I
In 2001, Dr. Ralph A. Rossum, Professor of American Constitutionalism at Claremont McKenna College, published Federalism, the Supreme Court, and the Seventeenth Amendment: The Irony of Constitutional Democracy.[1] He thereby joined an almost century-long dialogue over the efficacy and the merits of the alteration in our basic law that shifted the method by which United States senators are chosen, from selection by the state legislatures to direct popular election.[2] He also offered a startling thesis: the amendment, which was intended by its Progressive Era proponents to advance democracy and to eliminate one manifestation of corruption in government, fundamentally transformed the nature of American federalism by removing from the national government the primary institutional safeguard for the protection of the states as corporate entities. Rossum’s exploration led him first to review and critique efforts by the U.S. Supreme Court since 1976 to protect federalism through an array of constitutional doctrines; then to discuss the Framers’ view of federalism, and the Senate’s special role in protecting it, by drawing upon excerpts from the records of the Constitutional Convention of 1787, The Federalist Papers, case studies from the First Congress, and some landmark cases of the Supreme Court under the leadership of Chief Justice John Marshall; then to examine the forces that led to the adoption of the Seventeenth Amendment; and finally to review the subsequent expansion of national powers at the expense of the states, concluding with some recommended constitutional doctrines that, in his judgment, the Court should now adopt.
A work that attempts to cover so much constitutional ground inevitably invites scrutiny and hazards dispute. The present review will offer some of both. In addition to reexamining the relevant source materials that Rossum cites, I shall also consult some others, such as the observations of Alexis de Tocqueville, John Stuart Mill, and James Bryce. It is appropriate to state at the outset that the view taken here is generally critical of Rossum’s thesis, and to acknowledge that he has identified issues of fundamental importance to an informed grasp of the American regime and the nature of American politics, perhaps of politics per se. He has also, incidentally, highlighted the tunnel vision and naiveté of one strand of late nineteenth- and early twentieth-century Progressivism and thereby admonished against two features of reform movements—the familiar political phenomenon sometimes known as “the law of unintended consequences”” and a variation on this familiar theme that we may call “”the law of undetected (or misidentified) causes.””
II
In his first chapter, Rossum reviews and critiques recent efforts by the U.S. Supreme Court since 1976, and especially since 1992, to protect federalism—that is, to carve out a sphere of significant state autonomy—by means of an array of constitutional doctrines, all of them, in his judgment, flawed and inadequate. This critique of the Court’s conservative judicial activism and often confused jurisprudence is trenchant and generally convincing, but is so detailed and intricate as to warrant a separate review-essay. The context for this confusion, as he anticipates it in his introduction, is that “”these decisions reveal an activist Court that has utterly failed to appreciate that the original federal design it is so committed to protecting is no longer a part of our constitutional system, as it was fundamentally altered by the Seventeenth Amendment.””[3]
Rossum’s next three chapters discuss the Framers’ view of federalism, the Senate’s intended role in protecting it, and the mode of electing senators as the linchpin of securing that role. This inquiry provides a generally useful overview of the American Founding, as well as a challenging exercise in rediscovering the obscure and seeing the familiar in a different light. It entails examination of the records of the Constitutional Convention, The Federalist Papers, and other contemporaneous sources. It is to this material that I shall first turn my attention.
In order to assess Rossum’s core argument about the importance of the Constitution’s original method of senatorial selection, it is necessary to go back to the Virginia Plan as it was first presented to the Constitutional Convention on May 29, 1787. The relevant articles, as recorded by Madison, read:
3. Resd. that the National Legislature ought to consist of two branches.
4. Resd. that the members of the first branch of the National Legislature ought to be elected by the people of the several States every for the term of ; to be of the age of years at least, to receive liberal stipends by which they may be compensated for the devotion of their time to public service; to be ineligible to any office established by a particular State, or under the authority of the United States, except those beculiarly [sic] belonging to the functions of the first branch, during the term of service, and for the space of after its expiration; to be incapable of re-election for the space of after the expiration of their term of service, and to be subject to recall.
5. Resd. that the members of the second branch of the National Legislature ought to be elected by those of the first out of a proper number of persons nominated by the individual Legislatures, to be of the age of years at least; to hold their offices for a term sufficient to ensure their independency, to receive liberal stipends, by which they may be compensated for the devotion of their time to public service; and to be ineligible to any office established by a particular State, or under the authority of the United States, except those peculiarly belonging to the functions of the second branch, during the term of service, and for the space of after the expiration thereof.[4]
Certain phrases have been italicized to call attention to the Framers’ intent to make the two houses of Congress as different from each other as their common reliance on the principles of democratic republicanism—the absence of a hereditary aristocracy and the need for periodic or occasional elections—would allow.[5] Thus, the original proposal for the House of Representatives included provisions for mandatory rotation in office and recall. These were later removed, but the fundamental principle of dependence on the people and the expectation of a relatively frequently changing membership are evident here. By contrast, the original proposal for the Senate contains no limitation on a senator’s succeeding himself or any provision for recall. Other features of the Senate, though not stated in the resolution, seem to have been assumed: that it would be a smaller body than the House and that its members’ terms of office would be staggered in order to produce greater stability.
As originally proposed, the state legislatures would nominate senators, with the House of Representatives choosing from among the nominees. This was, of course, very soon simplified to election by the state legislatures.[6] This change is, I would suggest, less significant than it may at first appear, for the essence of this article is the phrase “”for a term sufficient to ensure their independency””—that is, their independence, once they are elected, from their electors, whoever those electors might be. In the original proposal, this meant independence from the House, upon whom the Senate was to function as a check, to which end a House membership that was constantly changing due to forced rotation would contribute. But even with the state legislatures doing the electing, the length of senatorial terms (first stipulated at seven years, later reduced to six in order to make for neater staggering of terms), in conjunction with the absence of recall, would necessarily render the senators independent of the state establishments during most of their term. This result would also follow from the common practice in the states of annual or biennial election of at least the lower legislative house, with consequently high rates of turnover in their membership. During the course of a single seven- or six-year senatorial term, the composition of the senator’s “”constituency”” might change between three and six times, thus diluting or even obliterating any obligation he might feel to the persons who had actually elected him, and making unknowable for most of his term the identities of those who would be deciding on his reelection. This principle of “”independency,”” and not the bare fact of election by the legislatures, is the true linchpin to the U.S. Senate’s originally intended constitutional role.
