Prison Segregation: Symposium Introduction and Preliminary Data on Racial Disparities

University of Michigan Law School University of Michigan Law School University of Michigan Law School Scholarship Repository University of Michigan Law School Scholarship Repository Articles Faculty Scholarship Prison Segregation: Symposium Introduction and Preliminary Prison Segregation: Symposium Introduction and Preliminary Data on Racial Disparities Data on Racial Disparities Margo Schlanger University of Michigan Law School, [email protected] Available at: https://repository.law.umich.edu/facarticles/461 Part of the Civil Rights and Discrimination Commons, Human Rights Law Commons, Law and Race Commons, and the Law Enforcement and Corrections Commons PRISON SEGREGATION: SYMPOSIUM INTRODUCTION AND PRELIMINARY DATA ON RACIAL DISPARITIES By Margo Schlanger* The Michigan journal of Race & Law's symposium, Inhumane and Ineffective: Solitary Confinement in Michigan and Beyond, was held on February 2, 2013, at the University of Michigan Law School; it brought together a dozen speakers on the topic of solitary confinement in three panels ("Isolation and Mental Health"; "Crisis in Michigan"; and "Strategies for Reform"). In keeping with the mission of theJournal, this brief Introduction offers some preliminary data that suggest that in many states the harsh conditions of solitary confinement are probably disproportionately affecting prisoners of color. I then introduce the two related papers published in this issue, each addressing an ongoing civil rights lawsuit seeking change.

Prisoner isolation-whether named solitary confinement, supermax custody, disciplinary and administrative segregation, extreme isolation, or any other of several extant labels-is as old as the modern prison, which is to say, only a bit younger than the United States.I The key milestone in its most recent history was the Marion lockdown, in 1983, when the Federal Bureau of Prisons responded to the separate murders of two of the correctional officers at the U.S. Penitentiary in Marion, Ohio, with what became a permanent lockdown; this ushered in the now prevalent model of the "supermax." For the next two decades or so, supermaxes were all the fashion,2 notwithstanding the case-made vividly by the two articles that follow this Introduction-that routine prisoner segregation is extraordinarily costly, damaging to prisoners, often inhumane by any reasonable understanding of the term, and frequently ineffective in terms of reducing misbehavior or increasing prisoner or staff safety.

In 2005, in one of the few national empirical studies of the use of supermax confinement, sociologist Daniel Mears tallied supermax facilities in forty-four states housing nearly 25,000 prisoners3-close to 2 percent of state and federal prisoners.4 Remarkably little is known about who is housed in these facilities. In that very report, Mears asked, but could not answer, "What are the characteristics (e.g., age, sex, race/ethnicity, prior record and length-of-stay, and behavior that led to supermax confinement) of inmates placed in supermax facilities and have these characteristics changed over time?"5 Prisoners' rights advocates often surmise that solitary confinement facilities-even more than prisons generally-likely house a disproportionately high share of Black and Hispanic prisoners. With colleague William Bales, Mears later produced the only study of which I am aware examining the race question in any depth. Mears and Bales examined Florida prisoners' supermax stints from 1996 to 2001; their study finds a notably higher rate for supermax placement of Black compared to non-Black prisoners.6 Similarly, in 1994, Maryland's isolation unit prisoners were just 9.9 percent White, compared to a system-wide population that was 22.5 percent White.7 On the other hand, a study of Washington State Intensive Management Units in 1999 found that "IMU residents were similar to all Washington prisoners in the proportion who were white (71%), but had a lower proportion of African Americans (18% vs. 23 %) and a higher proportion of Native Americans (7% vs. 3%)."8 The Florida and Washington data just described are over a decade old; the Maryland data are even older. We have even less information about more recent years. But the New York Civil Liberties Union (NYCLU) has described racial disproportion in New York in 2012. In the NYCLU's lawsuit attacking New York State's use of what the complaint labels "extreme isolation,"9 the most recent complaint alleges violations of the Cruel and Unusual Punishments and Due Process Clauses, and it includes the following demographic information: [Department of Corrections] data also reveals that black individuals are more likely to receive SHU sentences, and to receive longer SHU sentences, as compared to individuals of other racial and ethnic groups. For example, in June 2011, black individuals accounted for approximately 62% of the individuals held at Upstate and Southport correctional facilities, where individuals with the longest SHU sentences are generally incarcerated. In contrast, approximately 49% of the general prison population is black.'0 And the NYCLU's 2012 report, Boxed In: The True Cost of Extreme Isolation in New York's Prisons, provides corresponding information for the entire state; it compares state demographics to state prison demographics to the demographics of New York's Special Housing Units." In addition, a study of 2007 California parolees found that compared to the total parolee population, parolees who had spent time in a supermax Special Housing Unit were somewhat less likely to be either White or Black, but substantially more likely to be Hispanic.12 For this Introduction, I undertake to look a bit more broadly at recent data. The best sources of demographic information about prisoners are the various surveys and censuses conducted by the U.S. Department of Justice Bureau ofJustice Statistics (BJS). While no BJS publication directly addresses the issue, and no BJS dataset allows its full analysis, it is possible to glean something from the most recent BJS prison census, the 2005 Census of State and Federal Adult Correctional Facilities.'3 I present in Table 1, below, data derived from that census for seven state facilities. I also include, for comprehensiveness, the information from the NYCLU report. (Even so, the table covers only a very small portion of the nation's tens of thousands of supermax prisoners.)

