One of the most firmly established principles of constitutional law is "one person, one vote,"(1) meaning that legislative districts within each state must be equal in population. There are no dissenters from that proposition on the Supreme Court, and there have been none for decades. Legislatures, litigants, judges, and academics all accept the proposition. Yet as a matter of text and history, that proposition is almost certainly incorrect, and judicial enforcement of it has produced unintended results that are perverse from many different points of view.
In order to bring legislative districts as close to "precise mathematical equality" as possible, states must disregard preexisting political boundaries such as cities, townships, and counties. Adherence to these traditional boundaries was, historically, the principal constraint on creative districting, popularly known as "gerrymandering." Once freed from these traditional constraints by the Supreme Court's "precise mathematical equality rule," legislative line-drawers were able to draw maps to produce the results they desired, rendering elections less a reflection of popular opinion than of legislative craftsmanship. The problem has become particularly acute with modern computer districting software, which allows mapmakers to create imaginative districts with the precision of a surgeon. The results? Protection for incumbents, a tendency toward homogeneous -- and hence more partisan -- districts, racial and partisan gerrymandering, and ultimately, a widespread sense that elections do not matter.
Needless to say, these were not the objectives the Supreme Court thought it was pursuing when it embarked on its great adventure of solving the malapportionment problem. It behooves students of the political process to understand the mistakes that were made, and their results, if only as an object lesson that departure from the actual text and meaning of the Constitution, even in service of well-intentioned goals, can have unintended consequences.
This is not to say that the old system of grossly malapportioned legislatures was constitutionally proper, or that the courts were wrong to take steps to dismantle it. My point, instead, is that the Court adopted a legal theory for addressing the issue that was wrong in principle and mischievous in its consequences. More careful attention to constitutional text and history would have produced a better solution.
I. THE REDISTRICTING PROBLEM BEFORE BAKER V. CARR
Until the early 1960s, the federal courts played no role in legislative districting. By almost any measure of democratic legitimacy, however, the districting process was a disaster. Take the example of Tennessee, which was at issue in Baker v. Carr.(2) The Tennessee Constitution theoretically required the legislature to draw district lines in accordance with population, but there was no mechanism for judicial enforcement in state court.(3) From 1901 until Baker in 1963, the legislature simply left existing district boundaries in place.(4) Yet enormous population shifts had occurred during that time. In 1901, a majority of the American public still lived on the farm. Cities were small; suburbs, in that pre-automobile age, were nonexistent. But Tennessee employed the same district lines in 1960 that had been established in 1901. As a result, the suburbs and cities were grossly underrepresented. For example, rural Moore County, Tennessee, had a single representative(5) for 2,340 voters,(6) whereas urban Shelby County was allotted only eight representatives(7) for its over 312,000 voters(8) -- meaning about 44,000 voters per representative as compared to the 2,300 in Moore County. That is an astonishing disparity.
These disparities were not random; they were systematic. This style of malapportionment in Tennessee and elsewhere gave rural and agrarian interests a lock on legislative power, despite their minority status. Voters in rural districts having only 40% of the voting population elected 63 of the 99 members of the state House of Representatives(9) -- almost a two-thirds majority. …