We are accustomed to saying that there are 27 amendments to the U.S. Constitution, but when the Congress directs the Government Printing Office to publish official copies of that text, it employs greater precision. It inserts the following words immediately before the First Amendment: “Articles In Addition To, And Amendment Of, The Constitution Of The United States Of America.”These distinctions matter, for in truth, most of the 27 articles that follow the original Constitution do not amend its text but are either additions to it or amendments to Supreme Court decisions. Only three actually amend language found in the original text. The 12th Amendment alters Article II pertaining to presidential elections. The 14th Amendment amends the original text (eliminating the 3/5 clause) as well as a Court interpretation of that original text (reversing Dred Scott v. Sanford), and it adds to the original text by declaring that blacks are citizens, requiring, among other things, that states provide due process and equal protection. And the 17th Amendment alters Article I, Section 3 concerning the mode of electing the Senate.The 12th Amendment’s revision of the operation of the Electoral College was clearly necessary given the rise of political parties, and the 14th Amendment was necessary to redeem the Republic from the gravest flaw of its birth. The 17th Amendment’s provision for direct election of the Senate, however, was not necessary at all. It eliminated the Framers’ principal means of protecting federalism, and it was, in point of fact, a big mistake.It is rare to hear such a claim these days, and those few who make it are usually not taken very seriously. For example, when Alan Keyes, running for Senate in Illinois in 2004, called for the repeal of the 17th Amendment, Professor Lewis Gould of the University of Texas called Keyes “daffy.”Now, criticizing the 17th Amendment is not exactly a smart campaign technique, since even on a good day not one in a hundred college-educated Americans has the slightest idea what it is. But it’s not at all daffy. I will argue here that in 1913 the original federal structure of the American Republic, carefully and deliberately designed by the Founders, was radically altered, with important and ironic consequences for the evolution of American democracy ever since.What the Framers FramedThat most Americans do not really understand what the Framers intended—notwithstanding the neo-classical fad of unalloyed adulation for them in recent years—ought to be obvious, for we have a dead canary in our legal mine. Thirteen times since 1976 (and twelve times since 1992) the U.S. Supreme Court has invalidated federal laws in order to preserve what it has described as “the original federal design.” Why has this proved necessary? Did the Framers expect the Supreme Court to play this role, and if not, what did they expect?These are not rhetorical questions; investigation into the proceedings of the time turns up actual answers. The Founders certainly did not expect an activist Court to protect federalism. They expected constitutional structure itself, particularly the mode of electing the United States Senate, to protect it.1There are other structural protections of federalism in the Constitution—the states’ involvement in the election of the president by the Electoral College (Article II, Section 1) and in the amendment process (Article V) are two of them. The mode of electing the Senate was the structural provision on which the framers placed most emphasis, however. It is because the original federal design was amended out of existence in 1913 that the Supreme Court has had to defend the basic federal principle, and the frequency with which the Court has had to do so is itself related to the 17th Amendment.Following ratification of the 17th Amendment—as well as the 16th, which authorized the federal income tax and also dates to 1913—the power of the national government grew rapidly. It did so for many reasons, but the 17th Amendment was surely one of them. After 1913 the Congress enacted many measures adversely affecting states’ prerogatives that earlier Senates would never have approved. Not only have these post-17th Amendment congressional measures increased in number and intrusiveness, they have also over time become more abstract, discretionary and prescriptive, leading to what Theodore Lowi has called “policy without law.”2See Theodore Lowi, The End of Liberalism: Ideology, Policy, and the Crisis of Public Authority (W. W. Norton & Company, 1969), and Vikram David Amar, “Indirect Effects of Direct Election: A Structural Examination of the Seventeenth Amendment”, Vanderbilt Law Review 49 (1996), pp. 1360-89. No longer answerable to state legislatures, the Senate felt free to join the House in legislating on every social and economic as well as political problem confronting the nation, even if the resulting measures have amounted to little more than blank checks of authority to the Executive Branch and the Federal bureaucracy. The skyrocketing increase since 1913 in the ratio of bureaucratically mandated regulations to actual congressional law has been the result, one that has weakened not only the states but Congress itself.When the Court has been called upon to define the line between federal and state power, it has generally done so by invalidating congressional measures that would imbalance the federal system in favor of the national government. Yet, somehow, the Court has failed to understand the source of its own need for activism or the reason that its decisions have tended so strongly in one direction. This myopia is something of a puzzle, because it ought to be clear to any student of the Constitutional period that the Framers made the mode of electing the Senate the principal structural means not only for protecting the interests of the states as states, but also for identifying the line demarcating federal from state powers. That is why on May 31, 1787, early in the Constitutional Convention, the delegates rejected Resolution 5 of the Virginia Plan, which proposed that the “second branch of the National Legislature ought to be elected by those of the first”—in other words, that the Senate should be elected by the House of Representatives. Instead, on June 7 they unanimously accepted a motion by John Dickinson, seconded by Roger Sherman, providing for the appointment of the Senate by the state legislatures.The delegates were persuaded by Dickinson’s argument that the “sense of the States would be better collected through their Governments than immediately from the people at large”, and by George Mason’s observation that election of the Senate by state legislatures would provide the states with
some means of defending themselves against encroachments of the National Government. In every other department, we have studiously endeavored to provide for its self-defense. Shall we leave the States alone unprovided with the means for this purpose? And what better means can we provide than giving them some share in, or rather making them a constituent part of, the Nat’l Establishment?