Most of the controversy at the Convention regarding the Senate focused, as is well-known, on the formula for representation that would apply to it—proportionality to population versus equality among the states. This was the issue over which the Convention nearly ran aground, and which was finally resolved by the Connecticut Compromise, which stipulated “”that in the 2d branch each State shall have an equal vote.”” This proposition was formally presented to the Convention on July 5 and was approved on July 16.[7] Interestingly, a week later, on July 23, a motion to have the senators vote per capita—that is, as individuals rather than as state delegations—was proposed, very briefly debated, and passed 9–1. (Voting at the Convention was by state delegations.) Only one delegate, Luther Martin of Maryland, spoke against the substance of the motion “”as departing from the idea of the States being represented in the 2d. branch.””[8] Even the Virginia Plan’s ardent opponents, even those who insisted on equality in the Senate as necessary to protect the small states, seem (with the notable exception of Mr. Martin) to have viewed this prospective body not as a House of States, but as a self-directed second chamber of a national legislature, each of whose members should be able to exercise his own judgment, will, voice, and vote.
The absence of a recall provision may also imply that the state legislatures could not instruct their senators how to vote, or, in any event, that they could not effectively and immediately enforce any attempt to instruct them. The propriety of constituents binding members of the House of Representatives was debated in some of the early Congresses, and the objection then raised against it—that the practice sacrificed deliberative representative government to the fragmentation of narrow, partial views—seems equally applicable to the binding of senators by their state legislatures.[9] Also of interest in this regard is a proposed Anti-Federalist constitutional amendment that the first session of the House of Representatives rejected in 1789, in the course of considering what became the Bill of Rights, that “”the election of Senators for each State shall be annual, and no person shall be capable of serving as a Senator more than five years in any term of six years.””[10] This proposal, which would have subordinated the U.S. Senate to the state legislatures and transformed it into a near replica of the Confederal Congress, helps us to appreciate, by means of contrast, the substantial independence from those bodies that the Senate, as constitutionally established, was meant to have.
Rossum presents an impressive list of instances—from 1789 through the early twentieth century—in which state legislatures issued instructions to their U.S. senators. Sometimes such instructions were followed, sometimes they were disregarded, and sometimes the “”delinquent”” senator resigned his office.[11] This phenomenon may more closely reflect the assumptions, first of the Anti-Federalists who were selected by some states, and later of triumphant Jeffersonian and Jacksonian democracy, than the implicit design of the Framers.[12] As the Supreme Court was much later to characterize the Senate’s role in the appointment of inferior executive officers,[13] the state legislatures’ role in populating the Senate was to be limited to the front end. They could determine who the office holder would be, but, once chosen, do little to control his conduct in office.[14]
III
There is a pervasive assumption in The Federalist Papers that the general government will, at least in the then-near future, not concern itself with the states’ internal affairs. Thus, Madison says, in Federalist 10, on the subject of representation:
It must be confessed that in this, as in most other cases, there is a mean, on both sides of which inconveniences will be found to lie. By enlarging too much the number of electors, you render the representative too little acquainted with all their local circumstances and lesser interests; as by reducing it too much, you render him unduly attached to these, and too little fit to comprehend and pursue great and national objects. The federal Constitution forms a happy combination in this respect; the great and aggregate interests being referred to the national, the local and particular to the State legislatures.[15]
This and similar passages express a broad expectation of how the proposed general government is likely to behave, but it places no special importance on the selection process for U.S. senators as the reason for this expectation. Rather, the assumption seems here to be that Congress will stay clear of “”local and particular”” interests because it is the sensible thing to do, and perhaps because the people—and hence their elected representatives—are predisposed to favor their respective state governments.[16]
Federalist 39 considers the balance that the Constitution strikes between national and federal (i.e., confederation) principles, and concludes that it is a complex “”composition”” of both. “”The Senate,”” Madison says, “”will derive its powers from the States, as political and coequal societies; and these will be represented on the principle of equality in the Senate, as they now are in the existing Congress [under the Articles of Confederation]. So far the government is federal, not national.””[17] This statement, however, expresses no particular anticipation that the senators will behave as guardians of state “”sovereignty.”” It calls attention to one of several features of the Constitution—others being its mode of adoption, the method of selecting representatives and the president, the direct (unmediated) operation of its laws upon the people, and the amendment process—some of them just as “”federal”” as the mode of selecting senators, which, taken together, highlight its complex character and its irreducibility to a single principle.
In the papers leading up to the thematic discussion of the Senate, Publius drops some tantalizing but ambiguous suggestions about its supposed character. In Federalist 58, while answering the objection that the proposed Congress will fail regularly to reapportion itself, he says:
There is a peculiarity in the federal Constitution which insures a watchful attention in a majority both of the people and of their representatives to a constitutional augmentation of the latter. The peculiarity lies in this, that one branch of the legislature is a representation of citizens, the other of the States: in the former, consequently, the larger States will have most weight; in the latter, the advantage will be in favor of the smaller States.[18]
While the Senate is here portrayed as representing the states, the focus is not on the protection of state autonomy, but on the equality of state representation which gives the smaller states a relative advantage in that body. A few pages later, comparing the likely character of the two houses, he observes
that in all cases the smaller the number, and the more permanent and conspicuous the station, of men in power, the stronger must be the interest which they will individually feel in whatever concerns the government. Those who represent the dignity of their country in the eyes of other nations will be particularly sensible to every prospect of public danger, or of dishonorable stagnation in public affairs.[19]
Again, the emphasis here is on other qualities of the Senate: its size, duration, and conspicuousness, especially as a national presence on the international stage.
