There's No 13th Amendment Right to Abortion

A(n overdue) reply to Andrew Koppelman.

NOTE: A much shorter, more sedate version of this article was recently published by The American Conservative. If you just want the highlights of this article, in relatively temperate terms, start there.

Scene just after the Thirteenth Amendment passed the House, 31 January 1865.

Some scholars believe that a legal right to abortion is not just a good idea; they think the Constitution guarantees a legal right to an abortion.

These arguments, are, for the most part, very silly. They do not point to a constitutional provision that says, “The right to an abortion shall not be infringed,” because, of course, there isn't one, and no amendment saying so would have a prayer of passing today or at any time in the past. Instead, virtually all these arguments depend on "penumbras" or "substantive liberty interests" or "correlates of equal protection." These tools allow Professors of Law to scry unwritten rights in the Constitution in much the same way that Roman professors of haruspicy once scried omens in sheep intestines. This approach can discover anything in the Constitution a judge might desire to find—and, of course, always does. Under Roe v. Wade and Planned Parenthood v. Casey, American abortion law was built on precisely this type of wishcasting, plus a gluttonous amount of ipse dixit (the ancient legal maxim “because I said so”), and absurd abuses of the otherwise-sound principle of stare decisis.

Professor Andrew Koppelman (who last appeared on De Civitate just over a decade ago) is nearly the sole exception. He proposes that, in fact, the Constitution's text does contain a clause which, under its original public meaning, guarantees an abortion right. That clause is the Thirteenth Amendment, proposed and ratified in 1865, which reads:

Neither 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.

That’s right: according to Koppelman, the constitutional amendment that enacted Lincoln’s promise to make oppressed Black people “thenceforward and forever free” simultaneously ensured that unborn children would be thenceforward and forever oppressed.

Koppelman has been making this argument for nearly as long as I’ve been alive, in three separate papers: Forced Labor (1990), Forced Labor Revisited (2010), and Originalism, Abortion, and the Thirteenth Amendment (2012).¹ As Koppelman is the first to admit (in Forced Labor Revisited ‘10), few have ever taken his argument very seriously, even on his own side.² Nonetheless, Professor Koppelman suggests that two decades of everyone ignoring his argument probably means he’s right. He then adds:

Perhaps you, reading this right now, don't buy the argument. If so, I wish you would write to me and explain why. If there is a defect in the argument, no one has ever stated it in print. Hit me. I want you to.

I’ve been there myself, screaming into a silent void even though I’m clearly right. It’s maddening. No one should have to suffer like that.

Now, I am not a Professional Lawyer. I lack the time, resources, Westlaw subscription, credentials, and professional incentives to write a proper law review article for a peer-reviewed journal. (As long as this article looks, it’s not even as long as the first of Koppelman’s three papers on this topic.) My main legal qualification is my love of footnotes.³ However, since Koppelman has made this invitation, I’d like to offer some limited reply. Future writers might then be able to build on this, but, hopefully, this Blog Post provides a few useful takeaways, even in its current form.

I think this is important, because Koppelman’s is by far the best constitutional argument for abortion rights I’ve ever seen, and I’ve thought so since I encountered it in 2009. Koppelman’s argument for a constitutional right to abortion is the only one I’ve seen that’s actually an argument, not haruspicy.⁴ I think it’s wrong, and decisively so—but it is, at least, an argument.

Now that Roe has fallen, we can expect to see the “Thirteenth Amendment argument” making the rounds as abortion supporters are finally forced to try grounding their desire for a constitutional right to abortion in the actual Constitution, rather than… whatever Roe was doing. Koppelman himself notes the manifold failures of Roe in his introduction to Forced Labor ‘90.

Also, since no one has previously written an extended reply to Koppelman, I need to write one, or I will lose a side argument on Twitter.⁵

Koppelman’s Textual Argument (“Argument A”)

Professor Koppelman's primary argument, which he refers to as his “libertarian” argument, runs like this:

1. Involuntary servitude is unconstitutional. (from Amendment XIII)

2. The essence of involuntary servitude is “that control by which the service of one [person]⁶ is disposed of or coerced for another’s benefit” (quoting Bailey v. Alabama (1911)).

3. Abortion prohibitions mandate “forced pregnancy and childbirth.”

4. “Forced” pregnancy and childbirth “compel… the woman to serve the fetus,” which (from A2) constitutes that essence of involuntary servitude.

5. Therefore, abortion prohibitions mandate involuntary servitude. (from A3 & A4)

6. Therefore, prohibiting abortion is unconstitutional. (from A5 & A1)

Koppelman’s Original Meaning Argument (“Argument B”)

Koppelman makes a secondary argument, revolving around the “badges and incidents” of slavery, which he refers to as his “egalitarian” argument:

1. The Thirteenth Amendment, as it would have been originally understood, was intended to and did in fact grant Congress plenary power to outlaw not only “all forms of slavery and involuntary servitude, but also to eradicate the last vestiges and incidents of a society half slave and half free” (quoting 1968’s Jones v. Alfred H. Mayer Co.).

2. The “principal utility to the slaveholding class” of “black women of childbearing age” was their “ability to reproduce the labor force.” As such, they had no sexual freedom, even “through abstinence,” and were often “raped with impunity.” They were treated as “procreative vessels,” “reduc[ed] to sexual object[s]… to be raped, bred, or abused,” and “faced constant, coercive inducements to bear children.” The details,⁷ as always when one looks honestly at American slavery, are harrowing.

3. Given this history, “loss of control over... reproductive capacities” should be understood as “the state-enforced creation of the very indignity that enslaved women suffered.”⁸

4. Therefore, “loss of control over reproductive capacities” is a “badge, vestige, and incident of slavery,” which violates the original public meaning of Amendment XIII. (from B3 & B1)

5. Abortion prohibitions mandate “loss of control over reproductive capacities.”

6. Therefore, abortion prohibitions violate Amendment XIII. (from B4 & B5)

Is that a fair summary, Professor? One necessarily loses many subtleties and important distinctions when one summarizes 105 pages of law review articles into a dozen points, so I encourage you to read them yourself.⁹ I believe these claims are his bottom line, and I also think they are the strongest possible versions of his bottom line. This is a crisp, clear argument based on plenty of historical evidence, several important Supreme Court precedents, and the two guiding stars of sound legal construction: the literal meaning of the dead text, and the original understanding of that text by the public that debated and finally received it. Two cheers for Professor Koppelman.

Before coming to the meat of the Thirteenth Amendment argument, however, it will be profitable to consider two threshold issues: fetal personhood and violence.

The Argument Concedes Fetal Personhood

Let’s start here, because it’s short and sweet and perhaps too clever by half.

Professor Koppelman does not believe fetuses are persons in a moral sense. He concedes that, if the fetus “is in fact a person, a being with rights that others are bound to respect,” then it puts important parts of his argument under immense strain. Nevertheless, he insists that his argument “makes its constitutional case without any direct reliance on the position that a fetus is not a person.”¹⁰

Well, he’s right about that: his argument doesn’t rely on the position that the fetus is not a person. In fact, Koppelman’s argument relies on the position that the fetus is a person.

Koppelman’s argument (at A2) says that the essence of involuntary servitude is “that control by which the service of one [person] is disposed of or coerced for another’s benefit.” Specifically, he says that abortion prohibitions coerce the service of one person—the mother—for the benefit of another person—the fetus. Either the fetus is a person for the purposes of the U.S. Constitution (or at least the Reconstruction Amendments), or Koppelman’s entire Argument A collapses. If we accept that the fetus is a person for constitutional purposes, that keeps Koppelman’s argument alive, but it has enormous implications, given the Fourteenth Amendment's guarantee of a right to “life” and "equal protection" for all “persons”.¹¹

Koppelman recognizes these implications and tries to resist them. (We are about to go down a little bit of a rabbit hole. If you get dizzy at any point here, you can skip ahead to the next section.) Koppelman writes, in footnote 136 of Forced Labor ‘90:

Although it is because the fetus is recognized to be an entity separate from the woman that a law prohibiting abortion violates the Thirteenth Amendment, by compelling her to serve another’s private interests, this argument does not presuppose that the entity the woman is serving is a person. [In the seminal involuntary servitude case, Bailey v. Alabama,] Bailey, after all, was being compelled to serve a corporation, the Riverside Company, rather than a natural person. When Bailey was decided in 1911, the company was certainly protected as a “person” by the equal protection clause of the fourteenth amendment, see Santa Clara County v. Southern Pac. R.R., but this only shows that the concept of “personhood” is a complex one: an entity may be a person for some legal purposes but not others.

This defense does not strike me (the layman) as terribly persuasive. Indeed, it’s confusing, and, the more we look at it, the more confusing it gets.

It is true that there are associations that we count as “people” (or “juridic persons”) even though they aren't actually living, breathing individuals (“natural persons”). These associations include clubs, unions, corporations, trusts, monastic orders, and nation-states, among others. They are considered people under the law because the legal fiction of a “juridic person” is a useful, convenient way for actual people to exercise their rights and duties under law. Juridic persons are extensions of natural persons. As Blackstone explained in his famous Commentaries, several years before the letter “s” led its bloody war of conquest against the letter “f”:

…as the neceffary forms of invefting a feries of individuals, one after another, with the fame identical rights, would be very inconvenient, if not impracticable; it has been found neceffary, when it is for the advantage of the public to have any particular rights kept on foot and continued, to conftitute artificial perfons, who may maintain a perpetual fucceffion, and enjoy a kind of legal immortality… in order to preferve entire and for ever thofe rights and immunities, which, if they were grated only to thofe individuals of which the body corporate is compofed, would upon their death be utterly loft and extinct.

Thus, we do not (and cannot coherently) confer juridic personhood on, say, a bag of white rice, or the abstract concept of calculus. Personhood is for actual persons and associations of actual persons.

Koppelman seems to be saying in his footnote that the fetus is, for Thirteenth Amendment purposes, not a natural person but a juridic person. This seems incoherent. If the fetus is a juridic person—some kind of “association” of natural persons which collectively exercises rights on their behalf—what can this even mean?¹² If the fetus has limited personhood (in Koppelman’s argument) only because it is extending the rights of some community of natural persons… which natural persons are they?

Is Koppelman suggesting that the limited juridic fetal personhood he recognizes is a stand-in for the future natural person who (in Koppelman’s view) will come to exist only at birth? He seems to suggest this in an aside about Griswold at the end of footnote 136, where he contends (I think) that a State which banned contraception might be holding women in involuntary servitude to a future person.

This seems bizarre. To the best of my (layperson) knowledge, juridic personhood is exercised on behalf of currently living or deceased natural persons; I am aware of no case where juridic personhood is exercised on behalf of natural persons who have not yet come into being. The closest parallel I’m aware of is in inheritance law, but that runs counter to Koppelman’s argument: in inheritance law, the unborn child en ventre sa mere is “entitled to all the privileges of other persons,”¹³ but a (potential) child not yet conceived has no claim on anyone whatsoever, unless the potential child's prospective parents expressly impose such a claim on themselves (which is an exercise of their own rights, not the child’s).

Insofar as Koppelman’s argument is an originalist argument, it would seem that he must also contend that this odd view of fetal personhood was held by some plausible version of “the public” in the 1860s. This seems challenging. However, it is possible that I am not understanding his “aversive originalism” correctly.

Regardless, it seems that, for Koppelman’s argument to work at all, he must concede, as table stakes, that, from the time of conception¹⁴ forward, the fetus is a natural human person within the meaning of the Thirteenth and Fourteenth Amendments. Koppelman is very smart and more educated than I am. It’s possible he sees some way out of this quandary. But would his escape route cut through too much of the rest of his argument to hold it intact?

(Even if we accept Koppelman’s odd apparent view that fetuses are not natural persons, but limited juridic persons for constitutional purposes, this would still mean quite a revolution in abortion law. Juridic persons do not necessarily have a right to continued existence, but they have a variety of other rights and duties, including the right to have their interests represented in court and the right to equal protection under law. The pro-life movement could easily spend the next decade exploring the boundaries of those personal fetal rights in state courts.)

For most abortion arguments, the concession of fetal personhood would be fatal. As Roe v. Wade stated, “If the suggestion of personhood [under the Fourteenth Amendment] is established, the appellant’s case [seeking to establish abortion rights], of course, collapses.” However, Koppelman believes that his argument still holds up if the fetus is a person. I agree. The Thirteenth Amendment outlaws involuntary servitude universally. No person may hold another person in involuntary servitude. If abortion prohibitions hold mothers in involuntary servitude, then, as a legal matter, abortion prohibitions must fall—even though the mothers’ unborn children are persons. The concession that fetuses are persons puts large portions of Koppelman’s argument under severe strain, as Koppelman acknowledges. Other areas, where Koppelman depends on the ambiguity of fetal rights, are simply obliterated (like his claims about the burden of proof in Forced Labor Revisited ‘10). However, his core argument is untouched.

Before we move on, though, I wish to suggest mildly that the ease with which Koppelman intuitively recognized the fetus’s capacity to legally function as a natural person, building a hundred-page argument premised on fetal natural personhood and defending it for thirty years without ever really apparently noticing that he’d done that, is perhaps most easily explained by the possibility that the fetus actually is a natural person—and that, at least at a subconscious level, this is obvious to everyone with eyes to see, ears to hear, and a basic knowledge of modern embryology. It is very easy to accidentally slip into the idiom of talking about a fetus as a human person because, well, what else is she? A toenail? An alien? Legally speaking, her rights and duties are closely akin to those of a newborn.¹⁵ Whenever he considers the possibility of fetal personhood, Koppelman parades horribles at us,¹⁶ but this sort of fear has always led to the ugliest prejudices.

We should also note, in passing, that one of Koppelman’s motivations for developing the Thirteenth Amendment argument in the first place was to deliver a “constitutional case [for abortion rights] without any direct reliance on the position that a fetus is not a person.” He has succeeded in doing this, but only by conceding the position that the fetus is a person—at the very least a juridic person, and in all likelihood a natural person. This concession is a cost few pro-choicers are willing to pay. I think Koppelman has no choice but to do so, as must anyone else who advances a Thirteenth Amendment argument. However, for now, the core of his argument remains intact. We soldier on.

