You are using an outdated browser.
Please upgrade your browser
and improve your visit to our site.
Skip Navigation

It Could Be Worse

Our Undemocratic Constitution: Where The Constitution Goes Wrong (And How We The People Can Correct It)
By Sanford Levinson
(Oxford University Press, 233 pp., $28)

Though the United States iis a young country, Americans are prone to ancestor worship. We learn not merely to admire the Founding Fathers but also to revere the Founding Fathers, who are sometimes described as if they had been touched directly by God. We question our governors, including our judges, but we rarely question our founding document. We feel free to ridicule or to despise the decisions of the Supreme Court, but not those of the Founders (putting slavery to one side).

Toward the end of his life, Thomas Jefferson expressed some exasperation about this state of affairs. He complained of those who "look at constitutions with sanctimonious reverence, and deem them like the arc of the covenant, too sacred to be touched." He noticed with evident alarm that people were ascribing "to the men of the preceding age a wisdom more than human," and seeing "what they did to be beyond amendment." He proclaimed that the founding generation "was very like the present, but without the experience of the present." Insisting that the Constitution should be rethought every generation, Jefferson offered a plea to posterity: "Let us not weakly believe that one generation is not as capable as another of taking care of itself…. The dead have no rights."

James Madison had a fundamentally different view. Fearing the effects of popular passions, and insisting that the Constitution had been adopted under unusually favorable historical circumstances, Madison wanted the founding document to be a kind of fixed star, immunized from the vicissitudes of history and the pressures of endless public scrutiny. In his words, "frequent appeals to the public" would remove "that veneration which time bestows on every thing, and without which perhaps the wisest and freest governments would not possess the requisite stability." To be sure, Madison accepted the amendment process set out in Article V of the Constitution, which allows the Constitution to be changed after the approval of three-fourths of the states. But constitutional amendments are extremely difficult to produce; they are deliberately reserved, in Madison's own words, for "great and extraordinary occasions."

Over the course of American history, Madison's view has prevailed and Jefferson's has mostly dropped out of sight. Two years after the ratification of the Constitution, the document was in a sense completed through the addition of the ten amendments that constitute the Bill of Rights. Since then, the Constitution has been amended just seventeen times. The most important changes followed the Civil War, when the abolition of slavery was accompanied by a new set of rights and a significant transfer of authority from the states to the national government. During Franklin Delano Roosevelt's New Deal, the nation's understanding of both rights and institutions was again altered--but without the slightest change in the founding document. For more than two centuries, then, the basic design of the Constitution has proved remarkably resilient to any kind of major rethinking.

SANFORD LEVINSON TAKES JEFFERSON'S side against Madison. More particularly, Levinson believes that the Constitution is fundamentally undemocratic, and that a new constitutional convention is necessary to bring forward a better charter. Unlike most of those who have explored the question of constitutional change, Levinson focuses exclusively on the document's structural provisions, and does not investigate its provisions governing individual rights. Nor does he want constitutional change to occur through the existing provisions for amendment--provisions that are, in his view, hopelessly undemocratic and hence part of the problem, not the solution. In calling for a constitutional convention, Levinson wants the nation to reconsider the document not piecemeal, but as a whole. He offers multiple objections to the Framers' design of the three branches of government; he hopes that the sheer accumulation of objections will make the case for large-scale reconsideration.

Levinson's first complaint involves the legislature, and above all the composition of the Senate. Under the Constitution, every state, regardless of population, receives two senators for a period of six years. This is a conspicuous violation of the rule of "one person, one vote." Wyoming, with about 500,000 people, has the same number of senators as California, with about 35 million people. Nearly one-quarter of the Senate comes from states whose total population is merely 14 million--less than 5 percent of the nation. As of now, there are 55 Republican senators, 44 Democratic senators, and one independent (Vermont's Jim Jeffords). Stunningly, the Democratic minority actually represents more voters. In 2004, Barbara Boxer received more than six million votes in California, allowing her to retain her seat, and John Thune received less than 198,000 votes in South Dakota, allowing him to defeat Tom Daschle. And remarkably, the equal representation of the states in the Senate is the only provision of the Constitution that is made effectively unamendable by the document itself. (States can be deprived of equal representation in the Senate only with their consent.)

Does this matter? After all, the Constitution essentially preserves the idea of "one person, one vote" in the House of Representatives, where states receive representatives in accordance with their populations. But Levinson notes that under the bicameral system, the Senate is able to stop legislation. A large national majority determined to enact legislation might therefore be blocked by a small minority in the Senate. More concretely, Levinson claims that the composition of the Senate helps to explain the fact that the national government often, and perversely, redistributes resources from high-population states to low-population states. Consider the fact that in terms of per capita funding for homeland security, the citizens of Wyoming receive no less than seven times the amount received by the citizens of New York. (Do the citizens of Wyoming actually have more to fear from terrorists?) The composition of the Senate also ensures that racial minorities end up with a disproportionately weak voice, since small states tend to have unusually small nonwhite populations. All this, in Levinson's view, counts as an indefensible departure from democratic ideals.

