Saturday, July 7, 2018

Judge Kavanaugh's Curious Behavior in Garza

Before I properly begin this post, I'll start with a summary so that people know why they're reading it.

(1) In Garza v. Harganthe detained minor immigrant abortion case in which Judge Kavanaugh initially ruled, before being vacated en banc, that the government had eleven days to help a detained minor immigrant find a sponsor before potentially having to temporarily release her from detention without a sponsor to get an abortionJudge Kavanaugh declined to adopt a much broader anti-abortion argument that the government made.

(2) That argument was that even if the minor didn't find a sponsor to take her out of custody, the government would not unduly burden whatever abortion rights she had by barring her from getting an abortion so long as she refused to voluntarily depart the country, and that in any event, compelling the government to release her from detention to get an abortion would unlawfully force the government to facilitate or be complicit in abortion.  

(3)  That argument was the government's primary argument.  The argument Judge Kavanaugh adopted was an argument the government weakly advocated in its opening brief, never mentioned in its reply brief, and actually resisted at oral argument.

(4) At oral argument, Judge Kavanaugh completely understood the government's arguments and their priority, was obviously and deeply skeptical of the government's primary argument, and spent most of the oral argument attempting to talk the government into satisfying itself with a much narrower position it didn't want to settle for.

(5) Later that day, Judge Kavanaugh wrote an order, and several days later, wrote a dissent, in which he never mentioned the government's real position and all but denied that the government had taken the position it had taken—thereby avoiding the appearance of rejecting a broader anti-abortion position than the narrow one he was willing to accept.

With that said, I'll start this post with a personal anecdote.  Some years ago, the summer Judge Jeffrey Sutton and a Sixth Circuit panel were deciding a challenge to the Affordable Care Act's individual mandate, I befriended a number of liberal Sixth Circuit clerks.  On a balkanized court like the Sixth Circuit of, say, 2002–14, there can be a fair amount of distrust of judges across the aisle, at least on the part of staff if not the judges themselves.  And in those days, Judge Sutton was perhaps the preeminent candidate for a Supreme Court vacancy should President Obama lose reelection and a vacancy open up during his replacement's administration.  (At the time, my dark horse for the next appointment, which I was convinced would be made by Rick Perry, was recently outgoing Texas Solicitor General Ted Cruz.)

In this environment, and given Judge Sutton's excellent chances at elevation, at least a couple of my friends were convinced that Judge Sutton would find some way to avoid opining on the merits of the challenge to the mandate.  In that world, when the Democrats still held the Senate and would still have been able to filibuster a nominee if they lost it, the argument went that supporting the mandate might cost him the nomination (though there was still some hope that the Supreme Court would deem the challenge to the mandate so fringy that a Republican President wouldn't fault him for it), but opposing the mandate would certainly cost him confirmation.  

My friends were quite shocked, then, when Judge Sutton opined at great length that the mandate was constitutional—so shocked, in fact, that they initially insisted he really had ducked the merits by only holding that the plaintiffs' facial challenge failed because the mandate could be sustained in at least some applications, e.g., applications to people subject to state insurance mandates.  Which was all he held, but his dicta left no doubt that he rejected the activity/inactivity distinction at the heart of the challengers' argument, and were probably the most devastating critique of that distinction that any judge or lawyer put to paper during the entirety of the litigation.  

As persuasive as those dicta were, however, they failed to persuade Judge Sutton's former boss, Justice Scalia, or Justice Thomas, or Justice Alito, or Justice Kennedy, or even Chief Justice Roberts, who—never let it be forgotten!—completely embraced the activity/inactivity distinction and opined that a purchase mandate would have been unconstitutional had Congress enacted one.  (To be fair, maybe that's because the opinion was written almost entirely in rhetorical questions—seriously, there are twenty-five of themand sentences that began with "how strange" or "is it not strange.")  So it was that Judge Sutton's eloquent (if rhetorically unconventional) rejection of what once seemed a fringe theory became conservative apostasy, and so it is that today Judge Sutton is less of a Supreme Court contender than a junior member of his own court who has written exactly zero published opinions since her confirmation eight months ago.