Abortions Are Violent

Even if Koppelman’s argument is correct on every point from here on out, it would not justify abortion as it is typically practiced in the United States today. Libertarian arguments for abortion, including Koppelman’s, typically present it as something akin to an eviction. “You are no longer welcome on my property / in my body / to my labor, so I will now compel you to leave.” If you evict someone from your house and he freezes to death on the street that very night, well, that’s not your problem (says the libertarian). By the same token, if you evict a fetus from your womb and he can’t survive in the outside world, again, that’s sad, but also not your problem.

On the other hand, suppose you submit a written notice of eviction to an elderly tenant whom you know to be blind. He's also disabled, and unable to leave his second-story unit unless physically assisted on the stairs.¹⁷ When the tenant fails to vacate the premises (because he can't read the eviction notice, thus has no idea you've evicted him, and couldn't leave even if he did), you enter his room, inject him with deadly poison, then dismember him, finally chucking the bloody limbs of his corpse out an upstairs window without ceremony.

Even the libertarians are going to put you in jail for murder!

It is possible to imagine an abortion procedure that looks something like the libertarian dream, where abortion is “merely” a deadly eviction, not a murder. In such a procedure, the fetal “tenant” would be extracted with something like a C-section or induced labor and given some reasonable degree of medical care, even some token amount of futile care. The fetus, if and when she dies, would then be treated like a human being, death and birth certificates filed, funeral arrangements made (by the State if the mother is unwilling). Naturally, an important government and medical priority in this world would be to develop technology to push back the frontiers of viability to very early dates, eventually culminating in the old sci-fi fantasy of artificial wombs where unwanted children can gestate safely without inconvenience to their mothers.

Abortion in the United States does not resemble this. Surgical abortions are unbelievably violent. A dilation & extraction literally tears a fetus limb-from-limb and tosses them out one bit at a time. A dilation & curettage stabs the fetus to death, then dismembers her and “scoops” the mutilated corpse out. An aspiration abortion uses a vacuum to dismember the fetus. The kindest version of abortion we have uses digoxin to poison the fetus, killing her prior to dismemberment. This is typically only used on fetuses capable of directly experiencing pain, but abortion providers are reluctant to use it, because digoxin limits the resale value of the fetal body parts.

The closest thing we have to the libertarian “eviction model” of abortion is chemical abortion (called by advocates “medical abortion”) using misoprostol. Misoprostol induces labor, physically expelling the fetus from the uterus at a stage of pregnancy too early to survive—a straightforward “eviction.” Unfortunately, our chemical abortion protocol precedes the misoprostol with mifepristone, administered 1-2 days prior to the misoprostol. The mifepristone kills the fetus in the womb by cutting off its food and air supply prior to expulsion.

This is akin to murdering your tenant by refusing to allow DoorDash to come in so he starves and then also suffocating him by sucking all the air out of his room, then “evicting” the corpse on a stretcher. I’d rather be quietly suffocated than dismembered, I suppose, but you’re still going to libertarian jail for murder.

You can’t cut the mifepristone out of the protocol, either, because the whole point of chemical abortion is to allow mothers to “evict” their fetuses at home, on the toilet, with no need for any medical professionals to be around to care for the newborns when they come out. The fetuses therefore have to be killed in utero; if born alive, they would need medical care, and medical abortions could no longer happen at home.¹⁸

Abortion, as it is practiced today, is far more violent—and deliberately violent—than even the ugliest eviction process.

In my experience, libertarians typically wriggle out of this difficulty by retreating from it. They were trying to show that abortion is justified even if the fetus is a person. However, now that they see abortion is it is practiced today involves a great deal of unnecessary (and therefore unjustified) violence, they fall back to their real position that the fetus is not a person, so the violence doesn’t count.

However, Koppelman and others taking this approach are unable to make that retreat. As we’ve seen, the Thirteenth Amendment argument requires its supporters to accept the natural personhood of the fetus. Even if supporters continue to resist this conclusion, a major strength of the Thirteenth Amendment argument is that it is at least compatible with belief in the natural personhood of the fetus. Yet the version of abortion that the argument actually defends is not a version of abortion that exists in the present United States.

Nevertheless, it is possible to imagine a version of abortion that fits the Thirteenth Amendment argument, and we can still engage with that. Let’s finally get into the meat of the Thirteenth Amendment argument.

The Argument Proves Too Much

Your first reaction to the Thirteenth Amendment argument—Argument A especially—might be that it seems to prove too much. (That was John McGinnis’s reaction, as well as Michael Stokes Paulsen’s.)

Koppelman claims (in A1) that involuntary servitude is unconstitutional, and he claims (in A2) that involuntary servitude is defined as

that control by which the personal service of one [person] is disposed of or coerced for another’s benefit.

He provides some alternate definitions, all drawn from important court cases, to drive the point home: “involuntary servitude” is also defined as:

the control of the labor and services of one man for the benefit of another, and the absence of a legal right to the disposal of his own person, property and services (Plessy v. Ferguson)

and:

a condition of enforced compulsory service of one to another. (Hodges v. U.S.)

Koppelman claims that anything which fits these definitions constitutes involuntary servitude, that “coerced pregnancy” (FLR ‘10, p4) fits all these definitions, and that “coerced pregnancy” is therefore unconstitutional “involuntary servitude.”

However, we live in a society where able-bodied men can be conscripted into the military. Many of our grandfathers were conscripted. The military exercises absolute “control of the labor and services” over conscripts for the benefit of others, including their fellow citizens, the nation itself, the commander-in-chief, the military as a whole, or the conscript’s immediate superior officer. This appears to be “involuntary servitude,” if we accept Koppelman’s definition without further modification.

Koppelman resists this inference. In Forced Labor '90 (p519), he argues:

...the duties of citizenship, however burdensome, are inescapable conditions of freedom. A person cannot be a slave if he is as free as it is possible for a citizen to be. If "involuntary servitude" is defined as compulsory service to another private individual, as opposed to service to the state which guarantees civil liberty, then conscription for public duties simply falls outside the definition.

I don't think this quite grasps military service. If Lieutenant Smith orders Private Jones to get Lieutenant Smith a cup of coffee, purely for Smith’s benefit, Jones had better do it lickety-split or face prison for disobeying a direct order. One might argue that Smith's order is improper, because it does not “relate to military duty,” but good luck with that: the Manual for Courts-Martial (2019 ed., p331) states clearly that all orders are presumptively lawful and are "disobeyed at the peril of the subordinate" unless "patently illegal."

Let’s look at the structure of military service in light of Bailey v. Alabama (1911), a crucial Thirteenth Amendment case on which Koppelman heavily relies. His summary of the case is solid:

The case came to the Supreme Court as an appeal from a criminal conviction for fraud. Bailey, a black laborer, had accepted a fifteen dollar advance for signing a contract in which he agreed to work for a landholding corporation, the Riverside Company, for a year. Under the contract, he would earn twelve dollars a month, of which $1.25 would be deducted each month to repay the fifteen dollar advance. After about a month, Bailey left the job and refused to return to it. He then was prosecuted for defrauding the Riverside Company of fifteen dollars, convicted, and sentenced to 136 days of hard labor. While there was no evidence that he had intended to defraud the company, an Alabama statute provided that if one breached a service contract without refunding the money paid, fraud would be presumed. Under state rules of evidence, the accused was not permitted to testify about his intentions for the purpose of rebutting the presumption.

In reversing the conviction, Justice Hughes declared that “[w]ithout imputing any actual motive to oppress, we must consider the natural operation of the statute here in question, and it is apparent that it furnishes a convenient instrument for the coercion which the Constitution... forbid[s].” The thirteenth amendment, he concluded, “does not permit slavery or involuntary servitude to be established or maintained through the operation of the criminal law by making it a crime to refuse to submit to the one or to render the service which would constitute the other.”

We must consider the natural operation of the Uniform Code of Military Justice, the Manual of Courts-Martial, and the fundamental military presumption that orders are lawful. Without imputing any actual motive to oppress,¹⁹ the natural operation of these authorities working in concert clearly furnishes a “convenient instrument for the coercion which the Constitution… forbid[s],” and that instrument is indeed “maintained through the operation of the criminal law by making it a crime to refuse to submit.”

For much of our nation’s history, able-bodied men were also subject to the entire trinoda necessitas. This meant that the State could—and did—force men to leave their homes and livelihoods for some period of time to build roads and bridges, all without pay. In Butler v. Perry, the 1916 case upholding this practice under the Thirteenth Amendment, the rule was that Floridians had to devote six ten-hour days out of every year to uncompensated road-building or face criminal penalties. This appears to be “compulsory service for another’s benefit.”

Again, Koppelman tries to resist this, by arguing that “the duties of citizenship, however burdensome, are inescapable conditions of freedom. A person cannot be a slave if he is as free as it is possible for a citizen to be.”

Yet is a person under the trinoda necessitas as free as it is possible for a citizen to be? I’m a citizen, and I’m free, and I am not subject to the road-building draft. Nobody has been subject to the road-building draft in decades. Why? Cities and states decided to start paying for construction crews, instead of relying on the unpaid, involuntary labor of private citizens. The State could have done this all along. It chose not to. As the past hundred years have proved, compulsory road-building service is not and never has been an “inescapable condition of freedom.” Nonetheless, in Butler v. Perry, even as the trinodas was waning nationwide, the Supreme Court upheld it against a Thirteenth Amendment claim. It is also possible to imagine an all-volunteer military, or volunteer jury duty… yet here we are.

If we adopt Koppelman’s definition of “involuntary servitude” from Argument A2, these practices appear to be unconstitutional “involuntary servitude,” yet are widely accepted. The fact that they are apparent exceptions to the rule cannot be justified or explained away using the reasoning Koppelman supplies. It seems that, if (if!) Koppelman is correct about what the Thirteenth Amendment means, then the draft, the trinoda, and jury duty are all unconstitutional, and the Supreme Court has repeatedly gotten it wrong.

Compulsory Service to Private Individuals

The difficulties don’t end there, though. Many people, in many contexts, must render “compulsory service to another private individual” (emphasis mine).

If a man falls overboard, courts have held that a ship captain has a responsibility to stop the ship and attempt rescue, even if the rescue operations and resulting delay will incur enormous (perhaps crippling) expenses:

No matter what delay in the voyage may be occasioned, or what expense to the owners may be incurred, nothing will excuse the commander for any omission to take these steps to save the person overboard, provided they can be taken with a due regard to the safety of the ship and others remaining on board. (United States v. Knowles, N.D. Cal., 1864)

If the captain fails to do so, he can not only be sued civilly by the drowned man’s family, but will be guilty of criminal manslaughter.²⁰

If a railroad crossing guard decides mid-shift to resign his position and abandons his post (or does something tantamount to resignation, like falling asleep at his post), and someone dies on the now-unguarded rail crossing, he, too, will be guilty of manslaughter.

If someone accepts a duty of care for an unrelated elderly man (whether employed to do so or not), but then fails to care for that man every day, day in and day out, and the old man dies of neglect… yep, also manslaughter.

If a landlord promises an apartment water and heat, but then doesn’t do it, he can’t just pay the tenant damages; he has to actually perform the “compulsory service” on the tenant’s behalf. If he doesn’t? He can face criminal charges.

Koppelman’s definition of involuntary servitude is so sweeping that it includes all of these as instances of “involuntary servitude.” Indeed, Koppelman holds that initial consent to a service, or even a signed contract, doesn't matter. Any service becomes unconstitutional involuntary servitude, unless the service provider is able to quit “at any time” and “no law or force compels performance or a continuance of the service.”²¹ This point is important for Koppelman's abortion argument. Because, in his view, any service provider can change his mind at any time, a woman who consented to sex and/or to pregnancy can change her mind later. The moment she no longer wishes to perform voluntary service for her fetus, according to Koppelman, her pregnancy becomes involuntary servitude and she has a constitutional right to end it as quickly as possible.

Koppelman has to commit to this position, because, if he doesn’t, the Thirteenth Amendment argument for abortion rights really only applies to women who conceive children as the result of rape.²² That would not be a very useful argument for abortion-rights supporters, since only a tiny fraction of abortions are a result of rape! This commitment puts Koppelman in the position of claiming that railroad crossing guards who have to wait until end-of-shift to quit their jobs are victims of involuntary servitude during the remainder of that shift—and then he has to further explain why the courts haven't considered the Thirteenth Amendment applicable in such cases.

Koppelman attempts to explain these apparent exceptions to his rule in three ways. First, he contends that some of these duties are de minimis—that is, so trivial no reasonable person could complain about them. This does not fit very well with current federal circuit court precedent, which holds that, to violate the Thirteenth Amendment, “the temporal duration of the involuntary servitude need not be long; it can be slight.” Nor are these duties necessarily de minimis. Making someone remain at the scene of an accident is de minimis, sure. Making a lifeguard jump into the ocean to save a drowning man? Arguably. (Depends on the circumstances.) However, making someone provide daily care for an elderly man until he dies does not seem de minimis. Nor does cutting deeply into a sea voyage’s profits to conduct all-day rescue operations.

Second, Koppelman argues that all of these duties are terminable. A railroad crossing guard can end his employment through regular channels. A ship captain can quit once he’s back in port, or even stand down and transfer command to his executive officer. An elderly man under your care can be transferred to another guardian, or back into the care of the State. A landlord’s property can be sold. And, heck, if you don’t want to ever have to remain at the scene of a car accident, you can sell your car and stop driving. “In contrast,” Koppelman says (FL ‘90, p493), “unless abortion is permitted, one is not free at any time to abandon the status of being pregnant.”

