REAPPORTIONMENT, NONAPPORTIONMENT, AND RECOVERING SOME LOST HISTORY OF ONE PERSON, ONE VOTE.

• Author: Karlan, Pamela S.
• Position: Special Issue on Gerrymandering

TABLE OF CONTENTS INTRODUCTION. 1922 I. THE EVOLUTION OF APPORTIONMENT BEFORE 1920 1925 II. THE STALEMATE OF THE 1920S. 1935 III. THE IRONY OF 1929. 1941 CONCLUSION: FROM 1920 TO 2020. 1956 INTRODUCTION

I don't know whether it was Yogi Berra or Niels Bohr who first said it--if either of them did--but it's tough to make predictions, especially about the future. (1) That is certainly true about the next round of congressional redistricting--the process of drawing the boundaries of the districts from which members of the House of Representatives will be elected. In most states, control over drawing the lines will depend on the outcome of gubernatorial and state legislative elections in 2018 and 2020. And as has been true since the Supreme Court's decision in Wesberry v. Sanders, (2) the line-drawers will once again be crafting their maps under a set of legal constraints that has changed since the previous round. Mapmakers in the South and Southwest are free from the preclearance requirement and prohibition on racial retrogression that governed the last five redistricting cycles. (!) The Supreme Court has, yet again, refined the contours of its doctrine forbidding excessive reliance on racial considerations/ Depending on the Court's decisions in Gill v. Whit-ford nd Benisek v. Lamone, states may face new limits on the degree of permissible partisanship in the redistricting process. (5) The interactive effects of all these rule changes are hard to predict. And just as in prior decades, there are likely to be some unanticipated changes as well.

But when it comes to congressional reapportionment--that is, the process of allocating seats in the House of Representatives among the states (6)--there seems to be no real uncertainty. Since 1941, that process has been the paradigmatic "machine that would go of itself." (7) Sometime in early 2021, the President will "transmit to the Congress a statement" of each state's population "and the number of Representatives to which each State would be entitled under an apportionment of the then existing number of Representatives by the method known as the method of equal proportions, no State to receive less than one Member." (8) And roughly a fortnight thereafter, unless Congress acts to stop him, the Clerk of the House will send each governor "a certificate of the number of Representatives to which such State is entitled." (9) Reapportionment will be over; redistricting will be underway. (10)

It was not always so. In fact, the post-2020 round of reapportionment will mark the centennial of the most striking episode in the history of American reapportionment: Congress's failure, for the entire 1920s, to reallocate seats to reflect the census results. (11) The reasons for this failure, and the consequences of Congress's ultimate response, continue to shape our politics today.

Historians and political scientists have written excellent studies of apportionment that address the nonapportionment post-1920. (12) But none of these studies focuses directly on the doctrinal concerns that informed, and then flowed from, these developments. This Article aims to fill that space. I begin by describing the constitutional structure of apportionment, the questions the Constitution left open, and how those questions were resolved prior to 1920. I then turn to what happened in the 1920s and why. Finally, I explore the aftermath of Congress's ultimate solution, and how that solution set the stage for the Reapportionment Revolution of the 1960s. The story is interesting in its own right, but I also suggest ways in which the upcoming redistricting will present many of the same questions that the nation faced a century before.

1. THE EVOLUTION OF APPORTIONMENT BEFORE 1920

   The original Constitution provided for the apportionment of seats in the House of Representatives in Article I, Section 2, Clause 3:

   Representatives and direct Taxes shall be apportioned among the several States which may be included within this Union, according to their respective Numbers, which shall be determined by adding to the whole Number of free Persons, including those bound to Service for a Term of Years, and excluding Indians not taxed, three fifths of all other Persons. The actual Enumeration shall be made within three Years after the first Meeting of the Congress of the United States, and within every subsequent Term of ten Years, in such Manner as they shall by Law direct. The Number of Representatives shall not exceed one for every thirty Thousand, but each State shall have at Least one Representative. (13) Several things about the Apportionment Clause bear note. First, it clearly requires that seats in the House be allocated on the basis of population (14)--in contrast to the Senate, where each state is given two seats regardless of its size. (15) And inherent in the conjunction of apportionment "according to their respective Numbers," and the requirement for decennial "[e]numeration," is the expectation that those apportionments will change over time to reflect relative shifts in states' populations. (16)

   Second, while the Apportionment Clause sets lower and upper boundaries on the number of seats in the House--each state must have at least one, (17) so today the smallest House would have fifty members, and the total number cannot exceed something north of ten thousand! (18)--it leaves the exact number unspecified. Implicitly, determination of that number is confided to Congress and the political process.

   Finally, and most glaring to modern readers, in determining a state's representational base, the original Apportionment Clause counts slaves as three-fifths of a person for purposes of determining a state's population. (19) Today, when people think about the threefifths clause, they condemn it for the way in which it devalues the personhood of African American slaves. To be sure, the Framers deserve condemnation for denying the full humanity of Black people, but that is not why they inserted this particular clause into the document. Rather, the clause was designed to reduce the political power of the slave states relative to the free states, by discounting their slave population--who of course would have had no say in how the representatives apportioned to those states would be selected. (20) In any case, both the presence and the ultimate elimination of the three-fifths clause show how racial considerations and sectional competition have consistently inflected apportionment in America. (21) And today, observers invoke the three-fifths clause to criticize the practice of counting prisoners for apportionment purposes where they are incarcerated, rather than in the communities from which they came. (22) Prisoners, who are disproportionately persons of color, serve as electoral ballast in majority-white districts whose residents have very different interests from theirs. (23)

   But for all the Apportionment Clause does, there is one important issue on which it is silent: it does not specify the precise method for allocating seats beyond saying it should be based on the states' "respective"--that is, relative--populations. (24) Because state populations are not exact multiples of one another, "[n]o method of apportionment ... can assign representatives to the several states in exact proportion to their respective numbers." (25) There are many methods for "what to do about the fractions." (26) And the choice among them matters, because it determines which states get the last couple of seats up for grabs. (27) Different formulas may systematically advantage smaller states over larger ones or vice versa. (28) In the context of a particular census, different formulas may give the final seat to a safely Democratic state or a safely Republican one, thereby changing the partisan makeup of Congress. (29)

   Between the post-1790 census and 1920, Congress considered, and ultimately used, several different formulas. (30) Each time, the Members were acutely aware of the distributional consequences of choosing one formula over the alternatives. (31) Moreover, in every decade prior to 1920, not only did Congress pick the apportionment formula, but it also determined the number of seats to be apportioned, changing that number each time. (32) Perhaps predictably, with the exception of the post-1840 apportionment, Congress consistently increased the number of seats. (33) And between 1870 and 1920, it chose increases that interacted with the apportionment formula to ensure that no state actually lost a seat. (34)

   Article I, Section 4 of the Constitution--the Elections Clause--gives Congress the power to "make or alter" the "Manner of holding Elections" for members of the House. (35) Congress's first major exercise of this power came in 1842, when it required, after heated debate involving both constitutional and policy-based arguments, that states elect their representatives from single-member districts. (36)

   With the move to single-member districts, the question arose as to whether Congress should also provide direction on how districts should be drawn. (37) Initially, the Senate adopted a proviso "[t]hat such districts shall be, as nearly as practicable, equal in the number of their inhabitants." (38) Supporters of the proposal argued "that the general principle was so clear and obvious, that no one could deny its propriety": equal representation required same-sized districts. (39) And they warned that without such a requirement, each district might "contain any number" of inhabitants, with states manipulating district boundaries so that two parties with equal numbers of supporters might control very different numbers of districts. (40)

   Almost immediately, however, strong opposition emerged. Some of the opposition came from legislators who thought that, for what we would see as values of federalism, states ought to decide how to allocate their seats. (41) Others pointed to the impossibility of drawing equipopulous districts...