Of course, one Republican contender for elevation to the Supreme Court did exactly what my friends predicted Judge Sutton would do.  That would be Judge Kavanaugh, who avoided the merits in Seven-Sky v. Holder by voting to hold sua sponte that his court was ousted of jurisdiction by the Tax Injunction Act, a statute of rather dubious jurisdictional status (i.e., the Supreme Court had previously held the government could waive it) that was triggered, according to Judge Kavanaugh, by an obscure provision of the Internal Revenue Code that the attorney representing the government admitted at oral argument she hadn't even read.  (Judge Kavanaugh's dissent would cite this unbriefed provision thirty-five times.  That's ten more times than the number of times Judge Sutton would carpet-bomb his chances at elevation with pithy rhetorical questions like "If Congress may engage in the same type of compelling/conscripting/commandeering of individuals to buy products under the taxing power, is it not strange that only the broadest of congressional powers carves out a limit on this same type of regulation?")

It is possible—I don't even think it that unlikelythat along with Fourth Circuit Judges Diana Motz and James Wynn, the amicus the Supreme Court appointed to defend the position after the government adamantly disavowed it at considerable risk to its Taxing Clause defense, four tax professor amici and their counsel, two former IRS Commisioner amici and their counsel, and seemingly no other judge or lawyer in America, Judge Kavanaugh sincerely both believed that the Tax Injunction Act was jurisdictional and that it deprived him of jurisdiction.  After all, as it's turned out (though no one could have predicted this in 2011), between the 2014 election of a Republican Senate majority, the abolition of the filibuster for Supreme Court nominees, and the unexpectedly smashing success of the arguments against the individual mandate at the Supreme Court, Judge Kavanaugh's chances of Supreme Court nomination would be much higher today had he voted to reach the merits and invalidate the mandate.  At the time, though, Judge Kavanaugh's vote in Seven-Sky seemed like the cleverer contemporary equivalent of the time Justice Thomas told the Senate Judiciary Committee that he couldn't recall ever having expressed an opinion about Roe in a conversation with anyone.

Besides taking a fair amount of heat for not voting to strike the mandate down (although nowhere close to the disqualifying degree of heat Judge Sutton took for actually voting to uphold it), Judge Kavanaugh's presently being criticized in some circles for not taking enough of an anti-abortion-rights position in Garza v. Hargan.  In Garza, the Trump adminstration's Office of Refugee Resettlement refused to let an unlawful immigrant minor (Doe) in government detention out of detention to get an abortion. The district court ordered the government to let her out immediately.  Judge Kavanaugh, in what could have been a devastating blow to his nomination chances, was assigned to the panel hearing the appeal and forced to opine on its hot-button merits.

In a feat of Solomonic justice, Judge Kavanaugh crafted an order giving the government eleven days to wait for a sponsor to materialize who could take Doe out of detention and then take her to get an abortion, thereby absolving the government of any complicity in the matter.  Judge Kavanaugh's nominally per curiam order (nominally so because one member of the panel dissented and another, Judge Henderson, indicated her intention to concur on separate grounds) was summarily vacated en banc.  Judge Kavanaugh then dissented, elaborating on the reasoning that underlied his order; Judge Henderson separately dissented, writing that illegal aliens detained at the border had no abortion rights at all; and some people are now mad at Judge Kavanaugh because he didn't join her dissent and think the President should nominate someone else to the Kennedy vacancy because he didn't.

In defense of Judge Kavanaugh, some conservative lawyers have written that the government assumed Judge Henderson's position away (true), that it was unnecessary to decide whether detained illegal aliens had abortion rights in order to decide whether it unduly burdened Doe's putative rights to make her wait an additional eleven days to get an abortion (also true), and that Judge Henderson's dissent had nothing to do with abortion in particular, just the constitutional rights of illegal aliens generally (also true, though it's hardly as if this White House is disinterested in judicial nominees' position on the constitutional rights of illegal aliens generally).  So, Kavanaugh's defenders reason, what is the anti-Roe objection to Kavanaugh's behavior in Garza?

