Ned Foley replies to Earl Maltz response on originalism and gerrymandering:

Sometimes originalist inquiry leads to unexpected conclusions.  Texas v. Johnson, the flag burning case, serves as a good illustration of this.  Upon initial examination, it is easy to be dismissive of the claim that a statute punishing the burning of a flag is protected by the “freedom of speech” clause of the First Amendment.  Setting a flag on fire is not opening one’s mouth to speak, as any ordinary layperson would immediately observe.  And since the historical record shows approval for the suppression of political dissent among the Founders (Alexander Hamilton’s support for the Sedition Act of 1798 being one example), it would be easy to argue that the Founders themselves did not expect the First Amendment to protect flag burning.

Still, when one undertakes systematic analysis of what “freedom of speech” actually means as a principle added to the Constitution in the First Amendment, it turns out that the principle entails the right to criticize the government without fear of prosecution—and does so even though some of the Founders did not always abide by this truth or fully understand the implications of the principle that ratification of the First Amendment made part of the Constitution.  Likewise, even though the burning of a flag is obviously not the utterance of words, the principle that “freedom of speech” entails (as becomes apparent on further examination of it) the right to express dissent through symbols as well as words, including the symbolic destruction of the flag that visually depicts the government the protester is opposing.  Thus, a committed originalist necessarily comes to the conclusion that flag burning is protected by the First Amendment, as Justice Scalia did in Texas v. Johnson, and abides by this conclusion even at the cost of some personal discomfort (as Justice Kennedy openly expressed in his concurrence in that case).

A similar surprise occurs when one considers the specific issue of congressional gerrymandering from an originalist perspective.  I, for one, did not expect rigorous originalism to yield the conclusion that extreme congressional gerrymanders require judicial invalidation.  Influenced by Justice Scalia’s plurality in Vieth, I had expected that from an originalist perspective nothing was unconstitutional about partisan gerrymandering, which after all has existed as long as the first congressional elections (when Patrick Henry redrew a district in the hopes of defeating James Madison for a seat in the new House of Representatives).  Like Professsor Maltz, I also thought gerrymandering could be considered unconstitutional only if one adopted an avowedly non-originalist jurisprudence, like the kind that animated the one-person-one-vote doctrine of Reynolds v. Sims.

But that assumption was based on a faulty conception of originalism, one rooted in the idea that the law of a constitutional clause is to be discerned in how its authors originally expected the clause to operate.  Called “original expectation application,” this version of originalism has been largely discredited by scholars and jurists alike (including Justice Scalia) and replaced with “original public meaning”—the law of a constitutional clause is what the public as a whole understood those words to embody at the time of ratification.  This distinction is crucial.  It is what justifies decisions like Brown v. Board of Education and Loving v. Viriginia.  Did the authors of the Fourteenth Amendment expect that amendment to outlaw school segregation or laws prohibiting interracial marriage?  Of course not.  But the original public meaning of the phrase “equal protection” embodied a kind of anti-caste principle that, when understood for what it was, necessarily invalidated race-based discriminations regarding education and marriage.

Professor Maltz’s claim that originalism does not generate the conclusion that congressional gerrymanders are unconstitutional is mired in the since-repudiated “original expected applications” version of originalism.  His main piece of evidence is what delegates to the Constitutional Convention said to each other behind closed doors during the summer of 1787 in Philadelphia.  Not only do those particular passages fail to carry the implications that Professor Maltz ascribes to them—the fact that states might abuse the power to regulate the “time, places, and manner” of congressional elections, as Madison feared they would, hardly entails the conclusion that no such abuse could be too extreme to be unconstitutional in Madison’s eyes—but even if that’s an accurate understanding of what Madison himself thought on the particular question, it would not fix the “original public meaning” of the provisions that were actually ratified.  Madison’s own notes on the convention were not published until decades later, in keeping with the promise to keep the convention’s discussions secret.

The way to ascertain the original public meaning of the provisions in Article I that establish elections to the federal House of Representatives is to ask this simple question: is there any law that a state legislature might adopt that might so undermine the kind of election that Article I establishes to require nullification of that state law by the federal judiciary? The answer to this question is obviously yes, and not just because the U.S. Supreme Court has so held in two cases in which state legislatures attempted to impose term limits (or something similar) for elections to the federal House of Representations: U.S. Term Limits v. Thorton (1995) and Cook v. Gralike (2001).

In the second of my five essays last week, I posed a couple of hypotheticals to illustrate that the provisions of Article I for elections to the federal House of Representatives establish judicially enforceable limits that federal courts would be obligated to enforce if a state legislature violated them.  One hypo was that a state legislature would deem an incumbent Representative the winner of the election without bothering to actually count the ballots. The other hypo was that the state legislature would require a challenge to win two-thirds of the vote in order to unseat an incumbent.  Both of these hypos, to differing degrees, would contravene the most fundamental element of elections to the federal House of Representative, as established by Article I: that those elections must reflect the changing will of the state’s electorate on a biennial basis.

