Harvard Law School Professor Charles Fried has written a major article analyzing the Supreme Court’s late-June decision on the Patient Protection and Affordable Care Act. The article, which is scheduled to appear in an upcoming issue of the Journal of Health Politics, Policy and Law, and also in a volume to be published by Oxford, has been given a rare advance publication by SCOTUSblog, which posted it on August 2.

Fried—a noted constitutional law scholar, former U.S. Solicitor General, and a former Justice of the Massachusetts Surpreme Judicial Court—has weighed in on the Act numerous times over the past several years. In February 2011, he testified before the Senate Judiciary Committee at a hearing on the constitutionality of the individual mandate provision of the PPACA. At an HLS event in March 2011, he debated the constitutionality of the provision alongside HLS Professor Laurence Tribe ’66 and Georgetown University Law Center Professor Randy Barnett ’77, chief architect of the legal challenge to it. And this January, he served as counsel of record in an amicus brief filed with the Court supporting the constitutionality of the mandate.

The June surprises: Balls, strikes, and the fog of war

© Charles Fried, 2012

Chief Justice Roberts, in his confirmation hearing, famously analogized the role of the judge to an umpire, calling balls and strikes. Many of those relieved at the outcome of the litigation about the Patient Protection and Affordable Care Act (ACA) think the Chief Justice called a ball a strike, a strike a ball, but at the end of the day at least the batter was safe on base. [Ball 1 incorrectly called a strike, strike 1 incorrectly called a ball, balls 3 and 4 correctly called balls, strike 2 incorrectly called a ball: batter walks.] Though a dwindling number of Court watchers thought the mandate in the ACA might survive, the actual outcome of the case came as a surprise – a bag of surprises, indeed – to almost everyone.

Few had predicted that Chief Justice Roberts would be the only one of the Republican-appointed Justices to vote with what the media call the liberal four to carry the mandate over the top. Even fewer thought this would be on the ground that the mandate could be and thus must be justified as an exercise of Congress’s taxing power. Among constitutional experts the attack on the Medicaid expansion seemed even more ill-grounded than the Commerce Clause attack on the mandate. And most surprising of all, no one would have predicted that the Medicaid expansion would fail by a vote of seven to two, with Justices Breyer and Kagan joining the Republican majority.

Finally, even though the Chief Justice’s lead opinion rejected the Commerce Clause basis for the mandate in the same terms as his four Republican colleagues, those four did not join, or even so much as mention their Chief’s opinion on this issue, but instead published a highly unusual (though not entirely unprecedented) joint opinion which – whatever its motivation – had the appearance of a deliberate repudiation, not of the opinion but of its author. It is these four surprises that are the subject of this essay. In setting the stage for these surprises, I am afraid I must go over some ground that constitutional scholars would by now have plowed into a veritable dust bowl of commentary, but the lesson I draw may be a different one from the many that are being drawn from the unexpected dénouement of this most important case.

I. Background

I was thoroughly engaged in the case, though not as a partisan so much as a constitutional scholar. I had written about it; testified to the Senate Judiciary Committee; in three different fora debated Randy Barnett (the owner if not the inventor of the broccoli argument); was the counsel of record in an amicus brief for 104 health law professors; and in October 2010 in an interview on Greta Van Susteren’s Fox TV program offered to eat my hat on camera if the Court struck the Act down. As I testified to the Senate, I have doubts that the Act is good policy and am pretty sure that it will not improve the medical situation or the wallet of well-insured, upper-middle-class, well-connected persons like me.

But like most students of constitutional law I found the argument that it violated the Constitution bordered on the frivolous. After all, since 1937 the Court had not come even close to invalidating on Commerce Clause grounds a statute that was without question one of economic regulation. (The 2000 and 1995 Morrison and Lopez cases prove the point, because penalizing beating up a girlfriend or carrying a gun near a school – compared with offering meals at Ollie’s Barbecue – could not without a very long stretch be characterized as economic regulation.)

The Commerce Clause argument seemed so bad that neither the district court decisions by Judge Vincent in Florida and Judge Hudson in Virginia – written in extravagant terms that seemed to be dictated by right-wing talk radio – nor the fact that all the major Republican candidates for the presidential nomination competed to outdo each other in condemning particularly the mandate, nor even the polls showing that the public had been decisively turned against the mandate, had much moved me. But then a decision in the Eleventh Circuit upholding Judge Vincent and the skillful advocacy of Paul Clement shook my confidence – even after remarkable opinions by Judges Sutton and Silberman upholding the Act.

