U.S. Appeals Court Judge Neil Gorsuch delivering prepared remarks before a group of attorneys in Denver on Jan. 27. (David Zalubowski/Associated Press)

I’m delighted to hear that Judge Neil Gorsuch has been nominated to the Supreme Court; he and I clerked at the Supreme Court the same year (he clerked for Justices Byron R. White and Anthony M. Kennedy), so I got to know him then and have stayed in touch with him since. I think he’s brilliant, thoughtful and temperate; he has the intellect of Justice Antonin Scalia, but not Scalia’s occasional irascibility — he comes across more Roberts-like than Scalia-like, I would say. He also has, I think, a flair for the written word just as Scalia did, though it manifests itself in a somewhat different style.

Without doubt, he’ll be a conservative, and would maintain the court’s 5-4 mostly conservative philosophy. Just how conservative he will be, and how he will come out on particular matters, I don’t know. But rather than speculating, I thought I’d post an excerpt of one of Gorsuch’s opinions, one that I think well captures his manner and that involves an interesting and accessible subject: Yellowbear v. Lambert (10th Cir. 2014), a case on a prisoner’s religious freedom. You’ll see Gorsuch’s willingness to step back and explain the background principles behind the law, which I think would be especially helpful in Supreme Court cases, which are often read by nonspecialists.

I think you’ll also see some of Gorsuch’s approach to counterarguments, as well as his willingness to demand proof and not just assertion from the government, when the legal test calls for it. In any case, have a look and see for yourself:

Andrew Yellowbear will probably spend the rest of his life in prison. Time he must serve for murdering his daughter. With that much lying behind and still before him, Mr. Yellowbear has found sustenance in his faith. No one doubts the sincerity of his religious beliefs or that they are the reason he seeks access to his prison’s sweat lodge — a house of prayer and meditation the prison has supplied for those who share his Native American religious tradition. Yet the prison refuses to open the doors of that sweat lodge to Mr. Yellowbear alone, and so we have this litigation. While those convicted of crime in our society lawfully forfeit a great many civil liberties, Congress has (repeatedly) instructed that the sincere exercise of religion should not be among them — at least in the absence of a compelling reason. In this record we can find no reason like that….

Our story starts with Smith. In Employment Division v. Smith (1990), the Supreme Court held that the Constitution’s Free Exercise Clause does not exempt religious persons from the dictates of neutral laws of general applicability. The devout must obey the law even if doing so violates every article of their faith. When Smith was handed down, some worried that it upset existing free exercise doctrine dating back to Sherbert v. Verner (1963). In Sherbert and its progeny the Supreme Court had suggested that no law, not even a neutral law of general applicability, may “substantially burden” the exercise of religion unless that burden amounts to the “least restrictive means” of achieving a “compelling governmental interest.” … What protections Sherbert appeared to afford religious observances, Smith appeared ready to abandon.

Concerned with just this possibility, worried that Smith left insufficient room in civil society for the free exercise of religion, Congress set about the business of “restoring” Sherbert, at least as a matter of statute. It opened its efforts with the Religious Freedom Restoration Act of 1993. … Passed nearly unanimously, RFRA was (and remains) something of a “super-statute.” … It instructed that all forms of governmental action — state or federal — had to satisfy Sherbert‘s test or risk nullification.

But as it turned out, this marked only the opening lines in what proved to be a long dialogue between Congress and the Court. In City of Boerne v. Flores (1997), the Court held that RFRA stretched the federal hand too far into places reserved for the states and exceeded Congress’s Section 5 enforcement authority under the Fourteenth Amendment. As a result, the Court held RFRA unconstitutional as applied to the states, though still fully operational as applied to the federal government. …

Undaunted, Congress reentered the field soon enough, this time with the Religious Land Use and Institutionalized Persons Act of 2000. In RLUIPA Congress invoked not just its Fourteenth Amendment but also its Spending Clause powers to (re)impose Sherbert‘s balancing test on state action — though now state action in only two specific arenas, arenas in which Congress found the record of religious discrimination particularly clear and compelling. First, in the land use context, where churches are sometimes disfavored by local zoning boards because (among other things) church members are said to generate “too much” traffic or congestion or noise when they gather for communal expressions of faith. Second, in the prison context, where it is so easy for governmental officials with so much power over inmates’ lives to deny capriciously one more liberty to those who have already forfeited so many others. … This time Congress acted unanimously and this time the Court upheld its effort, at least against a facial challenge under the Establishment Clause. …

