Sometimes you just have to muddle through

Talevski and the victory of multi-factor tests

The Supreme Court is much more conservative than it used to be. As such, petitions asking the Court to overrule prior decisions have become hardy perennials on the Court’s docket. The Court deals with these cases in various ways. Sometimes it goes all the way and overrules prior decisions, as in Dobbs. Sometimes it limits old cases to their facts. And occasionally, not often, the Court accepts “liberal” decisions and applies them more or less according to how the prior version of the Court intended.

Health and Hospital Corporation of Marion County v. Talevski, issued on June 8, 2023, falls into the third category. In Maine v. Thiboutot, 448 U.S. 1 (1980), the Supreme Court, per Justice Brennan, held that plaintiffs could bring claims under 42 U.S.C. § 1983 to enforce federal statutes. In two subsequent bitterly divided 5-4 decisions, the Court held that federal statutes enacted under the Spending Clause were enforceable under § 1983. See Wilder v. Virginia Hospital Assn., 496 U.S. 498 (1990); Wright v. Roanoke Redevelopment and Housing Authority, 479 U.S. 418 (1987). (I’ll explain what all this stuff means further down.)

In Talevski, the Supreme Court granted a petition for certiorari asking the Supreme Court to either overrule these decisions or sharply limit them. The majorities in Wilder and Wright consisted of Justices Brennan, Marshall, Blackmun, Stevens, and White, so … the writing looked like it was on the wall. But the Court didn’t bite. In a 7-2 decision by Justice Jackson, the Supreme Court affirmed the Seventh Circuit’s decision allowing the plaintiff’s § 1983 claim, alleging a violation of the Federal Nursing Home Reform Act, to proceed.

At first glance, one would not have expected this outcome. The Thiboutot line of cases seem like the exact type of cases that would be triggering to the current Court. You’ve got Congress using its Spending Clause authority to impose its will on states. You’ve got plaintiffs suing state actors. You’ve got implied causes of action. You’ve got an indeterminate multifactor test. You’ve got a 5-4 opinion by Justice Brennan. Why stick to these liberal activist cases, and not only that, but actually let the plaintiff win?

Occasionally one hears speculation along the lines of “the Supreme Court wants to look like it’s sometimes sticking to stare decisis so it can feel free to overrule other cases it cares about more.” I find these theories to be unpersuasive and lazy. The Justices have life tenure; they can, and do, vote however they want.

In my opinion, the reason that the Court stuck with precedent is that there was no better option. It would be nice to have a Grand Unified Theory of the Spending Clause and § 1983, but unfortunately, it’s impossible. Justice Thomas attempts to offer such a theory, and God bless him for trying, but it doesn’t work. When a constitutional provision deliberately written to be ambiguous is combined with a statute containing a 150-year-old drafting error, efforts to make the law harmonious are doomed to fail. Unfortunately one has to muddle through, and if muddling is needed, the Court might as well stick with the muddling that went on before.

(Nursing homes, Spending Clause … Dall-E did a nice job on this one.)

Do or do not. There is no Spending Clause.

Let’s start with first principles.

The Supreme Court frequently refers to the “Spending Clause.” But if you open a digital copy of the Constitution and do CTRL-F for “spending,” you will be surprised to hear that annoying beep.

The “Spending Clause” refers to Article I, Section 8, Clause 1, which says this:

The Congress shall have Power To lay and collect Taxes, Duties, Imposts and Excises, to pay the Debts and provide for the common Defence and general Welfare of the United States.

Reading this language, a question comes to mind. Can Congress spend money on whatever it wants, even if the spending cannot be justified based on its other enumerated powers (such as its Commerce Clause authority)? Hmm…

On the one hand, if Congress can use tax revenue to “provide for … the general Welfare,” it sounds like Congress can spend money on the “general Welfare,” without grounding that spending in some other enumerated power.

Also, Congress is separately given the authority under Article IV to make “all needful Rules and Regulations respecting the Territory or other Property belonging to the United States.” Maybe “Property” encompasses money, which means Congress can spend money on whatever it wants? And, it seems like an inherent attribute of sovereignty that a government can spend its own money however it wants.