The answer to that question, I think, is very simple.  It's that Judge Kavanaugh (1) declined to endorse the much broader anti-abortion position the government did take, which was, unlike Judge Henderson's position on aliens, a specifically anti-abortion position, (2) cast doubt on that position in oral argument and, more subtly, in his en banc dissent, and (3) deftly avoided opining on the government's intensely controversial main argument, or even calling attention to his declining to opine on it, by not only neglecting to acknowledge in his order and dissent that it had been made, but all but denying that the government had made it.  

I am not the one, as the kids say, to hold (1) and (2) against Judge Kavanaugh; I happen to think that, given what the government did concede, the government's position was near-frivolously wrong under what Judge Kavanaugh pointedly referred to five times in his dissent as "existing Supreme Court precedent," and believe lower-court judges should follow Supreme Court precedent (especially the existing kind).  I am the one, however, to take issue with (3), which puts me in mind of Judge Kavanaugh's behavior in Seven-Sky.  And though I don't think that people should be faulting Judge Kavanaugh for failing to endorse cockamamie misreadings of precedents they don't like, the vast contingent of conservative lawyers who do believe in creatively misreading abortion precedent should know that Judge Kavanaugh hasn't displayed a propensity to do so—even though creatively misreading Supreme Court precedent might be his most singular trait.

With that said, what did the government really argue in Garza?  Judge Kavanaugh admirably answered that question about thirty seconds into oral argument:
JUDGE KAVANAUGH: So you seem to have three differential strands of arguments, if I could set them as I see them.  One is the facilitation argument, you don't want to be facilitating the abortion. Second, is that it's not an undue burden because she can return to her home country.  And the third is that it's not a undue burden because she could be released to a sponsor. 
That's exactly right, and correctly ranks the arguments in order of the weight the government placed on them.  To be precise, the government argued in its opening brief before the panel (technically a motion for stay pending appeal) that refusing to facilitate abortion did not unduly burden the abortion right as a matter of law; that the government would facilitate Doe's abortion by allowing her to temporarily leave its custody to get an abortion; and that it was not unduly burdening her right to obtain an abortion because her inability to get an abortion while in detention was a problem of her own making, as she could terminate her detention by voluntarily departing the country she'd illegally entered and caused to detain her, or, were she successful, by getting a sponsor.  

Since it was far from obvious that Doe would get a sponsor in time to get a legal abortion in Texas, where she was detained, and since getting one wasn't in Doe's control, the government naturally put vastly more weight on the first of her "two avenues to leave federal custody," as it called them.  Indeed, the government's short reply brief had as much to say about Doe's sponsorship possibilities as the government had to say in Seven-Sky about the provision of the Internal Revenue Code on which Judge Kavanaugh based his preferred jurisdictional holding:  that is, literally nothing.  

Instead, the reply argued that Doe had no right to obtain an abortion in her present position "because Ms. Doe is currently in HHS custody, and chooses to remain so rather than file a request to voluntarily depart the United States," that Doe was unlike federal prisoners that courts had held have rights to obtain abortions while in prison because, unlike prisoners, "Ms. Doe may elect to terminate her federal custody by filing a request for voluntary departure from the United States," and that "even if she is put to a difficult choice of choosing between leaving [sic; staying in] the United States and the ability to seek an abortion, that choice does not constitute an 'undue burden' that the federal government has placed in Ms. Doe's path."  It's little wonder, then, that a moment after summarizing the government's arguments, Judge Kavanaugh complained that "we're being pushed in a span of 24 hours to make a sweeping constitutional ruling in one direction or another," and suggested that "the sponsor option" might be "[an]other avenue[] to resolving [the] dispute short of that . . . that would solve the Government's objection."  