I don’t think that contrast is as clear he makes it out to be. A landlord who wishes to be free of his duty to provide water and heat can certainly go ahead and sell. All he has to do is evaluate which offers he can entertain (based on leases, city zoning code, etc.), talk with a lawyer, recruit potential buyers, succeed in negotiations with one of them, sign a purchase agreement with a willing buyer, manage the transition for tenants according to the terms of their leases (and more, if he’s a decent person), wait for the buyer to secure financing (this ain’t a single-family home transaction here, where the buyer’s already walking around with a pre-approved mortgage), hire a title clearance company, clear the title, transfer the title, and receive payment. If a landlord decides with absolute certainty that he is getting out of the landlording business as quickly as possible, and he makes this decision on the same day that his wife conceives a child in her womb, it is distinctly possible that the child will be born before he manages to transfer his property. The project will very possibly cost him thousands, if not tens of thousands, of dollars. If he cannot find a willing buyer capable of assuming his own legal responsibilities (no guarantee!), the project may fail altogether. If he doesn’t have the cash flow to pony up thousands or tens of thousands of dollars, he can’t even get started.

Throughout that time period, the landlord may be criminally liable for providing water and heat to all his tenants.

It is true that land ownership is, ultimately, terminable. But, contra Koppelman, so is pregnancy: wait nine months.²³

Finally, Koppelman argues that the Thirteenth Amendment forbids involuntary servitude, but does not forbid monetary damages for refusing to perform a service. In other words, if you make a contract and break the contract, Koppelman thinks the State can force you to pay monetary damages of some kind, because this leaves “the individual free to acquire the money in whatever way he prefers, [and] is a lesser deprivation of liberty than the obligation to perform specific work for a specific individual.” On the other hand, for Koppelman, it would violate the Thirteenth Amendment if the courts require you to perform the contract involuntarily, under the coercive threat of criminal penalty. (This is a common piece of received wisdom, though not universally held.) I don’t object to this distinction; monetary and criminal liability are indeed different.²⁴ But all the cases I mentioned above are criminal liability cases, so his defense simply doesn’t pertain here. (Koppelman deploys this argument mainly to defend child support orders and anti-discrimination laws in public accommodations.)

In short, Koppelman’s explanation for these “duty of care” laws is not persuasive. So, once again, we find that, if Koppelman’s definition of “involuntary servitude” in Argument A2 is correct, then “duty of care” laws are at least largely unconstitutional, and the courts have repeatedly gotten it wrong.

Duties of Parent and Child

Only now do we come to the largest difficulty with Koppelman’s broad definition of “involuntary servitude.” If the Thirteenth Amendment says, as Koppelman contends, that no person can ever be compelled to serve another person involuntarily (and can even abandon service after agreeing to provide it!), then what’s the deal with parents and children?

Children are obligated to serve their parents. A couple weeks ago, my four-year-old received her first regular household chore assignment: she has to wipe off the dining room table every night after dinner. She hates this chore and protests every time she is ordered to do it. She is not compensated for her labor (allowance doesn’t start until Age 7) and she wouldn’t want to do the job even if I paid handsomely (she doesn’t understand money and mostly thinks dollars are pretty). According to Koppelman, this is involuntary servitude; not only am I violating the U.S. Constitution every time I do it, but Congress can bypass federalism and take whatever direct action it deems necessary to ensure that I never make my daughter wipe the table again. Also, every other father who makes his child do a chore violates the Constitution. I understand Andrew Koppelman has several children; has he violated the Constitution by coercing his kids to, say, make their beds?

There is no support in American culture or law for the notion that this is unconstitutional. As James Gray Pope wrote a few years ago, “Parents enjoy plenary authority over their minor children… [N]obody would contend that children generally suffer a condition of slavery prohibited by the Thirteenth Amendment.”²⁵ There's not a lot of case law confirming that plenary parental authority is permitted by the Thirteenth Amendment, because (to quote Koppelman on a different point), the principle is so embedded in the bedrock of our law that "no litigant has challenged it in many years. The case... is strengthened, not weakened, by relying on a principle so fundamental that it is never questioned."

Yet if Koppelman's understanding of "involuntary servitude" is correct, we are all wrong, and children generally do suffer a condition of slavery! The labor I demand of my kids is, I think, de minimis (although my children would argue this strenuously), but many parents demand more of their kids, like farmers who require their kids to help out around the farm. Everyone agrees that parents demanding more than de minimis labor from your kids is just fine.

Meanwhile, my kids have no power to leave my service, except in truly exceptional circumstances. Their status and duties are not terminable. They cannot even abandon work I have assigned them and pay monetary damages instead; they will instead face domestic criminal penalties (five-minute time outs, deprivation of desserts, no Star Wars, et cetera.) (My four-year-old believes all these punishments violate her Eighth Amendment rights, but that’s another blog post.)

Parents, meanwhile, have a legal obligation to provide for their children. Period. If you are a custodial parent, and you don’t take minimally decent care of your kids, you’re going to jail, and, frankly, you deserve it. We are bound by law to the compulsory service of our children. This duty is binding even if we hate it, even if we’re very poor, even if we’re forced to sacrifice every comfort we have in the world to fulfill our duty. We are bound by law to serve our children even if it puts our physical and mental health at elevated risk. (Arguably, it routinely does just that!)

This is exactly the same burden laid upon pregnant parents in a polity where abortion is prohibited: parents are bound by law to take minimally decent care of their kids, even if they hate it, even if they’re poor, even it risks some injury²⁶ to our physical and mental health. Koppelman attempts to resist this, one last time, by arguing that parents who decide they don’t want to be parents anymore can terminate their parenthood by giving the kids up for adoption. However, what if there is no opportunity for adoption? (Perhaps your child is profoundly disabled and nobody wants to adopt him. Perhaps you are living in a time of lesser prosperity and/or greater fertility, such that national demand for adoptable children has been satisfied.) Koppelman's argument implies that, if you can't adopt your child out, the Thirteenth Amendment to the United States Constitution guarantees you the inalienable right to leave your child alone in the woods to die of exposure.

Perhaps this isn’t surprising, since Koppelman also believes that the Thirteenth Amendment to the United States Constitution guarantees you the inalienable right to not just allow your child to die in the woods, but to violently dismember her in utero. But I hazard that most U.S. courtrooms would regard a “constitutional right to expose infants” with considerable skepticism.

Once again, we must ask whether Koppelman’s definition of “involuntary servitude” is wrong. (If he’s right, then everybody else is wrong.)

Koppelman himself, to his credit, reconsiders the question of parental obligations in Forced Labor Revisited ‘10, recognizing his position as a weakness of his original piece. But his reconsideration makes no headway. He merely expresses ambivalence, then concludes by actually digging himself a little deeper into the hole:

I am sure, however, that the… claim… that parental obligations never raise a Thirteenth Amendment issue, is false.

I am not familiar with any authority to support Koppelman’s position here.

At this point in Forced Labor Revisited ‘10, Koppelman falls back to Argument B, seeming to abandon Argument A to the parental obligation critique.

I, however, am not quite ready to be done with Argument A. Professor Koppelman’s explanation of the Thirteenth Amendment is not looking so good at this point, but that raises the question: what does the Thirteenth Amendment mean? Let’s take a few minutes to build the Thirteenth Amendment back up on new and stronger foundations.

Servitude in the Early United States

When you put two words together into a phrase, the result is usually a phrase that means the sum of the two words. For example, the phrase “color television” means television wave transmitted with color data (or a television set capable of receiving those transmissions). People often think the same way about “involuntary servitude”: surely, it must mean any situation where a person serves without a choice.

Yet there are other phrases where putting two general words together means something quite different from what the words mean on their own. For example, the phrase “real property” feels like it should mean any property you own that is really your property, not imagined or borrowed or conditional—a chair, for example, or this keyboard I’m typing on. But that would be wrong: “real property” means land. A chair is never real property. (It might be “personal property.”)

“Involuntary servitude” was the second kind of phrase. It referred to a specific kind of labor contract. Because this kind of labor contract no longer exists, it is easy to misunderstand what involuntary servitude entailed, and how the institution related to slavery.²⁷

In the American British Colonies, and in the very early United States, there was a broad legal concept called “servitude.” You probably learned about “indentured servitude” in fifth-grade social studies. The idea of indentured servitude was that you could “enter service” with someone for a set number of years.

Sometimes this was done as an apprenticeship: you serve a blacksmith (or whatever) for seven years and then you have the skills to open your own blacksmith shop. Sometimes it was done to get to America: you’re a poor Londoner who can’t afford passage on a boat, but there’s a rich family that will pay for your boat ride if you serve them for two years, so you agree. Sometimes you’re a Downton Abbey character who has been bound and re-bound to a family “in service” (annually, suggests Blackstone), not just for your whole life, but for generations previous. The point is, this was a not-uncommon legal arrangement between a master and a servant, and it was facilitated by a big ol’ mess o’ laws.

If you were an indentured servant, you could not quit while your indenture lasted. If a person “bound to service” refused his master’s (legal) orders (or, worse, ran away), a typical punishment (p451) was to add one year to your indenture. It is understandable why indentured servants might want to run away: they tended to be worked very hard, because, until the indenture expired, they were slave labor! They had to work whatever hours they were instructed to work, on whatever task assigned, living on whatever provisions their masters made for them, without hope of compensation (except for the promise of eventual freedom and freedom dues). Indentured servants retained their rights to life and property, but they were understood by law to have temporarily forfeited their liberty.

Most indentured servitude was voluntary servitude. A servant and master would sign a labor contract spelling out all the terms of service (how long, what the servant would receive in exchange, et cetera), each acting (at least in theory) with full and informed consent. This contract could be sold between masters. Once the contract was signed, the servant was bound for the term of the agreement, even if he later changed his mind—but he had entered into the agreement voluntarily, so this was still considered voluntary servitude even after the servant changed his mind.

There were also cases of involuntary servitude. In these situations, the State would force individuals into servitude. Here are some examples of how a free citizen could be forced to work for years in slave-like conditions without compensation:

• A court could sell you into involuntary servitude in payment of debt (p103).

• Vagrants, vagabonds, and people found wandering who couldn’t give “satisfactory accounts of themselves” were sometimes assumed to be runaway servants, imprisoned, and then sold into servitude in order to pay the prison fees (p107), which, if you’re into evildoing, is a pretty neat trick.

• In colonial Virginia, mixed-race children were automatically bound to involuntary servitude until age thirty-one.²⁸ A similar rule held in North Carolina (p83-84).

• The children of poor people (or orphans) could be “set to work” in involuntary service, which could last until adulthood. On at least one occasion (p49), this involved kidnapping orphan children of London off the street and transporting them to Jamestown, where they were indentured out as “apprentices” until age 24.

• A court could order involuntary servitude in lieu of civil or criminal fines. A Pennsylvania law of 1701 fined housebreakers and arsonists four times the amount stolen or damage inflicted (rather a lot of money), and, if they couldn’t come up with the money, sale into involuntary servitude was mandatory (p102). A colonial North Carolina law fined people for harboring runaway servants, and, if they couldn’t pay, sold them into involuntary servitude to raise the money (p34).

• As we see from the above, criminals might be forced into involuntary servitude; this could include transportation from England into American involuntary servitude (p113). This was not the same as prison labor, where criminals do service for the State. In the colonial and early American periods, the State could actually hand criminals over to private citizens, who would keep them as involuntary indentured servants for a number of years for their own private benefit.

You may be thinking that there wasn’t very much difference between “involuntary servitude” and “slavery.” You’re not wrong! As one dissertation (p5) put it, “the legal distinction between White servitude and Black slavery did not occur for more than forty years after the system’s invention.” Pennsylvania’s 1780 law of gradual emancipation used “servitude for life” as a synonym for slavery. The U.S. Constitution’s infamous fugitive slave clause (requiring states to return runaway slaves from other states) applied with equal force against indentured servants.

However, as the institutions developed, there did come to be important distinctions between involuntary servitude and what Blackstone called “absolute slavery.” Abolitionists worked hard (p147-148; free draft available) to drive those critical distinctions home to the public in the face of an increasingly tyrannical²⁹ Slave Power:

• Involuntary servitude typically lasted seven years or less. Slavery was for life.

• An involuntary servant lost his right to liberty, but retained his rights to life and (to a limited extent) property. A slave had no rights before the law whatsoever. In practical effect, and sometimes by its plain text (p91), the law allowed masters to murder slaves with impunity.

• An involuntary servant was a civil person, with the ability to petition for redress in court if his master violated his rights. A slave was a species of livestock, not a civil person. Even where laws existed that (very theoretically) forbade masters from abusing their slaves, slaves themselves could not sue or testify in court against their masters, even if they were victims of such abuse.

• An involuntary servant was only bound to labor of some sort. A slave, as a chattel or livestock, could be used in any way whatever. Often, the rape of a slave was not a crime (p66-69).

• The children of slaves were also slaves, a deeply bizarre legal rule that everyone knew full well even at the time was nuts.

From this, we see that servitude (whether voluntary or involuntary) shared many features with slavery.³⁰ Like slavery, servitude (as it evolved in British America) described a “private economical relation,” a legal bondage, of a servant to a master. The bound servant had no right to resist the commands of his master, and many of his other rights (such as the right to marry) were restricted (for a slave, all rights were extinguished). The master was a private citizen unrelated to the servant.

Compulsory service to the State (via conscription or the trinoda) might be considered good or bad, but it was not considered involuntary servitude. Likewise, domestic relationships—the compulsory service of children to parents, parents to children, or wives to husbands—might be considered good or bad, but were not considered involuntary servitude (pp155-160).

On the other hand, debt peonage, of the sort dealt with by the Supreme Court in Bailey v. Alabama, would constitute at least voluntary servitude, and might constitute involuntary servitude if the design of the program were sufficiently biased against the servitor (as it certainly was in Alabama’s case).

The Careful Language of Amendment XIII

To the best of my understanding, the institutions I have just described were what the American People understood themselves to be abolishing when they first decided to outlaw “slavery and involuntary servitude.” That occurred, not in 1865, but in 1787, when Congress passed the Northwest Ordinance. This ordinance governed the territory spanning from modern-day Ohio to modern-day Minnesota, and it provided:

There shall be neither slavery nor involuntary servitude in the said territory, otherwise than in the punishment of crimes, whereof the party shall have been duly convicted.