In Federalist 59, while considering the hypothetical horror that some recalcitrant state legislatures might attempt to destroy the national government by declining to appoint senators, Publius says:
So far as that construction may expose the Union to the possibility of injury from the State legislatures, it is an evil; but it is an evil which could not have been avoided without excluding the States, in their political capacities, wholly from a place in the organization of the national government. If this had been done, it would doubtless have been interpreted into an entire dereliction of the federal principle; and would certainly have deprived the State governments of that absolute safeguard which they will enjoy under this provision.[20]
The phrase “”absolute safeguard”” is perhaps the strongest expression of the principle for which Rossum contends. But even here one should be mindful of the rhetorical context. The state power of appointment is susceptible to an obvious and undeniable abuse—refusal to cooperate—but this is a risk that must be tolerated, because the alternative, of lodging this power elsewhere, could readily lend itself to an even more unsettling “”interpretation.”” Moreover, it may be expedient, in order to discourage the states from neglecting their constitutional duty to select senators, to magnify, either actually or rhetorically, the stake that they stand to have in fulfilling that role. Finally, in No. 60, there is this observation:
In a country consisting chiefly of the cultivators of land, where the rules of an equal representation obtain, the landed interest must, upon the whole, preponderate in the government. As long as this interest prevails in most of the State legislatures, so long it must maintain a correspondent superiority in the national Senate, which will generally be a faithful copy of the majorities of those assemblies.[21]
The expectation expressed here is not that the Senate will defer to state autonomy, but that both it and the state legislatures will replicate the prevalent interests present in the constituent society. The overall effect of these several reflections is to create the anticipation that the relation of the senators to the state establishments that appoint them will be stated clearly and emphatically. This anticipation is strikingly disappointed in the immediate sequel.
The major discussion of the U.S. Senate’s functions occurs in Nos. 62 and 63. There, the connection between the Senate and the state legislatures that are to appoint them is stated with remarkable brevity and tepidness:
Among the various modes which might have been devised for constituting this branch of the government, that which has been proposed by the convention is probably the most congenial with the public opinion. It is recommended by the double advantage of favoring a select appointment, and of giving to the State governments such an agency in the formation of the federal government as must secure the authority of the former, and may form a convenient link between the two systems.[22]
Much more attention is paid to the other foci of these two essays: the senators’ qualifications; the states’ equality of representation; the number of senators and the length of their term of office; and the body’s powers. Much is said here of the need for a smaller, more stable,[23] second house of a national legislature, a body that will feel “”a due sense of national character,””[24] including a sensitivity to the opinions of other civilized nations. But there is no further topical treatment of the need to represent the states’ political establishments or to protect their distinctive interests as states via legislative appointment. One should always be cautious about interpreting a silence. It is, of course, possible that Madison fails to dwell on this function at this point because he regards it as trivially obvious or already established by the preceding argument. But given Publius’s general discursive inclination to be expansive and thorough, it is a most curious omission.
#page#
This pattern of minimizing or disconnecting the Senate’s indirect mode of election from its supposed role as special protector of state sovereignty is visible in the sequel argument as well. Thus, in No. 62, Madison remarks that “”the equal vote allowed to each State is at once a constitutional recognition of the portion of sovereignty remaining in the individual States, and an instrument for preserving that residuary sovereignty.””[25] John Jay, in his single paper on the treaty-making power, says of the appointment of the president and the Senate by “”select assemblies””:
This mode has . . . vastly the advantage of elections by the people in their collective capacity, where the activity of party zeal, taking advantage of the supineness, the ignorance, and the hopes and fears of the unwary and interested, often places men in office by the votes of a small proportion of the electors. . . . The inference which naturally results from these considerations is this, that the President and senators so chosen will always be of the number of those who best understand our national interests, whether considered in relation to the several States or to foreign nations, who are best able to promote those interests, and whose reputation for integrity inspires and merits confidence.[26]
And, in the discussion of the impeachment process, both houses of Congress are comprehended in Publius’s rhetorical questions: “”Is it not designed as a method of NATIONAL INQUEST into the conduct of public men? If this be the design of it, who can so properly be the inquisitors for the nation as the representatives of the nation themselves?””[27]
The Federalist‘s treatment of the Senate’s mode of election is, in sum, a nuanced and highly rhetorical exercise in reshaping and redirecting an easy but misguided first impression. In this respect, it may be distinguishable from other, less subtle, participants in the ratification debates whom Rossum cites, who, perhaps naively, took the nexus between the senators’ mode of election and expected conduct more at face value.[28] But on this score, Publius may have been too subtle by half. Insofar as some senators, especially in the early nineteenth century, may actually have regarded themselves as morally bound to follow the directions of their “”constituent”” state legislatures, to that very extent the constitutional reliance on the structural features designed to secure senatorial “”independency”” were trumped by the political spirit of deference to the states or by the ordinary virtue of gratitude. Contrary to Rossum’s contention, what this phenomenon betokens is not the primacy of structural controls, but the vindication of Tocqueville’s observation about the limits of formal rules in the face of a divergent set of social norms.[29]
IV
In the fourth chapter, Rossum offers in support of his thesis three case studies of important legislation adopted between 1789 and 1791 by the First Congress, in which sat several men who had either attended the Constitutional Convention or been active in the ratification process: the adoption of the Bill of Rights, the Judiciary Act of 1789, and the charter of the first Bank of the United States.
