Friday, April 29, 2022

Reflections on the Supreme Court, the Senate, and Sandy Levinson

Guest Blogger

This post was prepared for a roundtable on Reforming the Supreme Court of the United States, convened as part of LevinsonFest 2022—a year-long series gathering scholars from diverse disciplines and viewpoints to reflect on Sandy Levinson’s influential work in constitutional law.

Vicki C. Jackson[1]

Professor Levinson has been using his powerful scholarly voice for years now to call our attention to the increasingly hard to justify mal-distribution of the Senate and the way its democratic deficits are hard-wired into our present constitutional architecture.[2] This structural defect is related to the question of Supreme Court reform. For the disjunction between popular voting and presidential electoral success, which arises because of the mal-distribution of the Senate and its carry-forward to the number of votes in the Electoral College, contributes to the present unhappy situation, as does the malapportioned Senate’s role in the confirmation process.

The U.S. stands virtually alone, among constitutional democracies with well-established judicial review by independent courts, in providing neither for a retirement age nor for a limited term in office for its high court justices. Had one looked at this issue in 1921, the United States would have had company: At that time, Australia and Canada, countries that, like the United States, were influenced by the British tradition, provided judges with indefinite tenure during good behavior.[3] However, each of these countries amended their constitutions and adopted mandatory retirement ages for their federal judges later in the 20th century – 70 in Australia, 75 in Canada. All but one of the U.S. states now have term limits, or a mandatory retirement age, or both, for their high court judges.[4] During the New Deal period, and again in the 1950s, serious proposals were made (though not acted upon) for a constitutional amendment creating a retirement age for federal judges. Recently, norms about Senate confirmation have broken, increasing the risks of a confirmation process perceived as systematically skewing the Court and diminishing its legitimacy. Average tenures of the justices are increasing; and the haphazard nature of when vacancies occur have resulted in such powerfully important anomalies as President Carter having no appointments during his four-year term and President Trump (who lost the nationwide popular vote) having 3 – one-third of the Court – during his one four-year term.

Against this background, I suggest that the time has come to seriously take up proposals to limit the indefinite tenures of Supreme Court justices. Fifteen years ago I was disinclined to support proposed changes to tenure “during good behavior,” a provision that had so well secured an independent federal judiciary.[5] This past year I urged the Presidential Commission on the Supreme Court of the United States to take very seriously the benefits of some reform and to try to seize the moment to build a consensus in favor of change towards a more sensible system of appointment and tenure for Supreme Court Justices. The Commission did not do so but I hope a coalition will emerge to act.

The growing effects of the Senate’s counter-majoritarian character, together with recent breaches of political norms and conventions about Supreme Court appointments, are deeply concerning. Although recent abuses of the confirmation process – against the backdrop of the counter-majoritarian effects of the Senate – provide temptations to “Court-packing.” But sheer partisan court-packing bears high risks of further polarization, instability and delegitimation. Less partisan efforts to spread out appointments should be pursued, that will not dissolve the counter-majoritarian character of the Senate but help mitigate the way our haphazard and contingent tenure system for Supreme Court justices can magnify the counter-majoritarian effects of the Senate and the electoral college.

Elsewhere I discuss the potential benefits of mandatory retirement ages and 18-year single term approaches.[6] Here I discuss an alternative approach that Congress could constitutionally effectuate by statute and that might avoid at least some of the risks of being viewed as a naked partisan power grab inviting further partisan power grabs back.

Smoothing Out Appointments Though Conditional Changes in Court Size: One of the most compelling arguments for change in Supreme Court tenure derives from the very uneven distribution of opportunities for Presidents, elected at different times and from different parties, to make appointments. As the Constitution has become more oriented to the value of equal voting and equal representation over time, the importance of having democratically legitimate inputs to the Court has increased. Yet the current system makes vacancies depend on contingent, random events, and has the potential (however much it is or is not used) for sitting justices to decide which president chooses their replacement.

