Skilling: more blind monks examining the elephant.

• Author: O'Sullivan, Julie Rose

1. The Statute was Unconstitutionally Vague and the Court's "Limiting Construction" Constituted Improper Legislation II. "Honest-Services" Fraud as an Object Lesson in Why Congress Ought to be Charged with Defining the Scope of Criminal Law Most academics and practitioners with whom I have discussed the result in Skilling v. United States believe that it is a sensible decision. (1) That is, the Supreme Court did the best it could to limit the reach of 18 U.S.C. [section] 1346, which all nine justices apparently believed--correctly-was, on its face, unconstitutionally vague. (2) My friends reason that the Court had already tried once, in McNally v. United States, (3) to force Congress to "speak more clearly than it has" (4) in defining the outer limits of "honest-services" (5) fraud. Congress responded quickly and with little consideration with the supremely under-defined [section] 1346. In the over twenty years since the statute's enactment, the Courts of Appeals have been unable to come up with any unified limiting principles to contain its reach. The Skilling Court, evidently reluctant to again throw the matter back to Congress given that institution's previous default, and not satisfied with the Courts of Appeals' efforts, was determined to come up with its own "narrowing interpretation." (6) Thus, the majority deemed it appropriate to rewrite the statute to cover what it concluded was the "core" of the criminality the prosecutors had addressed in bringing [section] 1346 cases--bribery and kickbacks. (7) The Court comes up with narrowing constructions to avoid constitutional difficulties in many statutory interpretation cases, the argument goes, and this construction is one that many in the academic and practice communities believe is reasonable.

   My only quibble with this consensus lies in my conviction that what the Court did in Skilling is as patently unconstitutional as [section] 1346--and that its foray into legislation is not of only academic concern. It clearly accepted Congress' delegation of law-making authority and essentially promulgated a new statute out of the "dog's breakfast" that was pre-Skilling [section] 1346. (8) Some would argue that this is a good thing from a practical, if not an orthodox separation-of-powers, point of view. In what I hope will be taken as a back-handed compliment, I will focus on Professor Dan M. Kahan's long-standing arguments in this regard. Kahan favors administrative specification of the content of arguably vague criminal prohibitions, but he believes that if one has to choose between judicial gap-filling and congressional action, the former is preferable to the latter. Kahan has argued that the Court ought to come clean and simply acknowledge that it has long been engaged in interstitial lawmaking because Congress has declined to legislate with any specificity and "[a] criminal code at least partially specified by courts is both less costly and more effective than is a code fully specified by Congress." (9) Even were I prepared to join Professor Kahan in discounting the effect of judicial law-making on normative commitments to the "democratic accountability, notice and other rule-of-law values," (10) I disagree with his conclusion about the viability and attractiveness of this delegation of authority to federal courts to fill in the blanks in otherwise underspecified statutory schemes. It strikes me that the history of the honest-services fraud theory, which culminated in Skilling, presents a wonderful example of how criminal law ought not be made, whether viewed from an institutional, societal, or individual standpoint.

1. THE STATUTE WAS UNCONSTITUTIONALLY VAGUE AND THE COURT'S "LIMITING CONSTRUCTION" CONSTITUTED IMPROPER LEGISLATION

   All the justices concluded that [section] 1346, viewed on its face, was vague. If a statute is unconstitutionally vague, such that it does not give ordinary citizens fair notice and is susceptible to arbitrary and discriminatory enforcement, what should the Court do? The Skilling majority asserted that "lilt has long been our practice ... before striking a federal statute as impermissibly vague to consider whether the prescription is amenable to a limiting construction." (11) Thus, the majority contended, it "does not legislate, but instead respects the legislature, by preserving a statute through a limiting interpretation." (12) When one examines the nineteen precedents cited in support of this supposedly hallowed practice, (13) however, its legitimacy looks a lot shakier than the Court lets on--at least in this context. First, in only three of the nineteen cases cited did the Court directly address a vagueness challenge. (14) In the others, the Court employed its "constitutional avoidance" canon of construction to avoid potential First Amendment and other specific constitutional issues. Second, only ten of the cases involved a challenge to a criminal statute. (15) More to the point, in only one case hailing from 1954 did the Court actually adopt a limiting construction to avoid holding a criminal statute unconstitutionally vague. (16) In another, it found that even a judicially-created limiting construction could not save the vague criminal prohibition at issue--a prohibition that threatened First Amendment values. (17)

