1

Congress Is Not a Court

It was a bad meeting.
The Senate Judiciary Committee’s chief counsel asked when the majority leader planned to “bring up” the crime bill. Silence. Nothing. No response.
As we left, I asked, “What was that?”
“Moose.”
“Moose?” I inquired, with some perplexity.
“Guns,” the chief counsel grunted.
However politically naive I was at the time (and I was), even I knew that gun control meant a crime bill filibuster. It took longer to realize that “Moose” referred to hunters and, more specifically, hunters in Maine. The light clicked: the majority leader, George Mitchell, was up for reelection that year and did not want to bring up the bill lest it endanger his election. Of course, Senator Mitchell’s staff evaded the issue. Why? Because the calculus was more complex: two years later, after he was reelected, George Mitchell would bring up the crime bill, and eventually he would vote for it.
To lawyers trained primarily in the art of the courtroom and the appellate case, congressional politics has the whiff of incomprehensibility that I experienced upon hearing the term “moose” issue from the lips of the otherwise articulate and accomplished Judiciary Committee’s chief counsel. But it is precisely “moose” that explains the institutional rationality of the players’ actions: guns made the bill subject to a filibuster in a year when the majority leader, who represented Maine, was up for election. Put more bluntly, the leader was driven to time the crime bill based on what I will call the “electoral connection” and the “supermajoritarian difficulty.” Note, however, that this did not determine Leader Mitchell’s ultimate vote or even his policy position. It did, however, have a decisive impact on the legislative process.1
When it comes to Congress, contemporary legal theories about statutes risk irony, if not contradiction. Consider textualism, personified by Justice Scalia, the theory that judges should stop at law’s plain meaning, blinding themselves to how Congress writes statutes. One leading textualist describes Congress as “arbitrary” and “strategic” and its processes “tortuous” and “opaque.”2 If Congress is all that, why should plain meaning ever arise? Textualists’ opponents, purposivists, personified by Justice Breyer, reverse the irony. They idealize Congress. As their professors, Hart and Sacks, famously put it, members of the legislature are “reasonable men pursuing reasonable purposes.”3 But if Congress reasons so well, every statute should be plainly reasonable, not ambiguous enough to send judges running to the statute’s history as purposivists insist. Political scientists’ theories, known as positive political theory,4 fare no better,5 imagining Congress as either a devil incapable of making rational decisions or as an angel yielding efficient bargains. If the irrationality thesis were true, why would there be any statutes at all? If the bargaining were so efficient, why is passing legislation so difficult, and why are the bargains so difficult to divine?
Taken to extremes, these theories of what lawyers call “statutory interpretation”—how to read laws—adopt ideas of Congress capable of contradicting the theories themselves. If we take the textualists’ view of Congress-as-chaos to its extreme, then why should we look to text? A truly chaotic Congress could not create plain meaning. If we take Hart and Sacks’s “reasonable men acting reasonably” view of Congress, then purposivist theory risks similar irrelevance; a truly reasonable Congress will always make statutes plain. Finally, if we take the political scientists’ view of Congress to its logical extreme, then interpretation would be the least of our problems; government would be impossible or unnecessary. Put bluntly, leading theoretical opinions on how to read statutes—by judges, lawyers, and political scientists—have no coherent or realistic idea of Congress.6 In this book, I offer a minimalist theory that is both realistic and of use to lawyers and judges in statutory interpretation cases.
This is not a pedestrian enterprise; it is a matter of fidelity to one of the basic principles of democratic government—the power of the people, meeting through their representatives, to determine their own fate. No one believes that judges should exercise legislative power. In a democracy, the people’s representatives are responsible for legislation. Legislation deserves deference because of popular sovereignty. As Jeremy Waldron has so eloquently described it, legislation is the process by which citizens come to govern themselves in a world of wide and persistent disagreement.7 Statutes are the product of the people’s will, and even if that rosy picture seems too rosy, its opposite—rule by judges—is universally reviled.
Theories of statutory interpretation assume quite a bit about Congress, and it is time to focus on those assumptions. As Professor Jerry Mashaw has emphasized, “our positive beliefs about what is, powerfully constrain and shape our normative beliefs about what is good or desirable.”8 I begin with two meta-principles, which any theory of statutory interpretation should satisfy. First, as a descriptive matter, lawyers’ theories of statutes must be based on what we uncontroversially know about Congress, including its rules and incentives.9 Second, as a prescriptive matter, any proper theory must be consistent with basic democratic principles, such as majority rule.
Critics might wonder what is so new about the first principle. Realism in statutory interpretation traces to the 1930s. Its original claim was skeptical: Congress was incapable of forming a collective “intent.”10 My own view is that this objection—what I call “cheap intent skepticism”—is wrong. Groups like Congress act, and they make decisions; when we look for “intent,” we are looking for the meaning of those decisions, and that meaning can only be understood by looking at evidence of legislative context—not imagined minds—as I explain in Chapter 5. At the same time, I think it practically impossible for people to use the term intent without evoking this silly objection; therefore, as a pragmatic matter, I think Hart and Sacks were wise to try to come up with a different term. They chose purpose, which unfortunately suffers from many of the same problems as intent. I have chosen decision, as we use this term to describe judicial rulings—even when the court is deeply divided, the decision is merely plural, or soon overridden.11
