a willful violation of ß 5316’s reporting requirement. The forfeiture is thus imposed at the culmination of a criminal proceeding and requires conviction of an underlying felony, and it cannot be


How to Construct an Underclass 103

dissatisfied with the courts’ sporadic imposition of this punishment, Congress subsequently decided to make the sanction mandatory for all drug offenders, but did so by removing it from the courts and imposing it by legislation. One obvious question, then, is this: Does transferring the power to impose this sanction from a sentencing court to Congress change its nature, or is this just what the Constitution’s Bill of Attainder clause is designed to prevent? That clause prohibits Congress from singling out any individual or group182 and “meting out summary punishment for past conduct.”183 It guarantees that the judiciary, not the legislature, will impose punishment.184

imposed upon an innocent owner of unreported currency, but only upon a person who has himself been convicted of a ß 5316 reporting violation.

United States v. Bajakajian, 524 U.S. 321, 328 (1998).

The punitive goals of the legislation were also evident in Congressional debate. The House sponsor, Rep. McCollum, described the legislation as designed to make “a user accountable. . . . We need to put on the books something besides jail time which is not practical, to say to that young person, if you use and make that choice you are going to be held accountable, you are going to have to pay a price that is greater than you will want to pay.” 134 CONG. REC. H7074-02 (statement of Rep. McCollum) (daily ed., Sept. 7, 1988). Rep. Traficant said “it is time to stop turning the other cheek and start . . . taking an eye for an eye.” 134 CONG. REC. H10,710 (daily ed. Oct. 21, 1988) (statement of Rep. Traficant). In the Senate debate on the 1988 bill, Senator Gramm described federal aid ineligibility as a necessary adjunct to existing penalties because there was then too much “crime without punishment for people who are using drugs.” 134 CONG. REC. S15,966 (daily ed. Oct. 14, 1988) (statement of Sen. Gramm). He endorsed the bill as a way to strike at the “real drug kingpin . . . the drug user who “ha[s] put a drug thug at the door of every high school in America. . . .” 134 CONG. REC. S15,762 (daily ed. Oct. 13, 1988) (statement of Sen. Gramm). Other senators endorsed the bill as an attack on “drug users [who are] the ones to blame.” 134 CONG. REC. S17,303 (daily ed. Oct. 21, 1988) (statement of Sen. Dole) or on “greed soaked mutants who deal in illegal drugs.” 134 CONG. REC. S15,978 (daily ed. Oct. 14, 1988) (statement of Sen. Heflin).

182. Doe v. Selective Serv. Sys., 557 F. Supp. 937, 942-43 (D. Minn. 1983) (bill of attainder ban violated if a statute “clearly singles out an ascertainable group based on past conduct . . . [and] legislatively determines the guilt of this ascertainable group”); United States v. Lovett, 328 U.S. 303, 315-16 (1946).

183. Landgraf, 511 U.S. at 266 (citing United States v. Brown, 381 U.S. 437, 456-462 (1965)); see also Communist Party of the United States v. Subversive Activities Control Bd., 367 U.S. 1, 87 (1961).

184. Cummings v. Missouri, 18 L.Ed. 356 (1866) (defining a bill of attainder as a legislative act that inflicts punishment without a judicial trial).

The primary challenge for a bill of attainder claim against the Souder Amendment is an unfavorable precedent. See Selective Serv. Sys. v. Minn. Pub. Interest Research Group, 468 U.S. 841 (1984) (upholding a law denying HEA assistance to students who had failed to register with the Selective Service Board). There the Court found the sanction non-punitive and therefore outside the scope of the Bill of Attainder clause. Id. at 858. There are several important differences in the contours of the two sanctions, however.

In Selective Service, the Court described its method of determining whether a sanction is punishment for purposes of the Bill of Attainder prohibition as consisting of three necessary inquiries. First, does the statute fall within the historical meaning of legislative punishment? Id. at 852. Second, does the statute, viewed in terms of the type and severity of burdens imposed, reasonably further non-punitive legislative purposes? This further test is necessary “to ensure that the Legislature has not created an impermissible penalty not previously held to be within the proscription against bills of attainder.” Id. at 854. Third, does the legislative record evince a


104 The Journal of Gender, Race & Justice [6:2002]

This sequence is also helpful in establishing either of the Supreme Court’s alternative grounds for finding a violation of the Double Jeopardy Clause.185 The first examines Congressional intent: a finding that Congress expressly or impliedly intended to use the sanction as punishment conclusively establishes that it is punishment to which the double jeopardy clause’s ban on multiple punishment applies.186 On this issue the Supreme Court has stated that examining a parallel predecessor statute is “worth a volume of logic” in revealing Congressional intent.187 In this case the 1998

congressional intent to punish? Id. at 852-53. Under this three-fold test, a number of statutes excluding identified groups from certain occupations and professions has been found to constitute punishment in violation of the Bill of Attainder Clause. Id. at 852.