More generally, Levinson objects to the whole system of bicameralism on the ground that it gives rise to so many "veto points," allowing the democratic will to be thwarted. By its very nature, bicameralism makes it harder to enact legislation, simply because it allows measures to be defeated whenever one house is unwilling to approve of them. And even if we endorse bicameralism, must we accept the president's veto power, which has now become a crucial part of the law-making process? Levinson notes that at the founding, and for decades thereafter, the president exercised the veto power rarely, and only on the ground that he believed the relevant legislation to be unconstitutional. In the modern period, the president feels free to veto legislation simply because he dislikes it as a matter of policy. The United States has now had more than 2,500 presidential vetoes. Even that number understates the importance of the veto power, because its mere existence affects the legislation that actually is enacted.

The Constitution's provisions governing the legislature have another problem: they do not make adequate provision for emergency situations. What happens if a terrorist attack kills many members of the House of Representatives? The answer is that under the Constitution, their seats must be filled by a new election. But it takes a lot of time to do that. Levinson fears that without a constitutional contingency plan, the result is likely to be "centralization of basically dictatorial power in the president, if we are lucky."

LEVINSON BELIEVES THAT ARTICLE II, governing the executive branch, is also seriously defective. The initial difficulty is the electoral college, which makes it possible for the president to be selected without being the choice of a popular majority. It is widely known that in 2000 Al Gore received more votes than George W. Bush. But no fewer than nineteen presidents--nearly half!--have captured strong majorities in the electoral college despite failing to obtain majorities of the popular vote. In 1992, Bill Clinton was the choice of only 43 percent of voters; Richard Nixon did no better in 1968; Abraham Lincoln won nearly 60 percent of the electoral college in 1860 despite winning only 39.8 percent of the popular vote. Levinson argues that a direct popular election of the president would be much more sensible.

In so arguing, Levinson goes beyond an emphasis on historical anomalies. He contends that because of the winner-take-all system within the states, presidential candidates have a strong incentive to ignore most of the nation. In 2004, 99 percent of advertising expenditures were focused on only seventeen states; Ohio and Florida accounted for no less than 45 percent of total expenditures. Levinson thinks that the exclusive focus on "battlegrounds" greatly distorts public debate. Most of the nation's cities are not in those states, and hence presidential elections do not focus on the problems faced by the many millions of people in those cities. In late October 2000, anyone "living in the United States … would have believed that the most crucial issue facing the entire country was prescription drugs for the elderly." Levinson adds that the composition of the electoral college gives unfair advantages to states with small populations. In 2000, Gore won New Mexico and its five electoral votes, while Bush won Wyoming, Alaska, and North Dakota for a combined nine electoral votes; but New Mexico has a larger population than those three states combined.

Levinson has other objections to Article II. For one thing, the president is elected on the first Tuesday in November; but the president-elect does not assume office until January 20. Hence, a repudiated president will exercise the authority of office for ten weeks after his repudiation. More fundamentally, Levinson thinks that Article II makes it much too hard for the public to remove a sitting president from office. True, impeachment is possible for "high crimes and misdemeanors." But what if the president is incompetent, an evident disaster, or simply doing what most people believe to be a bad job? Why shouldn't the public be able to get rid of him? Among the world's constitutions, the U.S. Constitution is quite unusual in so strongly insulating the chief executive from removal while in office.

The difficulty of removing the president is especially troublesome in view of what Levinson sees as the Constitution's grant of excessive power to the presidency. Many chief executives, including Clinton, have acted as if they were not much constrained by Congress's power to declare war; they have felt free to use military force without any kind of declaration from Congress. Other presidents, including George W. Bush, have interpreted ambiguities in the document to allow the chief executive expansive power to protect the nation as he sees fit--as, for example, through detaining suspected terrorists and wiretapping by the National Security Agency. Levinson believes that the public should take a sustained look at the president's constitutional authority and impose greater restrictions on that authority.

Nor does Levinson spare the Supreme Court, objecting to the Constitution's grant of lifetime appointments. Human abilities often diminish over time, and the oldest justices, whose opinions were formed in the distant past, may be ill-equipped to understand current problems. Levinson argues that the United States should follow the practice of most nations and appoint members of the highest court for a fixed term--a step that might also reduce the political acrimony that now accompanies Supreme Court appointments.