Professor Maltz entirely fails to consider these hypotheticals and the implications they have for the problem of congressional gerrymandering.  Given that the original public meaning of Article I is that the biennial elections to the federal House of Representatives must be genuinely responsive to the changing preferences of the electorate, what if congressional gerrymandering becomes so effective that it utterly eviscerates this responsiveness, thereby making congressional elections entirely opposite of what they are supposed to be according to the original public meaning of the Constitution?  To be sure, gerrymandering involves a factual issue not present in either hypo: does gerrymandering actually eviscerate the responsiveness that is supposed to exist, or does it just reconfigure districts in a way that is equally permissible, requiring voters to register their preferences in the context of gerrymandered districts rather than some non-gerrymandered alternative?

Comparing gerrymandering to the two hypos, one can allow that not all instances of partisan redistricting would be unconstitutional in the same way that two hypos obviously are.  The fact that Patrick Henry drew a single district lines to favor James Monroe over James Madison would not, by itself, indicate a subversion of electoral responsiveness to render the particular district lines unconstitutional.  (This anachronistically labeled “Henry-mander” did not involve an egregious geographical distortion that the subsequent “Gerry-mander” did, which is a key reason why the malevolent practice is known by the latter name rather than the former.)  But if a state legislature grossly manipulates all the district lines in the state in order to maximize partisan advantage, and to prevent the electorate from pushing the incumbent party out of power after the electorate has changed the direction it wishes legislative policy to take, then there is a fair question whether the gerrymander in fact contravened the constitutionally required responsiveness for federal House of Representatives in a way similar to the two obviously unconstitutional hypotheticals.

Because facts are involved, there is a role for Congress to play.  The advantage of a “dormant Elections Clause” approach, like the “dormant Commerce Clause,” is that it gives Congress considerable latitude in setting the evidentiary and procedural rules for determining whether state laws undermine the electoral responsiveness that is supposed to occur, just as Congress can determine when state laws interfere with interstate commerce.  Professor Maltz is incorrect when he says that the explicit grant of power to state legislatures in the Elections Clause of Article I precludes the application of a “dormant Elections Clause” jurisprudence analogous to the dormant Commerce Clause.  Just the opposite is true.  If the Commerce Clause did not exist, states would have unfettered authority over interstate commerce: their plenary power over commerce preceded the Constitutional Convention.  Not true with respect to elections to the federal House of Representatives, which is a creature of the Constitution itself and did not predate the Convention.  The states have no antecedent, residual, or plenary authority over elections to the federal House of Representations.  They only have such power as Congress has chosen to let them have, pursuant to Congress’s ultimate authority under the Elections Clause. (The Supreme Court has been clear on this point, including in the term limits cases.)  So the idea that states might inappropriately intrude on congressional turf—in a way that the federal judiciary might need to intervene in order to protect that congressional turf from presumptively inappropriate state interference—is an idea much more suitable to the Elections Clause than to the Commerce Clause.  Thus, a “dormant Elections Clause” jurisprudence is much more sound from an originalist perspective than the longstanding dormant Commerce Clause jurisprudence.

But even if there were no room for circumstances in which Congress (pursuant to its own Elections Clause authority) might approve of a state legislature’s redistricting map that the judiciary invalidated for contravening the responsiveness principle inherent in federal House of Representatives elections under Article I, it would remain true that some state laws would be so egregious that they would need to be judicially invalidated.  The two hypos prove this point (as do the term limit cases).  Thus, the question remains whether a state legislature’s gerrymander is so extreme as to destroy the constitutionally required responsiveness of elections to the federal House of Representatives.  If so, then under Marbury v. Madison it cannot stand.

Undoubtedly, exercise of the Marbury duty in this context would be easier if the Constitution explicitly invalidated congressional gerrymanders.  But the Marbury duty would have been easier in New York Times v. Sullivan if the First Amendment, rather than just saying “freedom of speech,” spelled out explicitly that laws criminalizing seditious libel are unconstitutional.  Likewise, the Marbury duty would have been easier in Texas v. Johnson if the First Amendment explicitly covered flag burning as well as other forms of political dissent.  But, as in those cases, the Court’s obligation under Marbury is to discern the true original meaning of the provisions ratified as part of the Constitution and, once discerned, enforce that true original meaning—and to do so even if it leads to results inconsistent with the original expectations of some Founders.  Insofar as congressional gerrymandering destroys the responsiveness that the Constitution requires House of Representatives elections to have, Marbury requires the invalidation of that gerrymandering.  This conclusion is sound—it follows from proper originalist inquiry—no matter how unexpected it might have been to some at the time of the Founding, or that it still might be to some now.