That takes us to the nub of our case. Mr. Yellowbear, an enrolled member of the Northern Arapaho Tribe, seeks access to the prison’s existing sweat lodge to facilitate his religious exercises. The prison has refused. The prison’s sweat lodge is located in the general prison yard and Mr. Yellowbear is housed in a special protective unit (not because of any disciplinary infraction he has committed, but because of threats against him). Prison officials insist that the cost of providing the necessary security to take Mr. Yellowbear from the special protective unit to the sweat lodge and back is “unduly burdensome.” Mr. Yellowbear disagrees and seeks relief under RLUIPA. For its part, the district court discerned no statutory violation and entered summary judgment against Mr. Yellowbear. Mr. Yellowbear asks us to undo that judgment so that his case might proceed to trial.

At the end of the day, we find that’s exactly the relief we must provide.

I omit here most of the application of the law, which is important and readable but long — interested readers can see it here. Here, though, is one key passage:

Retreating to the argument it did pursue in the district court, the prison argues that its ban on Mr. Yellowbear’s access to the sweat lodge is warranted because moving him back and forth between the protective custody unit and the sweat lodge would require officials to “lock[] down a significant portion of the facility” to ensure Mr. Yellowbear isn’t placed in “contact with other non-protective custody inmates” who might seek to do him harm. And this ferrying business, the prison says, has cost and administrative implications. Of course, cost and administrative considerations are often intertwined and they certainly are here. This isn’t (again) a case in which the prison has evidence suggesting an inability to provide adequate security at any price. In the prison’s view, a lock down can supply an acceptable level of security — but only at some marginal cost it considers too high.

Two problems, it seems to us, arise here, preventing us from saying that the government has carried its summary judgment burden of establishing a compelling interest as a matter of law.

In the first place, the prison does not even attempt to quantify the costs it faces, let alone try to explain how these costs impinge on prison budgets or administration. Instead, the prison simply asserts, flatly and without more, that the marginal costs are “unduly burdensome.” But conclusory legalese … does no more to prove a compelling interest than post-hoc rationalizations unsupported by record evidence. RLUIPA’s compelling interest test is a strict one: Congress borrowed its language from First Amendment cases applying perhaps the strictest form of judicial scrutiny known to American law. That test isn’t traditionally the sort of thing that can be satisfied by the government’s bare say-so. …

To be sure, the Supreme Court has suggested that RLUIPA’s “compelling governmental interest” test holds an unusual twist in the prison context. Though RLUIPA uses the same linguistic formulation for prison and land use cases, though it uses the same language found in RFRA, the Court has told us that “context matters.” … So while courts do not usually afford much deference to the government when assessing whether its claimed interest is a “compelling” one in RFRA or even land use cases arising under RLUIPA … in the prison context the Supreme Court has instructed us to apply “due deference to the experience and expertise of prison and jail administrators in establishing necessary regulations and procedures to maintain … security … consistent with consideration of costs and limited resources” … How all this fits together raises some interesting questions.

But this much is clear and clearly enough for our purposes: the deference this court must extend the experience and expertise of prison administrators does not extend so far that prison officials may declare a compelling governmental interest by fiat. At summary judgment, the government must do more than assert an “undue burden” or a “compelling interest.” When weighing the existence of a compelling interest, the deference due prison administrators may be enough to nudge a questionable case across the line, but it doesn’t mean prison officials get to recognize compelling interests on their own. If that were the case, RLUIPA’s supposedly strict judicial scrutiny would become no judicial scrutiny at all. …

Neither is this the end to the problems confronting the government’s lock down rationale. It faces at least one more. Mr. Yellowbear has attested that prison-wide lock downs already occur on a daily and sometimes hourly basis to transport “specially housed” inmates — inmates in the geriatric unit, the women’s unit, and (like himself) the protective custody unit — to other parts of the prison, including the medical wing. Mr. Yellowbear has presented affidavits, as well, from other inmates stating much the same thing. All of these uncontested statements we must take as true at summary judgment. And they can’t help but suggest this question: If lock downs can be accomplished on a daily basis consistent with the prison’s administrative concerns to facilitate inmates’ medical needs, what compelling interest is served by refusing any lock downs ever to facilitate an inmate’s religious needs? Put differently, why is this religious exemption offensive to the prison’s putatively compelling no-lock-down interest when other secular exemptions are not?