The precise meaning of this provision was tried and tested over the following seventy years. It was incorporated into over a dozen state constitutions, the Missouri Compromise, and other provisions, until 1857, when (following a Roe-like legal theory) the Supreme Court struck down the Missouri Compromise in Dred Scott v. Sandford.³¹

During those seven decades, both voluntary and involuntary servitude effectively died out within the United States. The Revolutionary War had already severely disrupted the flow of indentured servants from abroad, Parliament banned exportation of servants soon after (p254), the cost of trans-Atlantic passage steeply declined, and regulations increased (p263), all of which put great strain on the system. “Debtors’ prisons” slowly disappeared, and with them their supply of involuntary servants. Robert Steinfield says (p29-30):

Until 1820… Americans continued to import large numbers of indentured servants and contract laborers whenever the international situation permitted it. But in 1820 the market in imported servants collapsed. Thereafter, between 1820 and 1830, relatively few adult white servants were imported, and after the early 1830s, none were… [A]dult white European servitude simply disappeared.

The North moved toward free labor, under which everyone worked for wages. The South moved toward slavery, which (from the master’s perspective) had all the advantages of ordinary servitude but with way fewer strings attached, and which (thanks to the law of hereditary slavery) didn’t require importation or recruitment to be permanent.

As this happened, the public meaning of the words “involuntary servitude” broadened somewhat. Without the original institution of involuntary servitude to compare it to, judges became more inclined to identify the language barring “involuntary servitude” with the institution formerly known as "voluntary servitude." That is, courts began to rule that a man who willingly sold himself into servitude (traditionally considered "voluntary servitude") might change his mind and walk away before the service was ended. If his master detained him further, it could, under certain circumstances, become "involuntary servitude."

During this transition, the circumstances that turned lawful service into unlawful “involuntary servitude” varied by state, as described by Professor Nathan Oman in his invaluable Specific Performance and the Thirteenth Amendment,³² but broadly comprised four judicial tests:

1. Did the servant enter the indenture in “a state of perfect freedom”?

2. Did the indenture provide proportionate compensation or consideration?

3. How long did the indenture last? Longer indenture meant greater court suspicion, especially given the traditional rules that indentures had to end before five or seven years.

4. Did the master “dominate” the servant, especially by asserting a right to physically harm or detain the servant outside legal channels? If so, that made courts very suspicious indeed.

This expanded interpretation of the original law does not seem to have been understood by the judges involved to be an expansion; they were trying to follow the text as they understood it. Some traditionally voluntary servitude was now at risk of being declared “involuntary” and voided… but, since the practice had already mostly died out in the meantime, this caused no great waves.

This growth in the meaning of “involuntary servitude” encompassed only contracts between masters and servants for labor. No source that I have consulted (including Koppelman) presents any evidence that the meaning of “involuntary servitude” went any further than this in 1865.

As Professor Kurt Lash details in his helpful working paper, Roe and the Original Meaning of the Thirteenth Amendment, the drafters of the Thirteenth Amendment consciously chose to rely on the tried-and-tested language of the Northwest Ordinance, rather than come up with new language freeing the slaves. They deliberately used precisely the same language as Congress had in 1787, intending to enact precisely the same rule of law as Congress had in 1787.

In my opinion, their understanding of the language had unconsciously drifted far enough from what Congress meant in 1787 that the same words probably meant something slightly different, but the involuntary servitude they abolished was still strictly that of indenture by servants to unrelated private masters for defined, extended periods.

Since the military draft, jury duty, compulsory road-building, public-accommodations laws, good samaritan laws, duty-of-care laws, landlord regulations, parental rights, child neglect laws, household chores, and anti-abortion laws do not pertain to indenture by servants to unrelated private masters, and do not establish anything that looks remotely like that relationship, they are not “involuntary servitude.”

Since debt peonage, the coolie system, and specific performance requirements in contracts can (at some times, under some circumstances) create conditions that involve or closely resemble the private indentures outlawed by the Northwest Ordinance and the Thirteenth Amendment, they might be voided by courts or restricted by Congress as involuntary servitude³³ —but not always! For example, in Robertson v. Baldwin (1897), the Supreme Court compelled seamen to fulfill their contracts against their will, ruling that they had entered the contracts voluntarily, so their service was voluntary.³⁴

This division comports nicely with our subsequent case law defining involuntary servitude, and requires none of the hoop-jumping and acrobatic flexibility displayed by Koppelman’s search to justify the Thirteenth Amendment’s “exceptions.” No justification is required, because the draft, landlord regulations, anti-abortion laws, and so on are not exceptions. They never fell within the Thirteenth Amendment’s scope in the first place, and can only be read into the Amendment by misunderstanding the Amendment’s historical meaning. Koppelman’s Argument A2 is not precisely false, but it is fatally incomplete in a way that causes Argument A4, A5, and A6 to fail in turn.

The reality is that human beings are not characters in an Ayn Rand book. We all have unchosen obligations, and we always will. The State has always enforced at least some of the most important unchosen obligations. I hope it always will, because the alternative—absolutist libertarianism—would be a hellscape. It is not servitude to be forced: to render aid to a man overboard; to rent a Black man a room at your hotel; to make a cake for a gay wedding;³⁵ to not kill a child by abortion or exposure. To make out that these are “servitude” requires a definition of “servitude” that did not exist when the Thirteenth Amendment was written.³⁶ (Even less could any of these be called “slavery” within the meaning of the Thirteenth Amendment.) Nor could you justify calling it “servitude” based on a fair reading of subsequent case law.

Oh, snap, I haven’t really talked about the case law yet, have I?

I now turn to Koppelman’s Argument B, which is a sophisticated, lawyerly argument based largely on the interpretation of specific acts of the 39th Congress, plus a body of Supreme Court case law that you, the average citizen, have never heard of, and which you have no reason to know (unless Koppelman’s Argument B is correct) (but it isn’t). If you were only here because someone shouted “unwanted pregnancy is involuntary servitude!” at you in a Reddit comment thread, you can be done at this point. This Blog Post has disposed of the obvious and, from here on out, will be delving into the esoteric.

Was the Civil Rights Act of 1866 Constitutional?

There is an argument that the Thirteenth Amendment bans more than just slavery and involuntary servitude themselves. According to this argument, the amendment also bans the “badges and incidents” of slavery—or at least allows Congress to ban those “badges and incidents,” at Congress’s discretion. Since this notion is core to Koppelman’s Argument B, let’s explore it (without committing to it).

The most important evidence for the “badges and incidents” theory is that, two seconds after passing the Thirteenth Amendment, most of the people responsible for passing it shook off their laurels and passed the Civil Rights Act of 1866, which did a lot more than simply ban slavery (emphasis mine):

Be it enacted by the Senate and House of Representatives of the United States of America in Congress assembled, That all persons born in the United States and not subject to any foreign power, excluding Indians not taxed, are hereby declared to be citizens of the United States; and such citizens, of every race and color, without regard to any previous condition of slavery or involuntary servitude, except as a punishment for crime whereof the party shall have been duly convicted, shall have the same right, in every State and Territory in the United States, to make and enforce contracts, to sue, be parties, and give evidence, inherit, purchase, lease, sell, hold, and convey real and personal property, and to full and equal benefit of all laws and proceedings for the security of person and property, as is enjoyed by white citizens, and shall be subject to like punishment, pains, and penalties, and to none other, any law, statute, ordinance, regulation, or custom, to the contrary notwithstanding.³⁷

Some, like Senator (and future Vice President) Thomas Hendricks (D-IN), objected: the Thirteenth Amendment banned slavery and involuntary servitude, but did not say anything about contracts, property, or other civil and constitutional rights.

Sen. Lyman Trumbull (R-IL), the guy who had introduced the Thirteenth Amendment, confidently proclaimed that, yes, this bill was authorized by his Amendment, and urged Congress in stirring language:

If the construction put by the Senator [Hendricks] from Indiana upon the amendment be the true one, and we have merely taken from the master the power to control the slave and left him at the mercy of the state to be deprived of his civil rights, the trumpet of freedom that we have been blowing throughout the land has given an uncertain sound, and the promised freedom is a delusion. Such was not the intention of Congress, which proposed the Amendment itself. With the destruction of slavery necessarily follows the destruction of the incidents of slavery. When slavery was abolished slave codes in its support were abolished also.

Those laws that prevented the colored man going from home, that did not allow him to buy or to sell, or to make contracts; that did not allow him to own property; that did not allow him to enforce rights; that did not allow him to be educated, were all badges of servitude made in the interest of slavery and as a part of slavery. They never would have been thought of or enacted anywhere but for slavery, and when slavery falls they fall also.

…If in order to prevent slavery Congress deem it necessary to declare null and void all laws with will not permit the colored man to contract, which will not permit him to testify, which will not permit him to buy and sell, and to go where he pleases, it has the power to do so, and not only the power, but it becomes its duty to do so.

As though to emphasize the force with which the Thirteenth Amendment Congress agreed with Trumbull, they passed the Civil Rights Act of 1866 as he requested… then passed it again, over President A. Johnson’s veto, by a two-thirds majority in both houses. This was the first time in American history that major legislation was passed over a presidential veto.³⁸

On the other hand, Kurt Lash and David Upham each offer good reasons to believe that the Civil Rights Act of 1866 could not, in fact, be justified under the Thirteenth Amendment, despite Trumbull’s bluster and Congress’s acquiescence. (Jennifer Mason McAward reports similar evidence but draws more equivocal conclusions from it.)

Rep. John Bingham (R-OH), who was largely responsible for the Fourteenth Amendment and strongly supported the goals of the Civil Rights Act, nevertheless voted against it, arguing along the same lines as Hendricks:

[I]n view of the text of the Constitution of my country, in view of all its past interpretations, in view of the manifest and declared intent of the men who framed it, the enforcement of the bill of rights, touching the life, liberty, and property of every citizen of the Republic within every organized State of the Union, is of the reserved powers of the States, to be enforced by State tribunals and by State officials acting under solemn obligations of an oath imposed upon them by the Constitution of the United States. Who can doubt this conclusion who considers the words of the Constitution: ‘The powers not delegated to the United States by the Constitution, nor prohibited by it to the States, are reserved to the State respectively, or to the people?’ The Constitution does not delegate to the United States the power to punish offenses against the life, liberty, or property of the citizen in the States, but leaves it as the reserved power of the States, to be by them exercised. . . . I am with [Mr. Wilson] in an earnest desire to have the bill of rights in your Constitution enforced everywhere. But I ask that it be enforced in accordance with the Constitution of my country. (Congressional Globe, 39th Congress, 1st Sess. 1291, 9 March 1866.)

Bingham’s objections pushed Congress away from justifying the Civil Rights Act as an exercise of its Thirteenth Amendment power. Congress instead tried to justify it under (variously) the Fifth Amendment’s due process clause, the Republican Form of Government Clause, the Comity Clause of Article IV, the Necessary & Proper Clause of Article I, the Naturalization Clause, Congressional war powers, implied powers under hated pro-slavery decisions like Prigg v. Pennsylvania, and/or outright rejection of Barron v. Baltimore (1848) (which had decided that the antebellum Bill of Rights did not apply against the states). Bingham and his allies thought none of these alternatives fit the bill, either, and insisted that the Civil Rights Act would not be constitutional unless the Constitution was amended (again). The entire debate is painstakingly reconstructed in Kurt Lash’s Enforcing the Rights of Due Process: The Original Relationship Between the Fourteenth Amendment and the 1866 Civil Rights Act.

Although, in the short term, Congress overrode Bingham’s concerns (variously citing all of the above constitutional clauses, not only the Thirteenth Amendment), Bingham’s interpretation was certainly a live view among the contemporary public. As Lash details in Roe and the Original Meaning of the Thirteenth Amendment, President Andrew Johnson’s administration—including the famous anti-slavery radical, Secretary of State William Seward—not only adopted Bingham’s limited understanding of the Thirteenth Amendment, but even convinced dubious slave states that it only did what it said: abolished slavery and involuntary servitude, no funny business.

The slave states, for their part, were largely unrepresented in Congress at this time, due (mainly) to their grave violations of the Republican Form of Government clause (see America’s Constitution, pp364-380). Nonetheless, they were still part of the public (albeit a largely treasonous one), and they seem to have more or less taken Bingham’s view of the amendment, with several of them adopting resolutions to that effect (p33). Moreover, it strains credulity that the slave states would have so easily adopted the Thirteenth Amendment if they shared Trumbull’s broad understanding of its text, rather than Bingham’s narrow one, especially since said slave states were busy passing Black Codes at this very moment in history. (The slave states really dug in their heels against the Fourteenth Amendment, which they—correctly—recognized as a grave threat to their entire system of apartheid.)

Congress certainly gave the appearance of buckling to Bingham’s concerns when, less than three months after passing the Civil Rights Act, they also passed Bingham’s Fourteenth Amendment. The Fourteenth Amendment contains several passages that strongly parallel the Civil Rights Act, including its declaration of citizenship and its due-process/equal-protection clauses. Bingham expressly intended the Fourteenth Amendment to enable a bill like the Civil Rights Act, and the Civil Rights Act was actually re-passed (with modest alterations) in 1870, once the Fourteenth Amendment had been ratified.

Professor Koppelman is, not unreasonably, a mite suspicious of modern-day scholars who claim to understand the original public meaning of the Thirteenth Amendment better than Lyman Trumbull. I look forward to reading Koppelman’s further development of the scholarship surrounding the Thirteenth Amendment’s original meaning, which could certainly benefit from his contributions. For the time being, however, it seems to me that the weight of evidence suggests that either:

• [A]: The 39th Congress lacked the power to pass the Civil Rights Act of 1866 (until the Fourteenth Amendment was ratified and cured the lack);

• [B]: The 39th Congress had the power to pass the Civil Rights Act, not based on the Thirteenth Amendment, but on one of the other provisions Congress mentioned in the debates; or perhaps even

• [C]: The 39th Congress had the power to pass the Civil Rights Act because the Thirteenth Amendment implicitly provided Congress with powers identical to those given explicitly by the Fourteenth Amendment—which means the Thirteenth Amendment provides no power beyond what the Fourteenth Amendment grants and is therefore useless as an alternative source for abortion rights.