The table includes all the facilities in the 2005 prison census that meet all of the following criteria: * Reported physical security as "supermax." * Reported 80 percent or higher share of facility prisoners as housed in maximum (or higher) custody. * Provided demographic data for 95 percent or more of prisoners. (For additional details on the data presented, see the Data Note at the end of this essay.)

Given the limited available information, the table is merely suggestive-but it does support a working hypothesis of current racialized impact for isolated confinement. In four of the eight columns (Arkansas, Colorado, Connecticut, and New York), non-White prisoners are substantially overrepresented in the highlighted facilities; statistical testing confirms that the difference is statistically significant. (In three of the other four-Massachusetts, New Jersey, Rhode Island-the small overrepresentation is not statistically significant; likewise, the tiny proportion of underrepresentation in Maryland lacks statistical significance. For details, see the Data Note.) Of course evidence of disproportion does not demonstrate racial discrimination; it is possible that whatever disproportion exists has other explanations. But whether or not the source the demographic impact of supermax and similarly isolated custody seems to me worthy of analysis. In short, it seems high time for corrections researchers to follow the lead of the Michigan journal of Race & Law and more systematically ask the race questionl4 irn this area. American jails and prisons are themselves vastly racially skewed in their populations, and what we are likely to find is an even more extreme skew for those who are on the receiving end of isolated confinement's harsh effects. is detectable bias, Looking more broadly at carceral isolation, it seems that the combination of fiscal constraints, litigation, popular protest, and the new focus within corrections on evidence-based programming may be turning the tide. Across the United States, advocates-former prisoners, prisoners' rights lawyers, human rights activists, and others-have been working together on a campaign that the American Friends Service Committee kicked off in 2008 under the title "Stop-Max." The campaign, somewhat coordinated via social media'5 and more traditional techniques, links litigation,16 lobbying and hearings,'7 press,18 public vigils, blue ribbon commission reports and projects,19 professional standards,20 and symposia21 like this one, all in service of reversing the trend from the 1980s till very recent years of increasing, and increasingly long-term, use of prisoner segregation units. In a number of states, advocacy in and out of the courts has led to sharp population reduction or even closure of supermax units; most recently in Illinois, the state Supreme Court has finally upheld the Governor's decision to shut down the Tamms supermax prison after a multi-year campaign.22 The two articles published in this issue of the Journal simultaneously underscore the pressing need for and the obstacles to reform. Elizabeth Alexander and Patricia Streeter's paper23 grows out of two different litigated reform methods: a longstanding injunctive class action litigation to reform what used to be the State Prison of Southern Michigan24 and a damage action brought by a single, devastatingly harmed prisoner.25 The article does not exude much optimism that either lever will bring about much change in Michigan. And yet, Alexander and Streeter admirably continue to do what they can using the methods available to them. It is worth noting, moreover, that even if lawsuits are not the best inducement to the kind of reforms they seek-better, more humane treatment of prisoners in segregation-litigation does have particular strengths to counter the precise problem they highlight: the near invisibility of segregated prisoners. The units in question are often called "prisons within prisons" and Alexander and Streeter's article demonstrates the most basic mechanics of what makes this label apt. Prison walls obviously serve an incapacitory function,26 but they simultaneously frustrate public oversight and accountability. Alexander and Streeter show how the quite literally thicker and more opaque walls that isolate prisoners within segregated housing grease a path to dehumanization and neglect. Litigation's discovery and access rights counter this awful tendency. Without litigation, the public would know almost nothing about what goes on inside segregated prison housing, which would make other advocacy methods even harder than they are.

