The Purpose and Powers of the Senate, Part XXXXIX: Political Arithmetic

             As mentioned at the very beginning of the present discussion, the ratification of the 21^st Amendment was an exceptional event in the history of the United States of America. Whereas every other amendment to the constitution, before and since, was submitted to the state legislatures for either approval or rejection, the Senate Joint Resolution that would eventually become the 21^st Amendment was specifically authorized by Congress to be discussed and then voted upon by special conventions summoned for that purpose in each of the extant forty-eight states. For a time during the debates which gave rise to said resolution, this feature was excised from the original draft and replaced by an equivalent clause mandating the more familiar method utilizing state assemblies. Such was the state of the resolution when it was voted out of committee on January 9^th, 1933. In fact, it was only after the defeat of a one-man filibuster on the part of Texas Senator Morris Sheppard and the resumption of the discussion at hand on February 15^th that Arkansas Senator Joseph Robinson insisted on putting forward an amendment to the relevant proposal restoring the state convention clause which the document’s originator, Wisconsin Senator John Blaine, had first proposed on December 6^th. After a lengthy and wide-ranging debate, Robinson’s amendment was comfortably approved by a vote of 45-15 and the repeal resolution was once more transformed into a kind of popular referendum on national prohibition.

Clearly, based on this back and forth over the period of December 6^th to February 15^th, the idea of calling special ratifying conventions was more than a little contentious among the membership of the upper house of the 72^nd Congress. It was part of Blaine’s original draft, was then removed during markup in committee, and was then reinstated as an amendment offered by Senator Robinson of Arkansas. Evidently, different groups of Senators had very different ideas about the proper mechanism by which the repeal of the 18^th Amendment ought to have been submitted to the people. And yet, based on the fact that the amendment ultimately responsible for shifting responsibility back onto the state conventions was approved by a substantial margin – 45-15, with 36 abstentions – most of those present felt it to be especially important. Why was this the case? Evidently, the supporters of repeal – who were proven to be in the majority – believed that an amendment to that effect was more likely to be approved by a series of special conventions than by the existing legislatures of the various states. But what brought them to this conclusion? Why did they imagine that the state assemblies were likely to quash their attempt at repealing Prohibition and what was the practical result of their successful effort to circumvent them?

It should perhaps come as no surprise that the answer to the first question has everything to do with the concept and practice of apportionment in the United States of America. More specifically, it has to do with the apportionment of legislative seats in the various state assemblies. In theory, under the terms of the constitutions of the various states, the principle of legislative equality stands as the paramount consideration for any distribution of legislative representatives. That is to say, every voter, regardless of where they live, should be able to exert the same effect upon the legislative process as every other voter living within the same state. The simplest way to accomplish this, of course, is to mandate a set ratio of voters to representatives. If electoral districts are drawn in such a way as to ensure that every legislator represents as near as possible to five thousand, or ten thousand, or fifty thousand people, every voter could theoretically lay claim to the same degree of access to their elected representative. Not only that, but every voter could also take comfort in their vote carrying the same weight as that of every one of their fellow citizens. The vicissitudes of regional geography – not to mention the ability of campaign donations to catch the attention of lawmakers running for reelection – will naturally skew the degree to which every voter can meaningfully interact with their assigned legislator or the extent to which certain districts or precincts are either underrepresented or overrepresented. But the ideal goal is always to create a baseline that is as flat as possible across an entire state. In practice, of course, this is much easier said than done.

While some states within the American republic come relatively close to achieving a condition of homogenous population density across rural and urban areas, such an even distribution of people and communities is most definitely not the norm. In most cases, the population of a given state varies widely between densely populated cities and sparsely populated countryside, the latter often interspersed with small knots of semi-urban density in the form of scattered towns and villages. Consider, by way of example, several different counties located in the State of New York. New York County – which jurisdiction is coextensive with the Borough of Manhattan – is extremely dense and completely urbanized, with a total population (circa 2020) of 1,694,251 and a population density of 74,780.7/square mile. Suffolk County, to the east on Long Island, is similarly populous – 1,525,920 – but far less dense – 1,637.0/square mile – being primarily suburban in nature with large population clusters spread across a series of moderately-sized towns and cities. Onondaga County, located in Central New York, is comparatively both less populous – 476,516 – and less dense – 600.0/square mile – though its county seat, Syracuse – 148,620 – is over three times as large as Suffolk’s own administrative center at Riverhead. And then, finally, there is Hamilton County in the so-called “North Country,” a jurisdiction whose population – 5,107 – and population density – 3.0/square mile – make it the smallest and least dense county in the whole of the United States east of the Mississippi River. Bearing in mind, once again, that not every state is so diverse in terms of its population distribution, New York nevertheless provides a vivid example of the kinds of problems which much immediately present themselves to anyone who would attempt to fashion a series of electoral districts in which every legislator represents approximately the same number of constituents.