Of these three, only the first clearly offers a prima facie example of the Senate acting as a guardian of state rights. The House of Representatives had approved and transmitted for the Senate’s consideration a list of seventeen articles of amendment. The Senate reworded some of these articles, consolidated some others, and rejected two outright, resulting in a list of twelve that were sent out for ratification by the state legislatures. Ten, comprising our familiar Bill of Rights, were ratified and added to the Constitution by December 1791.[30] One of the two articles that the Senate killed read as follows: “”The equal rights of conscience, the freedom of speech or of the press, and the right of trial by jury in criminal cases, shall not be infringed by any State.””[31] Representative James Madison had initially moved a slightly differently worded version of this proposal, and regarded it as “”the most valuable amendment in the whole list.””[32] Because the Senate at that time met in executive session, we have no record of their debates, and cannot know for certain why they acted as they did. But it seems plausible that they shared the sentiment of Rep. Henry St. George Tucker:
This is offered, I presume, as an amendment to the Constitution of the United States, but it goes only to the alteration of the constitutions of particular States. It will be much better, I apprehend, to leave the State Governments to themselves, and not to interfere with them more than we already do.[33]
Here, arguably, the Senate was acting, conformably to Rossum’s thesis, to preserve state autonomy.[34]
The other two examples are more ambiguous. The Judiciary Act of 1789 was, Rossum observes, “”crafted almost exclusively by the Senate,”” and subsequently approved overwhelmingly by the House without material alterations.[35] It performed the necessary function of filling the gap left by the Constitution regarding the detailed structure and procedures of the federal judiciary, and it did so, Rossum reports, in a way that satisfied Anti-Federalist Senator Richard Henry Lee, and that left Madison, who on this issue was an archnationalist, “”infuriated.””[36] The Act seems to be a careful and sensible piece of political compromise. On the one hand, it tilted in a nationalist direction by establishing a system of federal courts, including trial courts of first resort, below the U.S. Supreme Court, rather than leaving it entirely to the state courts to apply in the first instance acts of Congress and treaties to which the United States was a party. On the other hand, it placed a number of restrictions on the jurisdiction of the federal district and circuit courts, and it recognized “”the laws of the several States”” as “”rules of decision in trials at common law in the courts of the United States.””[37] Rossum offers a set of technical reasons for this outcome: the Senate opted to establish federal trial courts at all because the desirability of having such courts to apply admiralty law was universally acknowledged. But once the existence of such courts was conceded for this purpose, the Anti-Federalist objection to federal courts on the basis of their expense was no longer available to block the extension of their jurisdiction to some other areas of law. And the Virginia legislature had prohibited its courts from hearing cases that might arise under acts of the U.S. Congress. If other states followed this harmful precedent, federal laws might fail altogether to receive judicial enforcement.[38] But the premise of this line of reasoning is that there are some matters for which state judges, whether because they are unschooled or because they are susceptible to parochial bias, are not to be trusted. The existence of federal courts for these matters, while obviating the objection from expense, would in no way compel the addition of any other matters to federal jurisdiction. And the example of Virginia exposed the paramount importance of enforcing federal law, even if it meant defying state legislative judgments. The Senate, that is, arguably acted here primarily and emphatically as part of a sovereign national legislature. The further fact that the directly elected House, in which the Federalists predominated, made no attempt to materially change the Senate bill suggests that the package aptly reflected popular opinion as much as it might have reflected special solicitude for states’ rights.
By the time the Bank Act of 1791 came along, battle lines had shifted somewhat. Most notably, Madison was now arguing an anti-nationalist, strict constructionist line. One cannot shed the suspicion that constitutional arguments on both sides had become, at least in part, a façade for the pursuit of opposed sectional interests. In the course of presenting an account of the Act’s legislative history, Rossum apparently succumbs to circular reasoning. He quotes statements by Representatives William Loughton Smith and Elbridge Gerry in support of the bank bill, which had at that point already passed the Senate. First, Smith:
It would be a deplorable thing if . . . so enlightened a body as the Senate of the United States should, by so great a majority as were in favor of this bill, pass a law hostile to the liberties of this country, as the opposition to this measure have suggested the bank system to be.[39]
Next, Gerry:
The interpretation of the constitution, like the prerogative of a sovereign, may be abused, but from hence the disuse of either cannot be inferred. In the exercise of prerogative, the minister is responsible for his advice to his sovereign, and the members of either House are responsible to their constituents for their conduct in construing the constitution. We act at our peril: if our conduct is directed to the attainment of the great objects of Government, it will be approved.[40]
And Rossum’s gloss on both speeches:
Smith’s argument showed keen insight. The Senate, whose mode of election ensured the protection of the interests of the states as states, did not regard its reliance on implied powers to pass the Bank Act as a threat to the “”residuary sovereignty”” of the states. . . . The state legislatures were, of course, the constituents of the Senate, and, as Gerry made clear, it was for them to judge whether the Senate, through its use of implied powers, was serving “”the great objects of Government”” or jeopardizing the original federal design.[41]
But neither speech refers particularly to the Senate’s mode of election. Smith rather calls attention to that body’s general enlightenment, and Gerry’s reference to “”their constituents”” could just as well denote popular opinion, as filtered through the state legislatures, as the legislators personifying the states in their corporate capacity. The example that is supposed to demonstrate corporate representation ends up assuming its own conclusion.
What these actions of the First Congress, taken as a whole, seem to display is a pattern of sensible senatorial accommodation to state interests, and probably to the popular opinion of the day, but not subservience. Other than the rejected provision of the proposed Bill of Rights, no examples are given of the Senate blocking House-approved legislation in order to protect the states.[42] Prior to the Civil War, the Senate functioned as a vehicle to maintain the precarious regional balance between free and slave states. Viewed through the lens of this issue, the protection of state legislative autonomy became a subordinate matter. Whether, on the one hand, a senator favored expansive use of Congress’s power to regulate foreign and interstate commerce in order to discourage traffic in slaves or, on the other, to enforce the fugitive slave clause depended more on the degree of his region’s attachment to or abhorrence of “”the peculiar institution”” than on any consistent judgment regarding state rights. In the end, it was the still indirectly elected Senate of the Reconstruction Congresses that recommended the Civil War amendments, with their substantial explicit and implied inroads on state autonomy in the spheres of property in slaves, civil rights, and black suffrage.
