There are currently over a million people enslaved in the United States. Under threat of horrendous punishment, they cook, clean, and even fight fires. They do this not in the shadow of the law but with the express blessing of the Thirteenth Amendment’s Except Clause, which permits enslavement and involuntary servitude as punishment for a crime.

Despite discussions of this exception in law reviews, news reports, and Netflix documentaries, few commentators have recognized that this enslavement happens silently. No prosecutor, judge, or defense attorney tells convicted people that they will be enslaved as punishment for their crime. It is only once they are incarcerated that a prison administrator informs them they will be forced to work.

This Article uncovers how this state of the world has come to be. It argues that our current regime is one of administrative enslavement: a constellation of judicial and legislative choices that places the punishment of enslavement outside the scope and processes of our traditional criminal punishment structure and into the hands of prison administrators. This Article is the first to provide a taxonomy of the administrative-enslavement regime. It uncovers the weak jurisprudential underpinnings of that regime, and it surveys all fifty states’ and the federal government’s legislative implementation of the Except Clause. It concludes by utilizing this taxonomy to analyze administrative enslavement’s legal weaknesses as well as how the status quo might evolve in the face of growing attacks from states removing Except Clauses from their state constitutions.

The full text of this Article can be found by clicking the PDF link to the left.


In 2020, there were at least 600,000 slaves in the United States. 1 See ACLU & Univ. of Chi. L. Sch. Glob. Hum. Rts. Clinic, Captive Labor: Exploitation of Incarcerated Workers 5, 24, 47 (2022), [] [hereinafter Captive Labor] (estimating that, based on data from 2020, “at least 791,500 people incarcerated in U.S. prisons perform work as part of their incarceration” and 76.7% of those workers “are required to work” or “face additional punishment” (emphasis added)); see also id. at 112 n.170 (explaining the report’s methodology to arrive at the number of incarcerated people with work assignments). This 600,000 figure represents a minimum based on the number of incarcerated people forced to work under threat of punishment. More capacious definitions of slavery may more accurately capture the comparison between chattel enslavement and Except Clause enslavement. See infra text accompanying notes 38–40. They cooked. 2 Captive Labor, supra note 1, at 27–36 (categorizing types of prison labor). They cleaned. 3 Id. They did building maintenance and repair work. 4 Id. Some fought fires. 5 Id. at 30–31 (describing programs in thirteen states through which “[i]ncarcerated firefighters also fight wildfires”). And others, harking back to an age most thought long past, even picked cotton. 6 Daniele Selby, How the 13th Amendment Kept Slavery Alive: Perspectives From the Prison Where Slavery Never Ended, Innocence Project (Sept. 17, 2021), [].

These slaves, unlike many of their forebears, were not stolen from the coast of Africa or marked for this fate purely by dint of their birth. These people were enslaved by our criminal legal system: by prosecutors and judges empowered by our cities, counties, states, and nation. What’s more, they were almost uniformly enslaved by these carceral actors without a word that they were about to suffer this fate. 7 Some states have statutes that allow for an explicit sentence of hard labor, but these seem rarely used. See, e.g., Ohio Rev. Code Ann. § 5147.17 (2024) (allowing for sentences of hard labor alongside “the punishment of . . . imprisonment in the county jail or workhouse”); infra section II.A (overviewing the statutory placement and language of state provisions discussing prison enslavement). Indeed, it seems that even their advocates—their defense attorneys—made no mention that slavery was in their future.