Deborah Golden's article, about the recently filed class action27 against the Federal Bureau of Prisons (BOP) attacking its supermax facility in Florence, Colorado, demonstrates another key contribution of litigation-this one stemming from the adversary process. Golden highlights what looks like abundant (though as yet untested) evidence that the BOP is lying or mistaken in its repeated assertions that prisoners with severe mental illness are not housed in ADX Florence, the facility in question. There are few avenues outside litigation by which to hold the federal government to account in this stringent a way. As Golden makes clear, for example, facts can blur and slide in all but the most focused congressional hearings. So, litigation-initiated clarification and publication of facts is alone a vital contribution. In this case, moreover, it seems plausible that reform may be more accessible than in the Michigan ones. If Golden and her co-counsel can demonstrate to the court-perhaps even if they can demonstrate to BOP's leaders-that major flaws riddle the extant systems for keeping prisoners with severe mental illness out of ADX Florence, the result could be relatively simple, and therefore capable of implementation by litigation: buttress the screening systems, and house those prisoners elsewhere. These papers are sad testimonials to what we can currently hopewith far more optimism than was justified even five years ago-are the fading days of the American supermax era.

PRISON SEGREGATION: SYMPOSIUM INTRODUCTION AND PRELIMINARY DATA ON RACIAL DISPARITIES DATA NOTE This note explains the sources and choices embedded in Table 1, supra. 1. Supermax population:

For each column that is not New York, supermax unit demographics relate to a single facility. These are: * Varner Supermax (Arkansas) * Colorado State Penitentiary (Colorado) * Northern Correctional Institution (Connecticut) * Massachusetts Correctional Institution-Cedar Junction (Massachusetts) * North Branch Correctional Institution (Maryland) * New Jersey State Prison (New Jersey) * High Security Center (Rhode Island)

Each of these facilities constitutes a single observation in the U.S. Department of Justice's Census of Adult State and Federal Correctional Facilities, 2005.28 As explained in the text, together, these seven observations are the only ones that satisfy each of three criteria: * Reported their physical security as "supermax" (variable V23 == 1) * Reported 80% or higher share of facility prisoners as housed in maximum (or higher) custody (V94/V78 > .8) * Provided demographic data for 95% or more of prisoners (using V83 to V90)

The data for Colorado are confirmed by an analysis of the demographics of administrative segregation through that system more generally.29 For New York, data are not for a single facility but are rather system-wide; the figures provided describe all prisoners housed in a Special Housing Unit, according to information provided to the New York Civil Liberties Union (NYCLU) by the New York Department of Corrections and Community Supervision (NY DOCCS). In addition, these data are dated January 1, 2012, and are taken from NYCLU's Boxed In: The True Cost of Extreme Isolation in New York's Prisons.30 The NYCLU obtained the data from the NY DOCCS by use of New York's Freedom of Information Law.31 2. Prison population:

For all the columns except New York, overall prison population demographic and numerical data are from the Bureau ofJustice Statistic's 2005 prison census. I totaled the demographic figures by state, having first dropped all federal facilities, all private facilities, and all community corrections facilities. I excluded federal facilities because none of the profiled facilities listed is federal. I excluded private facilities because they frequently house a mix of prisoners for different states and the Federal Bureau of Prisons. I excluded community corrections because it seemed inappropriate to compare supermax incarceration to community-correction-type halfway houses and the like.