However, as I said, for the sake of exploration, we are going to accept (and put the best possible light on) Koppelman’s theory:

• [D]: The 39th Congress had the power to pass the Civil Rights Act because the Thirteenth Amendment provided Congress with the power (distinct from the powers granted by the Fourteenth Amendment) to abolish not only slavery and involuntary servitude themselves, but also to erase certain other “incidents” of slavery which gave “the trumpet of freedom… an uncertain sound.”

This theory is not only the basis of Koppelman’s Argument B1. It finds considerable support in various U.S. Supreme Court cases, and it appears to have been the view of a large portion of the 39th Congress itself, including Lyman Trumbull.

Rolling With It

Even a skeptic must admit that Trumbull’s thinking isn’t exactly bats, either. The central fact of slavery, the thing that distinguished it from ordinary servitude, was the fact that it rendered its victims civilly dead. It un-personed them. A servant (whether voluntary or involuntary) was still a person, still had rights, could still own property, and could still seek redress in court. A slave, in the eyes of the law, was a cow. Less than a cow, really; raping a cow would at least get a plantation owner executed for bestiality, but raping a (female) slave was (often) not even a crime.

Since the decisive element of slavery was that it extinguished civil personhood, it’s not outrageous to think that the abolition of slavery directly entailed civil personhood, and thus all the ordinary privileges and immunities that accrued to ordinary citizens. As Judge William Gaston wrote in State v. Manuel, a celebrated 1838 North Carolina Supreme Court decision that stayed influential (p8) through the turn of the century:

According to the laws of this State, all human beings within it who are not slaves, fall within one of two classes. Whatever distinctions may have existed in the Roman law between citizens and free inhabitants, they are unknown to our institutions. Before our Revolution all free persons born within the dominions of the king of Great Britain, whatever their color or complexion, were native born British subjects—those born out of his allegiance were aliens. Slavery did not exist in England, but it did exist in the British colonies. Slaves were not in legal parlance persons, but property. The moment the incapacity… of slavery was removed[,] they became persons, and were then either British subjects or not British subjects, accordingly as they were or were not born within the allegiance of the British king. Upon the Revolution, no other change took place in the law of North Carolina… Foreigners until made members of the State continued aliens. Slaves manumitted here became free-men—and therefore if born within North Carolina are citizens of North Carolina.

For many in the Congress that abolished slavery, to be free from slavery was to become a citizen; there was no middle ground (p442). Thus, an amendment banning slavery by that fact guaranteed citizenship, with all the attached privileges and immunities.³⁹ And, as Chief Justice John Marshall put it in upholding the First Bank of the United States:

We admit, as all must admit, that the powers of the government are limited, and that its limits are not to be transcended. But… [l]et the end be legitimate, let it be within the scope of the Constitution, and all the means which are appropriate, which are plainly adapted to the end, which are not prohibited, but consistent with the letter and spirit of the Constitution, are constitutional.

I don’t want to abuse this quote (as it has sometimes been abused), but is not the restoration of rights extinguished by slavery fairly within the scope of an amendment abolishing slavery? Wouldn’t that include the civil rights "to make and enforce contracts, to sue, be parties, and give evidence, inherit, purchase, lease, sell, hold, and convey real and personal property, and to full and equal benefit of all laws and proceedings for the security of person and property"? After all, in the American system, what use is any right—including the right not to be enslaved—if you don't possess the fundamental civil right to vindicate your other rights in court?

So the argument went. I am not entirely convinced, but it is reasonable on its face, it’s put down roots in our law, and this understanding seems to have been shared by enough of the public to raise questions for any originalist who disagrees with it. Koppelman’s “egalitarian” argument depends on it. We are therefore rolling with it.

Badges, Vestiges, and Incidents

The collection of rights and privileges that had been suppressed by (or because of) slavery came to be known as the “badges and incidents” of slavery (in part thanks to Trumbull’s influential speech, which used that language). Koppelman explains in Originalism, Abortion, and the Thirteenth Amendment (p1932):

Badges, incidents, vestiges, relics: the Amendment reaches all of these because they are associated in some way with slavery. …Property is familiarly regarded as a bundle of rights. Slavery is a bundle of disabilities. Each one of those disabilities is part of slavery and so raises Thirteenth Amendment concerns.

Debate about the exact contents of this “bundle of disabilities” began, in earnest, with the debates over the Civil Rights Act. I will now briefly trace the history of that debate in subsequent decades, omitting many related issues⁴⁰ along the way.

In United States v. Rhodes (1866), Justice Swayne (riding circuit on his own) ruled in favor of the Civil Rights Act of 1866—even though the Fourteenth Amendment had not yet been ratified. Opponents argued that both Black slaves and free Blacks were frequently deprived of their civil rights, and so depriving Black people of civil rights was not truly an incident of “slavery.”⁴¹ Justice Swayne was unpersuaded: “[Free Blacks] had but few civil and no political rights in the slave states. Many of the badges of the bondsman’s degradation were fastened upon them…. [Emancipation] was doubtless intended to reach further in its effects as to everyone within its scope.”

Blyew v. United States (1871) was a defeat for civil rights. However, the dissent claimed that the Thirteenth Amendment entitled Black people to access the justice system, and the majority didn’t disagree. (The case was lost on other, more technical grounds.)

In the Civil Rights Cases (1883), the full Court agreed that the Thirteenth Amendment allowed Congress to abolish a litany of what it called the “badges and incidents” of slavery: “Compulsory service of the slave for the benefit of the master, restraint of his movements except by the master's will, disability to hold property, to make contracts, to have a standing in court, to be a witness against a white person, and… severer punishments for crimes.” However, in 1875, Congress had also invoked the Thirteenth Amendment to abolish racial discrimination in public accommodations (hotels, taxis, bakeries, and so on). The Court rejected this as a bridge too far, emphasizing that the Thirteenth Amendment protected civil rights, not social rights.⁴² In the process, the Court also sided with the losers in United States v. Rhodes, ruling that, because free Blacks had not enjoyed rights to public accommodations before the War, public-accommodations discrimination was not a “badge of slavery” for emancipated Blacks after the War.

In Plessy v. Ferguson (1896), the Court refused to strike down a Louisiana law that required trains to be segregated. Plessy was mostly about the Fourteenth Amendment, but the justices also ruled that the Thirteenth Amendment could not apply here, because segregation did not “stamp” anyone with a “badge of inferiority.”

In Hodges v. United States (1906), the full Court finally heard a case directly about Congress’s power under the Thirteenth Amendment to ban the “badges and incidents” of slavery—and, surprisingly, rejected forty years of its own dicta to rule that the Thirteenth Amendment could only be used directly against slavery or involuntary servitude, not to address slavery’s “badges and incidents.” In effect, they held the Civil Rights Act of 1866 unconstitutional insofar as it relied on the Thirteenth Amendment.⁴³

Curiously, after doing this, the Hodges Court also accepted two more things into the “badges and incidents” canon: (1) a requirement to carry identification documents in order to walk freely on the street and (2) conspiracies to hinder the right to freely contract. Both these, the Court announced, were “badges and incidents” of slavery. It’s unclear why they bothered, though, since they had just ruled that the only “badges and incidents” that the Thirteenth Amendment might actually be able to help with were literal iron collars.

It took sixty years for “badges and incidents” to come roaring back from their interment in Hodges, but they did in Jones v. Alfred H. Mayer Co. (1968). In Jones, the Court overturned Hodges (a 62-year-old precedent) for being egregiously wrong (p442). They did it with zero consideration for broader doctrine or fundamentally changed facts, without even pausing for a moment to worry about the stare decisis implications.⁴⁴ Frosting on the cake, they killed Hodges in a footnote. Ice cold.

In doing so, Jones revived the Civil Rights Act of 1866 as an exercise of the Thirteenth Amendment. Its provision that guaranteed citizens “of every race and color” the right to “purchase, lease, sell, and convey property,” struck down in Hodges, sprang back to life. Jones held that a private property developer could not refrain from selling a house to a Black person solely because of his race.

Let us sum up, then, the badges and incidents of slavery identified by Supreme Court decisions and/or Congress and/or that speech by Lyman Trumbull:

• “compulsory service of the slave for the benefit of the master”

• “restraint of his movements except by the master's will”

• requirement to keep identifying documents on his person while traveling

• disability to hold, sell, buy, lease, or convey property

• disability to make contracts

• disability to access education (Trumbull)

• disability “to have a standing in court”

• disability “to be a witness against a white person”

• suffering “severer punishments for crimes”

There is much that could be said about this list. It is, for now, the canonical list of “badges and incidents” of slavery, but could perhaps be expanded by further jurisprudence or legislation.⁴⁵

However, I remind you (ad nauseam) that “badges and incidents” does not appear in the Thirteenth Amendment’s text. (Nor does “vestiges and relics,” a phrase Justice Stewart used in Jones.) To justify acting against badges, incidents, vestiges, or relics of slavery, Congress must tie them back to the original meaning of the Thirteenth Amendment’s text. Otherwise, we are no longer doing law, but haruspicy.

Here is my best attempt: the defining feature that separated slavery from mere involuntary servitude was civil death, which disabled and extinguished all the slave’s rights, privileges, and immunities. For the framers of the Thirteenth Amendment, to make a slave free was not merely to strike the shackles from his wrists, but to resurrect him from the dead and clothe him in the body of a free citizen. He became entitled to all the privileges to which (White) citizens had always been accustomed. The “badges and incidents” which disabled those privileges were thus obliterated. He was placed on equal footing with his (White) peers.

The list above fits with that concept rather nicely: each “badge and incident” inverts something that all White citizens could enjoy, but which slaves (because they were considered property, not persons) could not. Ordinary citizens can: move freely without identification; buy, sell, lease, and transfer; make contracts; act in court; and use public resources (including education). These are privileges and immunities spoken of and freely assumed since Corfield v. Coryell (C.C.E.D.Pa., 1823). It is therefore unconstitutional to deny those privileges and immunities (or any other generally enjoyed civil rights) to Black citizens, or any other group of citizens. To do so would re-impose the “badge of slavery” upon them.

There may be additional criteria before a legal disability can be justly called a “badge of slavery.” Jennifer Mason McAward’s Defining the Badges and Incidents of Slavery provides a good survey of the current state of jurisprudence and scholarship. (I am indebted to her work, even though she and I are not trying to do quite the same thing.) It is enough for now for us to observe that, in U.S. case law, a “badge or incident of slavery” deliberately disables a privilege or immunity to a discrete class of citizens, even while that privilege or immunity is deliberately allowed to and enjoyed by ordinary citizens.

(Remember that sentence. It will be on the test.)

The Badges and Incidents of Female Slavery

If God has bestowed beauty upon her, it will prove her greatest curse.  That which commands admiration in the white woman only hastens the degradation of the female slave. I know that some are too much brutalized by slavery to feel the humiliation of their position; but many slaves feel it most acutely, and shrink from the memory of it. I cannot tell how much I suffered in the presence of these wrongs, nor how I am still pained by the retrospect. My master met me at every turn, reminding me that I belonged to him, and swearing by heaven and earth that he would compel me to submit to him. If I went out for a breath of fresh air, after a day of unwearied toil, his footsteps dogged me. If I knelt by my mother’s grave, his dark shadow fell over me even there. The light heart which nature had given me became heavy with sad forebodings. The other slaves in my master’s house noticed the change. Many of them pitied me; but none dared to ask the cause. They had no need to inquire. They knew too well the guilty practices under that roof; and they were aware that to speak of them was an offence that never went unpunished.

—Harriet Jacobs, Incidents in the Life of a Slave Girl, chapter V.

Whenever I encounter someone who doesn’t seem to hate slavery deeply enough, I recommend the movie Django Unchained. Ordinarily, I find Quentin Tarantino’s over-the-top brutality cartoonish. When it comes to slavery, though, Tarantino is one of the only filmmakers who can tell the truth—precisely because of his over-the-top brutality. Slavery was cartoonishly evil. We all know that intellectually, but to really viscerally know it and hate it requires more than a photograph of Whipped Peter and an account of his escape. Django serves that purpose.

At least, it did for me. When I was a child, I was inclined to view the Confederacy as a fundamentally noble people who were undone by a tragic moral flaw—but whom we could still honor for the good in them. I had dress-up hats for both armies. By late high school, I recognized intellectually that slavery was not the Confederacy’s Achilles heel, but its foundation, yet I still wanted to honor our American brethren who had fought and died for the “Lost Cause.” When I walked out of Django, though, I finally understood, bone-deep, that the Confederacy was an evil empire whose destruction had been no less urgent than the overthrow of the Soviet Union.⁴⁶ I still have both hats, but my kids are only allowed to wear the blue "good guy" hats. That's the power of art.

Yet I think it’s significant that Django Unchained is principally about men. Broomhilda’s plight is sometimes implied, occasionally mentioned, but never centered. The movie is already a high R-rating. I doubt that an equally honest movie about the plight of slave women could be made less than NC-17, and I don’t know how anyone could get through watching it.⁴⁷ The violence that threatened female slaves was far more intimate, far more poisonous. Koppelman does not let us avert our eyes, nor should he:

Thomas Jefferson instructed his plantation manager in 1820, “I consider a woman who brings a child every two years as more profitable than the best man on the farm.”

…One North Carolina planter threatened to flog a group of women slaves to death, and when they asked what they had done, explained: “Damn you I will let you know what you have done; you don’t breed, I have not had a young one from you for several months.”

If a couple were separated by sale or death, each was expected to quickly find a new spouse. One ex-slave remembered: “A slave girl was expected to have children as soon as she became a woman. Some of them had children at the age of twelve and thirteen.”

…The future children of a slave were devisable in a will, just like the future rents of a piece of real estate.

…[W]hen [Rose Williams] was sixteen, her master moved her to the cabin of a male slave named Rufus. When Rufus came into her bunk, she fended off his advances with a poker: “De nex’ day de massa call me and tell me, ‘Woman, I’s pay big money for you and I’s done dat for de cause I wants yous to raise me chillens. I’s put yous to live with Rufus for dat purpose. Now, if you doesn’t want whippin’ at de stake, yous do what I wants.’ …Dere it am. What am I’s to do?