Very briefly, the argument that never before had Congress included as a premise for regulation of commerce a requirement that individuals engage in commerce (the broccoli argument), seemed both wrong and irrelevant. After all in the signal case of Wickard v. Filburn the Court had held that “the stimulation of commerce is a use of the regulatory function quite as definitely as prohibitions or restrictions thereon.” And even as far back as Gibbons v. Ogden Chief Justice Marshall had written that the Clause empowered Congress to lay down the rule by which commerce (not persons in commerce) could be regulated. And the ACA clearly did that. At any rate it was easily demonstrated that an overwhelming portion of the adult population was already as much engaged in the commerce of the health care system as was the wheat farmer in Wickard.

And finally, the energy behind the broccoli argument was about personal liberty, and no litigant ventured that the mandate violated the liberty clause of the Fifth Amendment (which, having the identical scope as the liberty clause of the Fourteenth, would have made a scheme like that of Massachusetts unconstitutional). Only Justice Thomas, who has consistently proclaimed a pre-New Deal conception of the Commerce Clause, was likely to vote to strike the Act down on that ground. Especially after the opinions of Judges Sutton and Silberman – two conservative stalwarts – the guessing in the halls of legal academe favored at least a six-to-three if not an eight-to-one vote upholding the Act, with the Chief Justice writing. Justice Kennedy, it was thought, would once again break ranks and the Chief, unwilling to yield the opinion leftward, would join and write.

The end of March brought the three days of oral argument, and the betting odds changed dramatically. The questions by the four Republican-appointed Justices who speak at argument were uniformly hostile. Justice Kennedy, generally the Court’s ideological center, early on asked: “I understand we must presume laws are constitutional, but even so, when you are changing the relation of the individual to the government in this, what we can stipulate is, I think, a unique way, do you not have a heavy burden of justification to show the authorization under the Constitution” – a question straight out of the Tea Party playbook. That “fundamental” change was worked seventy-five years before by the Social Security Act and fifty years before with Medicare. And Justice Scalia seemed more to bait and mock the Solicitor General than to probe the strengths of his arguments.

Such hope as there was for the mandate reposed now – precariously – on the Chief, and in some small “c” conservative instinct he might retain in the name of restraint. As for the argument that the mandate might be justified as a tax: if it had been frankly called a tax, like the Social Security or Medicare taxes, then only Justice Thomas would vote to strike it down.

That defense was argued by the Solicitor General almost as an afterthought. After all, Congress had for political reasons studiously avoided calling it a tax. And if it was a tax, maybe the whole dispute was premature because the Tax Injunction Act required waiting until the tax was due before it could be challenged. (And then there was the obscure, almost antiquarian question whether it was a “direct tax” constitutionally permissible only if apportioned among the states – a provision which the income tax avoided only by virtue of the Sixteenth Amendment specifically authorizing it.)

Then on June 28 came the decision. As a touch of comic irony, CNN and Fox News – violating the rule of “read on” and eager to be the very first with a news story available to all – reported that the Court had struck down the Act by a vote of five to four, apparently reading only so far as section two of the Court’s syllabus summarizing the decision. Five Justices voted that neither the Commerce Clause nor the Necessary and Proper Clause of Article 1, Section 8 authorized the mandate; five Justices – four from the dissent on the previous issue, plus the Chief Justice, again writing – voted that the mandate nevertheless was authorized by the Article 1, Section 8 taxing power and that the Tax Injunction Act did not deprive the Court of jurisdiction to rule on the point until after the tax was due (in 2014). And finally seven Justices voted that the threatened denial of all federal Medicaid funds to states that would not join in the significant expansion of Medicaid eligibility was so coercive as to constitute a form of duress that violated the states’ sovereignty and dignity, as announced in cases such New York v. United States and Printz.

The Chief Justice and the four Justices of the joint opinion diverged on the Medicaid expansion only to the extent that for the Chief Justice the supposed unconstitutional coercion of the cut-off invalidated only the cut-off, so that a state that refused to expand Medicaid eligibility could nonetheless continue to participate and receive federal funds at the pre-expansion level. As the joint opinion notes, this was a solution the government proposed “in two brief sentences at the very end of its brief.” (Slip op. at 47.) The joint opinion bridles at this suggestion, and in the only passage in which the joint opinions deigns to refer to the Chief Justice by name (at 46) rejects the government’s and the Chief’s conclusion and votes to invalidate the Medicaid expansion altogether, for participating as well as non-participating states.

Justices Breyer and Kagan did not dissent from the Chief’s opinion on this point, so that by drawing the appropriate Venn diagram of the opinions the Chief’s resolution on this point controls – although Justices Breyer and Kagan might arguably have produced the same result by joining Justice Ginsburg’s dissenting opinion rejecting the coercion argument altogether. Here were the surprises listed at the outset of this essay.