The Court found that law did not constitute a bill of attainder on three bases, at least two of which cut the other way in the Souder Amendment case. Selective Service, 468 U.S. at 852-59. First, the sanction denied the offender a mere non-contractual benefit which was not a disability historically associated with punishment. Id. at 853. The Souder Amendment, however, authorizes imposition of a sanction that was previously a punishment under the McCollum bill. Second, the Selective Service law sanctioned innocent as well as willful nonregistrants, “yet punitive legislation ordinarily does not reach those whose failure to comply with the law is not willful.” Id. at 855. The Souder Amendment, however, only applies to those convicted of a drug crime that requires scienter. Finally, in the Selective Service case eligibility could be restored “at any time simply by registering late” because the statute gave “nonregistrants 30 days after receiving notice that they are ineligible . . . to register for the draft and qualify for aid. . . . Conditioning receipt of Title IV aid on draft registration is plainly a rational means to improve compliance with the registration requirements.” Id. at 853. The Souder Amendment does not permit immediate restoration, although it does allow offenders to eventually become eligible by undertaking a much more arduous process: completion of a certified rehabilitation program.

185. In Hudson, the Court reiterated the two-stage procedure first enunciated in United States v. Ward, 448 U.S. 242 (1980).

A court must first ask whether the legislature, “in establishing the penalizing mechanism, indicated either expressly or impliedly a preference for one label or the other.” Even in those cases where the legislature “has indicated an intention to establish a civil penalty, we have inquired further whether the statutory scheme was so punitive either in purpose or effect,” as to “transform what was clearly intended as a civil remedy into a criminal penalty.”

Hudson, 522 U.S. at 99 (citing Ward, 448 U.S. at 248-49).

186. See, e.g., Turner, 207 F.3d at 427 (“If we determine that Congress intended the statute to be a criminal punishment of those convicted of drug-related felonies, our inquiry is at an end and the statute would constitute criminal punishment for purposes of the Double Jeopardy Clause.”).

The double jeopardy clause has been construed to prohibit multiple punishments as well as multiple prosecutions. See generally Kurth Ranch, 511 U.S. at 768 n.1 (citing North Carolina v. Pearce, 395 U.S. 711 (1969)). Although some justices on the Supreme Court would limit its scope to the latter. Compare the concurrences of Justices Stevens and Scalia in Hudson, 522 U.S. at 111 (Stevens J., concurring, citing Ursery, 518 U.S. at 267, and Hendricks, 521 U.S. at 346 as recent double jeopardy decisions that have “recognized that double jeopardy protection is not limited to multiple prosecutions”) and 106 (Scalia J., concurring).

187. Kennedy v. Mendoza-Matinez, 372 U.S. 144, 169 (1963). In Kennedy, the Supreme Court relied heavily on a predecessor statute to find that Congress intended its law revoking the citizenship of certain draft evaders as punishment:

[T]he objective manifestations of congressional purpose indicate conclusively that the provisions in question can only be interpreted as punitive. A study of the history of the


How to Construct an Underclass 105

education sanction is virtually identical to the admittedly punitive 1988 law (except for its elimination of judicial discretion), and Congress appears to have enacted the later enactment primarily as a way to insure that the punishment was uniformly rather than sporadically imposed.188 In fact, the government has used the 1998 law to impose HEA ineligibility on many earlier drug offenders who at sentencing had already been subject to, but avoided, the same sanction pursuant to the earlier statute. In doing so it has routinely transgressed a central guarantee of the Double Jeopardy Clause, which according to the Supreme Court “protects against the possibility that the Government is seeking the second punishment because it is dissatisfied with the sanction obtained in the first proceeding.”189

Despite this clear history, a court so inclined could rely on the absence of punitive language in the statute, and the assignment of the sanction to an executive agency,190 to find that Congress intended the Souder Amendment to be a non-punitive, regulatory measure. If so, the court must then undertake a second inquiry: whether the HEA ineligibility provisions are “so punitive in form and effect as to render them criminal despite Congress’s intent to the contrary.”191 This determination is to be made according to the indicia the Supreme Court first enunciated in Kennedy v. Mendoza-

predecessor of ß 401(j), which ‘is worth a volume of logic,’ N.Y. Trust Co. v. Eisner, 256 U.S. 345, 349 (1921), coupled with a reading of Congress’ reasons for enacting ß 401(j), compels a conclusion that the statute’s primary function is to serve as an additional penalty for a special category of draft evader. . . . The Senate and House debates, together with Attorney General Biddle’s letter, brought to light no alternative purpose to differentiate the new statute from its predecessor.

