Disagreeing Politely with Justice Souter on Jurisprudence

(How's that for a clickbait title??)

David Hackett Souter, former Associate Justice of the Supreme Court, died on Thursday. We learned this yesterday, and the encomia poured in. Everyone loved Justice Souter, well, everyone except the right-wingers convinced he'd betrayed them. He was, as everyone saw, a jurist of honor and integrity, but he was also just a mensch. Famously, he retired in 2009, a mere eighteen years after his appointment and just shy of his seventieth birthday, because he just didn't like being on the Supreme Court. He didn't like D.C.; he missed the quiet of New Hampshire, where he then continued to sit as a judge on the First Circuit for the rest of his life.

Many of the tributes to Justice Souter mentioned or alluded to a speech he gave at the 2010 Harvard commencement, shortly after his retirement from the Court. In that speech, he outlined a conception of constitutional jurisprudence – or, perhaps more precisely, he outlined why he rejects a conception of constitutional jurisprudence he called the "fair reading" model:

A claim is made in court that the government is entitled to exercise a power, or an individual is entitled to claim the benefit of a right, that is set out in the terms of some particular provision of the Constitution.  The claimant quotes the provision and provides evidence of facts that are said to prove the entitlement that is claimed.  Once they have been determined, the facts on their face either do or do not support the claim.  If they do, the court gives judgment for the claimant; if they don’t, judgment goes to the party contesting the claim.  On this view, deciding constitutional cases should be a straightforward exercise of reading fairly and viewing facts objectively.

This model, he suggested, gives rise to the charges that the Court is engaged in "judicial activism," in "legislating from the bench," in discovering rights not truly contained in the Constitution. To push back against those charges, then, Souter explained why constitutional judging is not nearly so simple as that model would have it, at least in the kinds of cases that get litigated up to the Supreme Court. In a nutshell:

Another reason is that the Constitution contains values that may well exist in tension with each other, not in harmony.  Yet another reason is that the facts that determine whether a constitutional provision applies may be very different from facts like a person’s age or the amount of the grocery bill; constitutional facts may require judges to understand the meaning that the facts may bear before the judges can figure out what to make of them.  And this can be tricky.

Souter then goes on to illustrate this view with stories of the Pentagon Papers Case, a.k.a. New York Times v. United States, and the School Segregation Cases, a.k.a. Brown v. Board of Education.

People love this sort of thing, this idea that judging entails complex choices between competing constitutional values. I... don't. In fact I think that the way of thinking Souter outlines in that speech, and to some extent the jurisprudential philosophy he displayed on the bench, is a serious weakness in liberal constitutional thought of the last half-century. To be clear, I think Souter was a very good Justice, and a good man, and (unlike some of the combatants in the Constitutional Theory Wars) I think his theory of jurisprudence is utterly sincere and in good faith. I just don't happen to agree with it. So, in the spirit of polite and respectful discussion, I wanted to go through his two case studies and explain why I think his analysis is unsatisfactory.

We can start with the Pentagon Papers case. The facts are fairly straightforward: the Pentagon Papers, a trove of classified, internal Pentagon documents detailing the thinking behind the Vietnam War, had been leaked to the Washington Post and the New York Times. For the youngsters in the audience, this was rather a big deal. The leak was known to have happened, and the newspapers were considering whether and how much of the material to publish, but President Richard Nixon was determined that the answer to that question should be "none." The administration asked for a court order forbidding the Post and the Times from publishing any of the Pentagon Papers, on the premise that publication would endanger national security.

The case, as Justice Souter says, went with shocking haste from the trial court to the Supreme Court, where the government lost by a vote of 6-3. The Court's decision is fascinating, not least because, since it was all decided so quickly, there was no time for the Justices to coalesce around one majority and one dissenting opinion as they normally would. Instead, every single Justice wrote their own opinion, along with the per curiam opinion of the Court itself. We can therefore see how all nine of the Justices at the time thought about the most fundamental questions of the First Amendment. The case stands for the proposition that, while perhaps not literally impossible, it is exceedingly hard for the government to show that so-called "prior restraint" is necessary and thus lawful.

Souter's interest in the case focuses on a rather infamous moment from oral argument. Erwin Griswold, then Solicitor General, was immediately confronted by Justice Hugo Black (no relation). Black is the greatest modern champion of the textual method of constitutional interpretation; relatedly, he was one of the great champions, along with William O. Douglas, of the "absolutist" view of the First Amendment. He was known to make great show of underlining the words "no law" in the text of the amendment, the implication being, obviously, that "no law" meant "no law," no matter how great a justification there seemed to be for it. This view was obviously awkward for Griswold and for Nixon, which led Griswold to say to Justice Black:

Now Mr. Justice, your construction of that is well-known, and I certainly respect it.  You say that no law means no law, and that should be obvious.  I can only say, Mr. Justice, that to me it is equally obvious that “no law” does not mean “no law,” and I would seek to persuade the Court that that is true.

