When decisions must be made or reviewed, deference often comes into play. Deference means suspending our own judgment in favor of someone other actor’s already-rendered judgment, agreeing to limit, color, or hold in abeyance some exercise of our own judgment or power because we recognize that the views of other actors may have relevantly superior claims over our own. See Horwitz, Three Facts of Deference, 83 Notre Dame L. Rev. 1061, 1065–66, 1072 (2007).
Courts are quintessential deferers to judgments of other legal actors. The most familiar of these are expressly labeled as deference: Chevron, Skidmore, and Auer deference, for example, by which courts defer to executive agencies. And superior courts, to some extent and on some questions, also defer to the judgments of inferior courts; that’s what standards of review are, de novo review being effectively no deference, while arbitrary & capricious is highly deferential. Similarly, levels of scrutiny can be cast as a species of deference, and so can stare decisis. These are all species of the same genus, and so described, nothing’s more common than for a reviewing court to accept the conclusion of another actor when they might—if magically substituted in to make the original decision—have made another decision.
Justice Scalia criticized the label “deference” as a “mealy-mouthed word . . . not necessarily meaning anything more than considering those views with attentiveness and profound respect, before we reject them.” Judicial Deference to Administrative Interpretations of Law, 1989 Duke LJ 511, 514. So for more illumination, let's look in just a bit more detail at two kinds of deference, Chevron and Skidmore.
Chevron deference teaches that “when an agency is authorized by Congress to issue regulations and promulgates a regulation interpreting a statute it enforces, the interpretation receives deference if the statute is ambiguous and if the agency’s interpretation is reasonable.”1 The court abstains from “deciding th[e] issue conclusively and authoritatively . . . [in favor of] merely deciding whether the agency's [interpretation] . . . is permissible.”2 This deference reaches an apogee
[w]hen Congress has explicitly left a gap for an agency to fill, [i.e. when] there is an express delegation of authority to the agency to elucidate a specific provision of the statute by regulation, and [in that circumstance] any ensuing regulation is binding in the courts unless procedurally defective, arbitrary or capricious in substance, or manifestly contrary to the statute.
United States v. Mead Corp., 533 U.S. 218, 227 (2001) (cleaned up). But it also applies to implicit delegations that are
apparent from the agency’s generally conferred authority and other statutory circumstances that Congress would expect the agency to be able to speak with the force of law when it addresses ambiguity in the statute or fills a space in the enacted law, even one about which Congress did not actually have an intent as to a particular result. . . . [In that circumstance], a reviewing court has no business rejecting an agency's exercise of its generally conferred authority to resolve a particular statutory ambiguity simply because the agency’s chosen resolution seems unwise, but is obliged to accept the agency's position if Congress has not previously spoken to the point at issue and the agency's interpretation is reasonable.
Id., at 229 (cleaned up).
Skidmore deference teaches courts to give leeway to the countless choices and interpretations of agencies that arise in the normal course of business. Given the “specialized experience and broader investigations and information available to the agency,” it may be that “where the regulatory scheme is highly detailed, . . . [the actor] can bring the benefit of [that] specialized experience to bear on the subtle questions in [a particular] case,” Mead, supra, 533 U.S. at 234–5 (cleaned up), the court puts a thumb on the scale. It gives artificial credit to the agency’s reasoned determinations.
This approach is flexible rather than mechanical. “The fair measure of deference to an agency administering its own statute has been understood to vary with circumstances, and courts have looked to the degree of the agency’s care, its consistency, formality, and relative expertness, and to the persuasiveness of the agency’s position.” Id., at 228 (footnotes and citation deleted); see County of Maui v. Hawaii Wildlife Fund, 140 S. Ct. 1462, 1474 (2020) (“we often pay particular attention to an agency’s views in light of the agency’s expertise in a given area, its knowledge gained through practical experience, and its familiarity with the interpretive demands of administrative need”). Thus, the level of deference “in a particular case will depend upon the thoroughness evident in [the actor’s] consideration, the validity of its reasoning, its consistency with earlier and later pronouncements, and all those factors which give it power to persuade, if lacking power to control.”3
Naturally, Skidmore has more traction when we review “a considered judgment . . . rendered on a full record [rather] than . . . [a] series of short opinions on isolated facts which contain no in-depth study of the problem,” Pittston Stevedoring Corp. v. Dellaventura, 544 F.2d 35, 50 (2d Cir. 1976) (Friendly, J.), or when we review the “informed [exercise of] discretion,” Turner v. Safley, 482 U.S. 78, 90 (1987), rather than one that is “not particularly thorough and . . . represents a stark, unexplained change in the [actor’s] position,” Nelson v. Great Lakes Ed. Loan Serv., 928 F.3d 639, 651 n.2 (7th Cir. 2019) (cleaned up), or that bears the marks of draconian or grossly disparate treatment, D’Alessio v. SEC, 380 F.3d 112, 124 (2d Cir. 2004).