For New York, as with the supermax data, general prison demographic data are from the NYCLU report, Boxed In, and are as ofJanuary 1, 2012. Using the same source allows maximum compatibility with the Special Housing Unit demographic data. For the overall prison population figure (which is not included in the Boxed In report), I rely on a NY DOCCS population document dated December 31, 2011, obtained and posted by the NYCLU.32 It is reassuring that the New York demographic figures from January 2012 are not terribly different from the 2005 Bureau of Justice Statistics prison census data that I used for other states. 3. Non-prison population data For all the non-New York columns in Table 1, the source of the non-prison population figures is the 2005 American Community Survey one-year estimate.33 I chose 2005 to align with the presented prison population data. The "White" row reports the estimated population, by state, of non-Hispanic Whites who listed only one race. The "Black" row reports the estimated populations of non-Hispanic Blacks who listed only one race-except in Rhode Island where it reports the estimated population of all non-Hispanics who listed Black as their race (whether it is the only race or one of two or more).

For New York, because prison population data are from December 31, 2011 and January 1, 2012, I used data from the 2011 American Community Survey, with the same race/ethnic categories.34 4. Comparison of demographics of supermax population and overall prison population.

To statistically test whether the difference between demographics of the profiled prisons and the rest of their state systems were (in 2005) statistically significantly different, I computed 95 percent and 99 percent confidence intervals for the proportion white in the rest of each relevant state system, and then observed whether the profiled supermax prison fit within or outside those confidence intervals. Table 3 presents the details. --- FOOTNOTES --- * Professor of Law, University of Michigan Law School. Thanks to Sonja Starr, John DiNardo, and (as always) Sam Bagenstos for helpful comments. 1.

Seegenerally MICHAEL IGNATIEFF, A JUsT MEASURE OF PAIN: THE PENITENTIARY IN THE INDUSTRIAL REVOLUTION, 1750-1850 (1978); ADAM JAY HIRSCH, THE RISE OF THE PENITENTIARY (1992). 2.

See, e.g., Daniel P. Mears, Evaluating the Effectiveness of Supermax Prisons, URBAN INST. ii (March 2006), available at http://www.urban.org/UploadedPDF/411326_supermax-prisons. pdf [hereinafter Mears, Evaluating Effectiveness]; Daniel P. Mears & Jamie Watson, Tovards a Fair and Balanced Assessment of Supermax Prisons, 23 JusT. Q. 232, 232 (2006); JOHN IRWIN, THE WAREHOUSE PRISON: DISPOSAL OF THE NEW DANGEROUS CLASS (2005); Daniel P. Mears, A Critical Look at Supermax Prisons, CORRECTIONS COMPENDIUM, 6-7, 45-49 (2005) [hereinafter Mears, Critical Look]; Chad S. Briggs et al., The Effect of Supermaximum Security Prisons on Aggregate Levels of Institutional Violence, 41 CRIMINOLOGY 1341 (2003); Leena Kurki & Norval Morris, Supermax Prisons, in CRIME AND JUSTICE: A REVIEW OF RESEARCH 385 (2001) (Michael H. Tonry ed.); Roy D. King, The Rise and Rise of Supermax: An American Solution in Search of a Problem?, I PUNISHMENT & SOC'Y 163 (1999); NAT'L INST. OF CORR., SUPERMAX HOUSING:

A SURVEY OF CURRENT PRACTICE (1997). 3. 4.

See Mears, Evaluating Effectiveness, supra note 2 at ii. JAMES J. STEPHAN, BUREAU OF JUSTICE STATISTICS, CENSUS OF STATE AND FEDERAL CORRECTIONAL FACILITIES, 2005, at 4 tbl. 3 (2008), available at http://www.bjs.gov/ content/pub/pdf/csfcf05.pdf (indicating that the confined prison population in 2005 was 1,375,975). 5.

Mears, Evaluating Effectiveness, supra note 2, at 48. 6.

See Daniel P. Mears & William D. Bales, Supermax Housing: Placement, Duration, and Time to Reentry, 38 J. CRIM. JUST. 545, 551 (2010). See generally Daniel P. Mears & William D. Bales, Supermax Incarceration and Recidivism, 47 CRIMINOLOGY 1131 (2009). 7.