This refers, of course, to the Thirteenth Amendment and its now infamous 8 Criticisms of the Thirteenth Amendment’s “Except Clause” have come from all corners, including popular media—see, e.g., 13th (Netflix 2016)—the legal academy—see, e.g., Michele Goodwin, The Thirteenth Amendment: Modern Slavery, Capitalism, and Mass Incarceration, 104 Cornell L. Rev. 899 (2019) [hereinafter Goodwin, Modern Slavery]; Alvaro Hasani, ‘You Are Hereby Sentenced to A Term of . . . Enslavement?’: Why Prisoners Cannot Be Exempt From Thirteenth Amendment Protection, 18 Barry L. Rev. 273 (2013); James Gray Pope, Mass Incarceration, Convict Leasing, and the Thirteenth Amendment: A Revisionist Account, 94 N.Y.U. L. Rev. 1465 (2019) [hereinafter Pope, Mass Incarceration]; Raja Raghunath, A Promise the Nation Cannot Keep: What Prevents the Application of the Thirteenth Amendment in Prison?, 18 Wm. & Mary Bill Rts. J. 395 (2009); Wafa Junaid, Note, Forced Prison Labor: Punishment for a Crime?, 116 Nw. U. L. Rev. 1099 (2022)—and public interest groups—see, e.g., Captive Labor, supra note 1; Selby, supra note 6. “Except Clause.” 9 Other commentators have called this portion of the Thirteenth Amendment the “Punishment Clause.” See, e.g., Goodwin, Modern Slavery, supra note 8, at 933; Pope, Mass Incarceration, supra note 8, at 1470. This Article uses the term “Except Clause” because it better encapsulates the current state of the world. The slavery and involuntary servitude discussed here is an “exception” to the norm, but it seems increasingly disconnected from the idea of “punishment.” Despite being billed as a wide-ranging prohibition on slavery, the Thirteenth Amendment states that “[n]either slavery nor involuntary servitude, except as a punishment for crime whereof the party shall have been duly convicted, shall exist within the United States, or any place subject to their jurisdiction.” 10 U.S. Const. amend. XIII, § 1 (emphasis added). The entire remainder of the Thirteenth Amendment states, “Congress shall have power to enforce this article by appropriate legislation.” Id. § 2.

The puzzle and the problem at the heart of this Article, though, is not the existence of prison slavery or involuntary servitude; that practice is clearly contemplated by the Thirteenth Amendment itself. Instead, what motivates this Article is the silent enslavement of hundreds of thousands of incarcerated persons in the country. One might think that the decision to enslave someone—particularly given this country’s history of violent and purportedly successful resistance to the institution of slavery, and within a criminal legal system that disproportionately ensnares the descend­ants of those whom the country historically enslaved—would be a somber one, made with deep thought and reflection. But instead, prosecutors, judges, and even defense attorneys seem to give this potentially momentous punishment no thought at all, despite its near-constant imposition.

Why is this? In a system in which defense attorneys and prosecutors litigate every arcane issue affecting the sentence a judge can impose, 11 See, e.g., Descamps v. United States, 570 U.S. 254, 257 (2013) (citing Armed Career Criminal Act, 18 U.S.C. § 924(e) (2018)) (explaining the categorical and modified categorical approaches to determining whether a past conviction qualifies to enhance a current sentence under the Armed Career Criminal Act). judges fiercely guard their discretion to impose individualized sentences, 12 See, e.g., Rachel Martin, A Federal Judge Says Mandatory Minimum Sentences Often Don’t Fit the Crime, NPR (June 1, 2017), [] (interviewing Judge Mark Bennett on his opposition to “mandatory minimum charging and sentencing guidelines for nonviolent drug offenses”); see also Kimbrough v. United States, 552 U.S. 85, 91 (2007) (holding that judges may depart from the guidelines based solely on a policy disagreement with them); United States v. Booker, 543 U.S. 220, 227 (2005) (holding unconstitutional the mandatory imposition of the United States Sentencing Guidelines). and something as miniscule as a five-dollar special assessment is men­tioned in the pronouncement of a sentence, 13 See 18 U.S.C. § 3013(a)(1)(A)(i) (2018) (“The court shall assess on any person convicted of an offense against the United States . . . the amount of $5 in the case of an infraction or a class C misdemeanor . . . .”). why does the fact that so many convicted defendants are about to be enslaved go unmentioned?

Past commentators have suggested that broader societal forces have pushed us here. Maybe capitalism is to blame, or racism, or the other systems that create the hierarchies within our society. 14 See, e.g., Michele Goodwin, The Thirteenth Amendment’s Punishment Clause: A Spectacle of Slavery Unwilling to Die, 57 Harv. C.R.-C.L. L. Rev. 47, 50–53 (2022) [hereinafter Goodwin, A Spectacle of Slavery] (arguing that prison slavery authorized by the Except Clause is an example of “the stunning insistence in law itself on the subordination of Black Americans” and suggesting ways to end it (citing Jones v. Mayer Co., 392 U.S. 409, 445 (1968) (Douglas, J., concurring))).