On the other hand, the Constitution doesn’t actually say that Congress can spend money on the “general Welfare.” It says that Congress can collect taxes, with the proviso that those taxes are used for the general Welfare. And isn’t the federal government supposed to be one of enumerated powers?

If you find this dispute difficult to resolve, you are not alone. The Constitution appears to be deliberately ambiguous on this issue, and people have argued about it since the early Republic. Alexander Hamilton and Justice Story’s Commentaries took the broader view, James Madison took the narrower view. Finally, in the famous case of United States v. Butler, 297 U.S. 1 (1936), the Supreme Court said that Hamilton’s view was correct: Congress’s power to spend money wasn’t limited by the direct grants of legislative power in the Constitution.

Justice Thomas’s Talevski dissent seems to disagree with the Hamiltonian view, although he doesn’t advocate overruling Butler and the stare decisis case for adhering to Butler is extremely strong. (Can you really say that Butler is egregiously wrong when Story’s Commentaries advocated for it in 1833?) But given that the Constitution is ambiguous on the basic question of what Congress can spend money on, and Hamiltonian and Madisonian views will always be with us, efforts to achieve consensus in Spending Clause disputes are off to a bad start.

I’d do anything for federal money (but I won’t do that)

So the government can spend money on whatever it wants. Can it give money to people with strings attached?

It sure seems like the government should be able to do this. If it can spend money on what it wants, it can control how its money is used, including controlling what recipients do with the money.

Also, it seems intuitive that the government should be able to make deals, in which it spends money in exchange for some concession. Deals, by definition, make both parties better off. Hooray for deals!

Should the government be able to give money to states with strings attached?

It should. The federal government frequently prefers to give money to states, and have the states administer the money, than administer the money by itself. For lovers of federalism, this seems good! If the federal government is banned from entering into deals with states, it will be forced to administer its own spending programs, and the Deep State will metastasize.

But this seems to give the federal government a lot of power. It can tell people: “If you don’t do exactly what we want, we will cut off all federal funds.” Given how much money the federal government collects and spends, people might not have any realistic choice but to obey. Maybe the political process would solve this problem, but maybe it wouldn’t. And the result is that Congress would effectively be able to regulate in a manner exceeding its enumerated powers.

Similarly, the federal government can tell states: “If you don’t regulate exactly as we want, we will cut off all federal funds.” States will have to obey, too, which means that the federal government will effectively be telling states how to regulate. This seems bad and even unconstitutional if you’re persuaded by the Supreme Court’s “non-commandeering” cases.

(Incidentally, under the Madisonian view of the Spending Clause, this problem still exists. Even if Congress can only spend money in furtherance of its Commerce Clause powers, that’s still a lot of money. It’s more than sufficient money for Congress to bend the states to its will, if Congress is willing to play hardball.)

So we have three options.

• Option #1: Congress can’t put any strings on money it spends. Can’t be right.

• Option #2. Congress can put any strings it wants on money it spends. Hello, Leviathan.

• Option #3. Some kind of position in the middle. Congress can impose some strings but it can’t Go Too Far.

Options #1 and #2 seem bad enough that in Butler the Supreme Court went with Option #3: the government can spend money with strings attached unless there are too many strings. In Butler the Supreme Court held, by a 6-3 vote, that the law at issue was “coercive”—too many strings.

Then came the famous Switch in Time that Saved Nine, where Justice Roberts flipped from conservative to liberal so that Roosevelt wouldn’t pack the Court. The next year, the Supreme Court heard Steward Machine Co. v. Davis, 301 U. S. 548 (1937), and, mirabile dictu, upheld a different Spending Clause statute by a 5-4 vote with Justice Roberts (and Chief Justice Hughes, the other swing voter) in the majority. The dissent thought the law interfered with state authority too much, the majority thought it didn’t.

Since then lawyers have been arguing a lot about what Goes Too Far. The most recent iteration of this dispute was NFIB v. Sebelius, 567 U.S. 519 (2012), where the Supreme Court held that although the feds could tax you for not eating broccoli, the Affordable Care Act’s threat to withhold Medicaid funding unless Medicaid programs were expanded was “coercive” and hence unconstitutional. (Justice Ginsburg’s dissent on this issue is excellent.)