Mears, Evaluating Effectiveness, supra note 2, at 17 (citing Nancy Moran, Maryland Correctional Adjustment Center: Allegation of Brutality and Allegation of Discrimination, PRISONERS AID ASSOCIATION OF MARYLAND, INC. (1994 . 8.

David Lovell et al., Who Lives in Super-Maximum Custody? A Washington State Study, FED. PROBATION J., Dec. 2000, at 33, 34, available at http://www.uscourts.gov/viewer.aspx? doc=/uscourts/FederalCourts/PPS/Fedprob/2000decfp.pdf. 9.

Second Amended Complaint at 1, Peoples v. Fischer, 1:11-cv-2694 (S.D.N.Y. Dec. 6, 2012), available at http://www.clearinghouse.net/chDocs/public/PC-NY-0062-0007.pdf.

SPRING 2013] 10.

Id. $ 33; see also id. T 1. 11.

N.Y. CIVIL LIBERTIES UNION, BOXED IN: THE TRUE COST OF EXTREME ISOLATION IN NEW YORK'S PRISONs 24 (2012), available at http://www.nyclu.org/files/publications/ nycluboxedinFINAL.pdf. 12.

Keramet A. Reiter, Parole, Snitch, or Die: California's Supermax Prisons and Prisoners, 1997-2007, 14 PUNISHMENT & Soc'Y 503, 550-51 (2012). 13.

U.S. DEP'T OF JUSTICE, BUREAU OF JUSTICE STATISTICS, CENSUS OF STATE AND FEDERAL ADULT CORRECTIONAL FACILITIES (2005): INTERUNIVERSITY CONSORTIUM FOR POLITICAL AND SOCIAL RESEARCH STUDY 24642 v.2 (Oct. 5, 2010), available at http://dx.doi. org/10.3886/ICPSR24642.v2.

TABLE 1: DEMOG1RAPHICS IN SELECTED SUPERMAX FACILITIES U.S. 2005 Ark. 2005 Colo. 2005 Conn. 2005 Md. 2005 Mass. 2005 N.J. 2005 N.Y.* 2011-12 R.I. 2005 Population (N) 1,038,363 11,416 14,272 17,928 16,626 9,297 23,176 55,197 2,632 1,825 4,293 White 66.8% 77.0% 71.9% 75.3% 59.1% 79.7% 63.0% 58.3% 79.1% 37.6% 52.6% 45.7% 29.4% 22.6% 45.5% 20.3% 48.3% _ 28.6% 35.5% 24.9% 41.0% 19.8% 41.9% Black 15.3% 3.5% 8.8% 28.4% 5.5% 12.8% 144% 6.3% 43.1% 45.7% 19.7% 42.9% 72.6% 27.1% 61.5% 49.5% 29.6% Supermax unit 1 70.7% 15.0% 43.7% 75.1% 34.2% 65.2% 59.0% 32.6% Hispanic 14.5% 4.7% 19.5% 10.9% 16.0% 1.4% 31.5% 27.0% 5.8% 0.1% 7.9% 15.3% 17.6% 10.9% 25.3% 17.7% 20.4% 0.4% 46.6% 43.9% 0.0% 22.7% 14.1% 24.7% 25.6% Other 6.8% 3.2% 3.0% 0.3% 02% 5.0% 4.9% 3.1% 0.7% 6.7% 4.6% 6.8% 2.2% 8.8% 0.5% 9.7% 3.6% 27% 1.7% 2.9% 0.5% 0.0% 2.0% 0.9% 17% 0.0% * New York data for the rows labeled "supermax unit" cover all prisoners housed on January 1, 2012 in Special Housing Units, statewide; New York non-prison population data are from December, 31, 2011.

In all other states, all data are from 2005. The profiled facilities are: Varner Supermax (Arkansas); Colorado State Penitentiary; Northern Correctional Institution (Connecticut);

Massachusetts Correctional Institution-Cedar Junction; North Branch Correctional Institution (Maryland); NewJersey State Prison; High Security Center (Rhode Island)

Sources (for details, see Data Note. infra):

Bureau ofJustice Statistics, Census of State and Federal Adult Correctional Facilities (2005)

U.S. Census Bureau, American Community Survey (2005)

U.S. Census Bureau, New York State Quick Facts (2010, 2011)

N.Y. Civil Liberties Union, Boxed In: The True Cost of Extretne Isolation in New York's Prisons (2012)

New York State Dep't of Corrections & Comm. Supervision, Daily Population Capacity Report (Dec. 31, 2011)

SPRING 2013] 14.

