Monday, April 22, 2019

Frankfurter’s Conception of Judicial Legitimacy – and Ours

Guest Blogger

Or Bassok

In a recent post, Neil Siegel describes a “major concern expressed during the partisan gerrymandering litigation before the Supreme Court over the past two terms”— “that the Court’s public legitimacy may suffer if it holds that federal courts may adjudicate the merits of political gerrymandering claims.” Siegel identifies this concern with “Justice Frankfurter’s dire predictions about the consequences for the Court’s legitimacy of subjecting malapportioned legislatures to judicial review,” and in particular, Justice Frankfurter’s dissenting opinion in Baker v. Carr.

This reading of Frankfurter is anachronistic. It imposes a contemporary understanding of judicial legitimacy on Frankfurter that he did not share. And the difference between Frankfurter’s understanding of judicial legitimacy in the early 1960s and our contemporary view is quite important.

In order to explain this difference, I track the revisions Frankfurter made to a pivotal sentence (which Siegel quotes) in the various drafts of his dissenting opinion in Baker v. Carr. Frankfurter wrote his dissent just as ideas about judicial legitimacy had begun to shift in the early 1960s. This shift in conceptions of judicial legitimacy has had a profound impact on how people think about the Supreme Court and its authority. At the end of this post, I briefly discuss some of the consequences of this important development.

Like many discussions of the Supreme Court’s legitimacy in recent decades, Siegel’s argument thinks of judicial legitimacy in terms of public support. He treats judicial legitimacy and public legitimacy for the Supreme Court as more or less interchangeable terms. Yet this equation is a recent development; it arose after the invention of public opinion polling that could reliably measure public support for the Supreme Court. Before the invention of this measuring tool, there was no independent source of evidence, considered reliable by all relevant players, of public support for the Supreme Court. As a result, judicial legitimacy could not have been understood in terms of public support, at least as we think of it today.

Before the invention of public opinion polling in the 1930s, the only tool considered reliable for measuring public support was elections. That is not to say that there were no impressionistic assessments of what the public thought of the Supreme Court. Yet, in the absence of public opinion polls, the equation so commonly made today between threats to judicial legitimacy and the decline of public support for the Supreme Court in opinion polls simply didn’t exist. The conceptual move from “judicial legitimacy” to “public legitimacy” to “public support for the Supreme Court in public opinion polls” was not yet possible. In any conflict between the Supreme Court and the elected branches, the President and Congress would rely upon their public support, based on the most recent elections. Except for unusual circumstances, when the Supreme Court itself was at the center of election campaigns, there was no way for the Supreme Court to know, based on a publicly accepted tool of measurement, that it also enjoyed public support.

The entrance of public opinion polls as a reliable metric and — starting in the 1960s — the regular measurement of the confidence of the public in the Supreme Court made public opinion polling an independent criterion of judicial legitimacy. The monopoly of elected institutions on the claim to legitimacy based on public support was broken. For the first time in history, it was now possible to view the Supreme Courts’s legitimacy in terms of public support.

In support of his argument about Frankfurter’s understanding of judicial legitimacy, Siegel quotes the following sentence from Frankfurter’s dissenting opinion in Baker v. Carr: “[t]he Court’s authority — possessed of neither the purse nor the sword — ultimately rests on sustained public confidence in its moral sanction.”  But Frankfurter had a very different idea than the one Siegel attributes to him – as did the other Justices who invoked the concept of “public confidence” in the judiciary before the 1960s.

In a recent article, I reveal that Justice Frankfurter’s drafts of his dissenting opinion in Baker v. Carr show that he changed this sentence twice. The first version of this sentence appears in a January 1962 draft and states that “[t]he Court’s power ultimately rests in sustained public deference to its moral authority.” In a draft dated February 1962, this sentence was changed again. The second version reads: “[t]he Court’s power — possessed neither of the purse nor the sword — ultimately rests on sustained public acknowledgment of its moral authority.” Eventually this language changes to “ultimately rests on sustained public confidence in its moral sanction.” (Frankfurter was a close friend of Walter Lippmann, who famously wrote on public opinion and who was well aware of the influence of the invention of public opinion polling.)

In this continually revised sentence, Frankfurter is paraphrasing Alexander Hamilton’s famous dictum from the Federalist No. 78. But Hamilton did not write that, without the purse or the sword, the judiciary merely has public confidence. Hamilton said that what the judiciary had was “merely judgment.”

