America's Unwritten Constitution: The Precedents and Principles We Live By (38 page)

But in these other areas, arguable overprotection of core rights did not threaten any counterbalancing citizen rights. Even had
Sullivan
doomed all libel law, the Constitution does not require that libel law exist; a state would be free to eliminate all libel law. Also, in the areas of segregation, expression, and incorporation, the Court’s opinions signaled that the justices understood the Constitution’s central meaning and were thus building implementational rules on a sound interpretational foundation.

The post-Warren Court’s deployment of separationist doctrine regarding church and state was different. At times, the Court seemed to misread the Constitution’s main meaning and to elaborate a vision of separation for its own sake rather than a vision of religious freedom and equality. Because of the justices’ misunderstanding of constitutional meaning and/or confusion about the proper relationship between interpretation and implementation, the post-Warren Court actually threatened Americans’ right to freely exercise their religion—a right expressly guaranteed by the Constitution. When Court rulings began to suggest that the Constitution would permit or even require that private religious schools be treated worse than otherwise identical private nonreligious schools, it became clear that the justices had veered off course. Implementation must subserve—not subvert—the core meaning of the written Constitution.

MODERN CRIMINAL-PROCEDURE CASES
—in particular, exclusionary-rule cases—have also veered off course, and here, the decisive wrong turns occurred on Earl Warren’s watch.

The problem with the exclusionary rule is not that it overprotects the core right to some degree. To repeat, some prophylactic overprotection in implementation of a constitutional right is necessary and proper. But the exclusionary rule is wildly out of sync with the relevant constitutional principles. On reflection, we should not be surprised by this fact, because the rule was not born as a traditional and proportionate judicial remedy—it
was always and remains today an outlandish judicial remedy bearing no proper relationship to the scope of the violation. The Fourth Amendment is about the violation of actual privacy and property that occurs during a search or seizure. Whether evidence of criminality is found in such a violation is wholly irrelevant. A proper remedy would address the rights of innocents. It would punish flagrant unconstitutionality more severely than mere error. It would protect against police brutality and governmental oppression even if such misconduct had no causal connection to a search for criminal evidence. The Warren Court exclusionary rule did none of these things.
¹²

Exclusion in America began not as a remedy rule, but rather as a rule about constitutional meaning—a rule deriving from a judicial interpretation that saw the Fourth Amendment and the Fifth Amendment self-incrimination clause as intimately interrelated. On this view, when a court excluded a defendant’s diary in a criminal case, the judge was not primarily remedying an antecedent Fourth Amendment violation that had occurred when the government had grabbed the diary. Rather, the judge was also—and more importantly—preventing the Fourth-Fifth Amendment violation about to occur in his own courtrooms were the diary to be read to the jury. Though it was a principled interpretation of the Constitution’s meaning, in the end this Fourth-Fifth-fusion view was demonstrably incorrect—indeed, preposterous—once the idea metastasized beyond diaries and personal papers to include stolen goods, murder weapons, and the like. Ever since the Court itself made that point about constitutional meaning clear in the 1966 blood-test case of
Schmerber v. California
, the exclusionary rule has been left without a principled legal leg—interpretational, implementational, or remedial—to stand on. Yet it still stands, in the name of
stare decisis
.

But why should a shaky rule that has lost its constitutional footing be perpetuated? We shall return to this key question in the concluding pages of this chapter.

CONSIDER, FINALLY, THE WARREN COURT’S
revolutionary one-person-one-vote rule. Here, too, we see arguable overprotection at work, at least initially. If the true constitutional rule governing voting rights derived
from the equal-protection clause, then the idea that each vote had to have exactly equal weight with every other would follow naturally. But this way of justifying
Reynolds
sits atop a faulty interpretive foundation. The equal-protection clause as originally written and understood was categorically inapplicable to voting.
Baker
and
Reynolds
were really republican-government-clause cases masquerading in equal-protection clothing.

