The Nemo Iudex Principle and the Presumptions of the Law

A Response to a Reader’s Comment

In response to my recent post against the idea that the separation of powers is a necessary concomitant of the rule of law, one of our readers, Graham, offered an interesting suggestion: I ought to have discussed the venerable legal principle nemo iudex in sua causa — no man may be judge in his own cause. Wrote Graham: “I certainly agree that the rule of law does not require separation of powers in every case, nor does it require American-style institutions. There are also cases where Americans would insist on separation of powers where this principle has no force - i.e., the prime minister being a member of parliament who helps make the laws he enforces. And none of this precludes departures from the nemo iudex principle in specific cases (as, per [Richard] Helmholz, often happens with natural law). I would tend to agree that administrative agencies' in-house tribunals are an appropriate example of such a case. But as a default rule, it seems a magistrate should not be judging the strength of his own case, and it seems like a principle that deserves discussion in a post about separation of powers.” The suggestion is well-taken and deserves a better response than a mere reply in a comment thread, and the response may, I hope, be of more general interest; hence this separate post.

The main reason I did not discuss the nemo iudex principle is that I do not believe it is actually implicated — not even as a default rule — by the legal contexts and institutions the post addressed, including the combination of powers in the administrative state. Indeed quite the reverse is true. To understand why this is so, we have to focus on the basic distinction between public and private interests, a distinction that our law draws in clear terms.

Take for example the situation in which - to mention a core case, recently controversial in American libertarian legal circles - federal administrative agencies combine the powers to make rules, initiate enforcement actions under those rules, and then adjudicate whether the rules have been violated, at least at the top level of the agency heads (note that initial formal adjudication is typically carried out by administrative law judges who are to some degree independent of the agency). In such cases, the agency is, in contemplation of the law at least, not judging in its own case. It is judging upon so-called public rights and is presumed to act in the public interest, as a delegated agent of the public. From the extrinsic standpoint of the political scientist or other external observer of the state, who is not unreasonably worried about administrative self-dealing, that legal presumption doubtless appears as, at best, an idealized legal fiction. But from the lawyer’s internal perspective, it is foundational to public law, and is reflected in an array of legal doctrines that underlie and structure the administrative state and indeed the whole constitutional order. Mr. Madison may have discoursed in some Federalist Paper or other about officials promoting the interest of their own branches, rather than the public interest, but our operative law has, right from the beginning and consistently, taken a different approach - one based not upon suspicion and a presumption of distrust, but on the legal presumption that officials act in the public interest unless the contrary is clearly shown. To question this presumption would be to undo much of the fabric of public law.

Some examples:

(1) The very reason Congress can authorize agencies to decide cases such as Federal Trade Commission v. Smith, or Securities and Exchange Commission v. Jones, as opposed to placing initial decision in the federal courts, is that the law calls them cases of “public right.”

(2) Under the “record rule,” a demanding standard applies to litigants who hope for discovery of decisional materials from agency officials, in order to supplement the administrative record; the challengers must make an initial “strong showing of bad faith or improper behavior,” which they can rarely do precisely because they would typically need discovery in order to do so. The standard, in other words, all but conclusively presumes that officials are acting properly and in good faith, in the public interest.

(3) In Carter v. Carter Coal Co., the Court observed that delegations to private parties are systematically suspect in a way that delegations to the executive are not, because the recipients of the latter are “official[s] or an official body, presumptively disinterested.” This approach followed the logic of one of the foundational American cases on delegation of authority, Schechter Poultry, which treated delegations of authority to private parties as especially suspect.

And so on. Now obviously agency action is usually (although not always) subject to eventual judicial review for legality, for the validity of agency factfinding, and for the rationality of the agency’s decision. But the ground of such review is not that there is some inherent or systematic reason for suspicion when agencies judge “in their own cause”, because in contemplation of the law, that is not what occurs.

In such cases, the idea that nemo iudex supplies the default rule is backwards. The law instead presumes that officials act disinterestedly within the scope of their official duties, or rather act in the public interest, and it takes the strongest evidence to the contrary to overcome that presumption. All these principles and doctrines are really just aspects of the venerable “presumption of regularity,” which has been part of our law from the outset, applies in both civil and criminal domains, “supports the official acts of public officers,” and dictates that“[i]n the absence of clear evidence to the contrary, courts presume that public officers have properly discharged their official duties.” This master principle of public law, which takes a variety of subsidiary forms, has ancient roots; Conor Casey has explained that a powerful presumption of regularity (of course expressed in varying but equivalent terms in varying times and places) is itself an inheritance and centerpiece of the classical legal tradition.