As Chief Justice Marshall said, so long ago, it is a Constitution we are interpreting….

This, to Souter, is a nuanced and sophisticated approach to constitutional law. Although Griswold lost the case at bar, Souter notes, his view of things more broadly certainly prevailed over Justice Black's, and Souter sees this as a happy development. Griswold "had pointed out that the First Amendment was not the whole Constitution.  The Constitution also granted authority to the government to provide for the security of the nation, and authority to the president to manage foreign policy and command the military." This shows, for Souter, that it is not only where the language of the Constitution is broad and open-ended that the "fair reading" model falls short. It also misleads us even where the text in question really does appear to be quite unambiguous, as in "no law":

It fails because the Constitution has to be read as a whole, and when it is, other values crop up in potential conflict with an unfettered right to publish, the value of security for the nation and the value of the president’s authority in matters foreign and military.  The explicit terms of the Constitution, in other words, can create a conflict of approved values, and the explicit terms of the Constitution do not resolve that conflict when it arises.  The guarantee of the right to publish is unconditional in its terms, and in its terms the power of the government to govern is plenary.  A choice may have to be made, not because language is vague but because the Constitution embodies the desire of the American people, like most people, to have things both ways.  We want order and security, and we want liberty.  And we want not only liberty but equality as well.  These paired desires of ours can clash, and when they do a court is forced to choose between them, between one constitutional good and another one.  The court has to decide which of our approved desires has the better claim, right here, right now, and a court has to do more than read fairly when it makes this kind of choice.  And choices like the ones that the justices envisioned in the Papers case make up much of what we call law.

Emphasis added; that sentence strikes me as the crux of the matter.

Now I have always been inclined to regard Griswold's comment to Justice Black as an embarrassment. To my mind it almost gratuitously places the case against an absolute reading of the Free Speech Clause on a weak footing. If you wish to quibble about the meaning of the text, say that "law abridging the freedom of speech" is rather narrower than it might immediately seem; that is not immediately absurd. "'No law' does not mean 'no law,'" on the other hand, is immediately absurd.

Part of what is going on here, of course, is a clash between modalities. Black, as I said, championed the method of text. Griswold, on the other hand, was coming at things from a prudential perspective, one he unsurprisingly shares with Justice Souter. Indeed, he invoked perhaps the most famous prudential maxim: that the Constitution is not a suicide pact. Which is fair enough! But to my mind, Griswold's case would have been stronger if he were willing to simply leave it in prudential terms, rather than asserting also that the prudential point of view is also to be found in the meaning of the text itself.

This gets at something very deep about how we talk about the Constitution. We think of constitutional arguments as being about the meaning of the text; we would say that Griswold was making a prudential claim about how to interpret the First Amendment. I think this is misleading (as, incidentally, does Philip Bobbitt, the original theorist of these "modalities"). It is better, in my view, to say that Griswold was making a prudential argument about how the Pentagon Papers case should've been decided. Consider the following two ways of representing the dialogue between Griswold and Black:

Black: When the First Amendment says "no law," it means "no law."
Griswold: No it doesn't.

And:

Black: The government ought to lose this case, because the First Amendment says Congress shall make no law abridging the freedom of speech.
Griswold: The government ought to win this case, because the Constitution is not a suicide pact.

To my mind, the second representation is much more satisfying all around. It gets to the heart of what both of them are saying. It only requires us to acknowledge that the text is not the whole of the Constitution, that there exist a plurality of modes of constitutional reasoning, that each of these modes are legitimate whether or not they are drawn from the text, etc. Of course these points are anathema to Justice Black, a textualist ideologue, but note how Griswold struggles with them as well! This idea that what we do when we decide constitutional cases is "read a text" is so pervasive that even prudential jurists implicitly adopt it.

Now on some level I am not unsympathetic to Souter's point. That constitutional judging is not a mechanical matter, that the judge is often faced with considerations that pull in opposite directions, that judgment must be exercised to resolve such cases: all this I concur in. But I still find Souter's vision unsatisfying in its specifics. He sees these conflicts, where they arise, as conflicts among "approved values." But the Constitution is not just a collection of approved values! Rather, it is a scheme according to which power is granted to a variety of different institutions.