At bottom, these doctrines reflect a reconciliation of ourselves to (and accommodation of) a discomfiting truth. Fact is, there’s not always a “right” answer to every question—or, at very least, there are questions on which “it is more important that the applicable rule of law be settled than that it be settled right.”4 Every actor in every legal system must perforce make myriad decisions daily that could be second-guessed. And we intuit that we shouldn’t be in the business of second-guessing. Why?
Sometimes a decision may so defy principled resolution and be so “inherently inexact,”Family PAC v. McKenna, 685 F.3d 800, 811 (9th Cir. 2012), as to be inescapably legislative. (That’s akin to the “weighing incomparables” situation I was describing in Blue Man Scoop.) In such situations, the opinion of a judge is no more “correct” than the opinions of any other actor in the system, and a reviewing judge who refuses to defer to the original choice is doing nothing more than substituting her own opinion for the opinion of another actor, adding perhaps heat but certainly not light. Such deference is maximal when comparable actors have reached comparable conclusions. See id. Why? Because unconscious parallelism suggests that the choice is, if not necessarily “correct” (for there is no “correct” in such matters), nevertheless reasonable.
Sometimes a decision may involve intricate technical questions in specialized disciplines where answers can be counterintuitive. In such situations, proper humility bids the judge recognize that there are things she does not know—or at least, knows less well than the specialists. Horwitz calls this “epistemic” deference, suggesting that “courts defer to other institutions when they believe that those institutions know more than the courts do about some set of issues, such that it makes sense to allow the views of the knowledgeable authority to substitute for the courts’ own judgment.” Horwitz, supra, at 1085. Facing a particularly abstruse or technical statute, deference is “compelled by necessity” insofar as “a court that tries to chart a true course to the [organic] Act's purpose embarks upon a voyage without a compass when it disregards the agency's views.” Ford Motor Credit Co. v. Milhollin, 444 U.S. 555, 568 (1980).
And “to the extent that deference to administrative views is bottomed on respect for agency expertise,” id. at 566 n.9, much the same can be said for deference to the military. The “complex, subtle, and professional decisions as to the composition, training, equipping, and control of a military force are essentially professional military judgments.” Gilligan v. Morgan, 413 U.S. 1, 10 (1973). Military matters are both textually and appropriately confided to the control of the professionals, who which have relevant expertise, and of the political branches, which have accountability, as compared to the courts, which have neither. See id.; accord Chevron USA v. NRDC, 467 U.S. 837, 865–66 (1984).
There are, moreover, any number of matters on which courts defer to executive actors (e.g. operating prisons, see Turner, supra; but see Johnson v. California, 543 U.S. 499 (2005)) or to legislatures (e.g. setting policy for immigration, Kerry v. Din, 576 U.S. 86, 97 (2015)) or state courts (e.g. in collateral procedings, Alexander v. Cockrell, 294 F.3d 626, 630 (5th Cir. 2002) (per curiam) for reasons substantially answering to institutional settlement. Yes, it’s emphatically the province and duty of the judiciary to say what the law is, Marbury v. Madison, 1 Cranch 137, 177 (1803), but when law runs out and the corpus juris offers no answer, why should the opinion of someone with the power to impose it control over the opinion of someone with accountability for it? At that point, you have to fall back on something, whether that’s a value of institutional settlement,5 or a presumption of private ordering,6 or anything else. Deference is one of the things to which we have recourse, and it is an appealingly self-abnegating one. Call this “positional,” “coordinate-branch,” or “legal authority” deference. Cf. Horwitz, supra, at 1080 ff.
All these factors can conspire together and may interlock. For example, foreign affairs, particularly intelligence and national security, are textually confided to a coordinate branch, involve decisions to which there is no single correct answer and in which delicate matters of judgment are intrinsic, which of their nature benefit from technical expertise and secret information known (and which can be known) to only a privileged few, and where the need of “[d]ecision, activity, secrecy, . . . despatch,” The Federalist, no. 70 (Hamilton), and finality vastly outweighs any improvement that second-guessing might, speculatively, engender.7 To armchair quarterback such decisions on anything but the weightiest considerations is foolishness.