Put differently, Graham’s idea, although apparently plausible and certainly worth discussing, slips too quickly between two very different problems — perhaps because both are lumped under the doctrinal rubric of “due process” in American law, although the law treats the two types of problems quite differently. One sort of problem involves situations in which officials decide upon cases in which their personal or private interests are implicated, especially if money is involved — the locus classicus in American law being Tumey v. Ohio, in which a village mayor personally received fees and costs from defendants upon conviction, but only upon conviction, in the mayor’s court. The risks of self-dealing in such cases are obvious; these really are cases in which the decisionmaker judges in his own cause, and the constitutional law of due process has rightly placed sharp restraints on this kind of act — with, however, conspicuous exceptions, as when legislators set their own salaries or judges decide on the constitutionality of putative reductions of their own salaries, a point to which I return shortly. The category of cases discussed above and in the original post, however, are quite different. They involve officials judging as delegated representatives of the public. In that type of case, even where agencies enjoy combined powers, and thus enforce and adjudicate under legal standards they themselves have created through rulemaking, the issue is not personal or private interest but a claim that officials will be biased by actions previously undertaken in their official capacities. Here both due process and associated bodies of law are far more forgiving; the Court has repeatedly allowed wide latitude for such arrangements. Unsurprisingly it has based that latitude on a version of the presumption of regularity, saying that claims of bias arising from the combination of powers in agencies generally fail to “overcome a presumption of honesty and integrity in those serving as adjudicators.”

The legal fallacy at issue here, the slippage between private and public interests and the resulting conflation of two entirely different aspects of due process, also infects the sort of arguments that Philip Hamburger has recently offered in an effort to show that deference to the reasonable legal interpretations of administrative agencies allows a kind of agency-self-dealing — in effect making one of the players into the referee. Hamburger sees this as a violation of due process. When an agency enforces public law against a private party, however, whether that law is statutory or regulatory, the law presumes the agency to act not in its own interest as a competitor, but as a referee for the polity as a whole against those who violate the rules that the polity, through its representatives, has laid down. (Where of course an agency happens to have genuinely proprietary and competitive economic interests as against private parties, rather than sovereign interests as a delegate of the public, the situation is different by virtue of Carter Coal). Hamburger short-circuits this axiom of public law with the tendentious claim that “when [courts] defer to agency interpretations of the law,” they are licensing a “systematic bias in favor of the government and against Americans, thus denying them the due process of law.” The claim is tendentious because agencies acting on the basis of otherwise-valid delegations are presumed by the law to speak and act on behalf of the very Americans who, through their representatives, created the delegations in the first place. In contemplation of the law, “the government” is us. Hamburger is also a skeptic of delegation, but that is an entirely different question and one that is conceptually and legally independent of the due process claim that Hamburger has offered. In short, in order to know whether the principle nemo iudex in sua causa is even implicated, one must be careful — more careful than Hamburger has been — to identify whose cause, exactly, the relevant official is judging.

Even were the nemo iudex principle implicated, that principle is hardly absolute, as I wrote some years ago. When it does apply at all, it is at best a defeasible consideration; like other legal principles, it has a dimension not only of scope but also of weight, and can be overborne by the local priority of other, more pressing principles in given contexts. Even in core cases that might well be thought to implicate self-dealing as to private interests, the constitutional order sometimes supposes that self-dealing is tolerable or even affirmatively desirable, as in the cases of legislators and judges voting (under various legal rationales) on their own salaries - the theory being either that the constitutional vesting of the power of self-dealing in such bodies is a necessary means of self-defense, or conversely simply that to entrust the final power to make such determinations in some other institution would allow too much scope for encroachment and institutional aggression.

More broadly, nemo iudex is a treacherous principle in any complex legal order, because violations of the principle are eventually unavoidable. Federal courts pass on the limits of their own jurisdiction, and the Supreme Court claims the power to say with finality what the law is, even the law bearing on its own power and constitutional status. In the end, the power to finally determine legal rights must be vested somewhere or other, and whichever person, body, or institution does so will necessarily be judging on the limits of its own power. Taken to its limit, nemo iudex would make sovereignty impossible. It is a useful principle for the design of subsidiary institutions within the legal system, but it is necessarily defeasible and cannot be fully generalized - a good servant, but a bad master.