On the Commerce power, the Chief’s opinion, while calm and methodical, is a veritable cascade of non-sequiturs and solecisms. Many of these have been noted by others and by Justice Ginsburg’s equally calm and methodical dissent. It is nonetheless worthwhile, in confronting the surprises to which this essay is addressed, to touch on a few of the Chief’s bad arguments. The foremost among these echoes the heart of the Tea Party-inspired detestation of this Act: the claim that if Congress can regulate individuals by requiring them to engage in commerce then “that is not the country the Framers of our Constitution envisioned. James Madison explained that the Commerce Clause was ‘an addition which few oppose and from which no apprehensions are entertained.’” (Slip op. at 23.) Professor Randy Barnett, the champion of the broccoli argument made this semi-hysteric and invalid point in his testimony to the Senate Judiciary Committee, declaiming that if the defense of the mandate is valid then we are all no longer “citizens but subjects.”

As I have indicated, Justice Kennedy early in the argument of the case, picking up the Tea Party banner with this silly slogan emblazoned on it, asked whether, technicalities aside – I suppose technicalities such as the taxing power – would not upholding the mandate work a fundamental change in the relationship of the citizen to the government. The Solicitor General, with superhuman restraint, let this pass, when he might have answered “Perhaps, Justice Kennedy, but that is a change worked seventy-five years ago by Social Security and fifty years ago by Medicare.” And of course the slogan ignores the fact that the mandate was first proposed by the conservative Heritage Foundation (whose amicus brief in this case trying to explain this fact away is almost comical in its tergiversations) and might well have been a feature in an analogous scheme proposed by President George W. Bush partially to privatize Social Security.

The Chief commits a similar constitutional solecism by harping on the supposed unprecedented nature of the mandate as somehow a symptom of its invalidity. (He does acknowledge that, of course, “there is a first time for everything.”) But the novelty of a Commerce Clause scheme proves nothing – though it does deprive a scheme of the safe harbor of an existing all-fours precedent. As economic circumstances change and economic projects and opportunities are devised to respond to them, regulation will necessarily take on new forms: new kinds of regulatory commissions and other mechanisms, including public/private and state collaborations, changing configurations of prohibition, requirement and incentive, the use of private causes of action, court enforcement by government agents or adjudication in the first instance by regulatory bodies.

Novelty is a matter of degree and description. Whether the requirement that certain categories of persons purchase health insurance coverage from commercial vendors is indeed a novel requirement has been the subject of considerable ingenious debate, with various analogies offered and refuted along the way. But the very scholasticism of this debate shows how irrelevant the sobriquet “novel” is to the question of validity. Doubtless any number of regulatory schemes when first proposed had some element of novelty about them, but the proper question has always been thought to be whether the scheme is indeed a regulation of interstate commerce.

It is the failing grade on this, the only relevant question, that doomed – and to my mind properly – the Gun Free School Zones Act and the private cause of action under the Violence Against Women Act. That the provision of health care and the modalities of payment for it are interstate commerce has never been doubted. That should have been the end of the story, which is why until only a couple of years ago the mandate (like the similar proposal of the Heritage Foundation) were debated only in policy, not constitutional terms.

Particularly eye-catching is the Chief’s fallacious refutation of the Necessary and Proper argument, which admits Chief Justice Marshall’s expansive reading of the word “necessary” in McCulloch v. Maryland, but resting instead on the word “proper.” Once again the Chief Justice acknowledged that the mandate was “necessary” in Chief Justice Marshall’s sense of convenient or adapted to the fulfillment of the regulatory purpose of making health insurance widely available without consideration of medical history or lifetime caps. Indeed, the mandate is much closer to being necessary to the success of the regulatory scheme even in the more restrictive sense urged by counsel to Maryland in McCulloch v. Maryland than many subsequent invocations of the clause.

For this reason the Chief Justice fell to arguing that though necessary to the regulatory scheme, the mandate was not proper. But certainly since the New Deal and thus for the last seventy-five years of jurisprudence under the clause – consider its invocation in relation to legislation under Section five of the Fourteenth Amendment – the conception has been that propriety is a matter of not contravening some distinct constitutional prohibition or principle. Put another way, questions of the power of Congress under Article I, Section 8 or any of the other power-conferring provisions of the Constitution (e.g., Article II, Section 2; Article III, Section 1) have been conceptualized as questions about how far the granted power extends; if the assertion of power is neither comprehended in the explicit grant nor necessary to such a power, then the grant runs out.

If the purported reach is too long a stretch, as in the VAWA, then the grant is not necessary and the power has run out. The propriety of the reach has been thought to be a question of whether the claim bumps up against an explicit or implicit constitutional barrier. So “necessary” extends the reach of a grant, but not beyond plausibility when as in VAWA it runs out. And “proper” is an affirmative block to any means of exercising a granted power contrary to constitutional command. Thus Congress’s compelled contribution to its mushroom marketing scheme in United States v. United Foods may have been marginally necessary to that silly scheme, but it was improper because it bumped up against the explicit command of the First Amendment.