Given Congress’s power to change the numbers of justices on the Court, possibilities (other than the frequently discussed 18-year staggered terms idea) could be considered, which could be implemented by statute. Specifically, Congress could provide for fluctuating Court membership, [7] or a “decoupling” of appointments from vacancies,[8] by guaranteeing one, or two, presidential appointments each four-year term, while letting the overall numbers on the Court fluctuate. Such approaches would spread out opportunities for appointment, albeit not as rigorously as the 18-year term limit proposal. A modest variation, similar to Terri Peretti’s,[9] would reflect the desirability that, in each four-year Presidential term, at least one appointment to the Court be made.[10] 

In the last 45 years, there have been four 4-year presidential terms that had no appointments to the Court.[11] A statute could authorize the conditional establishment of an additional seat on the Court (up to a limited total) in order to try to allow each four-year presidency to choose at least one justice. So, for example, it might provide that if by year two of a four-year term no vacancy in the 9-member Court had occurred, a tenth seat would be created to be filled by that Presidency. If the 10th seat were filled and a further vacancy occurred, say, during the 4th year of the same presidency, the number of authorized seats would revert to 9. If, however, no such vacancy occurred, and a new president was elected (or the sitting president was reelected), and by year 2 of that presidency no vacancy had occurred, an 11th seat would be authorized to be filled. Seats could revert down to the norm of 9 if vacancies occur, provided that the presidency at the time had had an opportunity to make one appointment. Such a mechanism could be capped (e.g. at 11, or 13) to keep the Court to a workable number. 

The advantages of such proposals are, first, that they could be enacted by statute. Second, they would improve on democratic inputs to the membership of the Courts (without requiring a constitutional amendment to depart from life tenure). Fluctuating proposals that do not allow the Court to grow too large could not guarantee as much regularity of democratic input as the staggered 18-year terms, but would nonetheless be an improvement over the status quo (assuming the Senate behaved more cooperatively, a problem across proposals).[12]             

Responding to Recent Abuses of the Confirmation Process? Longstanding structural aspects of the selection and tenure process for Supreme Court justices warrant attention now, with a view towards a long-term fix. But some calls for reform arise more from a felt need to respond to what are seen as abuses of the confirmation process in very recent years.[13] 

As is well-known, the Senate refused even to consider President Obama’s nomination of Merrick Garland in March 2016, shortly after the death of Justice Scalia in February, on the ground that it was within 8 months of a presidential election and the Senate should wait and “give the people a voice” in the selection of a new member of the Court.[14] However, in the Fall of 2020, when Justice Ruth Bader Ginsburg died, the Senate rushed to confirm the President’s nominee just weeks before the presidential election. The proffer of a quite novel reason not to consider the Garland nomination, combined with a complete failure to apply the reasoning to a situation in which logically it applies a fortiori, has created widespread concern that the confirmation process was abused. Moreover, as noted above, President Trump – who won the presidency with fewer popular votes than his opponent and who lost the November 2020 election in both the popular and Electoral College vote – was able to make three appointments to the Court during his single four-year term; President Jimmy Carter, who won a decisive popular majority in 1976, was able to make no appointments to the Court. 

These results reflect the randomness of when Supreme Court vacancies arise under our current system, and the counter-majoritarian aspects of the Senate as presently constituted (and operating now under quite broken norms of conduct), as well as the effects of the two-Senators per state rule on the Electoral College. The fluctuating Court proposals described above could help mitigate the randomness problem. 

Despite frustrations of the current moment, applying changes – whether of finite terms or age limits – to currently sitting justices would raise concerns for judicial independence and thus for the rule of law, especially if done by statute, which in the United States is much easier to enact than constitutional amendment. In countries where judges learn that they are subject to relatively easy removal on account of their jurisprudence, it is much harder to sustain judicial independence and the rule of law. Both, as Justice Breyer has recently emphasized, are precious and important.[15] Expanding the Court, without removing any existing justices, might thus be less of a threat to judicial independence and some scholars have called for such expansion.[16] But expanding the Court simply to give one specific President the power to fill seats, even when in response to abuses by the other party, invites those in the other party to engage in similar action when they return to power, and thus risks turning the Court into even more of a perceived political football. 