Or maybe we should look to the personal instead of, or in addition to, the societal. Perhaps there are psychological and social reasons for this phenomenon. All of these people—legislators, prosecutors, defense attorneys, judges—may simply want to think of themselves as good people, 15 See Carolyn Kaufman, Why Bad Guys Think They’re Good Guys, Psych. Today (Aug. 12, 2012),
why-bad-guys-think-theyre-good-guys (on file with the Columbia Law Review) (“By convincing themselves their behavior is moral . . . people can separate and disengage themselves from immoral behavior and its consequences.”).
and focusing on their role in enslavement makes that more difficult. 16 Indeed, ignoring or distorting the full consequences of one’s actions is far from a novel phenomenon in the context of American slavery. See, e.g., David Pilgrim, The Mammy Caricature, Ferris St. Univ. Jim Crow Museum (Oct. 2000),
HTMLS/news/jimcrow/mammies/homepage.htm [] (last updated 2023) (“From slavery through the Jim Crow era, the mammy image served the political, social, and economic interests of mainstream white America. . . . Her wide grin, hearty laugher, and loyal servitude were offered as evidence of the supposed humanity of the institution of slavery.”).
After all, even without considering enslavement, judges routinely remark that sentencing is the hardest part of their job. 17 See, e.g., Mark Scarcella, The Hardest Thing About Being a Judge? What Courts Say About Sentencing., Conn. L. Trib. (Feb. 19, 2020), (on file with the Columbia Law Review) (last updated Feb. 20, 2020) (“As judges are often quick to say, sentencing defendants ranks near or at the top of the most challenging parts of serving on the bench.”); Benjamin Weiser, A Judge’s Education, a Sentence at a Time, N.Y. Times (Oct. 7, 2011), (on file with the Columbia Law Review) (discussing Judge Denny Chin’s sentencing approaches and decisions).

But this Article is not about broader societal forces or carceral actors’ unspoken psychological motivations. It is about the legal regime that has enabled enslavement as default. Presumably, if the law said that at each sentencing the judge must announce whether a defendant was to be enslaved and explain the reasons for that decision, that is what judges would do. But our current legal interpretations require no such thing. This Article seeks to uncover what the law does require and to tell a thus-far unappreciated story of how it came to be that way.

What this analysis finds is not a bombshell or a smoking gun. Instead, it shows that our current system of prison slavery is built on the sorts of mundane processes and decisions that seem small and unimportant individually but, in the aggregate, create a regime that this Article calls administrative enslavement.

For nearly a century, the federal courts have almost uniformly stated that the only trigger necessary for the Thirteenth Amendment’s Except Clause is a conviction. 18 See infra section I.B (tracing the development of this broad Except Clause reading). The standard processes that apply to the taking of a plea or pronouncement of a sentence have no purchase here. 19 See infra section III.A. There is no requirement, for example, that a defendant be told that a conviction carries with it the loss of Thirteenth Amendment rights as part of the punishment or that a sentencing judge (or legislature) offer any reason for why that punishment is appropriate. 20 Cf. 18 U.S.C. § 3553(c) (2018) (requiring the court to “state in open court the reasons for its imposition of the particular sentence”). Indeed, there may not even need to be a statute on the books imposing the punishment.