That is not how the Chief Justice interpreted the propriety term. He first posited a constitutional limit to the otherwise necessary reach of the granted power and then argued that the mandate was improper (though undoubtedly necessary) because it exceeded that limit. The purported limit was not any distinct constitutional prohibition, but the very notion that the extension of congressional power just went too far. But this is double counting. Why did the mandate go too far? Because it reached into a domain in which otherwise only the police power of the states would be available. But that is true of most commercial regulations under the Necessary and Proper Clause – for instance, various congressionally mandated regulations of the terms of private employment, as in wages and hours regulation. After all, that is just the standard effect of the Supremacy Clause and preemption.

So to avoid this double counting or question-begging use of the impropriety notion, there has to be something particularly wrong about this necessary extension, and not just that it was an extension. And here the Chief Justice reached for the notion that the extension was improper not because it extended into a domain where otherwise only the states might regulate, but because there was something particularly inappropriate about this reach, and that was found in the notion that the regulation of medical systems and practice was a particular and traditional realm of state responsibility. (This kind of argument was invoked, particularly by Justice Kennedy, in both Lopez and Morrison.)

Now this is a very imprecise and untethered notion, but as applied in this case it is just wrong. One need only consider the massive intrusion of federal regulation represented by Medicare, Medicaid (both to be sure under the Spending Clause, but so what?), and the FDA to recognize how pervasive is the federal presence in the regulation of medical services. So it is one of the dangerous loose ends left dangling in this case that we have here a newly open-ended limit in terms of propriety to otherwise necessary features of congressional regulation. Nor can New York v. United States or Printz be invoked to impose such a constitutional limit (said to be implicit in the Tenth Amendment), as those cases rested on the improper “commandeering” of state authority, but the only commandeering going on here is of individual citizens (or, if you will, “subjects”) to whom this argument had never before been applied.

Of course, the real shadow of impropriety on everyone’s mind but studiously omitted from the argument and justifications is the supposed intrusion on individual liberty implicated in Congress’s scheme: the offense to liberty in requiring someone to enter the market and buy something from a nongovernmental purveyor. It is bewildering why it would be a greater offense to the liberty of the “subject” to compel someone to enter an equivalent or even more coercive and comprehensive scheme in which the benefit is supplied by the government and paid for by taxes. (Is it some kind of objectionable compelled association?)

This is why Justice Kennedy’s question in oral argument, faintly echoed in the Chief’s opinion, should be dismissed as empty Tea Party rhetoric. The Tea Party, of course, views many government schemes as similarly offensive to liberty and perhaps hopes at some point to see those annulled too. But the argument was not made because it would have had to be made under the Liberty Clause of the Fifth Amendment, and this would have carried over to the similar clause in the Fourteenth and therefore rendered any such a scheme enacted by a state, as in Massachusetts, similarly invalid.

But after marching, puffing, and wheezing up this mountain of fallacy, it is astonishing to read the Chief’s account of why the Court must nonetheless in spite of the Tax Injunction Act (“no suit for the purpose of restraining the assessment or collection of any tax shall be maintained in any court by and person”) uphold the mandate under Congress’s Article I taxing powers, a conclusion with which the Chief’s four Republican colleagues vehemently and convincingly disagreed, and the four Democrats with little comment and evident relief were prepared to join.

The Chief correctly and canonically set out the Court’s traditional doctrine, by which if Congress designates a measure a tax it must for constitutional purposes be taken as one, even if on an ordinary view it would seem more like a penalty for violation of a regulatory requirement. The mandate and the associated charge for failure to comply with it might be taken as a tax on a course of conduct: not obtaining health insurance though eligible to do so. Note that here to make this out as a tax the incident has to fall on inactivity, but that, the five ruled, was a fatal defect only when Congress’s regulatory Commerce Clause power is invoked. That the tax acted as an incentive to engage in the desired conduct was deemed not to be fatal.

Taxes quite often are used as incentives. Consider the case of heavy taxes on cigarettes. And once again such incentives operate to discourage activity, not inactivity. And the amount of the “tax” was not so great that it did not leave its targets a realistic option to fail to pay for the option not to comply – once again, a line depending very much on a matter of degree, although the possibility that many would exercise the tax option was quite realistic. (Under a similar scheme in Massachusetts, significant numbers choose to forego insurance and pay the “tax.”) But the cases on which the Chief relied – on gambling, on machine guns, on certain narcotic substances – were all cases in which the Court indulged Congress’s designation as a tax of what to the ordinary eye may have seemed like a penalty for violating a prohibition. In the case of mandate exactly the opposite was the case: Congress had deliberately and on evident political grounds declined to designate the mandate and the consequences for noncompliance a tax, even though it had assigned certain administrative aspects of the monetary exaction to the Internal Revenue Service and lodged some of the provisions in the Internal Revenue Code.