One of the most important features of this scheme is the relation between means and ends. This was the first great constitutional controversy in the nation's history, the fight between Hamilton and Jefferson, and then between Marshall and Jackson, over so-called "implied powers." The ultimate resolution of that fight, in McCulloch v. Maryland (1819), was that the federal government does possess implied powers; in other words, that it has a broad choice of means by which to effectuate its enumerated ends. As Marshall famously put it:

Let the end be legitimate, let it be within the scope of the Constitution, and all means which are appropriate, which are plainly adapted to that end, which are not prohibited, but consist with the letter and spirit of the Constitution, are Constitutional.

Emphasis in this sentence is typically put on "plainly adapted," or even just on "let the end be legitimate." But for our purposes the crucial point is that this is not the whole of the test. There is an entire second half: means which are prohibited, that are inconsistent with the letter and spirit of the Constitution, are not constitutional! Even when used toward a legitimate end. Seen in this light, there is no "conflict" between the Constitution's grants of power to the government to defend the nation and the First Amendment. They are easily synthesized: the government may not use laws restricting the freedom of speech, even as a tool to advance its legitimate ends, such as national security.

In other contexts this way of thinking is familiar. Prudential types like to say that, as no one value can be absolutely paramount over all others, there can be no true constitutional absolutes. But somehow you never hear this about, say, the right to trial by jury. No one, so far as I am aware, argues that the Sixth Amendment's guarantee of jury trial must really reflect an implicit prudential balancing test, that the government could defeat the Sixth Amendment upon a sufficient showing of vital national interest. We understand that it is simply in the nature of our government that it cannot employ this particular tool (criminal trial before only a judge, against the wishes of the defendant) even to accomplish its very real ends.

Of course this example is an interesting one, because the Constitution actually does provide an express provision allowing the government to avoid jury trials in cases of sufficient urgency: the Suspension Clause, which allows for the writ of habeas corpus to be suspended – and therefore for the government to detain people indefinitely without access to the courts – "when in Cases of Rebellion or Invasion the public Safety may require it." But I think this only proves my point. The Suspension Clause affords a specific, delineated means by which a particular value (security in times of true crisis) may be realized. But other means by which that value might be realized are forbidden. The government can, under very narrow circumstances, suspend the general rule against extrajudicial punishment, but it can never impose criminal punishment through a non-jury trial.

Now I do not mean to suggest that this is the only valid way of viewing the matter. The claims of prudence are not lightly to be discarded! But Souter's approach, it seems to me, reduces everything to just prudence, to just the balancing of different constitutional "values." In the process something is lost; the Constitution is flattened.

Now let's talk about the Brown case. I imagine that the reader is likely familiar with the basic history, but for good measure here is how Souter recounts it:

Brown ended the era of separate-but-equal, whose paradigm was the decision in 1896 of the case called Plessy v. Ferguson, where the Supreme Court had held it was no violation of the equal protection guarantee to require black people to ride in a separate railroad car that was physically equal to the car for whites.  One argument offered in Plessy was that the separate black car was a badge of inferiority, to which the court majority responded that if black people viewed it that way, the implication was merely a product of their own minds.  Sixty years later, Brown held that a segregated school required for black children was inherently unequal.

A fair enough summary, in my view. The problem comes in the next paragraph, where he makes his basic analytical claim:

For those whose exclusive norm for constitutional judging is merely fair reading of language applied to facts objectively viewed, Brown must either be flat-out wrong or a very mystifying decision.  Those who look to that model are not likely to think that a federal court back in 1896 should have declared legally mandated racial segregation unconstitutional.  But if Plessy was not wrong, how is it that Brown came out so differently?  The language of the Constitution’s guarantee of equal protection of the laws did not change between 1896 and 1954, and it would be hard to say that the obvious facts on which Plessy was based had changed, either.  While Plessy was about railroad cars and Brown was about schools, that distinction was no great difference.

The solution, he suggests, is that time had elapsed between Plessy and Brown; that although the objective facts of segregation had not changed in that time, those facts had taken on a different social meaning:

As I’ve said elsewhere, the members of the Court in Plessy remembered the day when human slavery was the law in much of the land.  To that generation, the formal equality of an identical railroad car meant progress.  But the generation in power in 1954 looked at enforced separation without the revolting background of slavery to make it look unexceptional by contrast.  As a consequence, the judges of 1954 found a meaning in segregating the races by law that the majority of their predecessors in 1896 did not see.

Again on some level I do not disagree that social meaning is essential, to equal protection cases in particular. Why are sex-segregated restrooms constitutional, where racially segregated ones are not? Why, because the latter carries a social meaning bound up in the inferiority of the "colored" folk who are forbidden to use the whites-only room, while the former does not. Has not, anyway: if this were to change, as it might very well with the advance of queer theory and the general deconstruction of gender categories, then perhaps the constitutional conclusion would change as well.