On June 20, James Wilson, a passionate nationalist, warned his fellow delegates that “a jealousy would exist between the State Legislatures & the General Legislature.” He observed, “the members of the former would have views & feelings very distinct in this respect from their constituents. A private Citizen of a State is indifferent whether power be exercised by the Genl. or State Legislatures, provided it be exercised most for his happiness.” On the other hand, “[h]is representative has an interest in its being exercised by the body to which he belongs. He will therefore view the National Legisl. with the eye of a jealous rival.”Wilson’s attack utterly failed, not because the delegates disputed his analysis but because they approved the outcome. Since they were committed to preserving the states as political entities, they found persuasive Mason’s assertions that the states would need the “power of self-defense” and that “the only mode left of giving it to them was by allowing them to appoint the second branch of the National Legislature.” Accordingly, on June 25, the Convention reaffirmed its previous decision to elect the Senate by state legislatures by a vote of nine states to two.”3Farrand, Records, 1: 408. The delegates were aware of other “modes” they could have incorporated into the Constitution that would have defended the interests of the states as states. To mention the four most obvious, they could have (1) specified that the Senate delegation from a state vote as a block, (2) made explicit provision for the instruction of senators by state legislatures, (3) allowed the states to recall their senators, and (4) required rotation in office.The service rendered to federalism by this mode of electing the Senate was repeatedly acknowledged during the ratification debates. In Federalist 45, James Madison declared that, since “[t]he Senate will be elected absolutely and exclusively by the State Legislatures”, it “will owe its existence more or less to the favor of the State Governments, and must consequently feel a dependence, which is much more likely to beget a disposition too obsequious, than too overbearing towards them.” In Federalist 46 he further noted that if the House of Representatives were to sponsor legislation that encroached on the authority of the states, “a few representatives of the people would be opposed to the people themselves; or rather one set of representatives would be contending against thirteen sets of representatives, with the whole body of their common constituents on the side of the latter.” The Senate, he assured his readers, would be “disinclined to invade the rights of the individual States, or the prerogatives of their governments.”In Federalist 59, Alexander Hamilton likewise emphasized that the appointment of senators by state legislatures secured “a place in the organization of the National Government” for the “States in their political capacities.” Had it been otherwise, he continued, “it would doubtless have been interpreted into an entire dereliction of the federal principle.” During the New York Ratifying Convention, Hamilton again explicitly connected the mode of electing the Senate with the protection of the interests of the states:
When you take a view of all the circumstances which have been recited, you will certainly see that the senators will constantly look up to the state governments with an eye of dependence and affection. If they are ambitious to continue in office, they will make every prudent arrangement for this purpose, and, whatever may be their private sentiments or politics, they will be convinced that the surest means of obtaining reelection will be a uniform attachment to the interests of their several states.
