by John Greenya
FIRST, CONSIDER THE CONSTITUTION. Judges, after all, must do more than merely consider it. They take an oath to uphold it. So any theory of judging (in this country at least) must be measured against that foundational duty. Yet it seems to me those who would have judges behave like legislators, imposing their moral convictions and utility calculi on others, face an uphill battle when it comes to reconciling their judicial philosophy with our founding document.
Consider what happened at the constitutional convention. There the framers expressly debated a proposal that would have incorporated the judiciary into a “council of revision” with sweeping powers to review and veto congressional legislation. A proposal that would have afforded judges the very sorts of legislative powers that some of Justice Scalia’s critics would have them assume now. But that proposal went down to defeat at the hands of those who took the traditional view that judges should expound upon the law only as it comes before them, free from the bias of having participated in its creation and from the burden of having to decide “the policy of public measures.” In place of a system that mixed legislative and judicial powers, the framers quite deliberately chose one that carefully separated them.
The Constitution itself reflects this choice in its very design, devoting distinct articles to the “legislative Power” and the “judicial Power,” creating separate institutions for each, and treating those powers in contradistinction. Neither were these separate categories empty ones to the founding generation. Informed by a hard-earned intellectual inheritance—one perhaps equal parts English common law experience and Enlightenment philosophy—the founders understood the legislative power as the power to prescribe new rules of general applicability for the future. A power properly guided by the will of the people acting through their representatives, a task avowedly political in nature, and one unbound by the past except to the extent that any piece of legislation must of course conform to the higher law of the Constitution itself.
Meanwhile, the founders understood the judicial power as a very different kind of power. Not a forward-looking but a backward-looking authority. Not a way for making new rules of general applicability but a means for resolving disputes about what existing law is and how it applies to discrete cases and controversies. A necessary incident to civil society to be sure but a distinct one. One that calls for neutral arbiters, not elected representatives. One that employs not utility calculi but analogies to past precedents to resolve current disputes. And a power constrained by its dependence on the adversarial system to identify the issues and arguments for decision—a feature of the judicial power that generally means the scope of any rule of decision will be informed and bounded by the parties’ presentations rather than only by the outer limits of the judicial imagination. As the founders understood it, the task of the judge is to interpret and apply the law as a reasonable and reasonably well-informed citizen might have done when engaged in the activity underlying the case or controversy—not to amend or revise the law in some novel way. As Blackstone explained, the job of the judge in a government of separated powers is not to “make” or “new-model” the law. Or as Hamilton later echoed, it is for the judiciary to exercise “neither FORCE nor WILL, but merely judgment.” Or again, as Marshall put it, it is for the judiciary to say (only) “what the law is.”
So many specific features of the Constitution confirm what its larger structure suggests. For example, if the founders really thought legislators free to judge and judges free to legislate, why would they have gone to such trouble to limit the sweep of legislative authority—to insist that it pass through the arduous process of bicameralism and presentment—only to entrust judges to perform the same essential function without similar safeguards? And why would they have insisted on legislators responsive to the people but then allowed judges to act as legislators without similar accountability? Why, too, would they have devised a system that permits equally unrepresentative litigants to define the scope of debate over new legislation based on their narrow self-interest? And if judges were free to legislate new rules of general applicability for the future, why would the founders have considered precedent as among the primary tools of the judicial trade rather than more forward-looking instruments like empirical data? And why would they have entrusted such decisions to a single judge, or even a few judges, aided only by the latest crop of evanescent law clerks, rather than to a larger body with more collective expertise?
In response to observations like these, Judge Posner has replied that “American appellate courts are councils of wise elders and it is not completely insane to entrust them with responsibility for deciding cases in a way that will produce the best results” for society. But, respectfully, even that’s not exactly a ringing endorsement of judges as social utility optimizers, is it? I can think of a lot of things that aren’t completely insane but still distinctly ill-advised (or so I try to convince my teenage daughters). And, respectfully too, wouldn’t we have to be at least a little crazy to recognize the Constitution’s separation of judicial and legislative powers, and the duty of judges to uphold it, but then applaud when judges ignore all that to pursue what they have divined to be the best policy outcomes? And crazy not to worry that if judges consider themselves free to disregard the Constitution’s separation of powers they might soon find other bothersome parts of the Constitution equally unworthy of their fidelity?
THIS FIRST POINT LEADS to a second. It seems to me that the separation of legislative and judicial powers isn’t just a formality dictated by the Constitution. Neither is it just about ensuring that two institutions with basically identical functions are balanced one against the other. To the founders, the legislative and judicial powers were distinct by nature and their separation was among the most important liberty-protecting devices of the constitutional design, an independent right of the people essential to the preservation of all other rights later enumerated in the Constitution and its amendments. Though much could be said on this subject, tonight permit me to suggest a few reasons why recognizing, defending, and yes policing, the legislative-judicial divide is critical to preserving other constitutional values like due process, equal protection, and the guarantee of a republican form of government.