But even when Judge Kavanaugh signaled he was willing to give the government a limited win on the "sponsor option," the government showed little to no interest in taking it.  The government's attorney initially conceded that "yes, if it could happen that would resolve the case," but noted that attempts to find Doe a sponsor thus far had been unsuccessful.  When Judge Kavanaugh continued to insist that "that option . . . is an option that solves her problem, it solves the issue, hasn't been explored," the government replied that "it's one that I would think if it was an option [emphasis in audio, see 4:08] that plaintiffs would have raised to identify to help [find a sponsor.]"  And when Judge Millett pushed back on Judge Kavanaugh's sponsorship solution, remarking that she assumed Doe had done everything she could to find a sponsor and had no control over whether she'd get one, the government, rather than attacking Millett's premise or defending sponsorship as a realistic out for the court, confidently replied that "voluntary departure is entirely within [Doe's] control," and that "[s]he can file a request for voluntary departure at any time and then she will be out of HHS custody."

As Marty Lederman's written, Judge Kavanaugh repeatedly signaled at oral argument that he didn't buy this argument.  At one point he skeptically asked the government if Texas could pass a law banning women who were in the country unlawfully from getting abortions "[o]n the theory that [they] can return to the home country"; at another point, he said that the government's assumption that Doe had abortion rights at all "does make it hard . . . for you [to] turn around and say 'yes, assume she has constitutional rights, but also leaving the country is an option,'" because "normally the answer to someone who has constitutional rights is not 'oh we can deny them, just leave the country.'"  

The answer to this seemingly devastating line of questioning is that, bizarrely enough, while the government was willing to assume Doe had a constitutional right to obtain an abortion, it did not assume or accept that she and other detained unlawful immigrants had the right to obtain an abortion in the United States.  Rather, the government's position was that it could bar her and other detained illegal immigrants from getting an abortion in the United States (the government certainly never conceded or assumed that it was constitutionally required to release her to a sponsor); all that it couldn't do was prevent her from voluntarily departing and attempting to get an abortion somewhere else.  

Though the government's position can't be understood any other way, Judge Kavanaugh would repeatedly assert in both his order and dissent that the government not only assumed Doe had a right to obtain an abortion, but assumed she had a right to obtain an abortion in the United States.  So I want to belabor why that's patently incorrect for a bit before coming to what Judge Kavanaugh said about it.  

It's true that the government never explicitly said that Doe only had a right to get an abortion in her home country (though it's also true that the government never intimated that she had a right to obtain an abortion in the United States).  Rather, what the government repeatedly said is that Doe's inability to obtain an abortion while in federal custody wasn't a government-imposed obstacle, but an obstacle she imposed on herself by staying in this country.  As the government put it in its opposition to Doe's en banc petition:
In other words, any alleged “obstacle” to Ms. Doe’s ability to obtain an abortion is by her own choice: she is in federal custody because she entered the United States illegally, and that custody is what she contends is blocking her ability to obtain an abortion. But Ms. Doe may elect voluntary departure to end her federal custody, which would eliminate the alleged “restriction” or “obstacle” of which she complains.
To say, though, that an illegal immigrant's choosing to immigrate to a country that forbids her from getting an abortion (as the United States forbade Doe from getting an abortion so long as she couldn't get a sponsor) is a "self-imposed obstacle" to getting an abortion assumes that the relevant baseline against which to measure government-imposed obstacles is, as the government crisply described it at page 17 of its opening brief, the "position she would have been in had she not illegally entered the United States."  (Similarly, the Court has held, in precedent that the government relied on, that blocking public hospitals from performing abortions doesn't unduly burden abortion rights because it merely puts women in the same position they'd have been if there were no public hospitals.)  But that is just to say that people in Doe's position only have a right to abortion under U.S. law to the limited extent that the government cannot block them from leaving the country and exercising whatever abortion rights they have elsewhere.  

To see this, suppose Houston argued that it could constitutionally ban non-residents from getting an abortion in Houston because the choice to run up against Houston's ban on non-resident abortions instead of going to Austin is merely a self-imposed obstacle to getting an abortion.  That argument might make quite a bit of sense.  But it assumes that the relevant baseline for assessing whether Houston unduly burdens a non-Houstonian's abortion rights is the position she would have been in had she not come to Houston.  That too might make sense, but Houston couldn't logically take that position and concede that non-Houstonians had a constitutional right to get an abortion in Houston, or against Houston-imposed obstacles to getting an abortion in Houston.  A ban on non-residents getting abortions in Houston is undeniably a Houston-imposed obstacle to getting an abortion in Houston.  Rather, Houston would have to be taking the position that non-Houstonians only have a constitutional right against Houston blocking them from getting an abortion anywhere at all, as it would if it jailed a pregnant non-Houstonian and refused to release her to get an abortion.