This same argument was also made repeatedly in the early days of the new republic. In a July 1789 letter to John Adams, Roger Sherman emphasized that “[t]he senators, being eligible by the legislatures of the several states, and dependent on them for reelection, will be vigilant in supporting their rights against infringement by the legislative or executive of the United States.” And Joseph Story in his Commentaries on the Constitution of the United States (1833) observed that one of the “main grounds” for the mode of appointing the Senate was that it “would introduce a powerful check upon rash legislation” and “would increase public confidence by securing the national government from undue encroachments on the powers of the states.”The Framers favored election of the Senate by state legislatures not simply because it was, as Madison put it in Federalist 62, “the most congenial with the public opinion”, or because it provided, in Hamilton’s words from Federalist 59, incentives for senators to remain vigilant in their protection of the “States in their political capacities.” They also favored it because it helped them sidestep what Madison described in Federalist 37 as the “arduous [task of] marking the proper line of partition, between the authority of the general, and that of the State Governments.”An episode at the very outset of the Convention is most telling on this point. On May 31, the Convention, meeting as a committee of the whole, had just taken up Resolution 6 of the Virginia Plan that proposed, inter alia, that “the National Legislature ought to be empowered . . . to legislate in all cases to which the separate States were incompetent.” In his notes of the deliberations of the Federal Convention, Madison reports that Charles Pinckney and John Rutledge “objected to the vagueness of the term incompetent, and said they could not well decide how to vote until they should see an exact enumeration of the powers comprehended by this definition.” Madison responded to their concerns by expressing his “doubts concerning [the] practicality” of “an enumeration and definition of the powers necessary to be exercised by the national Legislature.” Despite coming into the Convention with a “strong bias in favor of an enumeration”, he owned that, during the weeks before a quorum gathered in Philadelphia (when he and his fellow Virginia delegates drafted the Virginia Plan), “his doubts had become stronger.” He declared that he would “shrink from nothing”, including, he implied, abandoning any attempt to enumerate the specific powers of the national government, “which should be found essential to such a form of Government as would provide for the safety, liberty, and happiness of the community. This being the end of all our deliberations, all the necessary means for attaining it must, however reluctantly, be submitted to.”In this speech, Madison merely foreshadowed the argument he later developed in Federalist 51, namely, that the power of the new federal government was to be controlled not through an exact enumeration—i.e., through “parchment barriers”—but by “so contriving the interior structure of the government, as that its several constituent parts may, by their mutual relations, be the means of keeping each other in their proper places.” Nonetheless, his words were obviously reassuring, for the Convention voted at the conclusion of his speech to accept that portion of Resolution 6 by a vote of nine states “yes”, one state “divided.”The Convention shared Madison’s doubts about the “practicality” of partitioning power between the federal government and the states through an enumeration of the powers of the former; it spent almost no time debating what specific powers the federal government should have. Indeed, the only resolution pertaining to the powers of the federal government forwarded by the delegates to the Committee of Detail (the group charged with taking “the proceedings of the Convention for the establishment of a Natl. Govt.” and “prepar[ing] and report[ing] a Constitution conformable thereto”) stated only that
the Legislature of the United States ought to possess the legislative Rights vested in Congress by the Articles of Confederation; and moreover to legislate in all Cases for the general Interests of the Union, and also in those Cases to which the States are separately incompetent, or in which the Harmony of the United States may be interrupted by the Exercise of individual Legislation.
Not even when the Committee of Detail created out of whole cloth what ultimately became Article I, Section 8 did the Convention systematically scrutinize the powers enumerated therein. The delegates did not even object to the proposed Necessary and Proper Clause authorizing Congress “to make all laws which shall be necessary and proper for carrying into execution” the powers enumerated to it, and the reason is clear: Rather than rely on precisely drawn lines demarcating the powers of the federal and state governments, the Framers relied on structural arrangements to ensure that the national government would not abuse its vast powers, and that the federal design would be preserved.That the Framers sought to protect federalism on the basis of constitutional structure and the self-interest of senators, not through the intervention of the Supreme Court—the least democratic of the three branches of government—is clear simply from noting the placement, brevity and generality of the judicial article. To begin with, Article III, establishing the federal judiciary, follows