Deference, of course, is not obedience, see Horwitz, supra, at 1075–77, and deference is not absolute. Deference to state court fact-finding means a presumption of correctness so long as they have “fair support.” Alexander, supra, 294 F.3d at 630. Deference to an ALJ’s credibility determination requires that it be supported by “specific, cogent reasons.”8 Deference to administrative agency’s adjudications requires that its “factual findings . . . [be] supported by substantial evidence” and that its “depart[ures] from established precedent . . . [have] a reasoned explanation . . . .”9 A finding should have record support, of course,10 and in reviewing determinations based on findings, reviewing courts should defer to a departure from what those findings would allow only when the trial judge “at least explain[ed] her reasons.” Gettings v. Building Laborers Local 310, 349 F.3d 300 (6th Cir. 2003); see also Kisor v. Wilkie, 139 S. Ct. 2400, 2414 (2019) (Auer deference is unwarranted “when a court concludes that an interpretation does not reflect an agency’s authoritative, expertise-based, ‘fair[, or] considered judgment’” (alteration in original)).
As I’ve said before, American legal concepts—yes, even AdLaw—contain wisdom with application far beyond the law. I believe that deference is particularly appropriate when any or especially all of the following are true: When the actor is a specialist and is speaking on questions wholly within their specialty, when an actor is close to the action and their proximity and perspective gives rise to the inference that they have superior access to relevant information, and when an actor’s incentives strongly align with correct decisionmaking. For example, deference is my response to Ordinatio sacerdotalis (J.P.2 Ap. Ep., 1994). As an original matter, I might have reached a different conclusion as to the ordination of women to the presbyterate. Nevertheless, largely for reasons of positional deference and to some extent for reasons of epistemic deference, I give way to John Paul II’s contrary determination.
§
All of which brings us to what is happening in the Holy Land. Last week, Hamas—a Palestinian paramilitary organization that is the kinda-sorta government of the Gaza Strip—launched a massive missile strike against and ground invasion into Israel. This was no random impulse or reactive strike, it was done in the coldest blood imaginable: Hamas claims that the attack was two years in the making. I commend to you three helpful assessments, this by Caspian Report, and, sequentially, this and this by Simon Whistler at Warographics.
It is true (understatement, actually) to say that the region has complicated history. But only those willing to forget history are not doomed to repeat it. Imagine a gang of Lakota Sioux, claiming revenge for the Trail of Tears, murders 1300 innocent bystanders and takes more as hostages for torture. Assessing that action, which fact would we think more salient: That the Trail of Tears happened, years ago, or the fact that you just murdered and kidnapped hundreds of people, like, right now?
Whatever happened in the past, this is what’s happening now. Attempts to force the focus open wider than the most recent events are particularly unavailing because if we are to consider Israel’s past conduct, we would also be compelled to consider Hamas’ past conduct. (Hamas apologists tend to demand a wide but soft focus, bringing in everything exculpatory and nothing inculpatory. We can’t do that.) It is not at all clear that Hamas will benefit from such a calculation. Nor is it clear that context matters when the actions being contextualized cannot be justified by any any amount of context. “You murdered my brother” might excuse violence; “you murdered my cousin” doesn’t, and “your cousin murdered my cousin” is right out. Excessive emphasis placed on the conflict’s history is less often about explaining than it is about excusing—sometimes excusing what Hamas did, but often excusing the speaker from taking a position. “Well,” they say, “it’s a complicated situation.” It is, yes, but that’s not a get-out-of-hard-judgments-free card.
In a similar vein are the claims that, even stipulating that the focus is properly on what’s happening now, “we don’t have the full story.”11 This tends to go hand in hand with both-sides-ism, which, again, is usually pretext for not taking sides, which in turn, frankly, is usually a veiled refusal to take Israel’s side. Insisting that we ignore the fact that this attack was conducted by Hamas and instead, generically, “pray for peace in the region” and “urge restraint on all sides” strikes me as like insisting “all houses matter” when there is one specific house that is presently on fire, or “all lives matter” when it is black lives specifically that are under immediate threat. Reciting generalizing bromides at a moment dominated by a specific concern is nothing more or less than distraction and minimization. It’s a little like people insisting, “you’re not really pro-life unless you support a massive cradle-to-grave welfare state.” I half expect the speaker to pull a rabbit from a hat and shout “ta da!” when they finish that one. These aren’t arguments, they’re ploys. Rhetorical tricks. They’re attempts to use labels to distract from the core point (viz., don’t murder), boosting the level of generality and shifting to a different point by threading the label orthogonally through both as an axle.