Kennedy, 372 U.S. at 169-70, 183; see also id. at 183 n.35 (“The relevance of such history in analyzing the character of a present enactment is illustrated by the Court’s approach in Helwig v. United States, 188 U.S. 605, 613-19 (1902) wherein at considerable length it reviewed and relied upon the character of previous relevant legislation in determining whether the statute before it, which imposed an exaction upon importers who undervalued imported goods for duty purposes, was a penalty.”).

188. Rep. Souder intended his law to carry the punitive message that “actions have consequences, and using or selling drugs will ruin your future,” and his co-sponsor Rep. Solomon hoped the law would show “college students and high school seniors applying for college that illegal drug use is intolerable.” Sounder, supra note 40, supra note 142. In Fleming, the Court observed that where the legislature’s concern is “the activity or status from which the individual is barred, the disqualification is not punishment even though it may bear harshly upon one affected. The contrary is the case where the statute in question is evidently aimed at the person or class of persons disqualified.” Fleming, 363 U.S. at 614.

189. United States v. Halper, 490 U.S. 435, 451 n.10 (1989). The Court has since “disavowed” Halper on other grounds that do not undermine the vitality of this principle. See Hudson, 522 U.S. at 93.

We also note that if found to be punishment, this retroactive application would violate the ex post facto prohibition. The ex post facto clause bars infliction of a greater punishment than was provided by law at the time the defendant committed the crime. Lynce v. Mathis, 519 U.S. 433, 441 (1997); Calder v. Bull, 3 U.S. 386 (1798).

  1. See supra note 178.
  2. Hudson, 522 U.S. at 104 (citing Ursery, 518 U.S. at 290). According to the Supreme

Court, meeting the burden requires the “clearest proof” of the statute’s punitive effect. Id. at 100.


106 The Journal of Gender, Race & Justice [6:2002]

Martinez,192 and a reasonable case can be made that these factors collectively support a finding that the Souder Amendment is clearly punitive in its effect, as we elaborate in the margin.193 The Court has made only a few such findings, but it is notable that the grounds in some of those cases were no stronger than those that could be asserted here. For example, the Supreme Court made such a finding regarding a sanction that shared many of the indicia of punishment present in the Souder Amendment. In Dept. of Revenue of Montana v. Kurth Ranch, the Court held a state tax levied on the possession and storage of illegal drugs following the Kurths’ criminal

192. 372 U.S. 144, 168-69 (1963) (holding that forfeiture of citizenship for draft evasion in wartime was effectively a criminal punishment requiring all procedural guarantees afforded criminal defendants). The Kennedy factors are not exhaustive, Ward, 448 U.S. at 249, nor should any individual factor be considered controlling. Hudson, 522 U.S. at 101.

193. The “Kennedy factors” were reaffirmed in Hudson, 522 U.S. at 99-100. These factors and their application to the Souder Amendment are:

(1) Whether the sanction involves an affirmative disability or restraint: This is the factor that cuts most strongly against a finding that HEA ineligibility is punishment for double jeopardy purposes. The Supreme Court has already described that sanction as the “mere denial of a non- contractual government benefit” that involves no affirmative disability or restraint, albeit in a bill of attainder case. Selective Service Sys. v. Minn. Pub. Interest Research Group, 468 U.S. 841, 852 (1984) (citing Fleming, 363 U.S. at 603).

(2) Whether the sanction has historically been regarded as a punishment: As noted supra note 181, since the 1988 enactment of the McCollum amendment, HEA ineligibility sanction has been utilized as a criminal punishment and continues to be included in the U.S. Sentencing Guidelines. This distinguishes the historical record from that which the Court examined in 1984, when it held that withdrawal of financial aid to education had no historical pedigree as punishment in Selective Serv. Sys., 468 U.S. at 853, found as much in 1984.