Maybe Option #3 is the best option, but it isn’t particularly desirable either. The Supreme Court has devised its typical multi-part test: the condition can’t be “coercive,” it has to be sufficiently related (“germane”) to the spending, and it has to be “unambiguous.” None of these requirements is written in the Constitution and they are all beyond indeterminate and manipulable. Is an incentive strong, or too strong? Here’s what the majority in NFIB v. Sebelius says:

The Court in Steward Machine did not attempt to “fix the outermost line” where persuasion gives way to coercion. 301 U.S., at 591, 57 S.Ct. 883. The Court found it “[e]nough for present purposes that wherever the line may be, this statute is within it.” Ibid. We have no need to fix a line either. It is enough for today that wherever that line may be, this statute is surely beyond it.

Not helpful! The other requirements are just as bad. Sure, a condition is related to an expense, but is it related enough? The condition is clear, but is it really clear? Not promising for a coherent body of law.

Trust, but verify

Let’s say a spending condition satisfies this ambiguous multi-part test. The condition has to be enforceable somehow, right? The condition wouldn’t be very meaningful if recipients of federal funds could violate it with impunity.

That principle remains true when the recipient of federal funds is a state. The federal government has to have some way of enforcing the strings attached to the grant.

One way to do this is for the federal government to sue the states. But suppose the federal government wants to outsource this task to private citizens. In other words, private citizens can sue the state, on the theory that the state didn’t adhere to the federal strings attached to federal grants. Does the federal government have the authority to do so?

Yes. Under Article I, Congress has the power “to make all Laws which shall be necessary and proper for carrying into Execution the foregoing Powers, and all other Powers vested by this Constitution in the Government of the United States, or any Department or Officer thereof.” If we assume Congress has the power to spend money with strings attached, then a law providing a mechanism to enforce those strings is pretty much the paradigmatic example of a law that is “necessary and proper for carrying into Execution” the power to spend money with strings attached.

Also, while history is uneven on this, there are plenty of cases, from both the olden days and the ChatGPT era, in which third party beneficiaries can sue to enforce contracts. There’s nothing particularly novel about A suing B because B breached an agreement with C. And that’s essentially what happens when private citizens sue states for violating funding conditions imposed by the federal government.

Coming at it from a different direction, what’s the constitutional basis for saying that only the federal government can sue states for breaching a funding condition, and not a private citizen? There is none. The Constitution says nothing about anyone suing states for breaching funding conditions. If we say only the federal government can do it and not private citizens, we are making things up.

But this shouldn’t happen very often because we know from lots of modern Supreme Court cases that implied causes of action are dead. If Congress doesn’t explicitly say a plaintiff can sue, the plaintiff can’t sue. And when Congress makes block grants to the states, it rarely provides express causes of action for private citizens.

Unless …

Oops!

Unless Congress conferred a private right of action by accident.

After the Civil War, Congress enacted 42 U.S.C. § 1983. The purpose of this law was to allow citizens to sue states for violating the Constitution. Here’s what it says:

Every person who, under color of any statute, ordinance, regulation, custom, or usage, of any State or Territory, subjects, or causes to be subjected, any citizen of the United States or other person within the jurisdiction thereof to the deprivation of any rights, privileges, or immunities secured by the Constitution and laws, shall be liable to the party injured in an action at law, suit in equity, or other proper proceeding for redress.

Wait, it says “secured by the Constitution and laws.” “And laws.” How did “and laws” get in there?

This is a long story, but the short answer is, because some anonymous federal bureaucrat in 1874 messed up. Justice Powell’s dissent in Maine v. Thiboutot walks through this in detail; here’s Will Baude with an executive summary. As an executive summary to the executive summary, Congress enacted § 1983 in 1871 without the “and laws” language, and then decided to “codify” federal law in 1874 into a set of Revised Statutes, and the “codifier” accidentally added “and laws,” and Congress passed the Revised Statutes without realizing the mistake.

But hey, the text is the text. And if Congress didn’t like “and laws,” wouldn’t it have corrected the error over the past 149 years?

OK, so we’re stuck with “and laws.” What are the implications of that? Section 1983 applies to the “Constitution and laws.” We know that with respect to the Constitution, a plaintiff can file a lawsuit even though the underlying document—the Constitution—doesn’t create a private right of action. For example, an arrestee who was beaten by a police officer can sue the police officer under the Fourth Amendment, even though the Fourth Amendment doesn’t expressly confer a private right of action. Indeed, enabling such suits was the whole point of § 1983.