Hamas has played its card, and now it’s Israel’s turn. Israeli commanders are vowing to “demolish” Hamas. They disclaim negotiation and intend to “remove” Hamas from power. They say the strikes will be a “coordinated” “invasion.” What does that mean, concretely? What will happen next? I don’t know. But as you will likely infer from the above discussion about deference, I will defer to what Israeli commanders deem necessary.
After the Holocaust, the cry was, “Never Again.” I believe that. I also believe that it’s an ordering principle; centuries-long experience has proven that without a Jewish state, Jews will be vulnerable to an escalation of antisemitic hatred into antisemitic mass murder. Thus, Israel is the actualization of “Never Again,” its ne plus ultra if not its sine qua non. Thus, Israel must exist, and to exist, it must be capable of responding to the attacks of enemies who are now and have always been determined to destroy it, seeing Israel as “the Invaders,” see Hamas Cov’t 1988.
Even before last week’s assault, no one could think that antisemitic hatred was a thing of the past. Make no mistake: Israel’s enemies do not declare against Israel alone. They declare that “[o]ur struggle against the Jews is very great and very serious.” Ibid. (emphasis added). It’s oozing back into American culture, too, and “[a]ntisemitic incidents reached an all-time high in our nation in 2021.”12 You have the already-disgraced former American President dining with holocaust deniers, and a prominent member of his party claiming that Jews use space-lasers to start wildfires.13 You have the leader of Hungary spouting openly antisemitic rhetoric, decrying Jews as “[dis]honorable . . . [and] unprincipled; they are not national, but international; they do not believe in work, but speculate with money; they have no homeland, but feel that the whole world is theirs.”14 And lurking just below populist bloviations about “globalization,” we invariably find a long list of—it’s just a coincidence, we promise!—Jewish names. (NARRATOR: It was not a coincidence.)
I am as disinclined to second-guess those responsible for defending Israel as I am to second guess “Bomber” Harris and his masters as to Dresden. The so-called “debate about the morality of the bombing of Dresden” is much like the so-called “debate” over America’s bombing of Hiroshima and Nagasaki: Quite apart from any causal relationship, the armchair quarterbacks assume symmetrical knowledge that is usually absent in war and even foreknowledge if not omniscience that is always absent. Reading a cold record after the fact, we often fail to realize that we, unlike those who were involved, know how and when it worked out. Harris and Truman did not drop their bombs knowing victory was just around the corner. And their critics doubt, but do not know, that victory wasn’t in part the result of those bombs.
I don’t know whether Dresden was “necessary”; neither do the people who want to armchair-quarterback the decision. I have an opinion about whether Fat Man and Little Boy were necessary; that plus the ticket price at Fenway will get you into Fenway. Of more relevance: The people whose tuchuses—tuchi?—were on the line thought it was necessary. They were in a better position to assess its need and they had better incentives than I do to make the right call. Deference to their judgment is appropriate.
Some of what they will do is probably not what I would do. And they will, inevitably, be accused of going too far—no matter how far they go vel non. (That accusation will come from people who sub silentio support Hamas and it will come from people who are pontificating from very, very far away. But I hope that it will not come from Kyiv or its allies. The far-off pontificators would be appalled at the idea that we should armchair quarterback Ukraine’s response to Russian invasion! They should ask themselves why they think it’s different.) Nevertheless, giving the space to decide must, at least sometimes, include space to decide wrong.15 From the cozy safety of five thousand miles away and with no skin in the game, I decline to second-guess the decisions of Israeli commanders as to securing Israel’s safety. That latitude is not unlimited, but it is broad.
§
Notes & Queries.
Erratum: Last week’s newsletter left behind an artifact of hasty redrafting. I said: “She seems tough, competent, and driven; she is yes, receiving the Peridean Night Sisters’ dreams would have been somewhat luck-of-the-draw, but even if anyone could do that, Morgan figured out how to get Thrawn’s exact location . . . .” Recte, strike “she is.”
Last week’s newsletter was an interesting exercise. We had multiple travel obligations and a wedding to attend on Sunday. (Congratulations to Carrie and Josh.) Consequently, on Sunday morning, I had nothing written and no idea what to write. Driving to the wedding, my wife and I were listening to ATP and arguing about it, and it occurred, okay, this is the newsletter. Add some reflections on Ahsoka’s finale and you’ve got a brisk one, but I think a pretty good one.