(3) Whether the sanction comes into play only on a finding of scienter: A drug conviction is required to trigger ineligibility, and under the overwhelming majority of state and federal drug laws mens rea is a material element of the crime. The Supreme Court recently reaffirmed that “[t]he existence of a scienter requirement is customarily an important element in distinguishing criminal from civil statutes.” Hendricks, 521 U.S. at 362.

(4) Whether the sanction’s operation will promote the traditional aims of punishment, including retribution and deterrence: As noted, proponents supported both the 1988 and 1998 bills on the basis of their retributive and deterrent benefits. Supreme Court doctrine views retribution as a clear indication of punitive intent, but has recently given mixed signals on the significance of a statute’s deterrent purpose. In Bennis, 516 U.S. at 452, the majority found that civil statutes may also aim to deter, and that the forfeiture statute at issue served “a deterrent purpose distinct from any punitive purpose.” Yet in subsequent forfeiture case, the Court found that the statute’s deterrent purpose has “traditionally been viewed as a goal of punishment.” Bajakajian, 524 U.S. at 385.

(5) Whether the behavior to which the sanction applies is already a crime: The sanction only applies to those who have been convicted of a drug crime.

(6) Whether an alternative purpose to which it may rationally be connected is assignable to the sanction: We have noted other purported goals of the Souder Amendment, such as the promotion of safe campuses, but whether HEA ineligibility is a rational means for realizing those goals is doubtful. See supra notes 137-42.

(7) Whether the sanction appears excessive in relation to the alternative purpose assigned: Those challenging the Souder provision will argue that it makes a college education unavailable to many people convicted of no more than possession of a small amount of marijuana, a grossly excessive and punitive sanction in light of any legitimate governmental purpose.


How to Construct an Underclass 107

conviction was effectively a second punishment for the same offense because, inter alia, the tax was disproportionately high, had a deterrent purpose, was “imposed on criminals and no others” and “only after the individual [was] arrested for the precise conduct that gives rise to the tax obligation,” was levied by the same sovereign that criminalized the activity, and the revenue it produced could have been obtained in other ways.194

There is a second distinctive characteristic of the HEA ineligibility statute that should be at the forefront of any legal argument: withdrawing educational access to an offender is a truly extraordinary sanction in that it constitutes a deliberate and devastating roadblock to rehabilitation. To strip a person of an education is nothing less than an assault on her potential and dignity as a human being, and as a contributing member of the community. As this reality takes its toll on both offenders and society, we can hope that punishment by educational deprivation will eventually be recognized not only as punishment, but as a particularly unacceptable form of punishmentó one that is literally both cruel and unusual.195 When the Supreme Court found another exclusionary sanction, revocation of citizenship, unconstitutional under the Eighth Amendment, it condemned that sanction as constituting “the very antithesis of rehabilitation, for instead of guiding

194. Dep’t of Revenue of Montana v. Kurth Ranch, 511 U.S. 767, 780-83 (1994). In addition to these factors, which parallel aspects of the Souder Amendment, the two cases each include one unique factor. In Kurth Ranch, the Court found that Montana’s tax departed radically from typical tax laws by taxing contraband goods which might have been destroyed before the tax was imposed. Id. at 781. In the HEA ineligibility case the same sanction was previously applied as a criminal sentence for the same activities.

The discussion in Kennedy is also helpful here. Kennedy, 372 U.S. at 144. There the court found unconstitutional a law revoking the citizenship of certain draft evaders, distinguishing the sanction at issue from truly civil citizenship revocation laws: “The [truly civil] statutes . . . provided loss of citizenship for noncriminal behavior instead of as an additional sanction attaching to behavior already a crime, and congressional expression attending their passage lacked the overwhelming indications of punitive purpose which characterized the enactments here.” Id. at 170 n.30.

195. Unlike the constitutional prohibitions on double jeopardy, ex post facto laws and bills of attainder, all of which are expressly limited to criminal cases, the Eighth Amendment limits both civil and criminal sanctions. According to a Supreme Court case holding that the Excessive Fines Clause applies to civil forfeiture

[t]he purpose of the Eighth Amendment, putting the Bail Clause to one side, was to limit the government’s power to punish. The Cruel and Unusual Punishments Clause is self-evidently concerned with punishment. The Excessive Fines Clause limits the government’s power to extract payments, whether in cash or in kind, “as punishment for some offense.” “The notion of punishment, as we commonly understand it, cuts across the division between the civil and the criminal law.” . . . Thus, the question is not, as the United States would have it, whether forfeiture under ßß 881(a)(4) and (a)(7) is civil or criminal, but rather whether it is punishment. . . . For this reason, the United States’ reliance on Kennedy v. Mendoza-Martinez and United States v. Ward is misplaced. The question in those cases was whether a nominally civil penalty should be reclassified as criminal and the safeguards that attend a criminal prosecution should be required. . . . In addressing the separate question whether punishment is being imposed, the Court has not employed the tests articulated in Mendoza-Martinez and Ward.