Article I, establishing the legislative branch, and Article II, establishing the executive branch. By so arranging the articles, the Framers addressed each branch, in the words of James Wilson, a member of the Constitutional Convention and an original justice on the Supreme Court, “as its greatness deserves to be considered.”Further, Article III is only about a sixth as long as the legislative article, and only about a third as long as the executive article. Article I specifies in detail the qualifications of representatives and senators, the sizes of the two houses of Congress, the procedures they must follow, and the powers they are authorized or prohibited to exercise. Article II is likewise detailed in its discussion of the president’s qualifications, mode of appointment, powers and responsibilities. Article III, however, merely vests the judicial power of the United States in one Supreme Court of unspecified size and in “such inferior Courts as the Congress may from time to time ordain and establish.” It outlines no procedures the courts are obliged to follow, and it imposes no qualifications on judges—not even the requirement of citizenship!More specific proof that the Framers did not expect the Court to protect federalism is also evident. They understood that drawing a line between federal and state powers involves considerations beyond the Court’s legal capacity to pass judgment. To the extent that the Constitution authorized the Court to exercise the power of judicial review (and whether it did was itself a major question), it was only in those cases in which the popular branches had acted, in the words of Federalist 78, “contrary to the manifest tenor of the Constitution.” The Court was not to invalidate congressional measures in close ambiguous cases. Rather, as Hamilton makes clear in Federalist 78, it was to invalidate measures only in cases in which Congress’ disregard for “certain specified exceptions to the legislative authority” was akin to its passage of a bill of attainder or an ex post facto law. The Court could therefore not rule on decisions by Congress regarding where federal power ends and state power begins, for these were of a different character; they did not implicate “specified exceptions” to Congress’ legislative authority but merely involved prudential judgments, agreed to by a Senate elected by state legislatures, concerning the outer reaches of delegated congressional powers. Such decisions could never be held unconstitutional by the Court because they never could be regarded as clearly contrary to the Constitution’s “manifest tenor.”Quite apart from such considerations, the Framers did not expect the Court to protect federalism because they recognized that they could not make it in the Court’s self-interest to do so. As the Anti-Federalist Brutus had shrewdly remarked, it would never be in the self-interest of the Court to strike down federal laws encroaching on the “inviolable and residuary sovereignty” of the states, because “[e]very extension of the power of the general legislature, as well as of the judicial powers, will increase the powers of the courts.” The Framers made no effort to contradict Brutus’ assessment and thus concurred sub silentio, which is why they drafted Article III, Section 2 to make it decidedly contrary to the Court’s self-interest to interfere with Congress’ decisions concerning the line between federal and state power.One last point concerning the Framers’ intentions for the Court: As they did not see it in the self-interest of the Court to protect federalism, so also they did not think it in the best interest of federalism for it to be protected by the Court. The Framers wanted the people to have maximum flexibility to draw the line between federal and state powers where they wished. They recognized, as Madison argued in Federalist 46, that the people might “in future become more partial to the federal than to the State governments, . . . and in that case, the people ought not surely to be precluded from giving most of their confidence where they may discover it to be most due.” They were confident that such a “change [could] only result from such manifest and irresistible proofs of a better administration [by the federal government], as will overcome all [the people’s] antecedent propensities”; nevertheless, if such a change of public attitude did come about, they wanted to accommodate the people’s wishes to draw the line between federal and state power where their elected representatives wanted them, not where the Supreme Court might determine.The Framing Un-FramedThe Framers’ original understanding of how federalism would be protected succeeded admirably for the first century of the American republic. The measures Congress passed were understood to be consistent with the original federal design and as serving those interests that prompted the adoption and ratification of the Constitution in the first place. With Dred Scott the principal exception, they were understood similarly by the Supreme Court, as well.4During the entire period prior to the ratification of the 17th Amendment, the Supreme Court invalidated only seven congressional measures: Dred Scott v. Sandford, 60 U.S. 393 (1857); United States v. Dewitt, 76 U.S. 41 (1870); United States v. Fox, 95 U.S. 670 (1878); the Trademark Cases, 100 U.S. 82 (1879); the Employers’ Liability Cases, 207 U.S. 463 (1908); Keller v. United States, 213 U.S. 138 (1909); and Coyle v. Smith, 221 U.S. 559 (1911). An eighth statute considered