His sources provide more (p133-143):

“Marsters an’ overseers use to make slaves dat wuz wid deir husband git up, do as dey say,” [Reverend Ishrael] Massie recalled. “Send husbands out on de farm, milkin’ cows or cuttin’ wood. Den he gits in bed wid slave himself. Some women would fight an’ tussel. Others would be ‘umble—feared of dat beatin’. What we saw, couldn’t do nothing ‘bout it. My blood is bilin’ now, thoughts of dem times. Ef dey told dey husbands he was powerless.”

…Sam and Louisa Everett, both born near Norfolk, told of their master who frequently enjoyed entertaining his friends by forcing slaves to copulate publicly. “Quite often,” the Everetts reported, “he and his guests would engage in these debaucheries… Sometimes they forced the unhappy husbands and lovers of their victims to look on.”

In addition to all the horrors of male slavery, slave women⁴⁸ were: raped by their masters, raped by their peers, forced into marriages, forced out of marriages, forced to re-marry, coerced to bear children (even at dangerously young ages), torn away from those same children, raped some more, used for entertainment, and, above all—as our ever-so-noble Founding Father Thomas Jefferson succinctly explained above—used as breeding stock. When serial rape was naturally followed by conception, the victim could pay a further terrible price for the sins of the father (see esp. footnote 161).

Slave women were threatened in this way because they had neither legal nor actual power to resist. The law saw them, not as people, but as mere tools, for their master to dispose as he saw fit. Sexual violence exists in every society, especially in relationships with power imbalances, but White women bound to indentured service do not seem to have suffered to this extent, this often. How could they? A White servant—even one sentenced to involuntary servitude—had legal recourse (including witnessing against her master in court), she could not be forced into marriage, and her children were free. A free citizen could remove herself from the vicinity of an intractably dangerous man. Even today, many free women find it difficult, as a practical matter, to exercise these options; coercive partners are difficult and dangerous to escape. For the slave woman, however, every one of these options was disabled by operation of law.

It is therefore at least plausible to say (as Koppelman does in Arguments B2, B3, and B4) that any amendment purporting to abolish slavery must needs cure all these disabilities, bringing the woman slave to stand on the same ground as the woman citizen.⁴⁹ To be free of the sexual “badges and incidents” of slavery, a citizen in 1865 must not only have legal personhood, the right to court access, and all the other privileges and immunities list above, but must also have the legal rights:

• to consent to marriage

• to refuse sexual relations with someone she isn't married to⁵⁰

To this list, Koppelman would add the right to an abortion (Argument B5). This, he cannot do.

It is true that slaves lacked a legal right to procure abortions. However, so did citizens. No one in the year 1865, slave or citizen, woman or man, had a right to destroy the life of a child en ventre sa mere,⁵¹ and twenty-five of thirty-six states then in existence had already criminalized abortion at every stage of pregnancy (Dobbs, Appendix A; also Finnis & George), most with only a life-of-the-mother exception. The right to an abortion was simply not a privilege or immunity enjoyed by ordinary citizens at all. Therefore, its denial was not a disability of slavery. Abortion prohibition was simply a law, of the ordinary sort, promulgated to promote the common good (in this case, the protection of vulnerable unborn human lives).

Koppelman tries to escape this with an abstraction. The “badge of slavery” in this case, he suggests, is not the specific bundle of rights denied to slave women, but the general idea that slave women suffered a “loss of control… over their reproductive capabilities.”

This is an old lawyer’s trick. It is a fair move to bundle up the rapes and the forced marriages and describe them as a "loss of control over reproductive capabilities," which is a neutral description on its face. But, in our era, the phrase “control over one’s reproductive capabilities” is commonly used in a separate sense—as a euphemism for abortion rights.⁵² Koppelman, knowing you’re from our era and will naturally read the phrase to include the euphemism, is thereby able to smuggle a right to abortion into an abstraction that’s supposedly about the disabilities imposed by 19th-century slavery. It has no business being there. It's simple equivocation.

This is not, as Koppelman suggests (OA13, 1929-30), a question of reading the “paradigm case” broadly or narrowly, a kind of “fielder’s choice” in constitutional law. “Fielder’s choice” cases might very well exist: public accommodations discrimination might arguably be a “badge or incident” of slavery, depending on how broadly or narrowly you read the privileges and immunities of ordinary (White) citizens in 1865. Did they have a right to access all public accommodations, or did public accommodations always have a (largely dormant) right to refuse service? If the former, then ex-slaves must be given that right, too. This is an arguable question of the sort Koppelman is talking about.

But did ordinary (White) citizens in 1865 have a right to abortion? No. Broadly? No. Narrowly? Also no. Just, simply: no. To say that abortion rights fall beyond the scope of the Thirteenth Amendment is not to embrace the logic of Hodges (as Koppelman argues), but the logic of Lyman Trumbull.

Therefore, even if Argument B1 is granted, Argument B5 is not correct, and Argument B6 falls with it.

Conclusion

Using the Thirteenth Amendment, Andrew Koppelman advances a novel,⁵³ intelligent attack on laws protecting fetuses from violence. In his two-pronged assault, Koppelman first contends that pregnancy constitutes “servitude” of mother to child—and involuntary servitude if Mom decides at any time she doesn’t want to continue being pregnant anymore. Involuntary servitude is, of course, prohibited by the Thirteenth Amendment. Then, he argues that “loss of reproductive control” is a “badge and incident of slavery,” which can be outlawed under that same amendment.

The argument pays a high price up front: to invoke the Thirteenth Amendment against the fetus, the argument must concede fetal personhood. The argument must also deal with the apparently unjustified level of violence against the fetus in current abortion practice. Nonetheless, this Thirteenth Amendment argument constitutes the most serious constitutional challenge to abortion prohibitions I am aware of.

However, the argument fails. Understood in light of its original public meaning, the parental relationship of mother and child cannot be understood as “involuntary servitude” within the meaning of the Thirteenth Amendment—even though Koppelman is quite correct that pregnancy is horrible beyond all reason. Even assuming, for the sake of argument, that the “badges and incidents” of slavery are indeed vulnerable to Thirteenth Amendment regulation, the “badges and incidents” of slavery cannot fairly be understood to include the lack of a right to procure an abortion, because ordinary citizens did not have a right to procure an abortion at the time of the Amendment’s ratification. Since both the “libertarian” and “egalitarian” lines of the Thirteenth Amendment argument are turned back, the Thirteenth Amendment argument cannot yield a right to abortion.

At least, not in any way I can see. But, then, I am not a lawyer. I invite further development of this argument by either side of the question.

I do have more to say on the subject of so-called “compulsory pregnancy.” There is an ugly tendency by one side of this debate to say that abortion prohibitions create “compulsory pregnancy” or “forced birth.” This is a perverse distortion. Abortion prohibitions forbid only one pregnancy option: actions intended to cause the death of an unborn child (see, for instance, Texas’s law). Any option that ends (or prevents) a pregnancy without killing the child remains on the table. Leveling this critique at Koppelman’s argument would falsify Argument A3 and further weaken Argument B5.

However, my objection to the “forced birth” label will need to be developed at some length (the above paragraph certainly won’t convince anyone by itself). It is not specific to the Thirteenth Amendment argument, and besides, guys, I’m just frickin’ tired after writing all this… so I will pursue the “forced birth” question under separate cover at a later date.

In the meantime, I hope those of you who've stuck this out all the way to the end have learned a whole lot more about the Thirteenth Amendment than you expected when you woke up this morning!⁵⁴

---

Editor’s Note: As always with the really long articles, there’s probably typos or broken links that I missed in revisions no matter how many times I read it over. Please let me know about them in the comments.

1. The 2012 article is somewhat more interested in using Koppelman’s Thirteenth Amendment argument as a springboard into a larger methodological critique of originalism. This is an interesting critique, but I will not be dealing with it today. I will be working from my usual assumption: that Michael Stokes Paulsen is right about (almost) everything.

2. Since I started writing this article—which I first teased way back on May 3rd, and was clearly working on in mid-April—Kurt Lash has released the first-ever extended reply to Koppelman, although Lash dealt with only one aspect of Koppelman’s argument. We’ll discuss Lash’s piece later on. I’m grateful to him for writing it, because he knows a lot more than I do about the subject he addressed. I like to think that I add one or two useful points to his analysis.

3. Heh, made you look.

4. Koppelman’s is not the only argument for abortion rights that claims to be an originalist argument. It’s just the only one I find even initially persuasive. The other that comes to mind is Jack Balkin’s “living originalism” proposal, which trots out the 14th Amendment. I will leave the critique of Balkin's argument in the hands of the capable Michael Stokes Paulsen, writing in Does the Constitution Prescribe Rules for its Own Interpretation?, esp. pp878-881.

I suspect that Steven Calabresi and Livia Fine's Two Cheers for Professor Balkin's Originalism is a more thorough reply to Balkin, but I haven't finished reading it. In fairness, Thomas Colby thinks he has Paulsen licked in footnote 247 of The Sacrifice of the New Originalism. I disagree, but this is going too far afield from my actual topic today.

Some have also tried to ground their argument for abortion rights in the 9th Amendment. The most popular form of the argument, typified by Heidi Schreck’s anti-play, What The Constitution Means To Me, runs like this: the 9th Amendment clearly protects certain unenumerated “natural” human rights; We The People must get to decide what exactly those rights are; and We The People have decided, through Roe v. Wade, that one of these natural rights is the right to an abortion.

This is nonsense, which I’ve addressed elsewhere, which Michael McConnell has thoroughly undercut by reminding readers what “We the People” actually means in this context, and which has been deprived of its force by Dobbs anyway. After all, if judicial decisions stand for the determinations of “We the People,” the current Court’s determination that abortion is not one of those natural rights is just as efficacious as the Roe Court’s contrary decision. The judiciary occasionally withdraws putative individual rights, most notably when it overturned Lochner.

In full fairness to the Ninth Amendment, Justice William O. Douglas made a more sophisticated argument for a Ninth Amendment right to abortion in his concurring opinion for Roe v. Wade and Doe v. Bolton. After conceding that (emphasis mine), “The Ninth Amendment obviously does not create federally enforceable rights” (as the Heidi Schreck crowd wrongly believes), Douglas argues that the Ninth Amendment nevertheless protects rights that had already been recognized before the Ninth Amendment was ratified. Specifically, he argues, the Ninth protects “customary, traditional, and time-honored rights, amenities, privileges, and immunities.” That’s true enough, and is grounded in pretty rock-solid, respected precedents like Corfield v. Coryell (C.C.E.D.Pa., 1823).

The problem with invoking it in Roe, however, is obvious: abortion has never been a “customary, traditional, or time-honored” right. Abortion was precisely the opposite: at the time Roe was decided, abortion had been thoroughly (if not always efficaciously) outlawed under both American and (before that) English law for centuries. Douglas doesn’t deal with this problem at all, perhaps because he is acting under the influence of the very bad historian (and abortion advocate) Cyril Means, and has no idea how flatly his concurrence stomps all over its own premises. Douglas simply tries to analogize from actual “customary, traditional, and time-honored rights” (plus a few dumb Supreme Court precedents he was part of which claimed the same mantle) all the way to abortion, but the vast historical record showing directly that abortion is not a “customary right” obviates any need to draw strained analogies to other situations. Douglas’s argument is a better argument than Schreck et. al.’s “The Ninth Amendment protects everything I like.” But it’s still not a good argument.

5. I am not going to admit in main body text that the Twitter argument is what actually prompted me to finally put pen to paper, but LakeShowLeg was quite civil and earned a proper reply.

6. It says “man” in the original opinion, but both Koppelman and I agree that “man” here should be taken in its gender-neutral sense. That is, if it were Latin, it would be hominis, human person, not viri, male person.

7. See especially Originalism, Abortion, and the Thirteenth Amendment (2012), pp1938-1942, and also the final section of this Blog Post.

8. Argument B has evolved over the years. In Forced Labor ‘90, Koppelman argues that, because Jones v. Alfred Mayer Co. purported to expand the scope of Congress’s power to obliterate all the “badges and incidents” of slavery, the Thirteenth Amendment (as expanded by Jones) now allows the Thirteenth Amendment to be used against all analogous situations where a class faces “subordination… reinforced by a complex pattern of symbols and practices.” Because women are subordinated in a manner analogous to that of Blacks during and after slavery, the Thirteenth Amendment provides carte blanche to restructure society to eliminate that subordination, and abortion is a necessary component of doing that, especially because “sexism is as pervasive in the anti-abortion world view as racism was in the Southern peonage system.”

It is good that Argument B has evolved. I think every premise of that argument is wrong, and his comments on the anti-abortion movement we might politely call “the asinine prejudice of a grad school asshole.” Women’s subordination in our society is real, and patriarchy exists, but not to an extent that justifies an analogy to the plight of enslaved Blacks. Even if this were a close question, it would not be within the judicial power to determine it. Even if it were, constitutional amendments do not stand for vaguely analogous power structures unless they say so; we did not use the 19th Amendment (women’s suffrage) in order to secure adult suffrage by convoluted analogy and judicial fiat, but instead recognized that we had to pass the 26th Amendment (and did!).

Even if this subordination were accurately described and the constitutional analysis sound (there’s much more to be said about that, but see footnote 4, supra), abortion rights would not relieve the problems Koppelman raises, but would instead tend to exacerbate them, as my dear feminist mother (who taught me everything I know about standing up for the unborn) noted in a joint amicus brief in Dobbs. She co-signed that brief (organized by the incomparable Teresa Stanton Collett) with several hundred pro-lifers who don’t have a Y chromosome among them… and who thus know a hell of a lot more about womanhood than Young Andy Koppelman did. This primitive version of Argument B hardly even deserves the label “haruspicy.”

Happily, however, Koppelman evolved, and Argument B with him. In Forced Labor Revisited ‘10, Koppelman relies far more explicitly on an idea that had seemed marginal in Forced Labor ‘90: “loss of control over reproductive capacities” is not only analogous to slavery but partially constitutive of it. Originalism, Abortion, and the Thirteenth Amendment, while still gesturing toward the original “analogous” version of the argument, mainly serves to carry this revised “constitutive” argument further. The “constitutive” claim is stronger, presumably harder to prove, but it’s also the claim that breathes persuasive force into Argument B3.