This permissive interpretation of the Except Clause did not come about through any sort of grand doctrinal innovation but through the slow march of common law decisionmaking. In cases across the federal courts, judges faced primarily with zealous—indeed, relentless—pro se and imprisoned litigants made broad, unreasoned pronouncements about the Except Clause. 21 See, e.g., Draper v. Rhay, 315 F.2d 193, 197–98 (9th Cir. 1963) (“The Thirteenth Amendment has no application where a person is held to answer for a violation of a penal statute.” (citing Blass v. Weigel, 85 F. Supp. 775, 785 (D.N.J. 1949))). Those pronouncements then became the basis for courts throughout the country to dismiss challenges to enslavement-as-punishment, even when facing novel arguments. 22 See, e.g., Murray v. Miss. Dep’t of Corr., 911 F.2d 1167, 1168 (5th Cir. 1990) (per curiam) (finding no rights-based distinction between forced prison labor on public and private property). Narrower readings of the Except Clause occurred almost entirely in cases in which the plaintiffs were represented. 23 See Davis v. Hudson, No. 00-6115, 2000 WL 1089510, at *3 (10th Cir. Aug. 4, 2000) (unpublished table decision) (“[T]here might be circumstances in which the opportunity for private exploitation and/or lack of adequate state safeguards could take a case outside the ambit of the Thirteenth Amendment’s state imprisonment exception or give rise to Eighth Amendment concerns . . . .”); Watson v. Graves, 909 F.2d 1549, 1552 (5th Cir. 1990) (“We agree that a prisoner who is not sentenced to hard labor retains his thirteenth amendment rights . . . .”); Craine v. Alexander, 756 F.2d 1070, 1075 (5th Cir. 1985) (suggesting, in the context of a 42 U.S.C. § 1994 (2018) anti-peonage suit, that an imprisoned person might state a claim “by virtue of labor forced upon him by a custom or usage of the state that is, at the same time, outside the scope of a corrective penal regimen”). The common law, when combined with the realities of pro se and prison litigation, became a one-way ratchet to restrict the rights of imprisoned people.

This one-way ratchet has, in turn, allowed states and the federal government to create statutes and regulations that require all incarcerated people to be enslaved, most visibly through their forced labor. 24 To be clear, this Article does not mean to suggest a causal story about how the courts’ jurisprudence led to these statutes (or vice versa). Instead, it simply means that this jurisprudence and these statutes coexist in a way that allows for this particular structure to flourish. Though the Except Clause explicitly states that slavery and involuntary servitude are only allowed as “punishment,” nearly every federal and state provision regulating prison enslavement is contained within the portion of the code dedicated to prison administration. 25 This Article distinguishes provisions that call for a sentence of “hard labor” because while almost every imprisoned person can be forced to labor under the general prison slavery regimes on which this Article focuses, conviction under a statute calling for a specific sentence of hard labor is comparatively rare and so not relevant for the vast majority of imprisoned people. See infra section II.A. Functionally, what results is that none of the preconviction process usually attached to criminal punish­ment occurs for the punishment of slavery, and it is instead controlled almost entirely by prison administrators. 26 See infra sections I.B.2, II.A.2, III.A.

Administrative enslavement is this systemic, broad jurisprudential reading of the Except Clause combined with legislation transferring prison-slavery decisions into the hands of prison bureaucrats. Contrary to the usual notions of criminal punishment, the administrative-enslavement regime requires no notice that this punishment will be imposed, no explanation of why it is appropriate, and no decision by a judge or jury.

The rest of this Article proceeds in three parts. Part I introduces the Thirteenth Amendment, the Except Clause within it, and the commentary that has analyzed its role in our law and society. It does this with an eye toward the question: How have we gotten to where we are today? While most commentators focus on “big issues” to answer this question—race, capitalism, and maintaining the hierarchies of social and economic control those systems entail—this Article suggests that it is through small, mundane, and rarely noticed decisions that courts and legislatures have built the administrative-enslavement legal regime that allows these “big issues” to flourish. To highlight these decisions, Part I traces modern Except Clause cases to their origins. In doing so, it uncovers how the previous story told about these cases was incorrect and how the real story is much more troubling. Starting with bare statements and citations to largely inapposite precedent, the courts developed an Except Clause jurisprudence that slowly but surely constricted the rights of imprisoned people, typically in response to the pro se imprisoned litigants who brought challenges to their enslavement. The courts did so with little reasoning, often waving away novel pro se arguments in the process.

Part II shifts from the courts to the statute books. It reviews how prison labor has been enacted and regulated in all fifty states and in the federal code, and creates a taxonomy of those laws. What it finds is striking: Statutes in almost every jurisdiction in the United States treat prison slavery as a piece of prison administration as opposed to a criminal punishment. Prison-slavery statutes are located in parts of the code distinct from those that set out criminal punishments. What’s more, they do not empower the judiciary to impose this punishment; instead, they almost uniformly empower prison administrators. To the extent that the statutes mention punishment at all, it is through the lens of rehabilitation. Often, however, they state that incarcerated people should work for idleness-prevention and cost-saving reasons. Part II also discusses other statutory design features that, while currently dormant, will likely become relevant if the administrative-enslavement regime comes under attack. These are whether a prison-labor statute imposes labor through mandatory or permissive language and the (for now) rare statutes explicitly stating that some or all prison labor must be voluntary.