Levinson is aware that the amendment process seems to provide a democratic route for curing any democratic defects in the constitutional order, but in his view that process is plainly inadequate. Indeed, Levinson contends that Article V creates "what may be the most important bars of our constitutional iron cage precisely because it works to make practically impossible needed changes in our polity." Under the founding document, the Constitution can be changed in two ways. First, two-thirds of both houses of Congress may propose amendments, which become effective when ratified by three-quarters of the states. Second, the legislatures of two-thirds of the states can call a convention for proposing amendments, which also become effective when ratified by three-quarters of the states. Both processes are cumbersome and complex, ensuring that constitutional change may be more difficult under the U.S. Constitution than under any other constitution on the face of the earth. Often the need for approval from three-quarters of the states is a decisive obstacle. Three-quarters of the states, for example, have not been willing to approve reform of the electoral college.

Building on Jefferson's plea, Levinson believes that the Constitution's defects should be cured by means of nothing less than a new constitutional convention. In such a convention, the nation "would feel itself legitimately empowered, and psychologically free, to do what the framers of 1787 did, which was to look at all existing constitutions as well as the lessons derived from their own experiences." He would like the process to start with a nationwide petition campaign directed at Congress. He hopes that such a campaign might culminate in a healthy and reflective process, in which many millions of citizens engage in critical thought about their own institutions.

Most arguments for Constitutional change come from crackpots who badly misunderstand the Constitution. Levinson is emphatically not to be counted in that company. His critical discussion of the founding document is bold, bracingly unromantic, and filled with illuminating insights. He accomplishes an unlikely feat, which is to make a really serious argument for a new constitutional convention, one that is founded squarely on democratic ideals. Jefferson himself believed not only that old documents are likely to prove defective over time, but also that societies benefit from "a general attention to the public affairs." Levinson's call for a constitutional convention might be supplemented with the claim that it could promote a kind of civic education that might supply both immediate and long-term benefits to American democracy.

But in the end I am very far from persuaded. In some cases, the Framers probably had it right. In other cases, a constitutional amendment would probably be a good idea, but the problem is hardly serious enough to justify the turmoil of a constitutional convention. Might we try to imagine what such an event would actually be like, in the glare of the media and the blogosphere? Would true deliberation, and an excellent constitutional product, be likely? True, the composition of the Senate and the electoral college are major flaws--but there is a serious risk that a convention would produce a much inferior document, and that risk is best avoided. On this count, as on so many others, Madison, not Jefferson, is the proper guide.

Let us begin with those parts of the Constitution that are not especially hard to defend against Levinson's complaints. Is it really anti-democratic to allow the president to veto legislation, subject to a two-thirds override? After all, the president is himself elected, and whatever the problems with the electoral college, he is the most visible and in a sense the most accountable person in American government. The most veto-happy American president--contributing well over one-fifth of the total--was Franklin Delano Roosevelt, who was also one of the most popular presidents. True, the president's veto power increases the need for consensus before legislation may be passed, but it is hardly clear that this is a bad thing. We need to know whether this additional veto point does damage on balance. Levinson does not show that the veto power of the president, with his national constituency, decreases the democratic nature of American government or weakens the legislation that ultimately emerges.

Or consider the difficulty of removing a sitting president--a difficulty deliberately created by the Framers, who explicitly decided to forbid impeachment merely for "maladministration." The Framers sought to preserve the system of checks and balances by giving the president a degree of protection against legislative reprisal; and it is hardly clear that they were wrong. The president's relative independence decreases the risk that the nation will be destabilized, or that its political life will be distorted, by partisan efforts to remove its chief executive. The disgraceful (and unconstitutional) impeachment of Clinton provides a nice lesson about the risks of Levinson's proposal.

Nor is it at all clear that the Constitution would be better if "We the People" could amend it more easily. As Levinson is aware, the document provides a broad and general framework that allows American government to change over time, usually in response to democratic wishes. (In this way, Jefferson's desire for generational rethinking and refinement has not been entirely thwarted.) The twentieth century witnessed the rise of the administrative state, a massive strengthening of the national government, and significant changes in our understandings of constitutional rights. Some people deplore these changes. Yet it remains true that the Constitution does not freeze the nation; formal amendments are not necessary to alter key aspects of our government.

The Constitution certainly fixes certain features of our basic institutions, including those that Levinson deplores; but if formal amendments were made easier, we might find that the country would be rapidly engaged in a series of highly polarized and probably unproductive debates over such issues as flag-burning, same-sex marriage, the display of the Ten Commandments on public property, the future of affirmative action, and the use of the words "under God" in the Pledge of Allegiance. The history of proposed amendments in recent decades--often focused on unpopular decisions of the Supreme Court--strongly supports Madison's warning about the potentially harmful effects of public passions.