Perhaps there is some metaphysical sense in which it can both be true that (a) unlawful immigrant detainees have a constitutional right to obtain an abortion in the United States, and (b) the United States may constitutionally bar them from getting an abortion in the United States and force them to depart the United States to get an abortion.  I don't see one.  But in any event, to state in an opinion that the government assumed Doe and detainees like her had a right to obtain an abortion in the United States is both false in at least the sense that the government never positively assumed that much, and somewhere between extremely misleading and false in the stronger sense that the government probably should be understood to have taken the opposite position.

With that said, I find the following statements by Judge Kavanaugh rather puzzling.  First, later in the day he heard oral argument, at which time he perfectly understood that the government was arguing that it didn't violate whatever abortion rights Doe had to make her voluntarily depart the country to get an abortion, Judge Kavanaugh ended his per curiam order in this way: "We note that the Government has assumed, for purposes of this case, that J.D. – an unlawful immigrant who apparently was detained shortly after unlawfully crossing the border into the United States – possesses a constitutional right to obtain an abortion in the United States."  That assumption was never made; if anything, the government argued just the opposite.

Second, even after the government, in its opposition to en banc review, solely defended the panel's sponsorship-focused ruling on the ground that the government hadn't infringed Doe's assumed right to an abortion by forcing her to voluntarily depart the country to get one, see pages 8–12, Judge Kavanaugh continued to rather volubly insist that the government had assumed that Doe and similarly situated unlawful detained immigrants had a right to obtain an abortion in the United States:
All parties have assumed for purposes of this case, moreover, that Jane Doe has a right under Supreme Court precedent to obtain an abortion in the United States.  One question before the en banc Court at this point is whether the U.S. Government may expeditiously transfer Jane Doe to an immigration sponsor before she makes the decision to have an abortion.  Is that an undue burden on the abortion right, or not? 
. . .  
First, the Government has assumed, presumably based on its reading of Supreme Court precedent, that an unlawful immigrant minor such as Jane Doe who is in Government custody has a right to an abortion. The Government has also expressly assumed, again presumably based on its reading of Supreme Court precedent, that the Government lacks authority to block Jane Doe from obtaining an abortion. For purposes of this case, all parties have assumed, in other words, that unlawful immigrant minors such as Jane Doe have a right under Supreme Court precedent to obtain an abortion in the United States.
Third, in addition to never writing a word about the government's main argument for reversing the district court's TRO and giving ten pages of arguments for reversing it that the government either only weakly advocated or never* made (which can hardly be deemed an exercise in restraint, as his defenders of his handling of this case would claim), Judge Kavanaugh went so far as to suggest uncertainty about what the government might argue in the event that Doe didn't get a sponsor in his vacated order's eleven-day time-frame.  "[I]f no sponsor is expeditiously located," he wrote, "then it could turn out that the Government will be required by existing Supreme Court precedent to allow the abortion, depending on what arguments the Government can make at that point."  What arguments the government can make at that point?  How about practically the only ones it ever did make:  that making Doe leave the country to attempt to get an abortion did not violate whatever abortion rights she had, and that releasing Doe to get an abortion would force the government to facilitate abortion, which the court could not compel it to do.

* In the department of arguments the government really never made, Judge Kavanaugh claimed the government was "merely seeking to place the minor in a better place when deciding whether to have an abortion" by insisting she pursue sponsorship before getting an abortion, when the government barely advocated the sponsorship option and the only interests it asserted in its policy were promoting fetal life and avoiding complicity in abortion.  Not only wasn't the government "seeking to place the minor in a better place," it definitely wasn't "merely" seeking to place her in a better place; its reasons for insisting she get a sponsor or leave the country before she could get an abortion had everything to do with its not wanting her to get one.