While Parts I and II merely illuminate the current state of the world, Part III seeks to change it. To that end, it sketches a number of arguments that might end, or at least contract, the administrative-enslavement regime. It argues that administrative enslavement is constitutionally sus­pect on numerous grounds from both living constitutionalist and originalist frames. Turning to practice, Part III suggests how prosecutors and defense attorneys might use plea bargaining to disrupt administrative enslavement by allowing accused people to bargain to retain their Thirteenth Amendment rights. Finally, Part III looks toward the future to analyze how the courts, legislatures, and prison administrators who have created the status quo might seek to maintain it as administrative enslavement comes under attack.

This Article comes at a particular moment in history. After well over a century of constitutional stasis, we have allowed the peculiar institution 27 Kenneth M. Stampp, The Peculiar Institution: Slavery in the Ante-Bellum South (1956). —which most imagined dead and gone—to instead evolve and recapture hundreds of thousands of people in its grasp. 28 See supra note 1 and accompanying text (showing that more than 600,000 people working in state and federal prisons must work or be punished); see also Captive Labor, supra note 1, at 5 (explaining that, for those workers required to work, the alternative is “fac[ing] additional punishment such as solitary confinement, denial of opportunities to reduce their sentence, and loss of family visitation, or the inability to pay for basic life necessities like bath soap”). But change is fomenting. In 2018, Colorado voted to amend its state constitution to prohibit slavery and involuntary servitude totally. 29 Bill Chappell, Colorado Votes to Abolish Slavery, 2 Years After Similar Amendment Failed, NPR (Nov. 7, 2018), []. In 2020, Utah and Nebraska joined in this movement. 30 Nikki McCann Ramirez, Four States Banned Slavery on Tuesday. One Voted to Keep It . . . Sort Of, Rolling Stone (Nov. 9, 2022), (on file with the Columbia Law Review). And in 2022, Alabama, Vermont, Oregon, and Tennessee did, too. 31 Id. In many of these states, the votes to entirely abolish slavery and involuntary servitude were overwhelming. Tennessee’s measure passed with nearly eighty percent of the vote, 32 See Tennessee Amendment 3 Election Results: Remove Constitutional Language Allowing Slavery as Punishment, N.Y. Times, (on file with the Columbia Law Review) (last updated Dec. 13, 2022). and Vermont’s passed with nearly ninety percent. 33 See Vermont Proposal 2 Election Results: Prohibit Slavery in State Constitution, N.Y. Times, (on file with the Columbia Law Review) (last updated Nov. 17, 2022). Now is a time when the possibility of truly ending slavery and involuntary servitude is not only imaginable but seemingly likely. 34 Several members of Congress, to no avail thus far, have also pushed to alter the Thirteenth Amendment. See Scottie Andrew, Democratic Lawmakers Introduce a Resolution to Amend the 13th Amendment to End Forced Prison Labor, CNN (Dec. 3, 2020),
index.html []; Elizabeth Crisp, Lawmakers on Juneteenth Push for Change to 13th Amendment, The Hill (June 6, 2022), (on file with the Columbia Law Review).
Attacking, and ending, administrative enslavement is one important step toward that goal.

*    *    *

Before continuing, a note on terminology is warranted. This Article uses the terms administrative enslavement and prison slavery 35 The majority of this piece uses people-first language. Cf. Erica Bryant, Words Matter: Don’t Call People Felons, Convicts, or Inmates, Vera Inst. Just. (Mar. 31, 2021), [] (“[P]oliticians, media outlets, and more . . . still use harmful and outdated language like ‘convict,’ ‘inmate,’ ‘felon,’ ‘prisoner,’ and ‘illegal immigrant.’ There are better alternatives—alternatives that center a person’s humanity first and foremost.”). But occasionally, as in this introduction, it uses the term “slave.” This language highlights that, like chattel slavery before it, our current enslavement regime does create a status distinction between those people who fall within the Except Clause’s ambit and those who do not. Cf. Justin Driver & Emma Kaufman, The Incoherence of Prison Law, 135 Harv. L. Rev. 515, 525 (2021) (explaining the article’s choice of the term “prisoner” in part because “the term prisoner rejects the government’s appellations while underscoring that prisons are degrading spaces, where numbers replace names and humans live in barren cells”). While forced labor is perhaps the most obvious mark of this distinction, it is not the only one. See infra notes 43–46 and accompanying text. while also occasionally mentioning involuntary servitude. The choice to name this phenomenon “slavery” is intentional, as it accurately describes the system that is this Article’s subject. Nevertheless, there are several serious objections to this choice. Grappling with them explicitly will illuminate the relatively limited scope of this Article and the broad scope of the problems and systems it describes.