Austin v. United States, 509 U.S. 602, 609-10, 610 n.6 (1993) (citations omitted).


108 The Journal of Gender, Race & Justice [6:2002]

the offender back into the useful paths of society it excommunicates him and makes him, literally, an outcast.”196 So does any law specifically designed to exclude offenders from an education.


We conclude with a final comment on the utility of mounting these legal challenges. While retroactive enforcement may be quite vulnerable to legal attack, we do not underestimate the difficulty of prevailing on the more sweeping claims, nor the importance of asserting them nevertheless. For two decades now, court decisions have often tended to reduce social realities into doctrinal categories that obscure constitutional values and define away fundamental injustices. We have argued that denial of federal college benefits to drug offenders is punishment in violation of several constitutional guarantees, but that challenge may be blocked if courts choose to define such sanctions as “civil disabilities” that are not “punishment.” Similarly, we believe an equal protection claim should be brought to reclaim the equality of opportunity that is so dependent on educational access, but we recognize that many judges now view the equal protection clause as having little to say about caste or subjugation.197 Laws that create an uneducated underclass may be of no constitutional significance to those judges who share Justice Thomas’ view that “there is a moral and constitutional equivalence between laws designed to subjugate a race and those that distribute benefits on the basis of race in order to foster some current notion of equality.”198 Those consigned to the underclass by deliberate educational privation confront a body of constitutional law that in various ways has grown ill-suited to the democratic and egalitarian values it is assumed to serve. The ultimate defeat,

196. Trop v. Dulles, 356 U.S. 86 (1958). In Trop, the government argued that a serviceman who deserted in wartime had forfeited his citizenship. Finding that this was a punishment that “strips the citizen of his status in the national and international political community,” the Court held the sanction unconstitutional.

[T]he existence of the death penalty is not a license to the Government to devise any punishment short of death within the limit of its imagination. The basic concept underlying 8th Amendment is nothing less than the dignity of man. While the state has the power to punish, the Amendment stands to assure that this power be exercised within the limits of civilized standards. . . .

Id. at 99, 100. Trop was the first case applying the Eighth Amendment bar to a sanction that did not include physical punishment. Nor must the sanction issue from a sentencing court. See Estelle v. Gamble, 429 U.S. 97, 103 (1976) (applying the Eighth Amendment to deprivation suffered during imprisonment).

197. Cass R. Sunstein, The Anticaste Principle, 92 MICH. L. REV. 2410 (1994). Sunstein argues for an “anti-caste” interpretation of the Fourteenth Amendment and decries its replacement by the purely formal “notion that the law forbids unreasonable distinctions [that] . . . sometimes requires identical treatment in cases in which distinctions make sense, and ignores inequality when inequality is present. Sometimes the cause of equality requires people who are differently situated to be treated differently, and this is a major gap in constitutional doctrine.” Id. at 2455.

198. Adarand Constructors, Inc. v. Pena, 515 U.S. 200, 240 (1995) (Thomas J., concurring).


How to Construct an Underclass 109

however, would be to allow that legal system to deter its victims from even seeking justice. One project well suited to law reviews such as this one, and to committed academicians and practitioners generally, is to seek ways to reconnect legal doctrine to these values, and reclaim some legal space for their concerns.199

199. No one has expressed this hope more eloquently than Charles Black, responding to those who found the Supreme Court’s desegregation decisions constitutionally unjustified.

It seems that what is being said is that, while no actual doubt exists as to what segregation is for and what kind of societal pattern it supports and implements, there is no ritually sanctioned way in which the Court, as a Court, can permissibly learn what is obvious to everybody else and to the Justices as individuals. But surely, confronted with such a problem, legal acumen has only one proper taskñthat of developing ways to make it permissible for the Court to use what it knows. . . .

Charles Black, The Desegregation Decisions, 69 YALE L.J. 421, 427-28 (1960).