A further difficulty with this maneuver was the Tax Injunction Act (TIA): the Court in order to avoid the politically intolerable result of postponing decision for a couple of years and until after the election had to find a way of ruling that the mandate was a tax for the purpose of its being a valid exercise of Congress’s constitutional taxing power but not a tax for the application of the TIA. This was an exercise in interpretive jiujitsu – more reminiscent, in the words of Justice Rehnquist (as he then was) in another context, not of the work of “jurists like Hale, Holmes and Hughes, but escape artists like Houdini” – a maneuver in which the four other Justices concurred with little comment and on which the four dissenters poured scorn.

The Chief makes the delicate but correct general point that what is a tax for constitutional purposes need not be one for the purposes of the TIA, the former being in the ultimate control of the Court’s duty to “say what the law is,” the latter a matter of Congress’s free intent. But this truism does not do the job. The Court is not compelled to accept for constitutional purposes whatever Congress calls a tax, while Congress certainly may specify the coverage of its own enactments like the TIA. The trouble is that in this case Court is calling a tax what Congress would not, and there is therefore no indication of its intention as to the application of the TIA to the mandate, which it declined to call a tax.

But the Commerce Clause activity/inactivity argument is so artificial and strained that at the end of the day it may not be very constraining, easily gotten around by skillful drafting. The Medicaid expansion invalidation, however, has potential for cutting a broad swath through many programs hitherto seen as unassailable under the rubric of cooperative federalism. The Medicaid expansion provision, which required states to cover under their Medicaid programs persons whose income did not exceed 133% of the poverty line, was an integral part of the ACA, because only in this way could the near-universal coverage be achieved for persons too poor to purchase insurance even on a subsidized market. Initially the federal government would pay the full cost of this expansion, but after 2016 its contribution would decrease to some ninety percent.

There are innumerable programs in which the federal government invites the states to participate, funding all or part of the cost of the program. This device is crucial for two reasons: in some of them there is no or a contestable constitutional power granted to Congress to concern itself with a subject apart from its grant to “spend . . . for the general welfare” – education being one of the earliest such programs; and second, even if a modern view of the extent of federal regulatory power under the Commerce Clause might now grant such regulatory power to Congress over a particular subject matter, after the decisions in New York v. United States and Printz, it would seem Congress could not enlist unwilling state participation in a regulatory scheme even though it could regulate the scheme directly itself. The pre-1937 decision of United States v. Butler affirmed a longstanding interpretation to the effect that Article I, Section 8 conferred an authority to tax and spend for the general welfare not tied to any other power enumerated in the Constitution, but went on to strike down the use of that power in Roosevelt’s Agricultural Adjustment Act because it invaded the states’ supposed exclusive authority over agriculture, manufacturing, and mining. The latter part of that decision was abandoned in Steward Machine v. Davis. In South Dakota v. Dole the Court declined to invoke the anti-commandeering principle to preclude conditioning the provision of federal highway funds to any state that did not enact a drinking age of twenty-one. Only Justice O’Connor would have required that the condition on the funding be directly related to the activity for which the funding was granted – on this view Congress might condition highway funds by provisions detailing how those highway funds must be expended (e.g., requiring certain safety features in the construction of the highways) but not attach more remote conditions, like the drinking age.  (Justice Brennan also dissented, agreeing with Justice O’Connor that, because “regulation of the minimum age of purchasers of liquor falls squarely within the ambit of those powers reserved to the States by the Twenty-first Amendment . . ., Congress cannot  condition a federal grant in a manner that abridges this right.”) Chief Justice Rehnquist went no further than to note that the consequence of noncompliance was a “mere” five-percent reduction in the funds available.

In South Carolina v. Baker, however, the Court, with only Justice O’Connor dissenting, approved a provision denying tax-exempt status to any state securities not issued in prescribed, traceable, non-bearer form. And in the Solomon Amendment, Congress imposed a complete cut off all federal funding to any university if it denied equal access to military recruiters. The Court in Rumsfeld v. FAIR did not reach the draconic aspect of this provision, ruling that Congress was entitled to command such access directly, even if no funding at all was involved.

In the Medicaid expansion portion of the decision, the Chief Justice’s opinion ruled that the provision cutting off all federal Medicaid funds to a state that did not participate in the expansion was unduly coercive and therefore amounted to a violation of the anti-commandeering principle of New York and Printz. The Chief Justice treated this as analogous to an unconscionable provision in a contract or to imposition of a contract by duress, because the complete cut-off would work a drastic deprivation on non-acquiescent states, many of whose citizens depended on the old Medicaid provision and the non-compliant could only continue these accustomed benefits at unbearable cost to its own treasury.