Objections to calling the current regime “slavery” might come from two directions. First, one might argue that the Thirteenth Amendment’s Except Clause in fact only authorizes involuntary servitude, not slavery, thereby making the Amendment’s prohibition on slavery total. 36 For further discussion of the distinction between involuntary servitude and slavery, see, e.g., Andrea C. Armstrong, Slavery Revisited in Penal Plantation Labor, 35 Seattle U. L. Rev. 869, 882–86 (2012) [hereinafter Armstrong, Slavery Revisited] (“Whereas in cases of involuntary servitude the servant must justifiably believe there is no alternative other than service, in slavery there simply is no other alternative, as the law stands ready to enforce the obligation.”). That this potentially major interpretative question has gone largely uninterrogated by the courts for over 150 years is one example of the lack of thought, here in the form of doctrinal stagnation, that this Article suggests administrative enslavement has enabled. Ultimately, there are reasonable arguments on both sides, 37 For arguments that the Except Clause permits slavery, see Scott W. Howe, Slavery as Punishment: Original Public Meaning, Cruel and Unusual Punishment, and the Neglected Clause in the Thirteenth Amendment, 51 Ariz. L. Rev. 983, 989–90 (2009) (“The actual language purports to allow both [slavery and involuntary servitude], however, and there were no voices in Congress that proclaimed for it during the promulgation period any other meaning.”), and compare the language of the Except Clause to similar language in Iowa’s constitution. Iowa Const. art. I, § 23 (“There shall be no slavery in this state; nor shall there be involuntary servitude, unless for the punishment of crime.”). By contrast, James Gray Pope has found that Senator Charles Sumner had shifting views on the Clause. While Sumner at one time believed that it might countenance slavery, he “later opined that the Senators had ‘supposed that the [Clause] was simply applicable to ordinary imprisonment.’” Pope, Mass Incarceration, supra note 8, at 1476 (quoting Cong. Globe, 39th Cong., 2d Sess. 238 (1867) (statement of Sen. Sumner)). and the answer to this question—while potentially momen­tous for the lives of imprisoned people—does not alter the analysis of administrative enslavement.