The problem is just that, well... the specific claim about the social meaning of Jim Crow segregation in 1896 is obvious nonsense!! There was never any ambiguity, from the beginning, about what segregation meant. It came after the Thirteenth Amendment, and the Civil War, had abolished racial chattel slavery, and after the Fourteenth Amendment had directly forbidden the expressly unequal "Black Codes" adopted by the South immediately upon emancipation. It was a scheme by which the white majority in the South sought to preserve as much of their white supremacist social order as they could get away with!

Now, this view of the facts of segregation in 1896 raises a legal question: does the Fourteenth Amendment forbid this sort of thing altogether, or not? Plessy can be understood, if you like, to stand for the proposition that it does not; that it mandates only a kind of formal equality, and that as much real oppression as may be carried out consistent with that rule is permitted. Brown can be taken to stand for the opposite view, that when the Constitution says the equal protection of the laws it has in mind real equality, that any scheme whose purpose it to maintain "as much inequality as possible" must necessarily be illegal.

I think the latter view is rather clearly correct, among other things given the Fourteenth Amendment's place in the dialectic between emancipation and the Black Codes. Souter suggests that jurists in 1896 would have seen segregation as a great improvement over slavery. But that is not what the Fourteenth Amendment commands! No, it was the Thirteenth that abolished slavery; the Fourteenth very pointedly went beyond that, to say that the freedmen could not be kept in a position of inferiority full stop. But all of this is, as I say, a question of law, not of fact.

What changed between 1896 and 1954? It was not the social meaning of segregation. No, it was the political will to do something about it. In 1868, when the Fourteenth Amendment was adopted, the white South was largely excluded from political power. By the time Plessy was decided, Reconstruction had given way to Redemption, black voting rights had been curtailed, and even in the North the appetite of the white population to take a stand on civil rights had dwindled to nothing.

There was no real doubt that segregation offended the Fourteenth Amendment. I could suggest that you go read Justice John Marshall Harlan's famous dissent in Plessy, or my grandfather's short little article about the Brown case, to demonstrate this. But honestly reading the majority opinion in Plessy suffices to prove the point. It is terrible. The line about how, if segregation constitutes a badge of inferiority, it is only because black people choose to take it that way is asinine; as my grandfather put it, the "curves of callousness and stupidity intersect at their respective maxima." Plessy is just an act of nullification. We knew well enough what the Constitution meant, but the nation didn't actually want to follow the law, so they came up with a flimsy, insulting excuse not to.

And this, I think, illustrates another big danger with the kind of prudential approach Souter has in mind. Flattening everything to mere "values" is going to create the perpetual temptation to nullify. After all, is not popular self-rule one of the greatest of our constitutional values? So, when the populace really does not want to follow one of the Constitution's rules, is this not a "conflict of approved values"? Of course Souter would never have put it that way. But this idea that the task of constitutional judging is one of finding a reasonable compromise between, on the one hand, what the Constitution requires, and, on the other hand, popular sentiment to the effect of "but we don't wanna," is a hallmark of the prudential style, particularly as practiced by someone like Sandra Day O'Connor.

Souter is very plainly correct, I should reiterate, that many constitutional cases are not easy, cannot be decided in any mechanical or deterministic fashion. But Brown v. Board of Education is perhaps the single worst case in our entire history to choose to illustrate the point! Brown was one of the very easiest cases, as a matter of law, that the Supreme Court has ever been called upon to decide. Such a straightforward case made its way to that rarefied forum precisely because there had been so much political weight on the side of nullification for generations: actual practice, and Supreme Court doctrine to boot, had become completely divorced from what the Constitution had to say on the matter.

Part of the reason, I think, that Justice Souter's jurisprudence annoys me so much is that it is very very close to my own views; it is very very close to something I would find quite profound. The closing section of his speech, where he talks about the false hope of certainty, the need to embrace indeterminacy and the continual opportunities for the exercise of judgment: just so! That philosophy is, for instance, why I would reject out of hand proposals to have judging done by computer algorithms. For even if that could in some sense be done, it would be to surrender a major portion of our freedom as human beings.

But the Constitution is not just a "pantheon of values;" it contains not only aspirations but commitments. Those commitments themselves can be thought to reflect an aspiration of their own: the aspiration to govern ourselves according to law. And that aspiration, I suspect, cannot be made real without the principle that some things are forbidden to the People, no matter how good a reason they think they have, no matter how much they want them.