V
Rossum concludes his account of the Founding Era by examining three Supreme Court opinions of Chief Justice John Marshall. Two of them, McCulloch v. Maryland and Gibbons v. Ogden, are well-known landmark cases. The third (chronologically the first), United States v. Fisher, is largely overlooked but, Rossum convincingly suggests, deserves more attention than it has generally received.[43]
Fisher deals with a challenge to a provision of the bankruptcy statute that Congress enacted in 1797, which gave priority to the claims of the United States government against the estate of any deceased debtor who was indebted to the United States at the time of his death. Rossum calls attention to a particular passage in Marshall’s opinion for the Court:
Addressing the complaint that, by giving priority to the claims of the United States, the law will “”interfere with the right of the state sovereignties respecting the dignity of debts, and will defeat the measures they have adopted to secure themselves against the delinquencies on the part of their own revenue officers,”” Marshall bluntly asserted: “”[T]his is an objection to the Constitution itself. The mischief suggested, so far as it can really happen, is the necessary consequence of the supremacy of the laws of the United States on all subjects to which the legislative power of Congress extends.””[44]
Focusing on the phrase that I have emphasized, he then continues:
Marshall did not believe that “”mischief”” would result from giving the Congress free rein, for he implicitly trusted the Senate not to approve of such “”mischief.”” The senators were, he would later say, “”the representatives of the state sovereignties.”” Given the mode of electing senators, Marshall could confidently construe the Necessary and Proper Clause as conferring on Congress “”any means which are in fact conducive to the exercise of a power granted by the Constitution”” without worrying that this power would be abused.[45]
While Marshall’s language in Fisher is consistent with this explanation, it does not require it. Equally consistent is the supposition that Congress will be populated by prudent men, however they may happen to be chosen, who will be vigilant to prevent or to correct abuses of authority by officers of the United States.[46] Moreover, given that the defining characteristic of most of Marshall’s landmark cases is the quashing of state claims that clash with the Constitution or acts of Congress,[47] one should probably take with a box of salt his profession of doubt that the mischief of conflicting federal and state claims on debtors’ estates would really happen. Indeed, the phrase “”so far as it can really happen”” can be read as an expression of measured expectation.
Rossum’s accounts of McCulloch and Gibbons are generally clear and correct,[48] though sometimes overstated. Apropos of McCulloch, Rossum summarizes the Court’s interpretation of Congress’s Article I, Section 8 powers as follows: “”What Congress can do under its enumerated powers—i.e., what powers are delegated to it as opposed to reserved to the states—is a question for Congress alone to decide.””[49] This, despite Marshall’s perhaps grandiose prefatory paragraph:
In the case now to be determined, the defendant, a sovereign state, denies the obligation of a law enacted by the legislature of the Union, and the plaintiff, on his part, contests the validity of an act which has been passed by the legislature of that state. The constitution of our country, in its most interesting and vital parts, is to be considered; the conflicting powers of the government of the Union and of its members, as marked in that constitution, are to be discussed; and an opinion given, which may essentially influence the great operations of the government. No tribunal can approach such a question without a deep sense of its importance, and of the awful responsibility involved in its decision. But it must be decided peacefully, or remain a source of hostile legislation, perhaps of hostility of a still more serious nature; and if is to be so decided, by this tribunal alone can the decision be made. On the Supreme Court of the United States has the constitution of our country devolved this important duty.[50]
Again, although the famous rule of interpretation that the Court lays down is quite generous toward Congress, it does not totally renounce judicial limitation: “”Let the end be legitimate, let it be within the scope of the constitution, and all means which are appropriate, which are plainly adapted to that end, which are not prohibited, but consist with the letter and spirit of the constitution, are constitutional.””[51] Here are some six or seven grounds on the basis of which the Court could, however rarely, strike down an act of Congress.[52] On the particular matter of the senators functioning as agents of the state legislatures, Rossum himself records that “”several states, resentful of the federal presence it symbolized and the stiff competition it provided, took advantage of the public’s general hostility to the Bank and passed measures designed to regulate or prohibit its operations within their borders.””[53] And while the Court in Gibbons seems to allow Congress total discretion over which laws it may enact to regulate interstate commerce,[54] this discretion is predicated on a presumably justiciable determination that the particular law is indeed a regulation of interstate commerce.
In sum, what Rossum provides, in his review of early Congressional enactments and Supreme Court cases, is not a proof of the proposition that the election of U.S. senators by the state legislatures created a vital dependence of the former on the latter. Rather, he demonstrates, with only partial success, that assuming such a dependency may help to clarify some otherwise murky and ambiguous language in certain congressional speeches and Court opinions. And perhaps this is all he intends.[55]
VI
Apart from the aforementioned list of instances in which state legislatures attempted, apparently with mixed success, to instruct their senators on how to vote, Rossum provides no comprehensive account of the actual conduct of the Senate during the nineteenth century on the basis of which one could gauge the degree to which protecting state autonomy was an actual and ongoing concern.[56] This is an important gap in the narrative, which we can try to fill by consulting the comments of some notable observers of American politics of that extended era.