in Matter of Heff, 197 U.S. 488 (1905), could be added, but the Court explicitly overturned Matter of Heff in United States v. Nice, 241 U.S. 591 (1916). Over time, however, the public became increasingly dissatisfied with the indirect election of the Senate and unappreciative of the protection it rendered to federalism.While state ratification of the 17th Amendment came quickly and easily in 1912-13, its prerequisite adoption by the Congress took 86 years. The first resolution calling for direct election of the Senate was introduced in the House of Representatives on February 14, 1826. From then until the adoption of the 17th Amendment, 187 subsequent joint resolutions of a similar nature were introduced before Congress, 167 of them after 1880. The House approved six of these proposals before the Senate finally gave its consent. What explains this history?One factor was legislative deadlock over the election of senators, which was most often brought about when one party controlled the state assembly or house and another the state senate. While the potential for deadlock was always present, an 1866 act of Congress under its Article I, Section 4 power to prescribe the time and manner of electing senators turned potential into reality.The 1866 act was intended to assist the Senate in determining whether its members had been properly elected. The reason for it was that in 1857 a minority of the Indiana Senate, which had been in deadlock with the House, met with a majority (but not a legal quorum) of the House members and proceeded to elect a senator, and in 1866, the New Jersey legislature met in joint assembly and elected a senator who had received only a plurality of the votes cast (40 votes out of 81).To obviate such problems, the new law mandated that, at the beginning of a session in which a senator was to be elected, the two houses of the legislature were to meet separately and openly and, by voice vote, “name one person for senator.” If the houses did not elect the same candidate, they were required to meet daily in joint assembly and “take at least one vote until a senator shall be elected.”These requirements may have worked to prevent those without actual majority support from becoming senators, but it also revealed at the outset of each session what a small minority in one house would need to know to prevent, if it could not control, the election of any senatorial candidate. The 1866 law thus led directly to a chronic problem of deadlocks. For example, in 1885, the Oregon legislature failed after 68 ballots to elect a senator and eventually did so only in special session. Over the next quarter century deadlocks occurred in West Virginia, Louisiana, Montana, Washington, Wyoming, Kentucky, Oregon (again), California, Utah, Pennsylvania, Rhode Island, Colorado and Delaware. Delaware’s deadlock was the most embarrassing of all; it was represented by only one senator in three Congresses and was without any representation at all from 1901 to 1903. From 1885 to 1912, there were 71 such legislative deadlocks, resulting in 17 Senate seats going unfilled for an entire legislative session or more.These deadlocks sometimes led to the election of what George Haynes, the famous historian of the Senate, called “the darkest of the dark horse” candidates, and opened the way for excessive corruption as deadlocks were occasionally loosened by the lubricant of bribe money. Prior to the passage of the 1866 act, the Senate had investigated only one case of alleged bribery in the election of a senator. Between 1866 and 1900, the Senate was called on nine times to investigate alleged bribery in Senate election cases, and by 1912 that number had increased by another five. In the 59th Congress alone, 10 percent of the Senate’s membership was put on trial or subjected to legislative investigation.Instances of bribery and corruption were, in truth, few in number. Of 1,180 senators elected from 1789 to 1909, only 15 were contested on account of allegations of corruption. Only seven of these were denied their seats. Nonetheless, these instances were much publicized and proved crucial in undermining support for the original mode of electing senators.At the same time, the problems of deadlock and corruption affected attitudes within state politics. Beginning with what was then called a “memorial” from the California State Legislature on February 18, 1874, Congress received a total of 175 memorials from state legislatures urging the direct election of the Senate. State legislatures did more than merely send memorials to Congress. By 1912, 33 states had introduced preferential primaries, and most of them had exercised their power under Section V of the Constitution to call for a convention to consider amending the Constitution to provide for direct election of the Senate.In 1888, South Carolina was the first state to introduce the preferential primary, which democratized the election of senators in the same way that the election of the president had been democratized. Just as state members of the Electoral College were expected to cast their votes for the presidential candidate who had received the greatest popular vote in the state, so the state legislatures were asked to elect that candidate for the Senate who had received the greatest popular vote in the primary. The preferential primary, however, shared the same problem as the democratized Electoral College: that of the faithless elector. Since state legislators were not legally bound by the results of the primary, the