Unlike sophisticated fine-grained empirical theories of politics debated among political scientists, my aim is not to predict legislative strategies, policy outcomes, median voter results, or party influence. The model is realist, relying on uncontroversial, widely accepted facts about Congress and its members’ institutional incentives. This minimalist approach is deliberately parsimonious. It must first translate Congress to courts as a descriptive matter and then provide a means for courts to translate that understanding to statutory interpretation. Legislative decision theory—the core of this book—argues that, in all cases, legislative context is crucial to understand legislative meaning. Interpreters cannot understand that context fully, however, unless they have a basic sense of Congress 101.
Just as we are capable of theorizing about courts in general without reducing them to individual judges, we can analyze Congress as an institution without reducing it to its individual members. An opening caution for political scientists and economists: I reject the “view from nowhere”12 that politicians’ or judges’ “preferences can be specified independently of their institutional location.”13 One may readily concede, with many economists, that individuals are rational actors, but rationality may be influenced enormously by institutional context. The claim that individuals’ preferences are separable from those of institutions is wrong; preferences are endogenously formed. As Professor Graham Allison long ago explained, decisions are typically based on where you sit.14 As we will see, this view yields a theory of Congress starkly different from that held by either major school of statutory interpretation—textualism or purposivism. Congress is not a green-eye shaded scrivener nor a junior varsity court and, if it were, democracy would be in worse trouble than it is today.
My minimalist approach yields a strong lesson for lawyers and judges who embrace the concept of “plain meaning.” Ambiguity, not plain meaning, is to be expected of Congress. Any realistic approach to Congress concedes that the business of legislating among 535 persons will yield ambiguity. One need not posit bad motives, strategic behavior, or stupidity (standard scholarly claims about legislative behavior). Imagine trying to write a Supreme Court brief with 535 coauthors. Textual ambiguity emerges from Congress’s institutional structure, something no court can change. If this is true, then the search for plain meaning, which has driven much statutory interpretation theory for the past 20 years, is a diversion. Instead, we should focus on how to find the “proper” text (looking at evidence about how the text evolved), and in hard cases find the best evidence of legislative context to resolve ambiguity. Statutory interpretation must focus less on finding “plain meaning” and more on resolving statutory ambiguity.
It is sad, but true, that the search for “plain meaning” may be nothing other than an exercise of judicial fiat. As I explain in Chapter 2, it is fairly easy to find “two plain meanings” in every case; just contrast the best undisputed example (a prototypical meaning) with all logical examples of a text (an extensivist meaning). This does not imply that plain texts should not be honored; they must, and everyone agrees that they must. It does mean that we need to think harder about “plainness” and “ambiguity.” It also means that we need to worry that our decisions about what is “plain” are cognitively biased, blinding decision makers to conflicting texts, and that the “choice” of particular texts as “plain” may be question-begging, as I explain in Chapter 4.
The most important question for courts is how, not whether, to resolve ambiguity. Courts have two choices: they may use their own self-constructed tools of meaning (canons, precedents, asserted plain meanings), or they can look to the evidence of how Congress acted (legislative evidence). This book argues that courts must look to legislative evidence, even where “plain meaning” seems apparent, if for no other reason than to counter judges’ and lawyers’ likely cognitive biases. Put in theoretical terms, both textualism and purposivism have work to do: as I will show, textualists who ignore legislative evidence can pick and choose the wrong texts, and purposivists who ignore how Congress works can pick and choose the wrong legislative evidence. Although it is likely to seem oxymoronic for some, textualists and purposivists alike must learn how to understand the legislative record (what I call “legislative evidence”), but in a new, simpler, and more disciplined way, as described in Chapter 3.

The “Electoral Connection”

Cognitive science tells us that we often do not think clearly, and this is nowhere more true than when lawyers begin to think about Congress, otherwise known as the “sausage factory.” The “self-serving” bias tells us that we tend to think that success is our own, and failures are the fault of others. If a CEO is successful, this reflects his brilliant contributions; if he fails, it was the market’s fault. This should warn judges about attributing all problems in legislation to Congress. Similarly, the “overconfidence effect” should warn judges not to be so sure about what they think they know about Congress. People systematically overrate their knowledge; consider the 68 percent of faculty members who rated themselves to be in the top 25 percent of teachers.15 Institutions aggravate these biases in conditions of information scarcity. Because Congress is not taught as a subject in many law schools, individuals hear the same claims over and over again from their fellow judges and lawyers and professors: Supreme Court opinions “stunning,” academic articles “intriguing,” and legislation “sausage.” Given these biases, I begin from the other end of the spectrum, from very basic, uncontroversial, facts about Congress that are ofte...