Alexis de Tocqueville thought the principal institutional expression of the idea of state sovereignty to be the equal representation of the states in the U.S. Senate, but by 1835 he regarded this safeguard, which many of the Founders considered so essential, as a historical vestige:
All the states are young; they are close to one another; they have homogeneous mores, ideas, and needs; the difference that results from their greater or lesser size is not enough to give them very opposed interests. Therefore the small states have never been seen to join forces in the Senate against the designs of the great. Besides, there is such an irresistible force in the legal expression of the will of a whole people that when the majority comes to express itself through the organ of the House of Representatives, the Senate finds itself very weak in its presence.[57]
Turning his attention to the indirect mode of electing senators, Tocqueville calls it immensely significant, but not for the purpose of representing state interests. His vivid description of the U.S. Congress’s bicameralism merits quotation in full:
When you enter the House of Representatives in Washington, you feel yourself struck by the vulgar aspect of this great assembly. Often the eye seeks in vain for a celebrated man within it. Almost all its members are obscure persons, whose name furnishes no image to one’s thought. They are, for the most part, village attorneys, those in trade, or even men belonging to the lowest classes. In a country where instruction is almost universally widespread, it is said that the people’s representatives do not always know how to write correctly.
Two steps away is the chamber of the Senate, whose narrow precincts enclose a large portion of the celebrities of America. One perceives hardly a single man there who does not recall the idea of a recent illustrious [deed]. They are eloquent attorneys, distinguished generals, skillful magistrates, or well-known statesmen. All the words that issue from this assembly would do honor to the greatest parliamentary debates of Europe.
Whence this peculiar contrast? Why is the elite of the nation found in this chamber rather than in the other? Why are so many vulgar elements gathered in the first assembly when the second seems to have the monopoly on talents and enlightenment? Both nevertheless emanate from the people, both are the product of universal suffrage, and up to now no voice has been raised in America to assert that the Senate is the enemy of popular interest. Where, therefore, does such an enormous difference come from? I see only a single fact that explains it: the election that produces the House of Representatives is direct; that from which the Senate emanates is subject to two stages. The universality of citizens names the legislature of each state, and the federal constitution, transforming each of these legislatures in their turn into an electoral body, draws the members of the Senate from them. Therefore the senators express, however indirectly, the result of universal suffrage; for the legislature that names the senators is not an aristocratic or privileged body that draws its electoral right from itself; it depends essentially on the universality of citizens; it is generally elected by them every year, and they can always direct its choices by filling it with new members. But it suffices that the popular will pass through this chosen assembly for it to be worked over in some way, and it comes out reclothed in more noble and more beautiful forms. The men so elected, therefore, always represent exactly the majority of the nation that governs; but they represent only the elevated thoughts that are current in the midst of it, the generous instincts that animate it, and not the small passions that often agitate it and the vices that dishonor it.[58]
On this basis, Tocqueville recommends and predicts more extensive use of two-stage elections as “”the sole means of putting the use of political freedom within the reach of all classes of the people,”” without foundering “”on the shoals of democracy.””[59] And later still, when he takes up the growing sectional controversies of the 1830s, he mentions, as one would have to, not the indirect election of senators, but Calhoun’s and others’ championing of the doctrine of nullification as the proffered instrument of protecting states’ rights.[60]
A quarter-century later, John Stuart Mill debunked the general idea of two-stage elections. Taking his cues from the electoral college, as it had evolved by the mid-nineteenth century, he thought the likely outcome of such systems would be the reduction of the special electors to mere partisan ciphers, who would be chosen, not to exercise their own judgment, but to vote for one or another known candidate to whom they were pledged in advance. Alternatively, if the “”the primary electors”” (i.e., the ordinary voters) were to choose special electors without concerning themselves about who the ultimate officeholder will be, then
one of the principal uses of giving them a vote at all is defeated: the political function to which they are called fails of developing public spirit and political intelligence; of making public affairs an object of interest to their feelings and of exercise to their faculties. The supposition, moreover, involves inconsistent conditions; for if the voter feels no interest in the final result, how or why can he be expected to feel any in the process which leads to it? To wish to have a particular individual for his representative in Parliament, is possible to a person of a very moderate degree of virtue and intelligence; and to wish to choose an elector who will elect that individual, is a natural consequence: but for a person who does not care who is elected, or feels bound to put that consideration in abeyance, to take any interest whatever in merely naming the worthiest person to elect another according to his own judgement, implies a zeal for what is right in the abstract, an habitual principle of duty for the sake of duty, which is possible only to persons of a rather high grade of cultivation, who, by the very possession of it, show that they may be, and deserve to be, trusted with political power in a more direct shape. Of all public functions which it is possible to confer on the poorer members of the community, this surely is the least calculated to kindle their feelings, and holds out least natural inducement to care for it, other than a virtuous determination to discharge conscientiously whatever duty one has to perform: and if the mass of electors cared enough about political affairs to set any value on so limited a participation in them, they would not be likely to be satisfied without one much more extensive.[61]
The great exception to this rule is the U.S. Senate, for which Mill offers praise and an analysis comparable to Tocqueville’s, but with a significant additional wrinkle:
The case in which election by two stages answers well in practice, is when the electors are not chosen solely as electors, but have other important functions to discharge, which precludes their being selected solely as delegates to give a particular vote. This combination of circumstances exemplifies itself in another American institution, the Senate of the United States. That assembly, the Upper House, as it were, of Congress, is considered to represent not the people directly, but the States as such, and to be the guardian of that portion of their sovereign rights which they have not alienated. As the internal sovereignty of each State is, by the nature of an equal federation, equally sacred whatever be the size or importance of the State, each returns to the Senate the same number of members (two), whether it be little Delaware, or the ‘Empire State’ of New York. These members are not chosen by the population, but by the State Legislatures, themselves elected by the people of each State; but as the whole ordinary business of a legislative assembly, internal legislation and the control of the executive, devolves upon these bodies, they are elected with a view to those objects more than to the other; and in naming two persons to represent the State in the Federal Senate, they for the most part exercise their own judgement, with only that general reference to public opinion necessary in all acts of the government of a democracy. The elections, thus made, have proved eminently successful, and are conspicuously the best of all the elections in the United States, the Senate invariably consisting of the most distinguished men among those who have made themselves sufficiently known in public life. After such an example, it cannot be said that indirect popular election is never advantageous. Under certain conditions, it is the very best system that can be adopted. But those conditions are hardly to be obtained in practice, except in a federal government like that of the United States, where the election can be entrusted to local bodies whose other functions extend to the most important concerns of the nation.[62]
Like Tocqueville, Mill points to the equality of the states in the Senate as the safeguard of state sovereignty, and the high quality of the senators as the important consequence of indirect election.[63] An ironic implication of Mill’s argument is that insofar as national issues with which the Senate needs to deal may come to overshadow state and local issues, to that very extent will the ordinary functions of the state legislatures become obscured by their electoral function, and the legislators themselves become electoral ciphers, like their Electoral College counterparts—to the detriment of state self-governance![64]
#page#
A quarter-century later still, in 1888, Dr. (later Lord) James Bryce published the first edition of The American Commonwealth, a minutely descriptive work, of prodigious length, on American politics and society, which went through several revisions over the next three decades. (The 1914 edition came out shortly after the Seventeenth Amendment was adopted.) Bryce too offered similar, though not always consistent, observations about the U.S. Senate.