Consider – by way of a patently ludicrous hypothetical – what would happen if Hamilton County were to form the baseline, with one delegate to the state’s House of Representatives assigned to all of its five thousand inhabitants. How many representatives must then be assigned to Onondaga County? Approximately ninety-five. To Suffolk County? Three hundred and five. To the Borough of Manhattan? Three hundred and thirty-nine. Combined, under such a scheme, these four counties alone would send seven hundred and forty delegates to the state assembly in Albany, a number far in excess of its present lower house total of one hundred and fifty. Using the same ratio of constituents to representatives for every other county, New York would be left with a legislative body larger than any other in the world. As such an outcome would naturally present the state and people of New York with a whole host of logistical challenges – not the least of which would almost certainly be a glacially-slow pace of legislation – the number of voters assigned to a single lawmaker must necessarily be increased. But herein lies the real challenge. Increase the number so as to produce a relatively small body of delegates, cities maintain reasonably distinct representation while rural areas get grouped together into large, amorphous districts. Lower the number so as to ensure accurate representation, rural areas become more clearly represented while the number of delegates elected in the cities multiplies by factors of magnitude. Bearing the undesirable consequences of each of these extremes in mind, the only logical approach is accordingly to vary the ratio of constituents based on relative population density while staying as close to a predetermined average as possible. In order to keep the size of a given legislature more or less reasonable, of course, some decisions must be made that bring about a degree of representative imbalance.

Most states, fortunately, has been apportioned quite equitably of late. In New York State, for example, the City of New York is represented in Albany – as of 2022 – by a total of sixty-five delegates spread across New York County, Queens County, Kings County, Bronx County, and Richmond County. The total population of New York City being presently just shy of nine million people, each delegate sent from there to the lower house of the state assembly accordingly represents, on average, about one hundred and thirty-eight thousand people. When one considers that each delegate to this same body, on average, represents about one hundred and thirty-three thousand of the Empire State’s total population of twenty million, this would seem to be a perfectly reasonable arrangement. Just so, the rural 131^st District, comprising Ontario County and roughly half of Seneca County, contains something on the order of one hundred and thirty-two thousand people, while the 125^th District, encompassing all of Tompkins County and about one-fourth of Cortland County, contains around one hundred and sixteen thousand people. Clearly, the exact number of constituents per delegate can and does vary based on relative population density, but overall it would seem as though urban areas are represented on more or less equal footing with rural areas.

This has not always been the case, however. Prior to the 1960s, many states either required that legislative districts be drawn with equality of land area in mind as well as equality of population or simply neglected to redraw electoral district boundaries on anything approaching a regular basis. This prevalence of this tendency became painfully evident in the aftermath of the Supreme Court’s decision in Baker v. Carr (1962), a case brought by a resident of Shelby County, Tennessee named Charles Baker against the government thereof. According to the Constitution of Tennessee, the districts represented in the state assembly are to be redrawn every ten years so as to ensure that they each contain approximately the same number of people. By the early 1960s, however, as the aforementioned Mr. Baker had discovered, the relevant districts had not been redrawn since 1901, resulting in a high degree of malapportionment between various urban and rural areas. According to the Census of 1900, Shelby County contained approximately one hundred and fifty thousand people, at which point its representation in the state assembly was set. Some sixty years later, however, with county seat Memphis having become a major regional center of commerce and industry and the population having expanded to over six hundred thousand, the relevant Shelby County district still enjoyed the same representation as neighboring rural districts with only one-tenth the inhabitants. In asking for relief by way of a lawsuit, what Baker was essentially arguing that cases involving legislative malapportionment ought to be capable of being remedied by the courts. The State of Tennessee disagreed with this contention, holding instead that such questions were strictly political in nature, the result of which was that the Baker case eventually found its way to the Supreme Court of the United States.

Ruling 6-2 in Baker’s favor, the Court ultimately found that cases of malapportionment in fact could be remedied by the appropriate judicial authority, resulting in the establishment a new standard for the drawing of electoral districts – “one person, one vote” – and setting off a wave of reapportionment in the states over the course of the decade that followed. The scrutiny which this process consequently invited to the exact composition of the legislative districts of the various states unsurprisingly turned up more than its fair share of oddities and outrages. In Tennessee, for example, not only had Memphis become underrepresented in the state assembly as a result of population growth, but so too had Nashville, Knoxville and Chattanooga. At the same time – albeit somewhat more suspiciously – certain districts in the eastern portion of the state whose inhabitants had leaned toward the Republican Party since as far back as the Civil War were also found to be underrepresented compared to their counterparts in the central and western thirds of the state. The result was that the rural residents of a little less than two-thirds of the rural Tennessee held a functional monopoly on political power in spite of being in the minority in terms of the overall population.