The decision is problematic on many scores. It is worth noting that although the modern Court had twice – in Lopez and Morrison – ruled that a federal program exceeded the power of Congress under the Commerce Clause, it has never struck down or limited a Spending or Taxing Clause program. The invocation of Butler as if it were still good law is particularly remarkable, because it suggests that pre-1937 decisions and principles, long believed discredited and abandoned, might once again be invoked to limit Congress’s powers.

In straight contract cases the argument is regularly rejected when franchisees object to contractually allowable franchise termination or alteration, pleading their large prior investments in the franchise. Or imagine a commercial tenant who has a lease in commercial premises in which he has invested heavily and where he has lots of good will (e.g., a neighborhood restaurant). The lease is terminable after five years on sixty days’ notice and the lessor insists on a greatly increased rent. I submit that the tenant’s complaint that this constituted duress would get nowhere. The analogy of states to hapless consumers as in some stretchy unconscionability cases (e.g., Williams v. Walker-Thomas Furniture Co.) is obviously inapposite.

There is little reassurance in the Chief Justice’s citation to the language in Dole that emphasizes that the fund cut-off in that case was a mere five percent, and of federal highway funds at that – not one hundred percent of a massive budget item like Medicaid.

The Chief Justice quotes Steward Machine, the first case overruling the restrictions in Butler: while “[we] do not attempt to fix the outermost line” where persuasion gives way to coercion . . . [It is ‘enough for present purposes that wherever the line may be, this statute is within it.’” That may be good enough to say that a provision, as in Steward, is valid. The use of that off-hand rhetoric to invalidate an Act of Congress is deeply disturbing. It leaves open the suggestion that federal funding programs in transportation, education, public safety, and many other domains exist on a kind of ratchet: once a state has accepted them and its citizens have grown accustomed to them, they may not be amended or discontinued. An analogous argument under the Takings Clause was rejected in Bowen v. Public Agencies Opposed to Social Security Entrapment. Here indeed is a provision that nods in the direction of the Tea Party program of the “Constitution in exile,” whose ambition it is to unravel the fabric of the post-1937 welfare administrative state. What was truly astonishing was that two such accomplished administrative law scholars as Justices Breyer and Kagan joined this opinion, making this most mischievous argument a seven-to-two decision of the Court.

II. The Mysteries

The main mystery is why five members of the Court should have signed on to the weak and untethered broccoli argument. Cynical commentators charge this up simply to partisan politics: the Republican Justices wanted to hand President Obama a signal defeat to a cherished but unpopular program four months before the election, and incidentally dispensing Mitt Romney from the trouble of proposing an alternative to a statute he had pledged to repeal. This was another Bush v. Gore, a naked intrusion into electoral politics. Never mind that Bush v. Gore was on the merits (not the remedy) a seven-to-two decision, reacting in part to the Supreme Court of Florida’s blatant disregard of a previous unanimous order of the Supreme Court in Bush v. Palm Beach County Canvassing Board.

It is a challenge, but I choose to believe that the Justices were sincerely trying to do their job, calling – in the Chief Justice’s words – balls and strikes. They just got it terribly wrong. Why? Because they are human beings, with emotions and susceptibilities. In this case I believe they were carried away on a tide of hostility to a measure that looked like it might be the prelude to a major expansion of federal power, so that there just had to be something wrong with it. And the activity/inactivity line of the broccoli argument, by dint of constant repetition, looked like that might be what it was.

I have been a judge, and I am aware that one way one goes about deciding a difficult, disturbing case is to start with a feel, a gut reaction that there is something terribly wrong here. And now the intellectual and professional task is to try to locate what it is. A vague discomfort in the abdomen turns out to be an enlarged liver pressing on the descending aorta, and so on. This is a perfectly correct, perhaps inevitable way to proceed, but it demands discipline and questioning self-doubt. The seemingly eligible diagnosis must be subject to relentless skeptical probing, and the possibility must be entertained that the pain in the abdomen is just gas, so that what started seeming to be something dire turns out to be nothing at all.

This is a very good Court, with able, conscientious judges – none more so than the Chief – and that is how every day they do their work. This time the discipline failed them – it happens. Perhaps their distaste for the scheme, perhaps Justice Kennedy’s libertarian instincts (here entirely out of place), perhaps the intense partisanship prevailing in the city in which they work, perhaps the approach of the end of the Term and the shortness of time, the close and heated media scrutiny and expectations for the case threw them off. It happens. It did not happen to Laurence Silberman or Jeffrey Sutton, whose instincts and predilections are very similar to those of the five. To the five it must have seemed like a time for boldness, a chance to strike a signal blow for liberty, for limited government. The atmosphere in the courtroom, the tenor of the questioning, the unusual fixed gaze of the media for three days. It may just have overwhelmed them.