Like most of his generation, Hamilton understood judicial legitimacy as emanating from expertise. This belief in legal expertise is well reflected in the Federalist No. 81 where Hamilton writes about judges as “men selected for their knowledge of the laws, acquired by long and laborious study.” In this institutional design, expertise was opposed to public approval: The founders designed the Supreme Court to counter shifts in popular opinion that contradicted the Constitution. In this spirit, Hamilton wrote in the Federalist No. 78 that “the independence of the judges may be an essential safeguard against the effects of occasional ill humors in the society.” Expert knowledge was envisioned as the tool that would enable judges to guard the Constitution from public opinion.

All this has changed in our world. In recent decades scholars understand judicial legitimacy very differently. Take, for example, the closing lines of Pamela Karlan 2011 Harvard Foreword titled Democracy and Disdain:
“Alexander Hamilton was slightly off base when he wrote that the judiciary has ‘neither Force nor will but merely judgment; and must ultimately depend upon the aid of the executive arm even for the efficacy of its judgments.’ The judiciary must ultimately depend on the people.”

Hamilton was not “slightly off base.” He simply viewed the Supreme Court’s source of legitimacy differently than Karlan. It is not surprising that a contemporary scholar like Karlan views as virtually a truism that the Supreme Court must have public support to function properly, and therefore suggests that Hamilton got it (“slightly”) wrong. The link between public support for the Supreme Court and its ability to function properly has become so powerful in American constitutional discourse that it has become common sense. According to this picture, the political branches enforce the Supreme Court’s decisions not as a result of its expertise — as if they were the patient doing as the doctor ordered — but because of public support for the Supreme Court.

In my article I track the change in how Hamilton’s dictum has been paraphrased by the Supreme Court following the invention of public opinion polls. Justice Frankfurter no doubt contributed to this change in Baker v. Carr, but Frankfurter never believed that the Supreme Court’s legitimacy relied on public support and he would never have equated legitimacy with public support, much less public opinion polls. In his dissent in West Virginia State Bd. of Educ. v. Barnette (1943) Frankfurter emphasized that “[t]he Court has no reason for existence if it merely reflects the pressures of the day” and in Dennis v. U.S. (1951) Frankfurter stressed that “[c]ourts are not representative bodies. They are not designed to be a good reflex of a democratic society.”

Given the revisions that Frankfurter made to the sentence that Siegel quoted from Baker v. Carr, it seems that Frankfurter was trying to use the notion of “moral sanction” to square the circle: He sought to unite his progressive belief in the ability of experts to lead society based on their expert knowledge with his belief in democracy. For Frankfurter, “public confidence” rests on a normative (“moral”) standard, rather than merely on raw public opinion that may respond to populist whims. In this spirit, legal historian Brad Snyder speaks of “Frankfurter’s Jeffersonian faith in enlightened public opinion.” Frankfurter invokes not the aggregated opinion of the actual public, but rather a normative standard connected to an (imagined) enlightened public’s views. Frankfurter tried to create a normative criterion for assessing the Supreme Court’s legitimacy that would not be dependent on public opinion  but would still maintain a connection to the public through the idea that the public, once enlightened by expert knowledge, would follow the Supreme Court’s lead.

The same misunderstanding of Frankfurter has also been visited on his famous clerk Alexander Bickel. Under the spell of equating legitimacy with public support, political scientists have repeatedly argued that Bickel’s countermajoritarian difficulty is spurious. After all, political scientists have repeatedly shown that the Supreme Court’s decisions don’t actually counter public opinion as it is reflected in opinion polls.

Yet, there is a serious problem with this argument: Bickel simply did not understand the countermajoritarian problem as a problem of courts bucking public opinion as measured in public opinion polls. This contemporary way of thinking was beyond Bickel’s horizons. In his book The Least Dangerous Branch he writes “[m]ost assuredly, no democracy operates by taking continuous nose counts on the broad range of daily governmental activities.” (p. 17) Measuring public support for the Supreme Court had only begun its ascent when he wrote his famous book (published in 1962). Like Frankfurter, Bickel was not oblivious to the potential effects of the new polling technology on the role of the Supreme Court; he noted that “[s]urely the political institutions are more fitted than the Court to find and express an existing consensus – so long, at least, as the science of public sampling is no further developed than it is.” (p. 239)

Bickel saw the countermajoritarian difficulty as an inherent difficulty faced by an unaccountable Supreme Court, rather than a difficulty faced by a judicial institution that is un-responsive to public opinion as expressed in opinion polls. While there are connections between accountability and responsiveness, as long as Supreme Court justices are not elected, the only way to deal with their lack of political accountability is to justify the Supreme Courts’s countermajoritarian authority in other ways. And indeed, generations of scholars who understood Bickel correctly – including John Hart Ely, Ronald Dworkin and Bruce Ackerman (to name a few) – worked to devise a justification for the Supreme Court’s countermajoritarian authority.