A law’s underinclusiveness — its failure to cover significant tracts of conduct implicating the law’s animating and putatively compelling interest — can raise with it the inference that the government’s claimed interest isn’t actually so compelling after all. As the Supreme Court has said, it’s sometimes hard to see how a law or regulation can “be regarded as protecting an interest of the highest order,” as serving a compelling interest, “when it leaves appreciable damage to [the] supposedly vital interest unprohibited.” … Evidence that the prison grants secular exceptions more readily than religious exemptions to a putatively compelling policy can raise the inference, too, that its most compelling interest may actually be discrimination against, or at least indifference to, the religious liberties of incarcerated persons — precisely the scenario RLUIPA identified as too prevalent in our society and sought to redress. …

At the same time, it is important to acknowledge that inferences like these are not inevitable or irrebuttable. We know that few statutes pursue a single purpose at any cost, without reference to competing interests. … Given this, it would be odd if the mere fact that a law contains some secular exceptions always sufficed to prove the government lacked a compelling interest in avoiding another exception to accommodate a claimant’s religious exercise. If that were the case, the compelling interest test would seem nearly impossible to satisfy and RLUIPA’s suggestion that religious accommodations must sometimes give way to other competing governmental interests would appear close to a dead letter.

When exactly is (and isn’t) underinclusiveness enough to undermine the credibility of a claimed compelling interest? That question takes us into territory not yet fully charted, but at least this much we can say with confidence. A government can rebut an argument from underinclusion by showing that it hasn’t acted in a logically inconsistent way — by (say) identifying a qualitative or quantitative difference between the particular religious exemption requested and other secular exceptions already tolerated, and then explaining how such differential treatment furthers some distinct compelling governmental concern. …

Problem is, in this case the prison has made no effort (of any kind) to respond to Mr. Yellowbear’s underinclusiveness evidence. Mr. Yellowbear’s evidence about the regularity of secular lock downs dangles out there unanswered, inviting a reasonable fact finder to infer that the prison’s desire to avoid lock downs isn’t really as compelling as it insists, and that the prison may even harbor some animus or at least indifference to his religious exercise. …

So far the parties have contended at the level of absolutes. Mr. Yellowbear has sought some access to a sweat lodge. The prison has refused any access. But what happens if (or when) the discussion turns to questions of degree? What happens if the marginal burden on Mr. Yellowbear’s religious exercise becomes much less significant—the prison allows him periodic access, say, but not as often as his faith suggests—and the marginal cost to the prison much greater—because ever more visits mean ever more expense? Surely the relative strength of the two parties’ interests would appear quite different in that new light. Surely as well RLUIPA must account for that. Even accepting that the imposition on Mr. Yellowbear in this scenario might still qualify as a “substantial burden” (the government’s policy would still prohibit that which his faith compels), the change in the relative strength of the parties’ positions would reveal itself at later stages in the doctrinal analysis, with the government’s interests now arguably appearing more “compelling” and its policies more “narrowly tailored.” For now, however, these subtler (and admittedly more difficult) questions remain for the parties and district court to consider on remand. As litigated to us, the burden on Mr. Yellowbear’s religious exercise is high (no access of any kind, ever, to a religious exercise) and the cost to the prison left undefined by the record and thus presumably low. In these circumstances, we don’t doubt a reasonable trier of fact could find a RLUIPA violation.

The district court’s grant of summary judgment on Mr. Yellowbear’s RLUIPA claim seeking prospective injunctive relief … is vacated and this case is remanded for further proceedings consistent with this opinion.