Now let us move to the mystery of Chief Justice Roberts’s march up one side of the mountain and then down again on the other. I hold him to be a supremely intelligent and disciplined man, whose ambitions are for his Court and the honor of his Chief Justiceship. I think he was terribly torn: between the exhilaration of the moment, the intoxicating, raging exhilaration of his colleagues (I would exempt Justice Thomas from this: he only followed what he had been saying for more than a decade —and as usual he said it calmly, without rancor), and a nagging sense that something was not quite right. In the end, that led him to the precarious logic of the Tax Clause – all the more valid-seeming because precarious and implausible.

Principles of strategy, the balance of forces, from a great distance seem as if they might decide a battle, but generals thrust into the fog of war, the choking smoke of exploding shells, and gruesome death on all sides act almost at random, and the outcome is unpredictable. One thinks of the battle descriptions in War and Peace and in The Charterhouse of Parma: the participants in the battle rush about, perform feats of bravery or not, sometimes don’t know who they are shooting at, and only afterwards when the smoke clears do they (or the generals) know who won the battle, how, and why. That was Roberts’s situation in this case.

The cynics tell a different story, rather two different stories, stories of calculation and cowardice. In both stories the Chief is moved by a desire to save the reputation of the Court, and his own place in it, from an accusation of naked partisanship. To some his decision to uphold the statute on Tax Clause grounds is a case of cold feet. To others it was an act of supreme cunning: The Tea Party broccoli argument prevails, but the Court cannot be blamed for the practical outcome. And the potentially far more destructive Spending Clause ruling is left to lie about like Justice Jackson’s loaded weapon, available to wreak havoc on federal welfare programs. Finally, the Chief saves up (maybe acquires) some capital as an independent, principled jurist to spend on the upcoming affirmative action and DOMA cases.

I don’t believe it. I just see a serious man torn and groping to find his way. Consider the quite strained argument about why it was justified to rule on the broccoli argument at all, once the Chief had decided that the ACA could be sustained under the taxing power. If an important federal statute can be sustained on one ground, it seems gratuitous to go on to demonstrate why it would fail on another, novel, highly controverted and fraught ground, one which attracts a powerful dissent. The Chief’s explanation is that the justification under the taxing power was an interpretive stretch, given Congress’s assiduous avoidance of the tax label, but that nevertheless it was appropriate to stretch so as to honor the principles of judicial restraint and deference to a coordinate branch. I believe him.

There is a Frankfurterian, Jacksonian (Robert Jackson), even Rehnquistian legacy to which Roberts is an heir, that is extremely reluctant to strike down Congressional economic regulation. But this autobiographical note does not explain why, having granted the ACA this deference and saved it as an exercise of the taxing power, he went ahead and made up a Court of five Justices for the Commerce Clause argument which, once he had ruled the statute was valid as a tax, was no longer relevant to any work the Court properly had before it. To be sure, the dissenters, having rejected the tax argument, had every reason to go on to make the broccoli argument on the Commerce Clause, but Roberts had none.

It is as if, having crossed the tax bridge he was casting a wistful glance backwards to the dramatic argument he could have but did not feel in principle entitled to make. Again, the account of Roberts as judicial Machiavelli would have this as a shrewd way for him to have it both ways. I tend to see it not as a calculated maneuver but as the understandable impulse to lay bare one’s thinking, and if along the way this allows him to show that his heart really does beat in the right part of his chest, perhaps that was a temptation too hard to resist.

There is another, more dramatic indication that something unusual and untoward happened: the joint opinion of Justices Scalia, Kennedy, Thomas, and Alito, making the broccoli argument in terms – tone and rhetoric apart – hardly distinguishable from the Chief’s opinion on that count.

First, the jointness: the usual protocol going back to the days of the Marshall Court has a single designated author for an opinion – majority, concurring or dissent – which others join. (In earlier times and in England there was a practice of seriatim individual announcement, but not so now.) So it was in Marbury v. Madison, in Brown v. Board, in Lopez and Morrison. In signal cases the Chief will generally assign the opinion to himself. Historians, digging through the papers of former Justices, and Justices themselves delivering lectures on the functioning of the Court, explain how these opinions are circulated to the chambers in draft, with other Justices signaling their willingness to join, though sometimes they will condition that agreement on the author’s agreeing to changes. Though often robbing opinions of a certain individual flair, the practice has the great virtue of offering the world – and particularly the bar and lower court judges – a carefully negotiated, crafted, and therefore perhaps more stable and reliable statement of the law.