So if that’s true of the “Constitution,” that also has to be true of “laws.” By its plain text, § 1983 treats the “Constitution” and “laws” the same way. So it must be that a plaintiff can sue state actors for violating “laws,” even though there’s no private right of action in the underlying “laws.”

Not so fast, says Justice Thomas. He takes the view that plaintiffs can’t sue under § 1983 to enforce Spending Clause conditions. His theory is that Spending Clause conditions aren’t “laws,” and plaintiffs’ rights under those conditions aren’t “secured” by the laws because they’re “secured” by the States’ acceptance of the conditions.

This is a real stretch. Laws enacted under the Spending Clause are … just that. Laws. Justice Thomas observes that “the Necessary and Proper Clause is a natural candidate for the spending power because spending funds may be ‘necessary and proper for carrying into Execution’ the Federal Government’s enumerated powers.” I agree. The Necessary and Proper Clause authorizes Congress to “make all Laws which shall be necessary and proper for carrying into Execution the foregoing Powers.” “Laws.” Laws authorizing the spending of money are laws.

As for “secured” … if the law gives the plaintiff the right to sue the state, then the plaintiff’s right to sue the state is secured by the law. OK, the state has to accept the condition. But once the state accepts the condition, it’s the law that secures the right.

This is at least a reasonable position. I just don’t see how you can say it’s so egregiously wrong as to warrant overruling multiple binding precedents.

Justice Thomas says that if the Spending Clause “law” “secures” rights, it violates the non-commandeering doctrine. I don’t buy it. We wouldn’t get to the § 1983 question unless the condition was constitutionally permissible—i.e., clear, germane, and not overly coercive. And if that’s so, it’s not a violation of the non-commandeering doctrine for the federal government to impose the condition. And if that’s so, I can’t understand how it’s a violation of the non-commandeering doctrine if the condition is enforced by a private citizen rather than the federal government.

Oftentimes multiple Justices sign onto Justice Thomas’s originalist dissents. This time no one else did. This is not because his fellow Justices are wusses who are scared that the woke media will criticize them for overruling decisions by Justice Brennan. It’s because the argument doesn’t work, at least in my view.

Sailing the seas of uncertainty

We are back to a world of no good options.

• Option #1: Plaintiffs can’t sue under § 1983 to enforce federal laws; they can only sue to enforce the Constitution.

• Option #2. Plaintiffs can sue state actors under § 1983 every time they’re injured by the state actors’ violation of any federal law.

• Option #3. Some kind of position in the middle.

Option #1 conflicts with the statutory text. Section 1983 says “Constitution and laws.” Not “Constitution.” We can’t read “and laws” out of the statute. It’s there. Maybe by accident, but it’s there.

Option #2 conflicts with the statutory text too. Section 1983 permits suits to enforce the “rights, privileges, or immunities secured by the Constitution and laws.” So it shouldn’t be enough that the state actor violated a law: the plaintiff should have to show that the law created a “right” that the state actor violated.

So Option #3 is all we have left. We are going to have to devise a test that leaves us in the middle.

If we’re going to have to do this, we might as well look at prior Supreme Court cases that have attempted the same task, and they seem somewhat reasonable?

Gonzaga University v. Doe, 536 U. S. 273 (2002), holds that a Spending Clause statute isn’t enforceable under § 1983 unless it includes “clear and unambiguous” “rights-creating” language. Seems like a fair compromise. If Congress doesn’t create an express right of action, we may as well require it to almost create an express right of action in order to trigger § 1983’s accidental right of action.

Another line of cases, leading up to Fitzgerald v. Barnstable School Committee, 555 U.S. 246 (2009), holds that a § 1983 claim is in some cases implicitly precluded by a “comprehensive enforcement scheme that is incompatible with individual enforcement under § 1983.” Who could argue against that? It’s bad for things to be incompatible with other things.

As multipart tests go, this one isn’t so bad.

And so there were eight votes to stick with it.

I’m not saying we’re compatible, I’m just saying we’re not incompatible.

The final task in Talevski was to apply the test.