Whether rationalizing a decision to expand the membership on the grounds of redressing an abuse of the confirmation process would serve to contain future responses is at best uncertain; or whether, regardless of rationale, it would elicit “tit for tat” action further threatening the perceived independence of the court, are political questions on which reasonable minds might disagree. Experience elsewhere, however, raises concern about whether, once such a process gets started, it can be contained or would spread to create conditions in which further intrusions on the Court’s independence could result.[17] 

The broader context and the democratic deficit of the Senate: While it might thus be imprudent simply to expand the Court for the purpose of enabling a single President to make several appointments (however much this may be a justified response to past abuse), the broader context in which Supreme Court justices are selected reinforces my view that change is needed. 

The Senate is designedly not a population-based entity. But the degree to which its structure creates the risk of counter-majoritarianism is substantially greater than that in several other federal systems with which the U.S. is often compared and the degrees of counter-majoritarianism in Congress have increased over time. Indeed, some scholars think it possible that, given demographic change and the composition of the Senate, it might be that no Democratic president will be able to appoint a Supreme Court justice again in the foreseeable future.[18] 

At the founding, the greatest disproportion between the largest state and the smallest state was about 12.65 to 1 (Virginia being the largest by population, Delaware being the smallest).[19] Today, according to the 2020 census, the most populous state of California had something like 68 times the population of the least populous state, Wyoming.[20] Moreover, at the founding Virginia was the only state with more than ten times the population of any other state; today, by contrast, there are ten states with populations over 10 million, and six states with populations under 1 million.[21] Scholars writing in the 1990s found that the minimum percentage of the population that could elect a Senate majority has been getting smaller since the 1860s;[22] 2020 Census data suggests that today, the 26 smallest states, with 17.5% of the population, could in theory elect a Senate majority.[23] Although the Senate need not have any particular partisan skew, in recent decades it has skewed in a Republican direction.[24] 

It is an unstable situation for a party supported by a minority of the population to be able to control the Senate, frequently the Presidency, and the Supreme Court. The constitutional amending process may also be blocked by the combined effect of the Article V voting rules and the partisan state-by-state demographic skew. If citizens cannot look to elections, nor the courts, nor the amending process, to achieve a federal government that is in broad terms responsive to democratic views, what remains are methods that should trouble all who believe in the rule of law. Thus, even if reforms will only help at the margins, the time has come for reforming the Supreme Court so that it may bear somewhat more of an imprint of recently elected presidents, while at the same time preserving the judicial independence that it has achieved. Other changes may also be needed to the ways in which the Court exercises its jurisdiction, but it will be for the best if the Court itself re-develops modes of judicial restraint suitable to its position in our constitutional democracy. 

The kind of statutory change suggested above – for floating numbers of seats on the Court to enable every presidential term to make a minimum number of appointments – would be far more modest than those Sandy Levinson has argued for in recent years. But it might hold appeal to partisans on both sides; and some improvement, I tend to think, is better than none. 

I look forward to Sandy’s reaction – for he has been such a vigorous and influential force for serious rethinking of the U.S. Constitution. And long may he continue to be so! 

Vicki C. Jackson is the Laurence H. Tribe Professor of Constitutional Law at Harvard Law School. You can contact her at


[1] This short reflection is drawn primarily from Vicki C. Jackson, Statement to Presidential Commission on the Supreme Court of the United States (July 16, 2021).

[2] See, e.g., Sanford Levinson, Our Undemocratic Constitution 49-62 (2006).