Nevertheless, one-person, one-vote can be justified as a legitimate implementational device. True, the
Reynolds
rule arguably overprotected the constitutional principle at stake, but only after decades of judicial neglect and underprotection. Without some limit on malapportionment, a person’s right to cast a vote could be rendered utterly meaningless. For example, in a state composed of one hundred districts, could the government create fifty-one “rotten boroughs”—each with a single voter (say, the fifty-one most senior leaders of the incumbent party)—and relegate all other voters in the state to the remaining forty-nine districts? If this goes too far (and it surely does), and if Tennessee had surely gone too far in
Baker
, then where and how should judges draw the line in a principled way?
^*
Whatever its other flaws, the one-person-one-vote rule was a clean and workable implementational device.
¹³

Had the justices opted to openly rely upon the republican-government clause, several alternative sub-rules might have plausibly presented themselves. First, the Court could have chosen an approach akin to today’s Eighth Amendment jurisprudence, using the actual practice of the fifty states as a benchmark and proclaiming state practices that fell outside the mainstream to be unrepublican by contemporary standards. Today, applying this alternative would collapse into
Reynolds
, since all states now meet the one-person-one-vote standard. Even if states were now told that they are henceforth free to reject
Reynolds
, most would probably decline to do
so, because the voters themselves in most jurisdictions—along with leading politicians and opinion leaders—have come to embrace the idea of equally populous districts as a basic feature of political fairness.

Another imaginable alternative in 1964 would have been to allow voters in any given state, by a statewide initiative or referendum that itself would treat all voters equally, to authorize district maps that deviated from one-person, one-vote. If, at least once every census cycle, a state’s electorate had to bless any deviation from districting equality, there would likely be no systematic frustration of majority rule violative of the deep principle underlying the republican-government clause.

But what about minority rights? Suppose a 55 percent statewide majority of whites approved a malapportioned statewide map giving whites majorities in 90 percent of the unevenly sized districts. Such maps might violate the spirit of the Fifteenth Amendment, but prior to
Reynolds
that amendment had proved hard for judges to enforce on their own when confronting massive state disobedience. Also, unless the Court in 1964 had decreed that every district map had to be blessed by a statewide popular vote—an approach that would have obliged every state to institute a referendum or initiative process—there needed to be a doctrinal sub-rule specifying when such a popular vote would be required. Presumably the answer to this question would have been that a statewide popular vote would be needed only when a state was malapportioned. But when was that? When it departed from one-person, one-vote, of course! Our envisioned referendum rule was thus not a sharply distinct alternative to one-person, one-vote, but merely a softer variant that would have treated violations of one-person, one-vote as presumptively unconstitutional rather than unconstitutional per se.

If some sub-rules about the permissible size of voting districts were necessary in order to safeguard the basic right to vote, why weren’t sub-rules about the permissible shape of voting districts also necessary? In other words, once the justices decided to protect the basic right to vote in cases such as
Harper v. Virginia
and
Kramer v. Union Free School District
, and to buttress those right-to-vote rulings in the antimalapportionment cases of
Baker v. Carr
and
Reynolds v. Sims
, why did the Court stop there? Why didn’t the justices take the additional step of regulating political
“gerrymandering”—that is, the art of drawing district lines so as to favor the political group drawing the lines?

In
Reynolds
, Chief Justice Warren declared that “in a society ostensibly grounded on representative government, it would seem reasonable that a majority of the people of a State could elect a majority of that State’s legislators.…Since legislatures are responsible for enacting laws by which all citizens are to be governed, they should be bodies which are collectively responsive to the popular will.” Alas,
Reynolds’s
simple requirement that districts be of equal size fell short of guaranteeing that a majority of statewide voters would in fact control a majority within the legislature itself. Theoretically, a statewide minority faction supported by less than 26 percent of the voters could control the state legislature by winning a bare majority of ingeniously drawn districts and winning each district by a bare majority.