   I think these distinctions--based on the criminal character of the case and the type of constitutional challenge lodged--are important in assessing the legitimacy of the Court's remedy. As the Court has pointed out, "[t]he standards of certainty in statutes punishing for offenses is higher than in those depending primarily upon civil sanction for enforcement," and for very good reason. (18) "[T]he first principle" (19) of criminal law--the principle of legality--outlaws the retroactive definition of criminal offenses.

   It is condemned because it is retroactive and also because it is judicial--that is, accomplished by an institution not recognized as politically competent to define crime. Thus, a fuller statement of the legality ideal would be that it stands for the desirability in principle of advance legislative specification of criminal misconduct. (20) This foundational principle simply does not apply in civil cases.

   The vagueness doctrine's "connection to [the principle of] legality is obvious: a law whose meaning can only be guessed at remits the actual task of defining criminal misconduct to retroactive judicial decisionmaking." (21) Where a statute is vague--and thus by definition lacks ascertainable standards to guide citizens or law enforcement in the administration of criminal penalties--it is difficult to conceive how the Court can articulate such standards without legislating. Despite the Court's inexplicable reliance on the rule of lenity in declining to include undisclosed conflicts of interest in the newly articulated honest-services canon, this is not a case of ambiguity. The Court was not asked to elect between two equally plausible definitions of a term (or grammatical constructions of a phrase) embedded in otherwise reasonably articulated elements. Nor was it, as in other constitutional avoidance cases, making a choice among reasonably defined alternatives. For example, in First Amendment overbreadth cases, the Court is tasked with construing a clear statute that could, if not narrowed, impinge on protected speech. The Court was not, in short, asked to elect among two or even five alternative meanings. Rather, it was tasked with electing among a wealth of alternative means of defining and thus limiting the reach of a statute in which none of the big three of criminal statutes--the conduct, mens rea, and attendant circumstances--are defined with any particularity.

   Indeed, such were the smorgasbord of choices that required resolution to truly define the reach of this statute that even if one were to conclude that judicial definition of some vague statutes is permissible, this statute is not one that is susceptible to a reasonable narrowing interpretation. (22) As I have laid out at greater length elsewhere, this statute represents "vagueness on steroids," (23) in part because "when courts (let alone ordinary citizens) cannot agree on what conduct--attended by what mental state and what attendant circumstances--constitutes a crime, it is a vagueness trifecta." (24) The Court took these cases to decide three issues, although it ultimately failed to address them: must the government prove that it was reasonably foreseeable that the honest-services scheme could cause some economic or pecuniary harm to the victims (Black v. United States); (25) must the duty to disclose, the violation of which constitutes the "fraud," arise under state law (Weyhrauch v. United Stales); (26) and must the defendant intend to obtain private gain from the victim to whom honest-services are owed (Skilling). (27) But these are only three of the many questions that have split the circuits. Other critical questions include: what creates a duty of honest-services; what constitutes a breach of that duty; must a separate duty to disclose be found or is such a duty inherent in any case in which a duty of loyalty is found; what mental state need be proven and does the mental state differ, as most circuits hold, in public and private honest-services cases; must the government prove the attendant circumstances of the materiality of the breach, as well as the non-disclosure? (28)

   The Court was determined to strike out on its own and adopt a limiting construction that had not occurred to any of the many courts struggling with this statute for the past twenty years. (29) Instead of answering the questions presented and resolving at least three circuit splits, then, it decided to "rule in" two categories of conduct--bribes and kickbacks--while eliminating a third--undisclosed conflicts of interest. (30) On what principled basis can one argue that the Court is merely adopting a reasonable narrowing construction rather than...