Finally, careful to only decide the artificially narrowed case before him rather than the intensely controversial case the parties actually argued, Judge Kavanaugh reserved judgment on what would happen if a sponsor were not found by October 31, one week from the day he was writing, only lightly hinting at how he might rule then:
[I]f transfer does not work, given existing Supreme Court precedent and the position the Government has so far advanced in this litigation, it could turn out that the Government will be required by existing Supreme Court precedent to allow the abortion, even though the minor at that point would still be residing in a U.S. Government detention facility.  If so, the Government would be in a similar position as it is in with adult women prisoners in federal prison and with adult women unlawful immigrants in U.S. Government custody.  The U.S. Government allows women in those circumstances to obtain an abortion.  In any event, we can immediately consider any additional arguments from the Government if and when transfer to a sponsor is unsuccessful.
Besides the implicit skepticism of the government's unmentionable "additional arguments," what's particularly funny about this passage is the suggestion that it could somehow "turn out" that the government would be required to allow Doe's abortion under "the position the Government has so far advanced," as if the government hadn't taken the position all along that it could never be required to allow Doe's abortion so long as she was "residing in a U.S. Government detention facility."

As I said above, I certainly don't begrudge Kavanaugh's skepticism of the government's position.  Very briefly, if one assumes that Doe had a constitutional right under U.S. law to an abortion, it's very difficult to see how it can be limited to a right to not be prevented from going back to Central America to get one (or to be prohibited from getting one, as the case may be and apparently was).  A right under U.S. law to get an abortion is, almost by definition, a right to get an abortion in the United States (though the government didn't concede this), not a right against being prevented from getting an abortion anywhere on Earth.  The Constitution isn't even concerned with whether the government prevents women from getting abortions in Central America; whatever abortion rights it grants detained illegal immigrants, then, must be rights that can be enjoyed here.

As to facilitation, as I understand the Court's facilitation cases, they say that women don't have rights to government-facilitated abortions so long as the denial of government-facilitated abortions doesn't legally block them from getting non-government-facilitated abortions.  They don't say that the government can refuse to facilitate abortion when the only abortion legally available to a woman within the United States is a government-facilitated abortion.  That's why federal prisons have to "facilitate" abortion by allowing pregnant female prisoners to get them.  To say that in Doe's case there might have been a non-government-facilitated abortion available in Central America is irrelevant, because again, the Constitution does not guarantee a right to an abortion in Central America or apply there; whatever abortion rights it guarantees are both limited to the United States and enjoyable in the United States.

I would applaud Judge Kavanaugh, then, for implicitly rejecting the government's borderline-frivolous arguments.  All that I object to is his going to bizarre lengths to avoid acknowledging that they'd been made and to make it out that the government's arguments for reversal were something else almost entirely.  That, I think, is a strange way for a judge to act, but it is easily explicable.  Had Judge Kavanaugh explicitly rejected the government's arguments, and perhaps even if he had merely acknowledged their existence and explicitly chose to decide on narrower grounds, he would stand a lower chance of nomination to the Court today, and had he accepted them, he wouldn't have been confirmable.  As things are, he both appears to have gone as far as the government's litigating position allowed, thereby avoiding looking like an abortion moderate, while not going that far at all, thereby passing muster with moderates.  

Of course, I don't mean to say that's the only possible explanation of his conduct, which may owe as much or more to a sincere (if often hidden) minimalist streak as to his ambitions.  Giving the government some additional time to find a sponsor was a way to avoid unnecessarily reaching a "sweeping constitutional decision," as he said at oral argument.  But avoiding reaching the government's arguments is one thing; playing possum with them is something else.

As to the substantive merits of what Judge Kavanaugh did or didn't do, for those who do believe in creatively misreading (or "narrowing") the Court's abortion precedents, it should be noted that Judge Kavanaugh hasn't been reluctant to creatively narrow or misread other precedents in the past.  Most famously, though least egregiously, he opined that though the Supreme Court upheld the constitutionality of multi-headed independent agencies in Humphrey's Executor, and upheld the constitutionality of the single-headed Office of the Independent Counsel in Morrison v. Olson 8-1 only thirty years ago, the independent Consumer Financial Protection Bureau was unconstitutional because it had a single head.  