Cf Katharine T. Bartlett, Feminist Legal Methods, 103 HARv. L. REv. 829, 837-849 (1990) (analyzing the feminist method of "asking the woman question"). 15.

See Stop Solitary -Resources for Advocates, Am. Civ. LIBERTIES UNION, http://www. aclu.org/stop-solitary-resources-advocates (last visited Mar. 14, 2013); Stopmax, AM. FRIENDS SERVICE COMMITTEE, https://afsc.org/campaign/stopmax (last visited Mar. 14, 2013); News From a Nation in Lockdown, SOLITARY WATCH, http://solitarywatch.com (last visited Mar. 14, 2013); SUPERMAXED (Feb. 21, 2013), http://supermaxed.com. 16.

For a collection of the relevant cases, see Search Results, CIVIL RIGHTS LITIGATION CLEARINGHOUSE, general;caseCat-PC;searchlssues-263;orderby-caseState,%20caseName 2013). http://www.clearinghouse.net/results.php?saveRef=pl&search=source- (last visited Mar. 14, 17.

See, e.g., Reassessing Solitary Confinement: The Human Rights, Fiscal and Public Safety Consequences, Hearing before the Subcomm. on the Constitution, Civil Rights, and Human Rights of the S. Judiciary Comm., 112th Cong. (2012). 18.

See, e.g., Jeff Tietz, Slow Motion Torture, ROLLING STONE, Dec. 6, 2012, at 58; Atul Gawande, Hellhole, NEW YORKER, Mar. 30, 2009, at 36; Life in Solitary Confinement, NAT'L PUB. RADIO confinement.

(Jul. 26-28, 2006), http://www.npr.org/series/5584841/life-in-solitary- 19.

See, e.g., JOHNJ. GIBBONS & NICHOLAS DE B. KATZENBACH, CONFRONTING CONFINEMENT: A REPORT OF THE COMMISSION ON SAFETY AND ABUSE IN AMERICA'S PRISONS (2006), available at http://www.vera.org/pubs/confronting-confinement (report led to the Vera Institute's segregation reduction project, which is partnering with departments of corrections in Illinois, Maryland, and New Mexico to reduce use of segregation); see also Segregation Reduction Project, VERA INST. OF JUSTICE, http://www.vera.org/project/segregation-reduction-project (last visited Feb. 7, 2013). 20.

See, e.g., ABA STANDARDS FOR CRIMINAL JUSTICE: TREATMENT OF PRISONERS (2010), available at http://www.americanbar.org/publications/criminal-justice-sectionarchive/ crimjuststandardstreatmentprisoners.html; NATIONAL COMMISSION ON CORRECTIONAL HEALTH CARE, STANDARDS FOR MENTAL HEALTH SERVICES IN CORRECTIONAL FACILITIES (2008); AMERICAN PUBLIC HEALTH Ass'N, STANDARDS FOR HEALTH SERVICES IN CORRECTIONAL INSTITUTIONS VII.D (2003). 21.

Symposia have included the American Friends Service Committee's Stop-Max Kickoff in May 2008 (for details, see http://www.campusactivism.org/displayevent-2194.htm), and the New York Institute for the Humanities and the Institute for Public Knowledge at NYU's Should You Ever Happen to Find Yourself in SOLITARY, in November 2012 (for details, visit Should You Ever Happen to Find Yourself in SOLITARY, N.Y. INST. FOR HUMANITIES, http:// nyihumanities.org/event/should-you-ever-happen-to-find-yourself-in-solitary.). 22.

See Weems v. Appellate Court, 2012 WL 6163083, at *1 (Ill. Dec. 11, 2012) (upholding Governor Pat Quinn's decision to shut down Tamms by requiring dissolution of the preliminary injunction against that closure previously entered in American Federation of State County and Municipal Employees, Council 31 v. Weems). 23.