If
Reynolds
alone did not guarantee republican-government-style majority rule, neither did it ensure minority rights. Even in a state that was
Reynolds-compliant
, a minority group comprising 45 percent of the statewide vote could lose every single district, 55 percent to 45 percent, if each district were cleverly drawn so as to be a microcosm of the state as a whole.

Despite these problems, the Court was wise to stop where it did. Each of the four most salient subspecies of gerrymandering—racial gerrymandering, bipartisan gerrymandering, one-party gerrymandering, and incumbent-protective gerrymandering—implicated a unique cluster of constitutional considerations, and none of these clusters supported unilateral judicial intervention.

First, when governments have tried to fashion insidious district lines to disadvantage
racial
minorities, the modern Court has not faced a pressing need to develop its own implementational sub-rules based directly on the Fifteenth Amendment. Congress has already done much of the heavy lifting, via the 1965 Voting Rights Act and a series of subsequent statutory amendments. Enacted pursuant to Congress’s explicit enforcement authority under the Reconstruction Amendments, this landmark law has created an assortment of effective statutory tools—some to be wielded by courts, others by the Justice Department—to combat laws and practices that improperly dilute the voting power of racial minorities. This subspecies of gerrymandering highlights an important lesson: Federal courts are
not the only branch of government tasked with faithful implementation of the Constitution; nor are courts always the branch best suited to address every constitutional issue.

Consider next the category of
bipartisan
gerrymanders. In jurisdictions where the two major parties, Republicans and Democrats, have worked together to draw district lines that favor these two parties and freeze out all third parties, these bipartisan “collusions” have generally not violated the Constitution. Rightly read, the Constitution in fact sanctions a self-perpetuating and self-stabilizing two-party system. No elaborate Court doctrine is called for here, because the practice is constitutionally proper.
¹⁴

For different reasons,
partisan
gerrymandering designed to advantage one of the two major parties at the other party’s expense also calls for judicial restraint. To begin with, any judicial intervention would be messy in the extreme. Few, if any, easy, workable, and principled sub-rules present themselves as plausible scripts for a large judicial role to neutralize partisanship in the drawing of district lines.

By contrast, in pure right-to-vote cases, such as
Harper
and
Kramer
, the basic framework was easy enough to construct: All adult-citizen, non-felon residents are presumptively eligible voters. This is the group textually identified by section 2 of the Fourteenth Amendment, as updated by the later Woman and Youth Suffrage Amendments.
^*
It is also the lived-constitutional baseline suggested by actual modern practice in the fifty states. Some small questions have arisen at the margins—for example, how long a residency period may a state require?—but even here, actual practice and common sense have narrowed the range of plausible answers. Similarly, we have seen that the
Reynolds
rule offered a workable way to deal with malapportionment (although here, too, smallish questions at the margin needed to be addressed).
¹⁵

But no comparably clean sub-rule exists to regulate district shape. In a sense, all districting is gerrymandering. No district map is neutral. How can principled judges treat like cases alike when each district map seems utterly unique and not easily comparable to any other map in any other
state or census cycle? Perhaps the only clean approach would be to require each state to adopt some form of statewide proportional representation, but this audacious mandate would oblige every state to move beyond a single-member district system with deep historical roots almost everywhere in America.

The lesson here is that on some issues a court’s implementational sub-rules may slightly underprotect certain constitutional values, just as on other issues the rules slightly overprotect. In both situations, there exists a conceptual space between the abstract meaning of the written Constitution (in the domain of interpretation) and the doctrinal sub-rules promulgated and enforced by courts (in the domain of implementation). This space arises because of institutional considerations connected to the basic features of federal courts. When the Constitution goes to court, it needs to be translated into rules that courts qua courts can properly enforce. In this process of translation—when the supreme law of the terse text becomes the detailed court-law of judicial doctrine—areas of overinclusion and underinclusion arise.