9. Unless you are Professor Koppelman, in which case you presumably don’t need to.

10. Originalism, Abortion, and the Thirteenth Amendment (2012), pp1944-45

11. Note that the Fourteenth Amendment grants these rights to all persons, not just to all citizens. The Fourteenth Amendment states clearly that U.S. citizenship is conferred at birth, not at conception, but this cannot be read as a limitation on personhood. After all, recent immigrants have the same right not to be gunned down for sport as Koppelman or I have, even though they are not citizens. That’s because non-citizens (even non-citizens who immigrated illegally!) are still persons.

12. Corporations sole—corporations consisting of a single natural person—exist, largely to allow offices like “Secretary of State” to pass between officeholders without interruption. However, this does even less to advance Koppelman’s argument. If the “fetal office” (whatever that means) is a corporation sole (however that might work), the fetus would have to be the natural person comprising it. So we still end up with fetal natural personhood!

13. This particular quote is from Faulkner v. Commissioner of Internal Revenue, 41 BTA 875, quoting Cooper v. Heatherton, 73 NYS 14, quoting Thellusson v. Woodford, 4 Ves 227, relying on Long v. Blackall, 7 Term. R. 160, but you can find similar threads all over law, or just google “en ventre sa mere”.

Faulkner simply comes to mind because of my recent exploration of fetal tax law. Incidentally, if any pro-unborn lawyers are reading this, I can’t bring fetal taxation test cases, but maybe you can?

14. The “time of conception” is somewhat ambiguous. Does the human organism’s existence begin at the beginning of conception—first contact between sperm and egg—or at the end of conception, about twenty-four hours later, at syngamy, as is more commonly held? We need not resolve this question here, since in no event does a deliberate abortion take place until well after the conception process is entirely finished.

15. More Thelusson:

Let us see, what this non-entity can do. He may be vouched in a recovery, though it is for the purpose of making him answer over in value. He may be an executor. He may take under the Statute of Distributions. He may take by devise. He may be entitled under a charge for raising portions. He may have an injunction; and he may have a guardian. Some other cases put this beyond all doubt. [citing Clarke v. Clarke (1795) and Trower v. Butts (1823)].

Legally, what can a baby do that a fetus can’t?

(The answer is not “nothing,” but it’s a lot closer to “nothing” than really makes sense if unborn personhood is just an elaborate centuries-old legal fiction.)

16. I think especially here of Forced Labor ‘90’s Footnote 159, on page 517.

17. Don’t worry: he gets his food from DoorDash, a maid comes in on Tuesdays, his doctor does house calls, and he has a tight circle of friends online who write Callan fan fiction. Our blind tenant is trapped in his apartment but doing fine.

18. To be fair to abortionists, there are also major efficacy and maternal safety problems with misoprostol-only abortion; a living fetus does a much better job clinging to life and residence than an already-dead one.

Besides, even if a child survived a misoprostol-only abortion long enough to reach the toilet, she would not live long, and could easily escape notice altogether in all the other viscera. After all, this sort of abortion is strictly a first-trimester deal. (Live birth is quite possible if attempted later.) The largest possible baby in the first trimester is about the size of a peapod.

Still, it’s freaky enough that the other side goes to incredible lengths to obscure it. Check out these guys calling the dead baby the “product of expulsion”! (Yes, she’s a “baby” now cuz she’s been born.) Would they need to do such violence to the language if they didn’t suspect, deep down, whom they were doing violence to?

19. Incidentally, I don’t impute to the military any actual motive to oppress. I became fond of the military during my limited exposure to it while attending a military high school. My conclusion from that experience is that the military is what it thinks it has to be, does what it thinks it has to do, and tries its best to do it in a way that safeguards the rights of its members, those under its protection, and even its adversaries.

I just don’t think it’s sustainable to claim that military service doesn’t involve “involuntary servitude” in the same sense that Koppelman means it. There’s involuntary servitude even for volunteers, and much moreso for conscripts.

20. The case I linked, United States v. Knowles, predates the Thirteenth Amendment’s ratification by a few months. However, to my knowledge, nobody has suggested that U.S. v. Knowles was invalidated by the Thirteenth Amendment.

21. Those words from from Clyatt v. United States (1905), but Koppelman applies them to a vastly wider domain than Clyatt did. Clyatt was a peonage case about a certain specific form of labor contract. Its very next sentence is a direct refusal to apply this reasoning to more diverse situations:

We need not stop to consider any possible limits or exceptional cases, such as the service of a sailor, or the obligations of a child to its parents, or of an apprentice to his master, or the power of the legislature to make unlawful, and punish criminally, an abandonment by an employee of his post of labor in any extreme cases. That which is contemplated by the statute is compulsory service to secure the payment of a debt.

Koppelman simply does not heed the Court’s warning, applies the rule to all situations, and then tries to rationalize the remaining exceptions after the fact.

22. Koppelman, at various points, tries to expand the definition of rape and/or shrink the definition of genuine consent as a hedge against the possibility that his argument only works for rape victims. (The most direct and succinct iteration of this is in Originalism, Abortion, and the Thirteenth Amendment, Section V, pp1943-45.) These arguments mostly fail for the same reasons they failed when Judith Jarvis Thomson made the same arguments in her seminal “Defense of Abortion,” but, since it’s ancillary to Koppelman’s main argument, which I believe fails for other reasons, and since I’m not particularly interested in treating the unwanted children of rape differently from other unwanted children anyway, I’m not going to press the point in this article.

Koppelman is correct that, when a pro-lifer replies to the Thirteenth Amendment argument by pointing out that it only works in cases of rape, “The first thing [to] notice [is] how much it concedes” (Originalism, Abortion, and the Thirteenth Amendment, p1943). I think those concessions are as unnecessary to the Thirteenth Amendment argument as they are unfair to the children of rape.

23. More like 7.5 months, on average, since pregnancy is usually not detected until 5.5 weeks after the start of the last menstrual period (about 2.5 weeks after the child actually comes into existence), but nine months is a nice round figure.

24. Besides, I simply don’t have enough knowledge of contract law to speak of it more than glancingly without revealing myself as a fool.

25. Pope goes on to argue that there may be situations where parents so deeply abuse their plenary authority, and so profoundly act against the best interests of their children, that the Thirteenth Amendment could come into play. Although I find this idea plausible, it remains controversial and, at best, underdeveloped in federal courts, and it has no bearing on the Thirteenth Amendment argument for abortion rights.

26. There’s obviously a point where “some risk” becomes “imminent threat to life.” Parental duties generally end at that point, whether because an unborn child is dying and at high risk of putting Mom into deadly septic shock, or because a grown child has become violent and is coming at you with an assault rifle. In both cases, parents are legally permitted to act lethally if necessary for self-defense.

27. We have a similar difficulty with the Second Amendment, since colonial and early-American militias no longer exist. See generally Akhil Amar’s America’s Constitution: A Biography, pp322-362, a wonderful, highly readable book that I feel should be the textbook used in the first semester of any AP U.S. Government class.

28. Fun fact: this law gradually got worse, consigning mixed-blood people to hereditary involuntary servitude through several generations. Young Thomas Jefferson, six years before declaring Independence, tried to have these laws struck down, in the case Howell v. Netherland. He lost.

29. For examples of how the Slave Power became increasingly tyrannical—as well as increasingly certain of its own goodness—see The Civil War as Theological Crisis, by Mark Noll (2006), page 6, but also generally.

There is perhaps an interesting parallel in the modern movement from “safe, legal, and rare” to “shout your abortion,” but that’s probably best developed elsewhere.

30. I pass over the resemblance between the legal status of the slave (a human person deprived of all protections of law) and the legal status of the fetus (a human person both by biology and by Koppelman’s argument, yet deprived of all protections of law).

31. The Dred Scott court distinguished the Northwest Ordinance of 1787 itself by arguing that, as it was passed under the Confederation, when all states retained full sovereignty, it was not subject to the constitutional limitations that, in Chief Justice Taney’s “view,” forbade the Missouri Compromise. Then Taney does some awkward mumbling about how the Ordinance was repassed a few years later under the Constitution, unsuccessfully hand-waving it away. I really don’t have much patience for Taney.

32. I was also going to cite Akhil Amar’s America’s Constitution: A Biography on this point, just because I love plugging that book, but, frustratingly, his statement about the broadening meaning of “involuntary servitude” on p359 is vague and uncited!

33. Professor Lea VanderVelde, in The Labor Vision of the Thirteenth Amendment, uses a different interpretive methodology to argue for a somewhat broader reading of the Thirteenth Amendment. Her conclusions support broader economic liberties, such that you should probably replace “might” with “will” in the text accompanying this footnote.

However, VanderVelde, too, agrees that “No Congressmen claimed the term should apply to wives or children, relationships within the family which could be considered unequal and potentially abusive.” I note this not to lend support to her methodology—I take exception, for example, to her invocation of the ancient anti-originalist canard about the Air Force in The Thirteenth Amendment of Our Aspirations—but to observe that there is quite broad agreement across the interpretive spectrum that the Thirteenth Amendment responded to a particular kind of labor relation, not to familial relations. Thus, the Thirteenth Amendment ended neither child support laws (which are good) nor coverture laws (which are bad).

To the extent that VanderVelde’s argument in Our Aspirations culminates in a call for Thirteenth Amendment-based “reproductive rights,” I take her either to mean something other than a right to abortion (her text is not explicit) or, more likely, to endorse Koppelman’s argument (and thus to fail for the same reasons as Koppelman’s).

In a similar vein, Professor Dov Fox’s argument for abortion rights via the Thirteenth Amendment, in 2019’s Thirteenth Amendment Reflections on Abortion, Surrogacy, and Race Selection, explicitly follows Koppelman’s, and fails for the same reasons; as does James Gray Pope’s in Section 1 of the Thirteenth Amendment and the Badges and Incidents of Slavery.

34. Koppelman places great weight on showing that Robertson v. Baldwin is bad law (Forced Labor ‘90, pp523-526). After all, under Koppelman’s rule—any compulsion in the private service of another, even for a moment, even under contract, is unconstitutional involuntary servitude—Robertson should have been an easy case, and should have been decided the other way. Instead, Koppelman’s outcome lost 8-1.

It is therefore problematical for Koppelman that—despite his insistence that its holding is “bad law”—Robertson is still on the books. In fact, Robertson has (I am reliably informed) virtually no negative treatment on Westlaw. It’s routinely cited to this day. It’s racked up four new citations since I started writing this post in March, including an off-handed mention in the Supreme Court’s recent landmark decision in Bruen v. NYSRPA (gun rights).

I’m not particularly inclined to defend Robertson. Like many Lochner-era cases, it as pompous as it is sloppy. (Sorry, boys.) The majority decision is incredibly rude to sailors for no reason, calling them “deficient in that full and intelligent responsibility [of] ordinary adults,” and Harlan’s dissent is quite right that the majority’s long and winding road through Rhodian law is no more relevant to American law than Roe v. Wade’s digressions about the Persian Empire. Yet there a couple of things worth calling to our attention.

First, Koppelman concludes that Robertson’s central holding was implicitly overruled by certain sweeping declarations in Bailey v. Alabama, the 1911 case we discussed earlier (where the Court held that a Black debtor could not be held to peonage by a system slanted sharply against him). This seems tenuous, since Bailey mentioned Robertson in its text without critique. Bailey distinguished Robertson, but did not overrule it.

It seems easier to fit Robertson and Bailey together with one another—as well as the early American understanding of “involuntary servitude”—using the framework proposed in Nathan Oman’s Specific Performance and the Thirteenth Amendment: setting aside sweeping dicta on both sides, Bailey and its sequelae struck down debt peonage when the peonage was not entered into freely, did not pay proportionately, and, above all, featured the same “extreme domination that characterized the original understanding of ‘involuntary servitude.’” In Mr. Bailey’s case, the State itself had colluded in this domination, carefully designing a system to treat breach of contract as a criminal offense (rather than ordering damages, or even specific performance), amid a web of other laws designed to sap the Black man of his free labor.

By contrast, Robertson was, at least arguably, dealing with a system much less ferociously slanted against laborers. The seamen had made a contract and wanted out early. The Court noted they had freely entered the contract, passed over the other criteria in silence (including in Justice Harlan’s dissent), and ruled against the seamen. It seems economical to believe that the Bailey Court distinguished Robertson in no small part because the facts of Bailey were very distinct from those established in the record of Robertson.

Second, much negative commentary about Robertson centers on its assertion that:

It is clear, however, that the [Thirteenth A]mendment was not intended to introduce any novel doctrine with respect to certain descriptions of service which have always been treated as exceptional; such as military and naval enlistments, or to disturb the right of parents and guardians to the custody of their minor children or wards.

Koppelman reads this line with incredulity (p525):

[B]oth the Supreme Court and the lower courts have largely neglected this phrase, probably because it simply makes no sense. How can there be an exception that antedates the rule? In recent dictum [in U.S. v. Kozminski (1988)], the Court construed Robertson to say that “the Thirteenth Amendment was not intended to apply to ‘exceptional’ cases well established in the common law at the time…” …but this interpretation of the amendment must be mistaken, since it would exempt black slavery itself from the scope of the amendment. Slavery was “well established at the common law” of some states when the amendment was enacted.

You can hear his exasperation. It really feels like he would have used an exclamation point if legal journals allowed exclamation points. James Gray Pope makes a similar point in The Thirteenth Amendment at the Intersection of Class and Gender: Robertson v. Baldwin’s Exclusion of Infants, Lunatics, Women, and Seamen (p906-907):

By itself, this criterion [“services which have been from time immemorial been treated as exceptional…”] could not do the work of distinguishing exceptions from prohibited practices. The Thirteenth Amendment was enacted precisely to abolish practices that had been recognized in law from time immemorial. There could be no time-honored exceptions to a ban on slavery and involuntary servitude until and unless there was such a ban in effect. When Congress first provided for the forcible return of deserting seamen in 1792, for example, most states permitted slavery, indentured servitude remained lawful, and the Constitution provided for the return of any person “held to Service or Labour” who escaped across state lines. In that legal context, a seaman’s contract would have looked more like a short-term indenture—lawful in any trade—than an occupational exception to a general ban on slavery and involuntary servitude. On the criterion of origins in “time immemorial,” then, seamen’s contracts could not be distinguished from numerous other instances of coerced labor.