Elizabeth Alexander & Patricia Streeter, Mapping the Circles of Hell in Michigan, 18 MICH. J. RACE & L. 251 (2013). 24.

See Hadix v. Caruso (Hadix 1), 465 F. Supp. 2d 776, 778-95 (W.D. Mich. 2006). Stoudemire v. Mich. Dept. of Corr., No. 2:07-cv-15387, 2011 WIL 1303418, at *1 (E.D. Mich. Mar. 31, 2011) (describing the plaintiff's experience of entering prison with con- 25. trolled lupus and a treatable tendency to develop blood clots; by the time she left five years later, she had had both legs amputated and was at risk of losing an arm.).

SPRING 2013]1 26. Of course, that incapacitory function is exaggerated by a tendency to ignore withinprison crime. See, e.g., Kevin Bennardo, Incarceration's Incapacitative Shortcomings (Jan. 5, 2013) (unpublished manuscript) available at http://ssm.com/abstract=2191128, and sources cited id. at nn. 23 & 25. 27.

Complaint, Bacote v. Fed. Bureau of Prisons, No. 1:12-cv-01570-RPM (D. Colo.

Jun. 18, 2012), available at http://www.supermaxlawsuit.com/Complaint-and-Exhibits-Bacotev-Federal-Bureau-of-Prisons.pdf. For further information, see Cunningham v. Federal Bureau of Prisons, CIVIL RIGHTs LITIGATION CLEARINGHOUSE, http://www.clearinghouse.net/detail. php?id=12177 (last visited Mar. 22, 2013). 28.

U.S. DEP'T OF JUSTICE, supra note 13. 29.

See Maureen L. O'Keefe, Administrative Segregation From Within: A Corrections Perspective, 88 PRISON J. 123, 129-130, 133-134 (2008) (reporting demographics for the state prison system and all 1,207 administrative segregation prisoners in 2005). 30.

N.Y. Civil Liberties Union, supra note 11, at 22, 24.

SPRING; 2013]

TABLE 2: NEW YORK PRISON DEMOGRAPHICS: COMPARISON OF SOURCES Source BJS prison census, 2005 NY DOCCS, Jan. 1, 2012 Total prison population 58,097 55,197 White 20.3% Black 52.2% 49.5% Hispanic 25.3% Other 2.2% 2.7% 31.

The underlying documents are posted at N.Y. Civil Liberties Union, Library, BOXED IN, http://www.boxedinny.org/library/ (last visited Mar. 23, 2013). 32.

See N.Y. STATE DEPT OF CORRECTIONS & COMM. SUPERVISION, DAILY POPULATION CAPACITY REPORT (Dec. 31, 2011), available athttp://www.boxedinny.org/wp-content/ uploads/2012/10/2-daily-population-capacity-reports.zip (document is one of the included files). 33.

The data are available (by race/ethnic category, by state) at U.S. CENSUS BUREAU, http://www2.census.gov/acs2005/SPP/Race_AncestryHispanic Origin/ (last visited Mar. 22, 2013).

TABLE 3: STATISTICAL TESTING OF DEMOGRAPHIC DIFFERENCES, SUPERMAX AND OTHER PRISONS State Arkansas"

Colorado"

Connecticut"

Maryland Massachusetts New Jersey New York"

Rhode Island Supermax % white Other % white 95% conf. interval 99% conf. interval 28.6% 35.5% 24.9% 41.0% 19.8% 41.9% 53.6% 46.3% 29.8% 22.6% 45.5% 20.4% 48.5% 48.7% 43.7% 26.3% 18.7% 37.5% 16.2% 22.8% 36.1% 58.5% 48.8% 33.3% 26.5% 54.0% 23.6% 60.8% 46.8% 42.8% 25.0% 17.2% 34.0% 22.7% 28.0% 60.4% 49.7% 34.6% 28.0% 57.5% 26.1% 23.7% 68.9% " Difference in means is statistically significant at the .05 level. 34.

State and County QuickFacts, U.S. CENSUS BUREAU, http://quickfacts.census.gov/ qfd/states/360001k.html (select "demographic characteristics") (last visited Mar. 22, 2013); this information is derived from the ACS 2011 estimates.