Then there is the joint opinion. It is a very rare phenomenon. The Justices designate each of themselves as joint authors. Famous examples are the joint opinion of all nine Justices in Cooper v. Aaron, proclaiming the supremacy as law of the Supreme Court’s desegregation cases; the joint opinion of Justices O’Connor, Kennedy, and Souter in Planned Parenthood v. Casey, affirming the commitment of these three Justices to what they termed “the central holding” of Roe v. Wade; and the joint opinion of Justices Stevens and O’Connor in McConnell v. FEC, upholding the most important features of the 2002 Bipartisan Campaign Finance Reform Act (McCain-Feingold).

Formally a joint opinion has no greater weight than an opinion by a single Justice joined by others, but obviously the Justices designating themselves as joint authors are seeking to make some special point. In Casey it was that a change in personnel since the 1974 decision and appointments by two Republican presidents who had proclaimed their disagreement with Roe had not undermined the validity of that decision. In McConnell, perhaps the joint opinion was intended to emphasize the support from two separate wings of the Court for campaign finance reform. And the point in Cooper v. Aaron, in the face of concerted attacks on the authority of the Court, was obvious.

But here? One anodyne explanation is that the Chief may have switched sides and his tax ruling came as surprise, leaving the four dissenters little time until the end of the Term and forcing them to parcel out the different parts of the very lengthy (sixty-five pages as against the fifty-nine pages of the Chief’s opinion announcing the opinion of the Court) opinion.

That won’t do. It is entirely possible – indeed the internal evidence of the Chief’s opinion points that way –that the Chief in this important case had assigned the opinion to himself, written his broccoli analysis as the opinion of the Court, and then some time later became persuaded that the tax argument is possible and that, if possible, it was his duty to embrace it. But then the four could have joined the Chief’s broccoli opinion with which they were in virtually complete agreement, as well as Medicaid extension opinion, leaving them only to explain in dissent why they rejected the tax argument and therefore why, with the mandate gone, they would strike down the whole ACA, from which they thought the mandate was not severable.

And why joint? As I have shown, this is done only to make a special point. It may be a failure of imagination on my part, but I can think of no special institutional point being made and so am left concluding that the four wished to signal their utter condemnation of the Chief’s retreat from what they thought of as the path of righteousness. The jointness signals, if nothing else, the personal, individual commitment to the opinion, but there is no other ground for such a personal gesture than a sense of personal betrayal. That this is a gesture of contempt and repudiation shows up also in the fact that the four do not advert to the Chief’s opinion, although the largest part of what they write is in complete agreement with the largest part of what he writes.

If I am right about that perhaps we can reconstruct what happened from this internal evidence. The Chief voted initially to strike down the mandate and assigned to himself the opinion doing so. He may or may not have agreed that the mandate was inseparable from the rest of the ACA, although he at least thought the Medicaid extension provision was unconstitutional (as applied to states unwilling to go along with it). And then he thought again. And the Frankfurterian in him could not allow him to ignore the tax argument in order to condemn the mandate. This left his four colleagues high and dry – and, because this was the signal piece of legislation it was – with a feeling not just of disappointment, not just with disagreement, but with a sense of betrayal in the heat of battle. In that atmosphere and with time before the end of Term growing short, there was no time nor perhaps inclination, on the part of the Chief to reconsider his conclusion under the Tax Clause nor on the other hand whether to retain the now gratuitous Commerce Clause portion of his opinion – not even so far as just to say that had he not felt compelled to uphold the mandate under the tax power he would have agreed with the four on the Commerce Clause.

In the end, the Chief comes out the winner: not because he made the best arguments or even consistently adhered to his own principles, but because it is evident that in this most fraught, pressured, and political of cases, he tried to do the right thing and was willing to pay the price in the esteem of those with whom he was in closest political, doctrinal, and temperamental agreement.

By way of coda, I must address the mystery of Justices Breyer and Kagan joining the Chief’s most mischievous ruling that the Medicaid expansion provision could not be imposed on unwilling states by threatening a total cut-off of federal Medicaid funds. The cynical suggestion that this was a kind of pay-off to the Chief for saving the mandate borders on the paranoid. And yet one can imagine that as June 28 approached these emotionally and mentally exhausted Justices just lacked the energy to pick up the cudgels once again to deal further blows to the already bloodied Chief. And in those last fraught days Justices Breyer and Kagan might have been concerned to save as much of the expansion as possible by joining the Chief’s opinion, which would have preserved existing Medicaid participation for states unwilling to accept the expanded coverage. The joint opinion would have invalidated the expansion altogether. As I have noted, the present less radical invalidation result might have been achieved by joining Justice Ginsburg’s dissent, but less emphatically.