State of Oregon passed by initiative in 1904 what became known as the “Oregon system”, which 11 other states soon imitated. Under this system, a general election runoff was held between the major parties’ primary nominees for the Senate and candidates for the state legislature were “permitted” to include in their platform one of two statements regarding their views on the election of senators. “Statement No. 1” pledged the candidate to abide by the results of the general election, even if the winner was from a different party. “Statement No. 2” declared that the candidate would treat the results of the general election “as nothing more than a recommendation” and would vote according to his personal discretion. Most candidates for the state legislatures chose “Statement No. 1” for obvious reasons. Under such circumstances, the preferential primary coupled to a version of the Oregon system was tantamount to the direct election of the Senate.State calls for a constitutional convention constituted a high-risk strategy. Article V of the Constitution makes no provision for the manner of selecting and apportioning the delegates to such a constitutional convention, for the place of holding such a convention, for the rules of its proceedings, or for the scope of its authority. As a consequence, many argue that once a convention is called, there is no way to confine its deliberations; such a convention might consider itself authorized to propose other amendments to the Constitution as well—or even to propose an entirely new Constitution organized on completely different principles. Nevertheless, the states seemed willing to risk constitutional mayhem for the sake of securing the direct election of senators. By 1912, 28 of the then necessary 32 states to call for a convention had formally petitioned Congress for a convention, two more had done so informally, and Arizona and New Mexico, which were about to enter the Union, had expressed their support for such a convention.Of course, all of this activity on the state level was affected by broader movements in American politics. The Populist movement’s deep-seated suspicion of wealth and influence prompted it to portray the Senate as “an unrepresentative, unresponsive ‘millionaires club’, high on partisanship but low in integrity.”5David E. Kyvig, Explicit and Authentic Acts: Amending the U.S. Constitution, 1776-1995 (University Press of Kansas, 1996), p. 209. In the House, proponents of direct election such as William Calvin Oates (Democrat of Alabama) proclaimed a need to “awaken . . . in the Senators . . . a more acute sense of responsibility to the people.”6Election of Senators”, House Reports, 50th Congress, 1st Session, No. 1456, 2. Likewise in the Senate, John Hipple Mitchell (Republican of Oregon) criticized the Senate for being “a sort of aristocratic body—too far removed from the people, beyond their reach, and with no especial interest in their welfare.”7“Election of Senators”, Senate Reports, 54th Congress, 1st Session, No. 530, 10.When Populism waned, Progressivism waxed in its place, evangelizing belief in “the redemptive powers of direct democracy”, that is, the conviction that the solution to all the problems of democracy was more democracy. The people could be trusted to act for themselves; government was to be not only “of, by and for” the people but, as Woodrow Wilson put it, “through the people.”8Woodrow Wilson, The New Freedom: A Call for the Emancipation of the Generous Energies of a People (Doubleday, 1913), p. 55. Thus, William Jennings Bryan argued on the Senate floor that “if the people of the United States have enough intelligence to choose their representatives in the State legislature . . . they have enough intelligence to choose the men who shall represent them in the United States Senate.” Senator David Turpee agreed: However valid were the reasons for the Framers’ original mode of electing senators, the people at the end of the 19th century were “a new people living and acting under an old system.”In short, under the Progressives’ tutelage, people came to associate the election of the Senate by state legislatures with an outmoded, plutocratic constitution, and direct election of the Senate with reform, faith in the people and progress. Senators were no longer described, in the grandiloquent terms of Tocqueville, as “eloquent advocates, distinguished generals, wise magistrates, and statesmen of note, whose arguments would do honor to the most remarkable parliamentary debates of Europe.” Rather they were subjected to Beardian obloquy: “Some of them were political leaders of genuine talents but a majority possessed no conspicuous merits except the ownership of strong boxes well filled with securities.”9Charles A. Beard and Mary R. Beard, The Rise of American Civilization: Part Two (Macmillan, 1927), p. 559.
Woodrow Wilson was a particularly strong advocate of the direct election of the Senate. He argued that America’s constitutional forms were all associated with a mechanistic “Newtonian Theory” of politics that had been superseded by a modern “Darwinian Theory.” “The Constitution”, Wilson argued in his 1912 campaign book The New Freedom,
was founded on the law of gravitation. The government was to exist to move by virtue of the efficacy of ‘checks and balances’. . . . The trouble with th[at] theory is that government is not a machine, but a living thing. It falls, not under the theory of the universe, but under the theory of organic life. It is accountable to Darwin, not to Newton.