The Federal Nursing Home Reform Act provides that nursing homes receiving Medicaid funds are subject to certain conditions. The plaintiff—the representative of a now-deceased nursing home resident—alleged that the resident was abused in his nursing home, in violation of those conditions. The nursing home was operated by a state actor (the local government), so the plaintiff sued under § 1983.

Does this law include “clear and unambiguous” “rights-creating” language? It does. It repeatedly says that nursing home residents have certain “rights.” A statute that creates “rights” has “rights-creating” language.

Does the law contain a “comprehensive enforcement scheme that is incompatible with individual enforcement under § 1983”? Oh god. How are we supposed to figure that one out?

The problem here is that you can pretty much always say that § 1983 suits are incompatible with Congress’s enforcement scheme:

• If the law includes no private enforcement scheme, you can say that providing a private remedy is incompatible with Congress’s decision for the conditions to be enforced only by the federal government. (This is Justice Alito’s position in dissent.)

• If the law does include a private enforcement scheme, you can say that courts shouldn’t replace one carefully tailored private enforcement scheme with § 1983’s different enforcement scheme.

More concisely, if Congress made the policy decision not to expressly authorize § 1983 lawsuits, then § 1983 lawsuits will always in some sense be “incompatible” with the policy decision Congress did make. But we know that some § 1983 lawsuits have to survive—otherwise we’d be reading the “and laws” language out of § 1983.

It’s a head-scratcher.

The federal government’s solution to this quandary was to invite courts to examine the policy consequences of permitting § 1983 suits in every case. This resulted in the following exchange at oral argument:

JUSTICE KAGAN: Well, it does seem kind of a new idea. I mean, you know, this is a statutory interpretation question. We’re -- we’re trying to figure out whether one scheme is compatible with another scheme. You’re not looking at text. You’re not looking at structure. You’re not looking at history. You’re looking at, like, what you think Congress knew about the market composition. The -- and that -- that does seem -- Justice Kavanaugh said unusual. It seems unusual.

[FEDERAL GOVERNMENT’S LAWYER]: So, Justice Kagan, the best analogy we've been able to come up with is that in -- I'm hesitant to say this, but in this Court's ACCA cases, the Court has sometimes looked --

JUSTICE KAGAN: You should hesitate to say that. (Laughter.)

“ACCA” refers to the Armed Career Criminal Act. For those unfamiliar with this esteemed statute, one portion of ACCA was so incomprehensible that the Supreme Court struck it down as unconstitutionally vague in 2016. Another part of ACCA (which, in context, I think the government lawyer was referring to) is sufficiently obscure that it resulted in Justice Alito comparing the Court to Sabine Moreau, a Belgian driver who took a wrong turn and accidentally drove to Zagreb, Croatia. The government’s position did not prevail.

The Talevski majority instead held that a § 1983 suit is “incompatible” with an enforcement scheme if there were “indicia of congressional intent to preclude § 1983 enforcement.” It’s not enough to show that Congress didn’t provide a private remedy; if some provision that Congress did enact would be effectively nullified by the § 1983 claim, the claim is unavailable. This might happen if, for example, Congress provided an express right of action with particular limitations that a § 1983 suit would allow the plaintiff to circumvent.

In my view, the Court had no other choice but to adopt this rule. Every other option was either completely indeterminate or would foreclose § 1983 suits in every case.

And applying that test, the plaintiff had to win. The § 1983 suit wasn’t circumventing any express limitations on private enforcement, because a private enforcement scheme didn’t exist.

Wait, so why did the plaintiff win again?

Let’s circle back to where this post started. Why did the Supreme Court choose this case to stick with Justice Brennan, apply a multi-factor test, and rule for the plaintiff?

It’s because when history is a mess, fidelity to history requires doctrine to be a mess. The Constitution leaves the door ajar for Congress to impose spending conditions on states, and § 1983, by accident, leaves the door ajar for private plaintiffs to enforce those conditions. Shutting those doors would not be faithful to the Constitution or § 1983. Opening those doors too wide would lead to unacceptable consequences. So the Supreme Court had to leave them ajar. And because prior Supreme Court cases already did that, why overrule them? Legal tests that leave doors ajar are not very satisfying to the purist, but sometimes there are no better options.