Perhaps because he writes primarily for a British audience, Bryce initially emphasizes the American system’s federal features,[65] noting at one point that the early nineteenth-century senators, “”regarding themselves as a sort of congress of ambassadors from their respective states, were accustomed to refer for advice and instructions each to his state legislature.””[66] But the context of this remark makes clear that he regards this datum as of essentially historical interest:
The traditions of the old Congress of the Confederation, in which the delegates of the states voted by states, the still earlier traditions of the executive councils, which advised the governors of the colonies while still subject to the British Crown, clung about the Senate and affected the minds of the senators. . . . So late as 1828, a Senator after arguing strongly against a measure declared that he would nevertheless vote for it, because he believed his state to be in its favour.[67]
And he adds the following significant footnote, “”A similar statement was made in 1883 by a senator from Arkansas in justifying his vote for a bill he disapproved. But the fact that from early days downwards the two senators from a state might (and did) vote against one another shows that the true view of the senator is that he represents the people and not the government of his state.””[68] Again, when contrasting the Senate’s relation to the president with that of the seventeenth-century Privy Council to the monarch, he observes generally that “”there is all the difference in the world between being advised by those whom you have yourself chosen and those whom election by others forces upon you.””[69] But he does not take this potentially promising occasion to assert the Senate’s control by or accountability to the state legislatures.
Bryce’s thematic discussion of the Senate offers several relevant and interesting points: Indirect election “”has helped to make the national parties powerful, and their strife intense, in [the state legislatures]. Every vote in the Senate was so important to the great parties that they are forced to struggle for ascendency [sic] in each of the state legislatures by whom the senators were elected.””[70] The newly adopted (as of the 1914 edition) Seventeenth Amendment “”may add immensely to the expense falling on candidates, as well as to the labour thrown on them in stumping the state; and if it causes senators to be less frequently reelected at the end of their term, it will reduce the element of long political experience heretofore present in it more largely than in the House.””[71] The fact that senators vote as individuals means that “”the vote a senator gives is his own and not that of his state.””[72] Most significant, perhaps, for Rossum’s thesis, “”As the state legislatures sit for short terms (the larger of the two houses usually for two years only), a senator has during the greater part of his six years’ term to look for reelection not to the present but to a future state legislature, and this circumstance tends to give him somewhat more independence.””[73] Further, the staggering of senatorial terms produces a stabilizing effect, both on policy and on the ethos of the body itself, into which new members are assimilated, and thereby implicitly detached from their “”constituents””:
It is an undying body, with an existence continuous since its first creation; and though it changes, it does not change all at once, as do assemblies created by a singular popular election, but undergoes an unceasing process of gradual renewal, like a lake into which streams bring fresh water to replace that which the issuing river carries out. . . . An incidental and more valuable result [than enabling it to conduct or control foreign policy] has been the creation of a set of traditions and a corporate spirit which have tended to form habits of dignity and self-respect. The new senators, being only one-third, or less, are readily assimilated.[74]
Finally, under the heading of the relations between the two Houses, Bryce presents the following observation:
Questions relating to states’ rights and the greater or less extension of the powers of the national government have played a leading part in the history of the Union. But although small states might be supposed to be specially zealous for states’ rights, the tendency to uphold them has been no stronger in the Senate than in the House. In one phase of the slavery struggle the Senate happened to be under the control of the slaveholders while the House was not; and then of course the Senate championed the sovereignty of the states. But this attitude was purely accidental, and disappeared with its transitory cause.[75]
VII
Concerning the development of proposals between 1826 and 1913 for what eventually became the Seventeenth Amendment, Rossum presents a meticulous account that is likely to make his book the standard reference source on this subject. The overall picture he paints is of a Congress and a country that became progressively overwhelmed by other concerns that eclipsed and obscured virtually all regard for federalism. Prominent among these concerns were the occasional inconvenience of politically deadlocked state legislatures failing to elect senators, the exaggerated perception that the indirect election process was shot through with bribery and corruption, and, most importantly, the irresistible imperative to install democracy wherever possible.