This week’s cover image was photographed by Katangais and released under a CC BY-SA license.
Cornucopia.
Thrawn: A legend reborn. And, while we’re in a Star Wars mood, a defense of the opening movement of The Phantom Menace to which I can subscribe.
We finished Andor, by the way. It was good. I wouldn’t say it was so obviously better than The Mandalorian or Ahsoka as reviews have suggested, but it was good. One thought struck me: I may read one story beat very differently to how most people are reading it. Cassian, already on the run for a manslaughter and a straight-up murder, participates in what I will for brevity characterize as a robbery of the Imperial version of a Federal Reserve bank. He escapes with his loot, notching up some more murder in the process. The Empire, peeved by the robbery, announces a crackdown. Later, by sheer bad luck, Cassian is caught up in the crackdown. He’s living under an assumed name, Keef, and after being arrested for looking at a cop wrong, he receives a hefty sentence in a diabolical prison. I think that we’re supposed to take this as a demonstration of the Empire’s arbitrary (stochastic, even) cruelty. Here’s the thing, though: Neither the authorities who picked up Keef nor the judge who sentenced Keef realize that “Keef” is actually Cassian, the object of a KGB—uh, ISB—manhunt, certainly a thief, arguably a terrorist, and at very least a murderer several times over. But we know it. The arrest and sentence are grossly unfair and disproportionate to Keef’s crimes, but not to Cassian’s, and Keef is Cassian. They’re the same man. This strikes me as akin to nailing Capone on tax evasion or Trump on fraud. When someone is accused, society owes them due process. But when someone has in fact gotten what they deserve, society owes them little sympathy.
How Taiwan will foil China. (Yes, I know each side has claims on the word “China,” and I don’t care. Plain language. We all know what we’re talking about.)
Encino Motorcars v. Navarro, 579 U.S. 211, 220 (2016).
NLRB v. Food & Commercial Workers, 484 U.S. 112, 134 (1987) (Scalia, J., concurring) (cleaned up).
Skidmore v. Swift, 323 U.S. 134, 140 (1944); accord Bittner v. United States, No. 21-1195, at *14 (Feb. 28, 2023).
Burnet v. Coronado Oil & Gas Co., 285 U.S. 393, 406 (1932) (Brandeis, J., dissenting); cf. Bousley v. United States, 523 U.S. 614, 629–30 (1998) (Scalia, J., dissenting).
See Hart & Sachs, The Legal Process 3–4 (Eskridge & Frickey, eds. 1994) (constitutions allocate decisions to particular actors and “decisions which are the duly arrived-at result of duly established procedures . . . ought to be accepted as binding”); Fallon, Reflections on the Hart and Wechsler Paradigm, 47 Vand. L. Rev. 953, 964 (1994).
Cf. Easterbrook, Statute’s Domains, 50 U. Chi. L. Rev. 533 (1983).
Cf. Loving v. United States, 517 U.S. 748, 768 (1996); Department of the Navy v. Egan, 484 U.S. 518, 529 (1988); Parker v. Levy, 417 U.S. 733, 743 (1974); C.S. Air Lines v. Waterman S.S. Corp., 333 U.S. 103, 111 (1948); Youngstown Sheet & Tube Co. v. Sawyer, 343 U. S. 579, 645 (1952) (Jackson, J., concurring).
Pastora v. Holder, 737 F.3d 902, 907 (4th Cir. 2013).
NLRB v. Sw. Regional Council of Carpenters, 826 F.3d 460, 464 (D.C. Cir. 2016).
United States v. Lozada-Rivera, 177 F.3d 98 (1st Cir. 1999).
Just as predictable and a lot more meretricious were the attempts to blame it all on a reaction to recent statements by the Israeli Prime Minister as to possible Two State Solution maps—or, more fantastically yet, to insinuate that Israel itself instigated the attack based on the same statements. The former founders on Hamas’ claim that the attack took two years to prepare, and the latter is conspiratorial tosh for which there is zero evidence. (In the United States, nothing is more common after an incident than the conspiracy theory people claiming that it is a false flag. It is not just in America that we find the paranoid style.)
Houseman, Antisemitism at New All-Time High in US, American University College of Arts & Sciences News, Jan. 13, 2023.
Hagen, Antisemitism is on the rise, NPR, Dec. 1, 2022.
Soll, The Making of an Anti-Semitic Myth, The New Republic, April 10, 2019.
Cf. Fallon, supra, at 962.