[3] Const. Act 1867 § 99(1) (Canada) (superior court judges to hold “Office during good Behaviour,” removable only “on Address of the Senate and House of Commons”); Austr. Const. § 72 (Australia 1901) (judges removable only on “address” from both houses of the parliament in the same session and only for “proved misbehavior or incapacity”). Canada amended its constitution in 1960 to provide for mandatory retirement for all superior court judges at age 75, Const. Act 1960, 9 Eliz. II, c.2 (U.K.), an age that had by statute applied to Supreme Court Justices since 1927; under the 1982 Constitution Act, the structure and independence of the Supreme Court are protected. See Reference re Supreme Court Act, ss 5, 6, [2014] 1 SCR 433 (Can.). Australia amended its constitution in 1977 to require that its High Court justices retire by age 70. Austrl. Const. § 72, ¶ 2, amended in Constitution Alteration (Retirement of Judges), 1977 (Austrl.).

[4] As of 2014, the one exception is Rhode Island. See (citing an American Judicature Society publication of 2014).

[5] See Vicki C. Jackson, Packages of Independence: The Selection and Tenure of Article III Judges, 95 Geo L. J. 965, 1007-08 (2007) (suggesting that any changes life tenure could be detrimental to the overall system of adjudication in the United States given the dependence of many state court systems on popular elections of judges and thus urging “great caution” before departing from existing structures securing judicial independence).

[6] See Jackson, supra note 1.

[7] See especially Terri L. Peretti, Promoting Equity in the Distribution of Supreme Court Appointments, in Reforming the Court: Term Limits for Supreme Court Justices 435 (Roger C. Cramton & Paul Carrington eds., 2006) (guaranteeing one seat to be filled and limiting each presidential term to filling two seats); see also Philip D. Oliver, Increasing the Size of the Court as a Partial but Clearly Constitutional Alternative, in id. at 405 (guaranteeing two seats to be filled each presidential term). My proposal bears some resemblance to theirs; they both envision a combination of a certain number of guaranteed appointments per presidential term, coupled with a fluctuation up or down in the number of authorized seats. Oliver’s proposal is more ambitious and might lead to a much larger Court, which he argued was beneficial because a larger Court would diminish a sense of entitlement to a particular seat and the outsize (in his view) reputations of the existing nine. Oliver, at 410-11, I worry that increasing the Court’s size by too much would make its work much more difficult. Oliver’s suggestion that the Court could simply sit in panels, id. at 412, while supported by some comparative experience, might require a constitutional amendment, thereby undermining its value as a statutory proposal. Another proposal, for a gradual increase in the Court’s membership (to 17 or 19) -- from Jonathan Turley, Unpacking the Court, 33 Persp. on Pol. Sci. 3 (2004); see also Jonathan Turley, Destroying the Court to Save It, https/ -- would also allow expansion beyond what I think workable in the U.S. context and might require constitutional amendment.

[8] See Daniel Hemel, Can Structural Changes Fix the Supreme Court?, 35 J. Econ. Persp. 119, 121, 136-38 (2021) (arguing for scheme in which each President makes two appointments at beginning of their term--but Justices do not take their seats until after the presidential term, to diminish the “loyalty effect”). My suggestion above is a more modest departure from current practice than Hemel’s.

[9] See Peretti, supra note 7.

[10] See, e.g., Steven G. Calabresi & James Lindgren, Term Limits for the Supreme Court: Life Tenure Reconsidered, 29 Harv. J. L & Pub Pol’y 769, 778-81 (2006); see also Roger C. Cramton, Constitutionality of Reforming the Supreme Court by Statute, in Reforming The Court: Term Limits For Supreme Court Justices 348 (Roger C. Cramton & Paul Carrington eds., 2006) (stating that before 1970, “almost every president serving a four-year term received at least one appointment to the Court”).

[11] See U.S. Senate, Supreme Court Nominations (1789-present), (showing no Supreme Court appointments during President Obama’s second term; President George W. Bush’s first term; President Clinton’s second term; and President Carter’s term).