These scholars were not fools who somehow failed to recognize that the Supreme Court follows public opinion. Similarly, Bickel didn’t just somehow miss an article written in 1957 by Robert Dahl at Yale’s political science department, only a few blocks from Yale Law School. In his famous article, Dahl showed that the Supreme Court responded to the “dominant national alliance.” Many political scientists see Dahl’s article as the beginning of the body of research that shows that the Supreme Court decisions on publicly salient issues are normally in line with the results of public opinion polls. Yet, Bickel’s countermajoritarian difficulty was focused on accountability, not responsiveness, and nothing that Dahl showed could solve the problem that Bickel posed. In addition, Dahl himself did not refer to public opinion polls. He used the position of the legislative majority in Congress four years or fewer before a Supreme Court decision as a surrogate for majority will, since “scientific opinion polls are of relatively recent origin.” (p. 283). Dahl explained that “for the greater part of the Court’s history…there is simply no way to establish with any high degree of confidence whether a given alternative was or was not supported by a majority of adults or even of voters.” (p. 283-84).

This helps us understand why Bickel focused on the “passive virtues.” The reason the Supreme Court needed to adopt the passive virtues was not to conserve its public support (as measured by public opinion polls.) Rather, the point of the passive virtues is to defend the Supreme Court’s language of expertise. In Gerald Gunther’s words, Bickel insisted that the Supreme Court be 100 percent principled 20 percent of the time. Bickel actually feared what Siegel calls judicial statesmanship: judges corrupting the language of legal doctrine in order to ensure the Supreme Court’s public support. Bickel and Frankfurter saw the passive virtues as a defense against the Supreme Court corrupting its expertise in an attempt to curry favour with the public. Avoiding the “political ticket” helped ensure that doctrine would not be corrupted and the Supreme Court’s source of legitimacy – its expertise – would remain intact.

If we understand judicial legitimacy in terms of public support, then the criterion for assessing the Supreme Courts work, at least in highly visible cases, shifts from whether decisions are correct from the standpoint of expert knowledge to whether they have public support.

What expertise in constitutional law means may be (and often is) disputed, but one principle remains true for any field of expertise: the correct answer is determined by expert knowledge and not by public opinion. Expert knowledge must be determined by the community of experts according to the rules of their discipline.

The concept of public opinion may be complicated (e.g., does elite opinion matters more?); so too may the concept of expert knowledge (e.g., how do we decide when experts disagree?). Yet in areas governed by expertise (for example, monetary policy set by the Federal Reserve), whether the public supports or does not support a proposed action has no bearing on the correct answer (although, of course, the scientific measurement and prediction of public activity is relevant to setting monetary policy).  We do not rely on public opinion polls to tell us the best medical treatments or the best way to fly an airplane.

Hannah Arendt foresaw the dangers in shifting the source of the Supreme Court’s legitimacy from expertise to public opinion polls. Arendt characterized the American Supreme Court as “the true seat of authority in the American Republic” but unfit to power. This reflects her view that the Supreme Court’s source of legitimacy is expertise rather than public support. As I explain elsewhere, Arendt saw this as a vital feature of the Supreme Court. If this feature changed, the Supreme Court would be like the philosopher-king who depends on public support and corrupts his or her expertise. Whenever a philosopher-king tries to curry favor with the public, the result is inevitably a corruption of true philosophy. In like fashion, whenever the Court understands its source of legitimacy to stem from public support, its ability to engage in expert judgment will be corrupted.

A very good example of the problem is the “broccoli argument” used to attack the Obamacare individual mandate: Congress could not impose an individual mandate to buy health insurance because this would mean that Congress could force people to buy broccoli. Initially, this argument was ridiculed by almost the entire legal community. In terms of legal expertise, it was a non-starter. Mark Rosen and Christopher Schmidt, who wrote a scholarly article describing the evolution of the broccoli argument, noted that “[m]uch of the broccoli hypothetical’s power was in its very ridiculousness…so when challengers first floated it—at a time when few thought the constitutional challenge had much hope in the courts—it had a whiff of desperation.” Yet within a few years this argument had been adopted by the majority of the Supreme Court as a limitation on the Commerce Power.

The broccoli argument did not succeed by employing the tools of law as a language of expertise. Rather, Rosen and Schmidt explain that the argument succeeded because it rose in popularity. Public support encouraged the conservative Justices to adopt the broccoli argument as a legitimate—and powerful—constitutional argument. Once the legitimacy of the Supreme Court is viewed not in terms of its adherence to legal doctrine and expertise, but in terms of its public support, this development is no longer baffling.

Or Bassok is an Assistant Professor of Constitutional Law at the University of Nottingham. You can reach him by e-mail at Or.Bassok at

Older Posts
Newer Posts