Some of Levinson's points do make a persuasive argument for constitutional reform through the existing channels. To allow a period of preparation, there needs to be some gap between the election and the inauguration of a new president; but the current period may well be too long. The document should probably be amended to spell out procedures in case of a successful attack on members of Congress. With respect to the question of life tenure for Supreme Court justices, most nations operate on the assumption that the benefits of a fixed term are greater than the costs; they are probably right. But surely points of this kind cannot justify the radical step of convening a new constitutional convention.

Levinson does have two exceedingly strong complaints. The electoral college is clearly an anachronism, and as a matter of principle a democracy should elect its president by a popular vote. To be sure, it is not right to say that the electoral college favors small states; if a candidate wins a large state by one vote, he gets a much bigger leg up on winning the presidency than if he wins a small state by one vote, and in this way the current system actually favors large states. But as Levinson points out, it is especially unfortunate that millions of voters, even undecided voters, are ignored because they live in states whose majority will inevitably support one or another side.

A possible answer is that without the electoral college, presidential candidates would have an incentive to ignore the least populous states and to concentrate their efforts on New York, California, Illinois, Texas, and other highly populated places. And it is true that the electoral college requires the president to have a degree of trans-regional appeal, which may reduce political polarization. But presidents represent people, not geographical areas. If candidates focus their efforts on the most populated areas, is something really amiss? In any case, it is not at all clear that political polarization would be more serious if the president were elected directly. Whatever the best solution, it is very difficult to defend the electoral college in its present form.

Levinson's argument against the current composition of the Senate, the product of an old political compromise, is stronger still. Levinson is right to complain about a situation in which small numbers of people, concentrated in certain areas, have disproportionate political power. He is also right to object that the consequence of the Senate's composition is to ensure perverse forms of redistribution, typically to people in less populous states. It is ironic but true--and a definite point in Levinson's favor--that the provision of the Constitution that most deserves amendment is the only part of the document that it explicitly declares to be off-limits (unless the state consents to losing its equal representation, which is not likely).

Unfortunately, Levinson says very little about the risks of a new constitutional convention. Suppose that the entire Constitution were placed up for grabs. Would the new product be better than the current one? That would be most unlikely. More importantly, the document is not only the words that its Framers wrote. Its current meaning is a product not merely of the founding generations but of unfolding practices and interpretations, produced by countless people and developing over long periods of time. Some of those interpretations come from the federal judiciary, which is not a democratically elected body; but most of constitutional evolution, in the form of new understandings of the document's general phrases and frequent silences, has been produced politically, not judicially.

Here is the most important point. As we currently live it, the American Constitution is a creation not merely of Madison and Hamilton and their contemporaries, but also of Abraham Lincoln, Woodrow Wilson, Susan B. Anthony, Franklin Delano Roosevelt, Martin Luther King Jr., Ronald Reagan, and countless others (including anonymous voters at many stages in the nation's history). In a partial vindication of Jefferson's arguments, each generation has, in fact, contributed a great deal to constitutional understandings. If the existing constitutional framework is to a significant degree a democratic product, the democratic argument for changing it is weakened. And if the existing framework is a product of many minds operating over time, there is all the more reason to fear that any particular set of people, operating at any particular time and inevitably in the grip of current concerns, will make it worse rather than better.

There is another problem. If the United States had a constitutional convention, would it really focus on the composition of the Senate, the electoral college, the presidential veto, and the other issues that concern Levinson? Or would it instead rethink symbolic questions and constitutional rights--for example, by reducing the protections given to criminal defendants, restricting freedom of speech, increasing protection of property rights, weakening the separation of church and state, and entrenching some controversial views about foreign policy or the economy? It seems at least as likely that some such measures, rather than Levinson's structural proposals, would be the major results of a contemporary convention.

In calling for a constitutional convention, Levinson underplays the crucial fact that the Constitution was meant to create a republic, not a democracy. The original constraints on majority rule were deliberate; they were designed at once to promote deliberation and to protect liberty. In the words of Hamilton, the "jarring of parties" in the legislature was meant to "promote deliberation." Some of what Levinson deplores, as frustration of the majority's will, is best understood as an effort to promote a system of deliberative democracy. As Washington famously explained to Jefferson, the Senate was meant "to cool" legislation by providing a kind of deliberative check. The various veto points that Levinson dislikes can be understood as an effort to protect liberty as well as deliberation, by ensuring against the use of national power unless a broad consensus can be obtained on its behalf.

It is true that the Senate and the electoral college, as presently constituted, are not easily defended as ways of promoting deliberation or protecting liberty. But the argument for a new convention cannot be clinched by reference to those admittedly serious imperfections. Sanford Levinson has valuably shown that parts of America's founding document are seriously flawed, and he has demonstrated that both representatives and citizens should treat the document not with "sanctimonious reverence" but as the revisable product of fallible human beings. But nobody should be packing their bags for Philadelphia.

Cass R. Sunstein is a contributing editor.

By Cass R. Sunstein