I think this is a straightforward misunderstanding. The “exceptional” services listed in Robertson were not treated as “exceptional” to the prohibition of involuntary servitude ratified in 1865. They were treated as “exceptional” to the concept of involuntary servitude that began to develop in America starting not later than 1618, for reasons already discussed in the body of this Blog Post.

Personally, I would not have described these services as “exceptional.” That suggests that they do properly fall within the meaning of “involuntary servitude” except for an unprincipled exemption. I would rather have said that the listed services never fell within the original public meaning of the phrase “involuntary servitude,” a legal term-of-art which encompassed a particular form of labor relation, at all. (This is also how I would rephrase Justice Brown’s later comments on the freedom of speech and the right to bear arms, at p281-282.) Of course, language about “original public meaning” would not be available for another eighty years or so, so I cannot blame Justice Brown for not using it.

This view has the disadvantage of colonizing Justice Brown with originalist thinking, but the advantage of actually making a lick of sense.

Third, let us suppose, for a moment, that Koppelman is correct that the central holding of Robertson—seamen can be compelled to perform their service contracts—has been implicitly overturned (or at least ought to be). I agree that’s a live possibility. Not only are various statements in the Peonage Cases pretty sweeping, but the case of seamen trying to get out of a lengthy contract in poor conditions when they’ll be under the near-absolute control of their captain arguably does fall within the original public meaning of “involuntary servitude.”

Even if the so-called “seamen exception” central to Robertson was wrongly decided and/or has been overturned, Robertson’s reasoning—that is, that there are some relationships which fall inside the modern plain-English meaning of “involuntary servitude” but outside the colonial/Civil War legal public meaning of “involuntary servitude”—has wormed its way deep into our case law. To my knowledge, no final ruling has ever questioned Robertson’s proposition that the parent-child relationship falls outside the scope of the Thirteenth Amendment in all or virtually all cases. (For the edge cases, see Pope’s Intersection of Class and Gender generally.) The most recent major Thirteenth Amendment case to reach the Supreme Court, U.S. v. Kozminski (1988, 487 US 944), heartily agreed with Robertson (by name) that the parent-child relationship was not subject to Thirteenth Amendment scrutiny. Pope notes (p903) that “Robertson has historically served to block jurisprudential development by preserving the domestic sphere as a zone where services can be coerced free from Thirteenth Amendment scrutiny.”

Although Pope argues that this consensus is ripe for reconsideration, and suggests that courts have begun to nibble on its edges, the fact that this consensus exists in our law is a thorny problem for Koppelman. Remember that Koppelman is precisely attempting to bring the Thirteenth Amendment into the domestic sphere—and claiming a Thirteenth Amendment right to destroy the life of one’s own child is not exactly nibbling at the edges! An awfully large consensus stands in his way.

35. I do think Jack Phillips should have won Masterpiece Cakeshop, and on much broader grounds—but those grounds are contained in the First Amendment, not the Thirteenth.

36. Here is an interesting, mostly irrelevant, but perhaps illustrative sidebar about the “meaning” of a legal text, and specifically the text “involuntary servitude.”

In the Thirteenth Amendment (and similar contemporary state legislation), “involuntary servitude” refers to private service. The phrase does not seem to include prison labor at all. The exception clause in the Thirteenth Amendment (“except as a punishment for crime…”) allows inmates to be “sold” out to private parties. If that clause had never passed, then criminal labor could not have been sold to private parties… but the State could still compel prison labor directly, within prisons. (Since the Thirteenth Amendment did include the exception clause, the State could do either, and, with considerable brutality, did both over the following century.)

However—this is the interesting bit—we in 2022 have largely lost the sense of what the words “involuntary servitude” originally meant. Here in 2022, it is popularly understood that prison labor is involuntary servitude, and that erasing the “exception clause” of the Thirteenth Amendment would render prison labor unconstitutional. This has led several states (including Alabama, Colorado, and Nebraska) in the 2010s and 2020s to pass state constitutional amendments stripping the “exception clause” out of their state versions of the Thirteenth Amendment, with the intention of outlawing prison labor altogether.

According to originalism, their state versions of the Thirteenth Amendment used to follow the 1860s meaning of “involuntary servitude,” and therefore had nothing to do with prison labor. However, since the public amended those amendments in the 21st Century, the amended text now follows the 21st Century meaning of “involuntary servitude.” Their state constitutions previously said nothing about prison labor, but because the enacting public clearly understood that the new amendments would affect prison labor, they did. The text “Neither slavery nor involuntary servitude shall not exist within this jurisdiction” does not abolish prison labor if the text was adopted in the 1860s… but, if adopted in the 2010s, it does. Prison labor is legitimately outlawed in Alabama, Colorado, and Nebraska.

Because “involuntary servitude” now encompasses something much broader in these revised state constitutions than it did when it was used in their 19th-century laws, it is possible that Professor Koppelman’s arguments for abortion rights would fare better in those states. It is also possible that jury duty is now unconstitutional in those states. I leave such questions to others. I merely find the re-enactment of identical text with divergent meaning really interesting. (Although probably somebody somewhere has already used it as the basis of an article to discredit originalism.)

Ryan Williams argues that the identical texts of the Fifth and Fourteenth Amendments bear different meanings for similar reasons in his provocative (albeit now decade-old) “The One And Only Substantive Due Process Clause.” I can’t vouch for (or against) his case because I haven’t finished reading it. But the divergent meaning of “involuntary servitude” is getting put into law right now, which is pretty neat.

37. The rest of the Civil Rights Act describes how this sweeping statute will be enforced, and by what penalties. It is interesting reading, but not relevant to our inquiry.

38. It bears mentioning here that the veto override only worked because the Radical Republicans had ejected Sen. John Stockton (D-NJ) from the Senate on legally questionably grounds—and they seem to have done so precisely so as to reach the two-thirds threshold for veto overrides. Had Stockton voted, the overrides would have failed.

39. There’s that Fourteenth Amendment language leaking through from 1868 again!

Other inherent rights of citizenship suggested during the 39th Congress (though not enshrined in the Civil Rights Act) included: the right to go about freely (that is, without documentation or impediment); the right to marry (an opposite-sex member of the same race); the right to preach; the right to educate; the right to be educated; the right to have and raise a family; the right to bear arms. All are among the basic privileges and immunities that eighteenth- and nineteenth-century Americans believed synonymous with freedom. (For citations to the 1866 Congress, see James Gray Pope’s Section 1 of the Thirteenth Amendment and the Badges and Incidents of Slavery, pp433-451.)

However, these particular civil rights were not included in the Civil Rights Act. Some would raise their heads again in later jurisprudence; others would not.

Nevertheless, the idea that a free person might not have these rights was fairly shocking to many Northerners. Forthrightly evil legislatures in the South were trying to invent second-class citizenship in real time, largely out of whole cloth, and Congress had its work cut out keeping up with them.

40. There is a pretty interesting debate about the role of Section 2 in the Thirteenth Amendment (“Congress shall have the power to enforce this article by appropriate legislation”). Specifically: whether Section 2 creates any power beyond enforcement of Section 1’s single self-executing clause, whether and to what extent the “badges and incidents” doctrine depends upon Section 2 rather than Section 1, and what degree of discretion Congress has to answer those questions for itself.

There is also a live debate about whether the Thirteenth Amendment protects all discrete groups equally, or reaches only racial discrimination, or reaches only anti-Black discrimination specifically. A good place to start getting familiar with some of these discussions is Jennifer Mason McAward’s Defining the Badges and Incidents of Slavery, from which I have cribbed generously in writing this blog post. Among many other things, she covers Cruikshank (1874—the solo “justice riding circuit” decision, not the disastrous SCOTUS decision) and LeGrand v. United States (1882), which deal directly with the question of whether non-slaves or non-Blacks can be protected by the Thirteenth Amendment.

Meanwhile, many lawyers are eager to unleash the Thirteenth Amendment as a new fount of constitutional haruspicy, and have presented long arguments to that effect. Quite a lot of current Thirteenth Amendment scholarship is about the potential for what Jamal Greene calls “Thirteenth Amendment optimism,” although his idea of optimism is very different from my own.

41. I suppose the argument here is that, instead of being incidents of slavery, they were incidents of sheer poisonous anti-Black racism, but the people making this argument had the good sense not to admit that directly in court filings.

42. Today’s federal public accommodations laws are based on the Interstate Commerce Clause, not the Thirteenth or Fourteenth Amendments, which is how they evaded this holding. State public accommodations laws require no special justification, so long as they do not violate federal law, because states have plenary power.

43. As we have seen, Lash and Upham provide pretty good reasons for thinking that this outcome was legally correct, despite leading to significant injustices against Black citizens. The Thirteenth Amendment doesn’t say anything about “badges and incidents,” therefore it grants no power over them, therefore Rep. John Bingham was right and the Civil Rights Act of 1866 was indeed unconstitutional in the absence of the Fourteenth Amendment.

For its part, the Hodges Court didn’t fret very much about the text or its original meaning. They seemed principally concerned about the reach of the Thirteenth Amendment if they upheld the Civil Rights Act. It is evident from their decision that they had lost contact with any theoretical foundation for the “badges and incidents” theory, and thus any limiting principles within the “badges and incidents” theory. Seeing Congress’s purported power to regulate the “badges and incidents” of slavery as an essentially plenary power, the Court felt it had no choice but to strike it down.

After all, Congressional plenary power would destroy the whole constitutional order, effectively erasing states. On the other hand, limiting that plenary power to “the African race” would turn Blacks into special permanent wards of Congress “like the Indian tribes”. Neither seemed like something the Thirteenth Amendment could possibly do by mere implication, so they rejected the whole “badges and incidents” theory. A more clearly defined version of the theory might have fared better.

44. If it’s not clear, this is a backhanded comment on the fake stare decisis in the Dobbs v. Jackson dissent, not a dig at Jones. On the contrary, I applaud Jones for having the courage to put its idea of justice over stability in a constitutional case!

45. …or by someone noticing a line in a controlling opinion that I missed.

46. Whether the same must be said of an empire that allows—even encourages—the wholesale slaughter of its own children in the name of modern false idols, I leave as an exercise for the reader.

47. If there is such a movie out there, let me know in the comments!

48. Male slaves were also victimized in some or all of these ways, but surely even the most far-gone gender nominalist will agree that slave women were in greater peril from these particular harms.

49. One might further argue that the abolition of slavery must bring the woman citizen to stand on the same ground as men citizens. For example, coverture laws, which disabled the legal standing of specifically married women, seem, at first impression, to be vulnerable to a “badges and incidents of slavery” attack. However, even if you can sustain the “badges and incidents” theory in the face of historical criticism, the “wall of separation” that the Thirteenth Amendment appears to impose around parent-child and husband-wife relationships (as we discussed in our survey of involuntary servitude) may present an insurmountable obstacle to this attack.

Elizabeth Hearn goes even further out on this limb by arguing that the Thirteenth Amendment justifies the Violence Against Women Act. (I have not read her argument or VAWA, so I can’t vouch for against it, but it’s an interesting notion.)

None of this will do, however, for a right to abortion, as we will discuss momentarily.

50. As a moral matter, everyone should have the right to resist sexual relations. However, legally speaking, it was generally agreed in the 1860s that marriage granted permanent sexual consent. The right to refuse intercourse to one’s husband, specifically, was not established until the late Twentieth Century.

Meanwhile, the right of males not to be raped is still not fully established throughout the Western World. Progress is slow!

For our purposes, the “badge and incident of slavery” is lifted as soon as one has the same legal rights against rape as an ordinary citizen has—even if those rights remain morally inadequate.

51. In footnote 197 of Forced Labor ‘90, Koppelman claims common-law protection for early abortion. However, as Robert George and John Finnis explained in their Dobbs amicus, even where abortion was sometimes semi-tolerated, it was still to “no lawful purpose,” nobody would have called it a “right,” and everyone recognized the legislature’s power to interdict it indirectly (or, when practical, directly). Legislatures did, indeed, punish early abortions indirectly (e.g. through felony-murder prosecutions of abortionists whose patients died), right up until science and policing had advanced enough that they felt compelled to start prohibiting early abortions directly. (George and Finnis more recently expanded on / recapitulated this in Indictability of Early Abortion c. 1868, in response to Aaron Tang’s unpersuasive critique.)

Put another way: my home state of Minnesota didn’t have a law against revenge porn until 2016. Does that mean that Minnesotans had, from time immemorial, a right to create and share revenge porn, and the state stripped that right away from us just five years ago? Of course not; the legislature always had the power to legislate against revenge porn, and everybody knew it. Toleration of a crime does not make it a right.

As the Dobbs majority persuasively (if concisely) illustrated, history is clear: even where grudgingly and partially tolerated, antebellum abortion pre-quickening was always seen as unlawful, never as a right.

52. It is a euphemism, and not simply a change in the meaning of the phrase, because abortion has nothing to do with control over one’s reproductive capacity. Anyone who procures an abortion has already reproduced. For better or worse, once you have a child, questions of “reproductive capacity” are in the rearview mirror, and the question becomes what you’re going to do with your offspring. To deny this is to deny a century of utterly uncontested science. Vaccine denial, while ignorant, is far less ignorant than this.

53. It’s odd to call a 32-year-old idea “novel,” but his argument has played so little role in mainstream abortion debate that it still feels novel and plenty of people are still encountering it for the first time.

54. I wish to thank Professors Kurt Lash and Michael Stokes Paulsen for generously reading a draft of this article (over Christmas break, no less) and suggesting a few improvements, principally in my treatment of the Civil Rights Act of 1866. I’d never communicated with Prof. Lash in my life, and I haven’t seen Paulsen in probably a decade at this point, so it was particularly kind of them to take a look at this manuscript after I threw it over their transoms in the middle of Christmas grading. Any remaining flaws are entirely mine.