Connecticut had apparently witnessed a similar drift towards malapportionment over the course of the 19^th and 20^th centuries, principally as a result of the terms of its 1818 constitution. According to the relevant provisions of the document in question, each town was to have at least one representative in the state assembly at the same time that no town was permitted to claim the right to more than two. In consequence, as the state industrialized and certain specific towns became home to textile mills that attracted increasingly large workforces from the surrounding countryside, the balance of power between urban and rural voters began to shift decisively in favor of the latter. By the early 1960s, the Republican Party had come to dominate state politics by way of its popularity among the residents of the state’s many small towns, each of which possessed at least one seat in the state assembly, while Democrats were mostly confined to a handful of large cities like Hartford and New Haven, none of which, regardless of their comparatively large populations, could claim more than two seats in the state assembly. 

On the other side of the country, California was revealed to contain one of the most egregious examples of deliberate malapportionment in the whole of the American republic. At some point in the 1920s, Californians had approved a scheme of legislative representation whereby seats in the lower house of the state assembly were drawn according to population while seats in the upper house were drawn with particular attention to existing county boundaries. Specifically, no county was to be divided up for the purpose of representation unless it was to be split equally into multiple districts with one seat each and no more than three counties were to be merged in order to form a single district. The intention of such a scheme, it seemed, was mainly to prevent the rapidly growing Los Angeles County – already the hub of the nation’s burgeoning film industry and containing some forty percent of the state’s total population – from becoming the single dominant force within the state’s complex internal power dynamics. The result, by the early 1960s, was an almost comical imbalance within the state senate between the numerous seats representing small, rural counties – the smallest of which, Alpine County, had less than three hundred permanent residents – and the single seat representing the expansive City and County of Los Angeles – the total population of which, circa 1960, was something on the order of six million. 

Returning back to New York for one final example, one finds a similar case – if somewhat less extreme in its outcome – of competition for influence between urban and rural interests resulting in the emergence of an extremely lopsided power structure. In 1894, in the midst of an era of Democratic Party dominance within a rapidly growing New York City, a convention was called for the purpose of rewriting the Empire State’s constitution. But while such occurrences were far from uncommon – conventions had previously met and approved amendments in 1801, 1821, 1846, 1867, and 1872 – the outcome of this specific meeting was far from what one might describe as customary. As the state had, since its most recent legislative apportionment, increased significantly in population, the assembled delegates naturally set about adding to the number of seats in both the upper and lower houses of the New York State Legislature. The state senate, specifically, was increased in size from thirty-two seats to fifty while the state assembly was expanded from one hundred and twenty-eight seats to one hundred and fifty. Notwithstanding the fact that much of the aforementioned population growth had been concentrated in New York City, however – or indeed, exactly because of it – a coalition of upstate Republicans conspired with a group of rural Democrats to ensure that New York City received fewer of these new seats than its population would otherwise have indicated. A deal was even struck with the city’s few Republican delegates to gerrymander a handful of safe seats for them in exchange for their cooperation. The result – approved by the people in a referendum vote by a margin in the tens of thousands – was the calculated disenfranchisement of New York City voters for a period extending for several decades.

            What this all amounts to, of course, is that for a significant period from the beginning of the 19^th century through at least the 1960s, many – perhaps even most – state assemblies in the American republic gave far too much weight to rural populations and far too little to their urban-dwelling counterparts. In some cases, to be sure, this was entirely accidental. In Connecticut and Tennessee, legislative districts had been drawn most recently at some point in the 19^th century, leading to significant imbalances as, over the course of the 20^th century, what had been small county towns expanded tremendously under the influence of industrialization and its accompanying labor draw. In other cases, however, such imbalances were quite intentional. As in California and New York, rural interests sometimes conspired to minimize the political power otherwise conferred on growing urban centers by drafting and enforcing redistricting rules that underrepresented urban voters at the behest of a number of comparatively small rural communities. As the assembled members of the United States Senate set about considering the mechanism by which the repeal of the 18^th Amendment might properly have been accomplished in January and February of 1933, the sad truth of such developments doubtless flashed through their minds. Many of these same men having risen to positions of national prominence by way of service in state government, they doubtless possessed firsthand knowledge of just how skewed state power structures could be in favor of rural over urban populations. Their choice, therefore, between allowing the state assemblies to ratify whatever draft amendment they managed to approve or else granting that same privilege to special conventions summoned for that purpose was surely known to them to in fact be a choice between allowing either rural interests or urban interests to determine the fate of the 18^th Amendment.