Consider if we allowed the legislator to judge. If legislatures were free to act as courts and impose their decisions retroactively, they would be free to punish individuals for completed conduct they’re unable to alter. And to do so without affording affected individuals any of the procedural protections that normally attend the judicial process. Raising along the way serious due process questions: after all, how would a citizen ever have fair notice of the law or be able to order his or her affairs around it if the lawmaker could go back in time and outlaw retroactively what was reasonably thought lawful at the time? With due process concerns like these would come equal protection problems, too. If legislators could routinely act retroactively, what would happen to disfavored groups and individuals? With their past actions known and unalterable, they would seem easy targets for discrimination. No doubt worries like these are exactly why the founders were so emphatic that legislation should generally bear only prospective effect—proscribing bills of attainder and ex post facto laws criminalizing completed conduct—and why baked into the “legislative Power” there’s a presumption as old as the common law that all legislation, whether criminal or civil, touches only future, not past, conduct. (“ ‘Every tin horn dictator in the world today, every president for life, has a Bill of Rights,’ said Scalia . . . . ‘That’s not what makes us free; if it did, you would rather live in Zimbabwe. But you wouldn’t want to live in most countries in the world that have a Bill of Rights. What has made us free is our Constitution. Think of the word “constitution;” it means structure.’ . . . ‘The genius of the American constitutional system is the dispersal of power,’ he said. ‘Once power is centralized in one person, or one part [of government], a Bill of Rights is just words on paper.’ ”)
Now consider the converse
situation, if we allowed the judge to act like a legislator. Unconstrained by the bicameralism and presentment hurdles of Article I, the judge would need only his own vote, or those of just a few colleagues, to revise the law willy-nilly in accordance with his preferences and the task of legislating would become a relatively simple thing. Notice, too, how hard it would be to revise this so easily made judicial legislation to account for changes in the world or to fix mistakes. Unable to throw judges out of office in regular elections, you’d have to wait for them to die before you’d have any chance of change. And even then you’d find change difficult, for courts cannot so easily undo their errors given the weight they afford precedent. Notice finally how little voice the people would be left in a government where life-appointed judges are free to legislate alongside elected representatives. The very idea of self-government would seem to wither to the point of pointlessness. Indeed, it seems that for reasons just like these Hamilton explained that “liberty can have nothing to fear from the judiciary alone,” but that it “ha[s] every thing to fear from [the] union” of the judicial and legislative powers. Blackstone painted an even grimmer picture of a world in which judges were free to legislate, suggesting that there “men would be[come] slaves to their magistrates.”
In case you think the founders’ faith in the liberty-protecting qualities of the separation of powers is too ancient to be taken seriously, let me share with you the story of Alfonzo De Niz Robles. Mr. De Niz Robles is a Mexican citizen, married to a U.S. citizen, and the father of four U.S. citizens. In 1999, he agreed to depart the country after being apprehended by immigration authorities. For two years his wife tried without luck to secure him a spousal visa. At that point, Mr. De Niz Robles decided to return to the United States and try his own luck at applying for lawful residency. In doing so, though, he faced two competing statutory provisions that confused his path. One appeared to require him to stay outside the country for at least a decade before applying for admission because of his previous unlawful entry. Another seemed to suggest the Attorney General could overlook this past transgression and adjust his residency status immediately. In 2005, my colleagues took up the question how to reconcile these two apparently competing directions. In the end, the Tenth Circuit held that the latter provision controlled and the Attorney General’s adjustment authority remained intact. And it was precisely in reliance on this favorable judicial interpretation that Mr. De Niz Robles filed his application for relief.
But then a curious thing happened. The Board of Immigration Appeals (BIA) issued a ruling that purported to disagree with and maybe even overrule our 2005 decision, one holding that immigrants like Mr. De Niz Robles cannot apply for an immediate adjustment of status and must instead always satisfy the ten-year waiting period. In support of its view on this score, the BIA argued that the statutory scheme was ambiguous, that under Chevron step 2 it enjoyed the right to exercise its own “delegated legislative judgment,” that as a matter of policy it preferred a different approach, and that it could enforce its new policy retroactively to individuals like Mr. De Niz Robles. So that, and quite literally, an executive agency acting in a faux-judicial proceeding and exercising delegated legislative authority purported to overrule an existing judicial declaration about the meaning of existing law and apply its new legislative rule retroactively to already completed conduct. Just describing what happened here might be enough to make James Madison’s head spin.