The first of these concerns was arguably a serious technical problem that could have been adequately addressed by an equally technical solution. The culprit here was an 1866 federal law that required a state’s senators to be elected by a majority of the total number of state legislators, either by concurrent vote of each house or, that failing, by a joint vote of the two houses combined into a single body. This requirement of an absolute majority produced legislative deadlocks, owing to divided party control of the state legislatures or other causes, seventy-one times between 1885 and 1912.[76] A law that allowed election by plurality vote, either in the first instance or after a specified number of ballots had failed to produce an absolute majority, would have provided a sufficient remedy. But limited, technical solutions could not satisfy the totalistic ideological demand for direct democracy. Similarly, Rossum, citing Todd Zywicki, maintains that only a tiny fraction of senatorial elections were demonstrably tainted by bribery and corruption.[77] Related to this charge was the populist characterization of the Senate as an aristocratic “”millionaires’ club,”” with the implication that admission could be purchased through bribery.[78]
Under the sway of the populist ideology of class conflict and the Progressive belief in “”the redemptive powers of direct democracy,””[79] some states, beginning with South Carolina in 1888, adopted one or another version of “”advisory”” primary elections, with the intent of binding the state legislatures to popularly preferred senatorial candidates. Rossum notes that “”by 1912, thirty-three states had introduced the use of direct primaries, and twelve states had adopted some form of . . . the ‘Oregon System.’ “” Under this system, an “”advisory”” general election was held between the major parties’ primary nominees, and candidates for the state legislature would be expected either to vote for the winner or to declare publicly, as part of their own election campaigns, that they did not consider themselves bound by the general election results.[80] But Rossum underestimates the significance of this datum to his general thesis. The U.S. Senate that finally acquiesced in the proposal that became the Seventeenth Amendment was already a substantially democratized body. That is, the amendment, rather than constituting a sharp break with the federal principle, was the culmination of a gradual process of drifting away from it that had by then been in the works for a quarter-century.
Congressional consideration of direct election proposals during this period was generally responsive to growing popular opinion, including the opinion of an ever increasing number of state legislatures. It was also, as Rossum acknowledges, largely devoid of serious discussion of the federalism issue to which he attaches paramount importance. He finds only three exceptions in “”the voluminous record””: Rep. Franklin Bartlett and Sen. George F. Hoar in the Fifty-third Congress (1893) and Sen. Elihu Root in the Sixty-first Congress (1911).[81] (The case of Senator Hoar is equivocal. Rossum cites both a floor speech, in which Hoar expressly invoked the federalism rationale for indirect election, and his report on behalf of the Senate Committee of Privileges and Elections, a laundry list of objections to the proposed constitutional amendment which does not even mention state autonomy.[82]) One particularly telling index of this low level of concern is the repeated attempt by Southern members of successive Congresses to link direct election to a proposal to give the states total control over the manner of conducting senatorial elections—that is, to remove Congress’s constitutional power to “”make or alter such [State] Regulations.””[83] As several Northern Congressmen noted, this stipulation would give the Southern states carte blanche to thwart black registration and voting. What states’ rights meant for these Southerners was not general legislative autonomy, safeguarded by an appointed Senate, but the states’ ability to set racially exclusionary voting standards.[84]
VIII
Rossum concludes his substantive argument by reviewing the development of a number of constitutional doctrines, including that of dual federalism, as judicially constructed substitutes for the superseded institutional safeguard for state autonomy that indirect election of the Senate had supposedly provided. The rhetorical thrust of this chapter, as with his opening chapter, is a critique of some of the Supreme Court’s recent decisions. But his generally correct overview of this familiar story[85] is colored by two exaggerated assumptions.
The first such assumption is a kind of constitutional application of Ockham’s Razor, which excludes redundant remedies. That is, because the Senate, as originally constructed, was intended to be the institutional guardian of states’ rights, the federal judiciary, enforcing constitutional limitations on the delegated powers of Congress, could not also have been intended to perform this function. This conclusion, of course, does not follow. One has only here to recall both Marshall’s language at the start of his McCulloch opinion, quoted above,[86] and that Hamilton’s classic explanation of judicial review presents it as an antidote, not only to “”serious oppressions of the minor party in the community,”” but also to “”dangerous innovations in the government.””[87]
The other assumption has more to do with the attitude toward the judiciary of the Reconstruction Congress that drafted the text of the Fourteenth Amendment. Like Raoul Berger a generation ago, Rossum reads Section 5 of the amendment, which vests in Congress “”power to enforce, by appropriate legislation, the provisions of this article,”” as implicitly discountenancing, if not outright precluding, judicial enforcement.[88] To be sure, the Supreme Court’s prestige had suffered a severe blow as a result of the discredited Dred Scott decision,[89] and some language in an important post–Reconstruction Era case, Ex parte Virginia, suggests that, but for Section 5, “”there might be room for argument that the [Fourteenth Amendment’s] first section is only declaratory of the moral duty of the States.””[90] On the other hand, the Supreme Court had, by 1867–68, when the Fourteenth Amendment was drafted and approved, been substantially reconstituted by President Lincoln’s appointment of five new justices. As to the issue of the amendment’s justiciability, I offer the following observation, which I had occasion to make some years ago, regarding a “”crucial change”” that took place in what was destined to become Section 1:
Representative Bingham’s original proposal read, “”The Congress shall have power to make all laws which shall be necessary and proper to secure to the citizens of each State all privileges and immunities of citizens in the several States, and to all persons in the several States equal protection in the rights of life, liberty, and property.”” When this provision was discussed in Congress, the objection was voiced that whatever this Republican Congress might do pursuant to such authorization a future Democratic Congress might undo. Bingham’s empowering text was thereupon referred to committee, where it was changed to the current language, which simply declares the inviolable existence of privileges and immunities of national citizenship. This change had the effect of placing the civil rights which the Thirty-ninth Congress wished to affirm beyond the reach of a future hostile majority by rendering them “”self-enforcing.”” That is, by giving these privileges and immunities Constitutional status, Congress gave the task of enforcing them to the courts, no matter who might happen to control the Congress at any given time.[91]
While it is relatively easy to pinpoint the (at least temporary) demise of dual federalism,[92] locating its beginning is a bit more elusive. Rossum seems to suggest that it and related constitutional doctrines came, at least implicitly, in response to the Seventeenth Amendment:
