Fifty Years of Reapportionment
by George W. Liebmann
In late 1964, I was a young associate in a Baltimore law firm, which I had joined while waiting to serve six months in the Army Reserve. I was allowed to cut my teeth on insurance subrogation cases, and when I returned from my adventures at Ft. Jackson and Ft. Holabird had begun to wonder why I had entered the law. My deliverance came when a partner, Lawrence Rodowsky (later a judge of the Maryland Court of Appeals) came down the hall and told me I had been assigned to work on a Maryland reapportionment case. “Which side are we on?”, I inquired. “The Eastern side,” he replied.
Maryland at the time had not fallen into the clutches of the “political scientists,” comfortable persons of assured if moderate means intent on conferring the blessings of equality and mediocrity on everybody else. Maryland still thought of itself as “the Free State”; the memory of Governor Albert Ritchie’s and H. L.Mencken’s revolt against National Prohibition still lingered on. (See Prohibition in Maryland: A Collection of Documents (Baltimore: Calvert Institute, 8 West Hamilton Street, Baltimore, MD 21201,2012) The Eastern Shore, for which I was a new and improbable champion, no longer was entitled to have one of its citizens rotate into the Governorship every third election, as under the Constitution of 1851, but still enjoyed 9 of 29 seats in the state Senate. Kent County, with 15,481 people enjoyed one seat, as did Montgomery County with 492,000. Unlike today’s anonymous legislators, the State Senators of that period are still remembered, such worthis as E. Brooke Lee, John A. Cade, William S. James, and James Clark; lengthy bridges across the Susquehanna and Choptank Rivers are named after two of them. All but 4 of Maryland’s 24 subdivisions had been established by 1789. (Maryland Geological Survey, The Counties of Maryland (1907), 426. The only serious lapse from the tradition of separate county representation in the State Senate was a provision of the 1867 Constitution giving Baltimore City six seats rather than one. One of these was occupied in 1964 by the far from anonymous Sen Harry Mc Guirk, best remembered for his admonition to an overly verbose colleague: “You’re beating a dead horse dead!”
Maryland’s polity was distinguished by two other unusual features. Along with Georgia and Mississippi, it retained a “county unit” system for the election of Governors. This assured that in the election of 1962. the successful Democratic nominee was J. Millard Tawes, a temperate man from one of the smallest and poorest of the Eastern Shore Counties (per capita income $797) rather than George P. Mahoney, a demagogic paving contractor from Baltomore County (per capita income $2167). After the demise of the county unit system, Mahoney secured the Democratic nomination in 1966 with a campaign against public housing: “your home is your castle: protect it.” He was defeated by a Republican, one Spiro T. Agnew, whose Democratic supporters styled themselves “Agnewstics.”
The other unusual feature of Maryland government was the role of county delegations as local legislatures. “Every one of the other 47 states has one or more standing committees to which the local bills of all counties must be referred.” (Maryland Legislative Council, Research Report 23, at 17 (1944)) This system was upheld in Salsbury v. Maryland (346 U.S.545(1954)) (differing rules of evidence) and Mc Gowan v. Maryland (366 U.S. 420, 427, 537 (1965), see also South Carolina v. Katzenbach, 383 U.S. 301, 328(1966)) (differing Sunday closing laws). Following reapportionment, the larger counties acquired county executives, including Anne Arundel County’s Joseph Alton (jailed in 1974); Baltimore County’s Spiro Agnew (allowed to plead nolo contendere, 1973) and Dale Anderson (jailed in 1974) and Prince George’s County’s Jack B. Johnson (jailed in 2010).
Our resistance to this brave new world was of course unsuccessful, though we had some fun along the way. We eagerly reprinted a reform proposal put forward by several of the plaintiffs just before Reynolds v. Sims that would have retained separate county representation, as well as the post-Reynolds congressional testimony of such liberal luminaries as Attorney General Robert Kennedy, Dean Jefferson Fordham and Robert Mc Kay, former Solicitor General Lee Rankin and Maryland Senator Joseph Tydings, who, in order to stave off constitutional amendments, offered pious assurances that Reynolds v. Sims would be construed to give wide discretion to the states to preserve separate county representation. Our delightful adversary, the late Alfred Scanlan of Shea and Gardner (best known as the long-time counsel for the United States Catholic Conference) paid my role in this brief a nice back-handed compliment: “He’s kookier than the kooks I’m representing!”
We were of course defeated by a 5-2 vote in the Maryland Court of Appeals, The lengthy dissenting opinion by Judge Wilson Barnes quoted at length from our brief. It proved to be more than the usual tombstone for a lost cause, hence this article.
The majority opinion, though invalidating separate county representation even in an expanded 53-member Senate because of 6 to 1 disparities in size between districts, upheld it in a 142-member House, in which the largest district had 29,666 people and the smallest 15,481. It noted that “five subdivisions having 75.3% of the population will select 106 delegates or 74.7% of the membership. . . To us this demonstrates an honest and sincere apportionment founded primarily and principally upon population.” (Hughes v. Maryland Committee, 241 Md. 471(1965), cert. denied 384 U.S.950(1966)) Similar indulgence to separate county representation where the ability of a popular majority to rule was not in doubt was displayed in several other cases immediately following Reynolds v. Sims. (White v. Denman, 34 F.R.D.252 (D. Mass. 1963), appeal dismissed 329 U.S.28(1964)(separate representation for Nantucket and Martha’s Vineyard); Jackman v. Bodine, 44 N.J. 414(1964); Id., 44 N.J. 453(1965); Schaefer v. Thomson, 240 F. Supp. 247 (D.Wyo. 1964)(upholding 2 ½ to 1 disparity))
In 1974, in the second round of reapportionment litigation, the Maryland court, not surprisingly, was confronted with the sort of gerrymandering predicted in the dissenting opinions of Justice Harlan and Frankfurter. (In re Legislative Redistricting, 271 Md. 320 (1974)) It appointed a special master, former Chief Judge Hall Hammond, who sought to limit gerrymandering by using two mathematical tests of compactness set forth in an affidavit by an obscure University of Maryland political science professor, Parris N. Glendening which I had procured in my new role as plaintiffs’ counsel. (Later, as Maryland’s Governor, Glendening became a shameless gerrymanderer in his own right). The plan under attack was propounded by the Governor pursuant to a new constitutional provision (Article III,Section 5) reposing reapportionment powers in him, which bade fair to make Maryland not merely a ‘strong governor’ state but a virtual dictatorship.
The third round of reapportionment litigation in 1982 was less controversial. It involved a plan put forth by our former client Governor Harry Hughes which avoided gerrymandering by combining counties for purposes of Senate representation, the once-mighty Eastern Shore thus being reduced to two multi-county districts, each with one Senator. (In re Legislative Districting, 299 Md. 658(1982))
The fourth round of reapportionment in 1992 involved a plan put forth by Governor William Donald Schaefer. It was upheld by a majority which held its nose in the process. (Legislative Redistricting Cases, 331 Md. 574(1993)) A two-judge dissenting opinion by Judges John Eldridge and Robert Bell resurrected Judge Barnes’ dissenting opinion of 27 years earlier and through it, our brief, alluding to Maryland’s long-established counties and traditions of local legislation and citing the Supreme Court’s little-noticed decision in Brown v. Thomson,462 U.S.835(1983) upholding, partly on standing grounds, a disparity of 3.19 to 1 in apportionment of the Wyoming House, noting that it would require districts including 46.65% of the population to elect a majority of the house, and that there was no systematic discrimination.
In the fifth round of litigation, in 2002, the Maryland court speaking through Chief Judge Bell in a 6 to 1 opinion (Matter of Legislative Districting, 370 Md. 312 (2002)) invalidated Governor Glendening’s plan, invoking the state constitutional principle of ‘due regard’ for local subdivisions, which it declined to view as a “secondary requirement,” and observing that the plan “had twenty-two inter-jurisdictional or shared senatorial districts, an increase of four.” The plan propounded by the court reduced the number of inter-jurisdictional districts to 14.
Finally, in “Round 6″ in 2013, the Court of Appeals, again speaking through Chief Judge Bell, upheld a plan providing for thirteen border crossings (In re 2012 Legislative Districting, 436 Md. 121(2013)), one fewer than under the plan the court itself had propounded ten years earlier’
It will be seen that the Maryland court, like many others, has endeavored to use state “compactness” and “respect for local subdivisions” provisions to attempt to limit gerrymandering, but that it has been not very successful in doing so.(See R. Pildes, Diffusion of Political Power and the Voting Rights Act, 24 Harv. J. of Law and Public PolIcy 119,137-39 (2000))
As Professor Michael Mc Connell has pointed out (M. Mc Connell, The Redistricting Cases, 24 Harv. J. of Law and Public Policy104 (2000)), when seats are allocated among subdivisions according to the method of equal proportions constitutionally mandated for the U.S. House of Representatives, “The rounding off process is purely random and both the largest and smallest districts are likely to have small populations.” In states with many counties and a relatively small legislative body it would be necessary to combine small counties into single districts [to avoid] systematic and significant overrepresentation of rural voters.”
The Supreme Court has applied two different numerical tests: the ‘theoretical majority’ and the ‘argument from extremes.’ With respect to both approaches, the conventional percentages are 10% or 15%. In other words, a 15% approach applied to ‘theoretical majorities’ would require that districts with at least 42 ½% of the population be necessary to elect a majority of the legislative house, a 10% rule would require 45%. Both the 29 member and 53 member Maryland Senate plans allowing separate county representation would fail these tests, largely because of the unusual variations in density of population in Maryland (under the 53-member plan. 37% of the electorate could elect a Senate majority), but the Maryland, Massachusetts and Wyoming lower houses would pass it without difficulty as would those of many other states.The Maryland Court of Appeals majority in the 1964 case noted that the 24 least populous districts under the plan in which each county had at least one delegate had 49.5% of the population and elected exactly half the delegates.
The ‘argument from extremes’ makes no sense at all. It would invalidate the Massachusetts reapportionment previously discussed where the only districts in a large house that deviated from the numerical norm more than slightly were those for two islands. The argument overlooks the fact that underrepresentation, not overrepresentation is the vice to be cured. (Comment, 48 Marquette Law Rev. 516, 520-21 (1965); R. Hanson, Courts in the Thicket, 12 American Law Review 51, 77-78; White and Thomas, Urban and Rural Representation and State Legislative Apportionment, 17 Western Pol.Q.724, 728(1964))
The numbers used are essentially fatuous. Population is not the same thing as registered voters, or votes cast. See Evenwel v. Abbott, No.14-940 on the Supreme Court’s current docket and Burns v. Richardson, 384 U.S.73 (1966). Populations shift, sometimes drastically, between reapportionments.
Worse still, legislatures can be composed of substantially equal subdivisions and still be profoundly undemocratic or perversely constituted. This has been the experience of the Third and Fourth French Republics, the Weimar German Reichstag, and the present Israeli Knesset. Proportional representation is perfectly consistent with even the most exacting equality standard, yet it produces fanatical splinter parties, geriatric major parties, and weak coalition governments. At-large elections are also consistent with a strict population standard; in the wake of Reynolds v. Sims, the Illinois House in 1964 was constituted on this basis, though the majority party provided a nominal opposition by allowing 59 of the 177 candidates to run unopposed. The total effect of the Supreme Court’s wonderful reform, said to be Chief Justice Warren’s proudest act, was the creation of state legislatures that are economically more conservative and socially more liberal than their predecessors, and an almost permanent Republican majority in the federal House of Representatives. Minority, urban, and low-income voters have little reason to celebrate these results; indeed, the last four rounds of reapportionment litigation in Maryland have involved gerrymanders designed to produce six districts in which Baltimore City voters are dominant even though on a population basis Baltimore City is only entitled to five.
The whole argument over apportionment is divorced from the more important questions concerning the recruitment of legislators: how are persons of independence and integrity to be obtained, and what is the appropriate relation of local to state governments. The function of legislatures under modern conditions is less as initiators of legislation but as checks on the executive and on popular passions stimulated by the mass media. That is the theory of our American polity, which was conceived not as a pure democracy but as what Aristotle called a “mixed government.” Were the case otherwise, there would be no Supreme Court to make mischief; its discretionary and increasingly political decisions are the worst malapportionment of all, even if it included graduates of law schools other than Yale and Harvard and even if it accorded Staten Island equal representation with the other New York boroughs. (The Court includes the following New York borough representatives: Ginsberg (Brooklyn), Scalia (Queens), Sotomayor (The Bronx) and Kagan (Manhattan). On the Court’s hostility to Staten Island, see Board of Estimates v. Morris, 489 U.S.688(1989); R. Briffault, Voting Rights, Home Rule and Metropolitan Governance: The Secession of Staten Island as a Case Study in the Dilemnas of Local Self-Determination, 92 Colum.L.Rev.775(1992)) As was said by Justice Woodbury in the first reapportionment case to reach the Court, Luther v. Borden, judges are “irresponsible and unchangeable for life;” the reapportionment power would be “one more dangerous (in theory at least) than the worst of elective oligarchies in the worst of times.” 48 U.S. (7 How.) 1, 53 (1849)
The new fashionable remedy for this is the creation of ‘nonpartisan’ redistricting commissions, chosen perhaps to insure political purity and innocence, from the populace at large by sortition. This cure is as bad as the disease. These inexpert commissions will inevitably fall into the hands of the “political scientists.” See the dissenting opinions in Arizona State Legislature v. Arizona Independent Redistricting Commission, 576 U.S. (2015). They are the people who got us into this mess; they are not the people to get us out of it. For reasons stated by the late Pendleton Herring, government by Muldoon is almost invariably preferable to government by Goo-Goo. P. Herring, The Politics of Democracy: American Parties in Action (1940). The political scientists will inevitably worry about the Voting Rights Act, and will try to construct “majority-minority” districts. The trouble (not the only trouble) with these is that Blacks are not the only, or even the largest ‘minority’ and the Voting Rights Act refers also to “language minorities [and] American Indian, Asian American, Alaska Natives, or of Spanish heritage.” (42 USC 1973 l(c)(3))The result will be legislatures constituted like Noah’s Ark and as disfunctional as deliberative bodies or agencies for compromise as the United Nations General Assembly.
I agree with Prof. Mc Connell that the right remedy can be supplied by the Court, through abandonment of the ‘argument from extremes’ and automatic validation of plans allocating seats according to the method of equal proportions among subdivisions or constitutionally fixed groups of subdivisions where doing so does not permit “the systematic frustration of the will of the majority”. In order for this to happen, a legislature must give it the opportunity to do so, by enacting two plans: the first a principled plan based on county representation and the second a conventionally gerrymandered plan, which “shall not become effective” if the first plan “is declared valid. This course was followed by the Maryland General Assembly in 1965. (241 Md. at 476). Use of pre-existing and permanent subdivision boundaries limits gerrymandering to the lines drawn within more populous subdivisions and greatly limits its possibilities. If a numerical test is needed, 42 ½% or, at most, 45% of the populace to control a legislative majority will do nicely. Because a neutral and rational formula is being applied, it is difficult to justify involvement of the Voting Rights Act, which requires “proof in each and every case that racially polarized voting patterns continue to exist in the specific jurisdiction whose electoral structure is being challenged.” (R. Pildes, Diffusion of Political Power and the Voting Rights Act, 24 Harv. J. of Law and Public Policy 119, 132 (2000)). It is not necessary to accept Professor Mc Connell’s suggestion that the Republican Form of Government clause be added to the court’s arsenal; conventional rational basis doctrine under the Equal Protection clause achieves this result, as Justice Stewart asserted. (Lucas v. 44th General Assembly of Colorado, 372 U.S. 713,at 754(1964)(dissenting opinion))
The writer makes no apology for the conclusion of his abortive Petition for Certiorari of 50 years ago:
“The issues presented in the. . . second round of cases which have begun to appear on the Court’s docket relate less to the allocation among areas in a state and more to the nature of the relationship of a representative to his constituents. The issues now relate not to whether the predominant forces in the community will be able to make their weight felt, but rather to whether representatives are to retain significant and continuous relationships with their districts.
“Demands for mathematical perfection in apportionment and the creation of fluctuating constituencies crossing community boundaries may result, not in more representative and effective legislatures, but rather in making legislatures overly dependent on the favor of political machines and pressure groups, agencies of mass communications, and the executive branch of government.”(Petition for Certiorari, Hughes v. Strong, October Term 1965, 28. See R. Hanson. The Political Thicket (1966), 122: “Mobility has created neighborhoods of strangers and jurisdictions without traditions. . . In this context, alienation from the political process, nonparticipation in the official system of politics or support for mass movements as an anti-politics reaction appears to be increasing in metropolitan areas.”)
Posted in: Judiciary and Legal Issues, State and Local Politics