What did all this mixing of what should be separated powers mean for due process and equal protection values? After our decision in 2005, Mr. De Niz Robles thought the law gave him a choice: begin a ten-year waiting period outside the country or apply for relief immediately. In reliance on a judicial declaration of the law as it was, he unsurprisingly chose the latter option. Then when it turned to his case in 2014, the BIA ruled that that option was no option at all. Telling him, in essence, that he’d have to start the decade-long clock now—even though if he’d known back in 2005 that this was his only option, his wait would be almost over. So it is that, after a man relied on a judicial declaration of what the law was, an agency in an adjudicatory proceeding sought to make a legislative policy decision with retroactive effect, in full view of and able to single out winners and losers, penalizing an individual for conduct he couldn’t alter, and denying him any chance to conform his conduct to a legal rule knowable in advance.
What does this story suggest? That combining what are by design supposed to be separate and distinct legislative and judicial powers poses a grave threat to our values of personal liberty, fair notice, and equal protection. And that the problem isn’t just one of King George’s time but one that persists even today, during the reign of King James (LeBron, that is).
AT THIS POINT I can imagine the critic replying this way. Sure, judges should look to the traditional tools of text, structure, history, and precedent. But in hard cases those materials will prove indeterminate. So some tiebreaker is needed, and that’s where the judge’s political convictions, a consequentialist calculus, or something else must and should come into play.
Respectfully, though, I’d suggest to you the critics’ conclusion doesn’t follow from their premise. If anything, replies along these lines seem to me to wind up supplying a third and independent reason for embracing the traditional view of judging: it compares favorably to the offered alternatives.
Now, I do not mean to suggest that traditional legal tools will yield a single definitive right answer in every case. Of course Ronald Dworkin famously thought otherwise, contending that a Herculean judge could always land on the right answer. But at least in my experience most of us judges don’t much resemble Hercules—there’s a reason we wear loose-fitting robes—and I accept the possibility that some hard cases won’t lend themselves to a clear right answer.
At the same time, though, I’d suggest to you that the amount of indeterminacy in the law is often (wildly) exaggerated. Law students are fed a steady diet of hard cases in overlarge and overly costly casebooks stuffed with the most vexing and difficult appellate opinions ever issued. Hard cases are, as well, the daily bread of the professoriate and a source of riches for the more perfumed advocates in our profession. But I wonder: somewhere along the way did anyone ever share with you the fact that only 5.6% of federal lawsuits make it all the way to decision in an appellate court? Or that, even among the small sliver of cases that make it so far, over 95% are resolved unanimously by the courts of appeals? Or that, even when it comes to the very hardest cases that remain, the cases where circuit judges do disagree and the Supreme Court grants certiorari, all nine Justices are able to resolve them unanimously about 40% of the time? The fact is, over 360,000 cases are filed every year in our federal courts. Yet in the Supreme Court, a Justice voices dissent in only about 50 cases per year. My law clerks reliably inform me that’s about 0.014% of all cases. Focusing on the hard cases may be fun, but doesn’t it risk missing the forest for the trees?
And doesn’t it also risk missing the reason why such a remarkable percentage of cases are determined by existing legal rules? The truth is that the traditional tools of legal analysis do a remarkable job of eliminating or reducing indeterminacy. Yes, lawyers and judges may sometimes disagree about which canons of construction are most helpful in the art of ascertaining Congress’s meaning in a complicated statute. We may sometimes disagree over the order of priority we should assign to competing canons. And sometimes we may even disagree over the results they yield in particular cases. But when judges pull from the same toolbox and look to the same materials to answer the same narrow question—what might a reasonable person have thought the law was at the time—we confine the range of possible outcomes and provide a remarkably stable and predictable set of rules people are able to follow. And even when a hard case does arise, once it’s decided it takes on the force of precedent, becomes an easy case in the future, and contributes further to the determinacy of our law. Truly the system is a wonder and it is little wonder so many throughout t
he world seek to emulate it.
Besides, it seems to me that even accepting some hard cases remain—maybe something like that 0.014%—it just doesn’t follow that we must or should resort to our own political convictions, consequentialist calculi, or any other extra-legal rule of decision to resolve them. Just as Justices Sotomayor and Kagan did in Lockhart, we can make our decisions based on a comparative assessment of the various legal clues—choosing whether the rule of the last antecedent or one of its exceptions best fits the case in light of the particular language at hand. At the end of the day, we may not be able to claim confidence that there’s a certain and single right answer to every case, but there’s no reason why we cannot make our best judgment depending on (and only on) conventional legal materials, relying on a sort of closed record if you will, without peeking to outside evidence. No reason, too, why we cannot conclude for ourselves that one side has the better of it, even if by a nose, and even while admitting that a disagreeing colleague could see it the other way. As Justice Scalia once explained, “[e]very canon is simply one indication of meaning; and if there are more contrary indications (perhaps supported by other canons), it must yield. But that does not render the entire enterprise a fraud—not, at least, unless the judge wishes to make it so.”