[12] The 18-year term proposals assume a reasonably cooperative Senate, a point suggesting the need to reconsider the entire appointment process – sources of nominees, selectin mechanisms, voting rules to select etc.

[13] These concerns are reinforced by jurisprudential trends that contribute to developments that undermine the representative quality of our democratic institutions. See Shelby County v. Holder, 570 U.S. 529 (2013) (invalidating Section Four of the Voting Rights Act at a time when the Republican Party is pushing for legislation to suppress voting); Rucho v Common Cause, 588 U.S. ___ (2019) (precluding the possibility of federal courts addressing excessively partisan districting); Citizens United v. FEC, 558 U.S. 310 (2010) (overruling earlier judicial decisions and limiting legislative efforts to control the role of wealth in elections).

[15] See, e.g., Stephen G. Breyer, Scalia Lecture, The Authority of the Court and the Perils of Politics, Harvard Law School (April 7, 2021).

[16] See, e.g. Michael Klarman, Foreword: The Degradation of American Democracy—And the Court, 134 Harv. L. Rev. 1, 250-51 (2020) (raising the possibility of Democrats in 2021 expanding the Court by four members to “unpack” the Court in response to Republican abuse of the confirmation process and describing “reciprocal hardball” as “the only adequate remedy for Republican court packing”).

[17] See Tom Gerald Daly, ''‘Good’ Court-Packing?,” paper presented at the Centre for Comparative Constitutional Studies (CCCS), Melbourne Law School (8 June 2021) (discussing experience in Argentina and Turkey); see also Joshua Braver, Court packing: An American Tradition?, 61 B.C. L. Rev. 2747 (2020) (arguing that past changes in the size of the U.S. Court had plausible bases (in factors like need for additional circuit judges) that were less likely to provoke tit-for-tat responses than proposals to add multiple seats all at once).

[18] See Klarman, supra note 15, at 236 (“If sparsely populated states continue to vote mostly Republican, then the thirty percent of Americans who pick seventy percent of the senators would virtually guarantee Republican control [of the Senate], which would be a legitimacy crisis of massive proportions. For example, such a Senate might never again confirm a Democratic President's nominee to the Supreme Court.”).

[19] See; see also Vicki C. Jackson, The Democratic Deficit of United States Federalism: Red State, Blue State, Purple?, 46 Fed. L. Rev. 645, 650 n. 20 (2018). This proportion assumes that all persons count. IF one compares the state populations based on the 1790 Census numbers of “free white men” over age 16 (populations of 110,936 to 11,783), the ratio was under 10:1. With special thanks to Jonathan Gould for the references and information in this and the next two footnotes.

[21] See sources cited supra notes 18 and 19.

[22] See Frances E. Lee & Bruce I. Oppenheimer, Sizing Up the Senate The Unequal Consequences of Equal Representation 11 (1999); see also Jackson, Democratic Deficit, supra note 18, at 650 (arguing that the composition of the Senate created a democratic deficit; based on 2016 Census Bureau projections, “the 52 Republican Senators sitting in early 2017 could be said to represent states including roughly 144 million Americans, while the 48 Democratic-caucusing Senators could, on the same basis, be said to represent a substantial majority of about 178 million”).

[23] See U.S. Census Bureau, Table 2. Resident Population for the 50 States, The District of Columbia, and Puerto Rico: 2020 Census (2021), With thanks to Sam Weinstock for his help on this note.

[24] See Lee Drutman, The Senate Has Always Favored Smaller States,. It just Didn’t Help Republicans Until Now, Five-Thirty Eight (July 29, 2020), (explaining how over recent decades of demographic and political change it has come to be the case – and may be the case in the future -- that “Republicans now hold a majority of Senate seats while only representing a minority of Americans”); see also supra note 21.

Older Posts
Newer Posts