Moreover, Progressives regarded constitutional forms as evidence of the Framers’ lack of confidence in the people. As Wilson argued, the Framers were “willing to act for the people, but . . . not willing to act through the people. Now we propose to act for ourselves.” Alexander Hamilton came under particularly heavy fire: Because he relied on constitutional forms and not simply on the people, Wilson branded him a “great man, but, in my judgment, not a great American.”As the prospect of a constitutional convention over the issue of the Senate’s election loomed, and after ten Republican Senators lost their seats in 1910 for opposing the direct election of the Senate, the new 62nd Congress finally approved the 17th Amendment on May 12, 1912. The vote in the Senate was 64 to 24, 238 to 39 in the House. The Amendment was ratified by the requisite three-fourths of the state legislatures in less than eleven months and declared to be a part of the Constitution in a proclamation by the Secretary of State on May 31, 1913.What is particularly noteworthy about the adoption and ratification of the 17th Amendment is the absence of any serious or systematic consideration of its potential impact on federalism. The popular press, the party platforms, the state memorials, the House and Senate debates, and the state legislative debates during ratification focused almost exclusively on expanding democracy, eliminating political corruption, defeating elitism and freeing the states from what they had come to regard as an onerous responsibility. Almost no one paused to weigh the consequences of the amendment on federalism itself. Almost no one asked what the Framers had in mind in 1789, or sought to understand their design.Only three exceptions are apparent in the voluminous record. One was Representative Franklin Bartlett, a Democrat from New York, who argued during the 53rd Congress that the interests of the states as states could only be preserved by keeping the senators as representatives of state governments. He fully appreciated that “the Framers of the Constitution, were they present in this House to-day, would inevitably regard this resolution as a most direct blow at the doctrine of State’s rights and at the integrity of the State sovereignties; for if you once deprive a State as a collective organism of all share in the General Government, you annihilate its federative importance.”The other two exceptions were in the Senate: George F. Hoar, a Republican from Massachusetts, and Elihu Root, a Republican from New York. On the Senate floor during the 53rd Congress, Senator Hoar defended the indirect election of the Senate, declaring that the “state legislatures are the bodies of men most interested of all others to preserve State jurisdiction. . . . It is well that the members of one branch of the Legislature should look to them for their re-election, and it is a great security for the rights of the States.” After quoting approvingly from Story’s Commentaries that election of the Senate by the state legislatures “would increase the public confidence by securing the national government from any encroachments on the powers of the states”, Hoar continued:
The State legislature will be made up of men whose duty will be the administration of the State authority of their several State interests and the framing of laws for the government of the State which they represent. The popular conventions, gathered for the political purpose of nominating Senators, may be quite otherwise composed and guided. Here, in the State legislature, is to be found the great security against the encroachment upon the rights of the States.
In the 61st Congress, Senator Root argued against direct election of the Senate on the same grounds—if the sovereignty of the states was to be preserved, the original mode of electing the Senate had to be preserved. Root reasoned, too, that if the 1866 law was at base responsible for the problem of deadlocks, and if deadlocks were the seedbed of corruption, then the sensible thing to do was to fix the 1866 law, not to amend the Constitution.But almost no one listened. Few worried about altering constitutional structure because they embraced the Progressive notion that the Constitution is a living organism that must constantly adapt to an ever-changing environment. They believed in and celebrated the “Darwinian” adaptability of the Constitution and the evolution of its principles. In the glare of the Progressives’ white-hot confidence in the justice and superiority of simple majoritarian rule, the Framers’ arguments for relying instead on a more complex, mitigated democracy were of no avail.Alas, no one mentioned or heeded Tocqueville either, who had observed shrewdly that “men living in democratic ages do not readily comprehend the utility of forms”—Tocqueville’s word for constitutional structure. Indeed, he wrote, such men “commonly aspire to none but easy and present gratifications” and, since the “slightest delay exasperates them”, they are “hostile to forms, which perpetually retard or arrest them in some of their projects.” As Tocqueville argued, however, “this objection which the men of democracies make to forms is the very thing which renders forms so useful to freedom. . . . Their chief merit is to serve as a barrier between the strong and the weak”—in the case of federalism, between the national government and the states. Indeed, concluded Tocqueville, “Forms become more necessary in proportion as the government becomes more active and more powerful”, so that “democratic nations naturally stand more in need of forms than other nations, and they naturally respect them less. This deserves most serious attention.”Tocqueville thus pointed, prophetically as it turned out, to the possibility of irony. The 17th Amendment was ultimately intended to make the Constitution more democratic, but instead, by contributing to the vast growth of the national government and the federal bureaucracy, it has produced on balance a less democratic result. Thus an amendment intended to promote democracy even at the expense of federalism has undermined both.Federalism was originally protected both structurally and democratically—the Senate, after all, was chosen by popularly-elected state legislatures. American federalism today is protected neither structurally nor democratically; the ratification of the 17th Amendment means that the fate of traditional state prerogatives depends entirely on either congressional sufferance (what the Court calls “legislative grace”) or on whether an occasional Supreme Court majority can be mustered. A majority of the Supreme Court is perfectly willing to protect federalism by invalidating what the popular branches have enacted, but it is unwilling to protect federalism by returning areas of policymaking to the states. (To do that it would have to repudiate its own decisions holding that the 14th Amendment incorporates most of the provisions of the Bill of Rights and makes them applicable to the states.) The result is anti-democratic as well as anti-federalist: As University of Texas law professor Lino A. Graglia points out, the passage of the protection of state autonomy into the hands of the Court is impossible to justify because it means “the loss not only of federalism but also of the rights of representative self-government, the removal of power not only from the states but from the ordinary political process.”10See “United States v. Lopez: Judicial Review under the Commerce Clause”, Texas Law Review 74 (March 1996), p. 726. Judging by how the Framers saw the role of the judiciary, they clearly would be appalled by the present condition of the American federal democracy.Just below surface of their reverence for American democracy, many Americans sense that something is not quite right about the way that democracy is functioning today. So they ask, what’s the problem? Too many special interest groups? Too much money awash in an unfettered system of political fundraising? Too much bureaucracy? Too much regulation, or too little? Most of us can’t put our finger on the problem, but what unites most people’s discontent is the sense that individuals, unless they have lots of money or connections, have no meaningful access to the political process except, perhaps, once every four years.One reason for that feeling is that the government that matters most is far away: It’s the Federal government, with its massive and impenetrable administrative apparatus. And yet most of us instinctually believe in subsidiarity, in the common-sense principle that the best solutions to most problems are those nearest their source, those solutions based on local knowledge and endorsement through genuine citizen participation of some kind.Federalism is subsidiarity in institutional form, and American democracy today is characterized by neither. Few American voters know who their statehouse representatives are and most could care less, for the perfectly sound reason that state government is responsible for little of consequence in most people’s lives. If those statehouse representatives were again responsible for electing U.S. Senators, voters would probably care a great deal more about them. And, clearly, ordinary citizens would have greater access to their U.S. Senators though their state representatives than they do today—far more than they have had since 1913 and far more than most citizens can hope to get through the ward structures of our political parties.Of course, this is not what the Progressives anticipated when they urged adoption of the direct election of senators. Instead it reflects well what Wilson himself said of stupidity: “If a man does not know enough to know what the consequences are to be to the country, then he cannot govern the country in a way that is for its benefit.” Indeed. So would the repeal of the 17th Amendment fix all that is awry in American politics? Of course not. But it’s by no means a daffy idea to think it might help. 1There are other structural protections of federalism in the Constitution—the states’ involvement in the election of the president by the Electoral College (Article II, Section 1) and in the amendment process (Article V) are two of them. The mode of electing the Senate was the structural provision on which the framers placed most emphasis, however.2See Theodore Lowi, The End of Liberalism: Ideology, Policy, and the Crisis of Public Authority (W. W. Norton & Company, 1969), and Vikram David Amar, “Indirect Effects of Direct Election: A Structural Examination of the Seventeenth Amendment”, Vanderbilt Law Review 49 (1996), pp. 1360-89.3Farrand, Records, 1: 408. The delegates were aware of other “modes” they could have incorporated into the Constitution that would have defended the interests of the states as states. To mention the four most obvious, they could have (1) specified that the Senate delegation from a state vote as a block, (2) made explicit provision for the instruction of senators by state legislatures, (3) allowed the states to recall their senators, and (4) required rotation in office.4During the entire period prior to the ratification of the 17th Amendment, the Supreme Court invalidated only seven congressional measures: Dred Scott v. Sandford, 60 U.S. 393 (1857); United States v. Dewitt, 76 U.S. 41 (1870); United States v. Fox, 95 U.S. 670 (1878); the Trademark Cases, 100 U.S. 82 (1879); the Employers’ Liability Cases, 207 U.S. 463 (1908); Keller v. United States, 213 U.S. 138 (1909); and Coyle v. Smith, 221 U.S. 559 (1911). An eighth statute considered in Matter of Heff, 197 U.S. 488 (1905), could be added, but the Court explicitly overturned Matter of Heff in United States v. Nice, 241 U.S. 591 (1916).5David E. Kyvig, Explicit and Authentic Acts: Amending the U.S. Constitution, 1776-1995 (University Press of Kansas, 1996), p. 209.6“Election of Senators”, House Reports, 50th Congress, 1st Session, No. 1456, 2.7“Election of Senators”, Senate Reports, 54th Congress, 1st Session, No. 530, 10.8Woodrow Wilson, The New Freedom: A Call for the Emancipation of the Generous Energies of a People (Doubleday, 1913), p. 55.9Charles A. Beard and Mary R. Beard, The Rise of American Civilization: Part Two (Macmillan, 1927), p. 559.10See “United States v. Lopez: Judicial Review under the Commerce Clause”, Texas Law Review 74 (March 1996), p. 726.