And while fully clarifying the distinction between involuntary servitude and slavery in this context is beyond the scope of this Article, it is worth briefly highlighting that the Article’s focus on forced labor is, in some ways, artificial. While forced labor for the benefit of another has always been at the core of American slavery, the institution included other pathologies that our current carceral system replicates. 38 See, e.g., Gabriel J. Chin, The New Civil Death: Rethinking Punishment in the Era of Mass Conviction, 160 U. Pa. L. Rev. 1789, 1790–93 (2012) (discussing how collateral consequences of a conviction create a new civil death); Alexandria Gutierrez, Sufferings Peculiarly Their Own: The Thirteenth Amendment, in Defense of Incarcerated Women’s Reproductive Rights, 15 Berkeley J. Afr.-Am. L. & Pol’y 117, 123–24 (2013) (connecting the lack of abortion rights for imprisoned women to “chattel breeding” in slavery); Priscilla A. Ocen, Punishing Pregnancy: Race, Incarceration, and the Shackling of Pregnant Prisoners, 100 Calif. L. Rev. 1239, 1245 (2012) (connecting the shackling of incarcerated pregnant people to “Black women’s subjugation during slavery” and other past eras of punishment); Dorothy E. Roberts, The Supreme Court, 2018 Term—Foreword: Abolition Constitutionalism, 133 Harv. L. Rev. 1, 34 (2019) [hereinafter Roberts, Abolition Constitutionalism] (describing how “[a] large body of social science literature explains criminal punishment as a form of social control of marginalized people”). For that reason, this Article does not identify a precise number of people that our carceral system has enslaved. At a minimum, the hundreds of thousands of people currently forced to work while incarcerated seem clearly within the Except Clause’s ambit. 39 See supra note 1 and accompanying text (showing that more than 600,000 people working in state and federal prisons must work or face punishment). Briefly comparing incarcerated people’s experiences to how the Thirteenth Amendment protects people who have not been convicted of a crime illuminates why this figure appears to be an appropriate minimum. The Thirteenth Amendment’s protection against slavery and involuntary servitude, though containing other labor protections, most prominently takes the form of an ever-present option to quit. See Pollock v. Williams, 322 U.S. 4, 17–18 (1944) (“The undoubted aim of the Thirteenth Amendment as implemented by the Antipeonage Act was not merely to end slavery but to maintain a system of completely free and voluntary labor . . . . [I]n general the defense against oppressive hours, pay, working conditions, or treatment is the right to change employers.”); James Gray Pope, Contract, Race, and Freedom of Labor in the Constitutional Law of “Involuntary Servitude”, 119 Yale L.J. 1474, 1478–79 (2010) [hereinafter Pope, Contract, Race, and Freedom of Labor] (“One of [the unenumerated Thirteenth Amendment] rights, the inalienable right to quit work, is so prominent in our constitutional consciousness that it tends to overshadow other possibilities.”). While there are a few exceptions, see infra section I.C, as a general matter you cannot be punished if you refuse to work for someone. That is not to say you will not face consequences, including dire ones—perhaps you will lose some government benefits that have work requirements, receive a negative reference, or simply no longer have the means to provide for yourself—but you cannot be forced to work for any employer by the state or a private entity. By contrast, whether, how, and for whom imprisoned people work is decided overwhelmingly by prison administrators, and if those people refuse to do their assigned work, they will suffer a variety of punishments, often including solitary confinement. See Captive Labor, supra note 1, at 5–6. But a more capacious comparison between chattel enslavement and Except Clause enslavement might suggest that everyone who is incarcerated, or perhaps everyone who is on parole or probation, or has been convicted of a crime, has experienced the sort of status-based degradation of their place in civil society that previously marked those who were chattelly enslaved.

Second, one could argue that referring to the current regime of forced prison labor as enslavement belittles the experience of those who suffered through chattel slavery. I am particularly sensitive to this possibility, but I believe that referring to our current system as slavery is correct for three reasons. First, while chattel slavery may have been a particularly evil and extreme incarnation of slavery, it is not the only practice that warrants that label. 40 Indeed, African chattel slavery is not the only form of slavery that has existed on these shores. See, e.g., Hudgins v. Wrights, 11 Va. (1 Hen. & M.) 134, 135–36 (1806) (describing how in 1679 “an act passed declaring Indian prisoners taken in war to be slaves”). Slavery in various forms has existed in numerous cultures throughout human history. Even today, individuals, organizations, and governments fight against forced labor practices across the world that are rightly labeled slavery despite contours that differ from chattel enslavement. 41 See, e.g., Program to End Modern Slavery, U.S. Dep’t of State, [] (last visited Nov. 2, 2023); see also Nathan J. Robinson, The Clintons Had Slaves, Current Affs. (June 6, 2017), [] (noting how in attempting to draw fine distinctions, “‘involuntary servitude’ immediately begins to sound like little more than a euphemism for slavery, and many of the situations that modern anti-slavery advocates would consider to be slavery . . . do not necessarily include” the total intergenerational domination of chattel slavery).

Relatedly, this Article uses the term “slavery” here because courts have attempted to use the depth of the evil of chattel enslavement to constrict the Thirteenth Amendment’s reach. Because even practices that fit well within the label “involuntary servitude” were not “akin to African slavery,” the courts have allowed them to continue. 42 See Butler v. Perry, 240 U.S. 328, 332 (1916); see also infra section I.C (discussing the “exceptional” and housekeeping exceptions to the Thirteenth Amendment).

Finally, and perhaps most importantly, the term slavery is used here because it is a term that numerous imprisoned people have used to describe their experiences, 43 See, e.g., Kevin Rashid Johnson, Opinion, Prison Labor Is Modern Slavery. I’ve Been Sent to Solitary for Speaking Out, The Guardian (Aug. 23, 2018), [] (“I see prison labor as slave labor that still exists in the United States in 2018.”); Mitch Smith, Prison Strike Organizers Aim to Improve Conditions and Pay, N.Y. Times (Aug. 26, 2018), (on file with the Columbia Law Review) (“Much of the recent activism has focused on inmate pay, which can range from nothing at all in states like South Carolina and Texas to, at best, a few dollars for a day of hard labor in other places. Prisoners frequently refer to it as ‘slave labor[]’ . . . .”); Daniele Selby, How a Wrongly Incarcerated Person Became the ‘Most Brilliant Legal Mind’ in ‘America’s Bloodiest Prison’, Innocence Project (Sept. 17, 2021), [] (quoting Calvin Duncan, who was exonerated after twenty-eight years of incarceration, as saying: “When people say this is modern day slavery—this ain’t no modern day slavery” and “[t]his shit is slavery”); Daniele Selby, A Mistaken Identification Sent Him to Prison for 38 Years, But He Never Gave Up Fighting for Freedom, Innocence Project (Sept. 17, 2021), [] [hereinafter Selby, A Mistaken Identification] (describing Malcom Alexander’s experiences at Angola prison in Louisiana, stating that “[i]t was like you see in old pictures of slavery” and that “[w]e even had a quota we had to meet at the end of the day” (quoting Malcom Alexander)); Jailhouse Lawyers Speak, @JailLawSpeak, Twitter (Apr. 24, 2018), [] (demanding “[a]n immediate end to prison slavery” as a condition of ending a prison strike). experiences which too often reflect those of chattel enslavement. Indeed, their descriptions, which invoke traumas beyond merely being forced to work, 44 For example, one seemingly large difference between prison slavery and chattel slavery is its effect on families of those enslaved. But these may be differences of degree, not of kind. While chattel slavery was fiercely intergenerational, empirical studies have consistently found that having a parent imprisoned increases the likelihood that a child will also be imprisoned at some point in their life. See Albert M. Kopak & Dorothy Smith-Ruiz, Criminal Justice Involvement, Drug Use, and Depression Among African American Children of Incarcerated Parents, 6 Race & Just. 89, 92 (2016) (reviewing studies describing the notable impact parental incarceration has on their children’s criminal justice involvement). But perhaps more drastically, the two systems have similar family separation dynamics. Professor Dorothy Roberts has explained how the criminal legal and child welfare systems intersect and overlap to remove children from the care of incarcerated, disproportionately Black mothers and to place them into state-run and state-sponsored foster care. See Dorothy E. Roberts, Prison, Foster Care, and the Systemic Punishment of Black Mothers, 59 UCLA L. Rev. 1474, 1491–99 (2012). accord with the conception of slavery put forward by Professors Jack Balkin and Sanford Levinson as “more than simply being free from compulsion to labor by threats or physical coercion. Rather, the true marker of slavery was that slaves were always potentially subject to domination and to the arbitrary will of another person.” 45 Jack M. Balkin & Sanford Levinson, The Dangerous Thirteenth Amendment, 112 Colum. L. Rev. 1459, 1484 (2012). Though this Article focuses overwhelmingly on forced labor, it should not be lost that labor is only one way that the ever-present threat of domination manifests for convicted people. 46 See, e.g., Chin, supra note 38, at 1790–93 (“A person convicted of a crime, whether misdemeanor or felony, may be subject to disenfranchisement (or deportation if a noncitizen), criminal registration and community notification requirements, and the ineligibility to live, work, or be present in a particular location.” (footnotes omitted)); Sharon Dolovich, Cruelty, Prison Conditions, and the Eighth Amendment, 84 N.Y.U. L. Rev. 881, 891 (2009) (arguing that “[t]he state, when it puts people in prison, places them in potentially dangerous conditions while depriving them of the capacity to provide for their own care and protection” and so creates “cruel” prison conditions when it violates its “ongoing duty to provide for prisoners’ basic human needs”); Sharon Dolovich, Exclusion and Control in the Carceral State, 16 Berkeley J. Crim. L. 259, 261 (2011) (using the rise of life without parole sentences and supermax confinement to explain how “exclusion and control has emerged . . . as the animating mission of the carceral project” (footnote omitted)).