I still don't get why Judge Kavanaugh thinks single-headedness meaningfully distinguishes the CFPB from the SEC, or why he thinks that Morrison isn't fatal to his argument from single-headedness.  But I can say that while he purports to faithfully follow Morrison and merely find it completely distinguishable, he also writes that today there is "nearly universal consensus . . . that Justice Scalia [the lone Morrison dissenter] had been right back in 1988 to view the independent counsel system as an unwise and unconstitutional departure from historical practice and a serious threat to individual liberty" (emphasis added).  For the proposition that there is near-universal consensus that a near-unanimous thirty-year-old Supreme Court precedent was wrongly decided, he cites only a law school magazine's quotation of Justice Kagan's meaningless remark in a law school Q&A that Scalia's dissent in Morrison "was one of the greatest dissents ever written and every year it gets better."  (This in the context of a discussion of legal writing, not Justice Kagan's views about removal power. It is a really well-written dissent.)  

Better yet, immediately after asserting that the independent counsel system was unconstitutional, Judge Kavanaugh clarifies that "[i]n this section of the opinion, I am addressing the historical practice of how independent agencies are structured. A separate question is whether Morrison v. Olson constitutes a judicial precedent on the question of whether a single-Director independent regulatory agency is constitutional."  That way, it's clear that he's only saying that the Office of Independent Counsel was unconstitutional as an historical matter, not as a legal one, given the binding precedent that says it wasn't.  Or something like that.  (Perhaps Morrison has been overruled in the court of history.)

Then there's Chevron.  I've written at great length about Judge Kavanaugh's distortions of the major-questions exception to Chevron before.  But to briefly summarize, in King v. Burwell the Court fashioned, for the first time, a major-questions exception to Chevron, under which the Court won't defer to agencies on really major interpretive questions and will instead decide for itself whether to uphold agency interpretations.  The agency still can win, as it did in King; there isn't even a thumb on the scale against the agency.  All that changes when the major-questions doctrine applies is that the agency's interpretation is reviewed de novo.  

Prior to King, the Court routinely gave Chevron deference on major questions, but said in a number of cases that Congress is unlikely to "hide elephants in mouseholes," that is, hide authorizations of massive regulatory power in especially obscure or unassuming statutory provisions.  And it said that in cases where an agency claims Congress has done so, the mismatch between those mousehole-like provisions and the agency's elephantine claims of regulatory power can be so great that it is unambiguous, under Chevron's first step, that the agency lacks the power it claims.

From these cases, Judge Kavanaugh fashioned a new doctrine completely of his own invention that he called the "major rules doctrine," which he would claim was just a better name for the major questions doctrine.  It's not, which is why he had to give the doctrine a new name.  Under Judge Kavanaugh's major rules doctrine, Congress can never be understood to authorize a "major rule" unless it does so unambiguously.  

This doctrine, not to put too fine a point on it, has nothing to do with anything the Supreme Court has ever said about the major questions doctrine.  What the Court has said is that (1) Congress can authorize profoundly major rules in profoundly ambiguous and even arguably textually foreclosed ways (see King), but that sometimes agencies won't get deference on the major question of whether Congress did (hence the name of the doctrine), and (2) sometimes when an agency claims that Congress gave it authorization for an elephantine rule in an obscure mousehole of a provision, it will be unambiguous that it didn't because of the implausibility of that kind of obscure authorization.  The Court has never said that if Congress is truly ambiguous on the subject of a major rule, the agency writing the major rule automatically loses.  Indeed, to say so would be literally nonsensical, because there are some subjects on which anything that an agency does would be a major rule (like the question in King, or how to classify ISPs, which was the question before Kavanaugh in the opinion where he launched the major rules doctrine).

Finally, there's Judge Kavanaugh's take on my beloved Marks doctrine, which again so badly misdescribes the doctrine in question that he had to literally rename it to talk about his misdescription of it.  According to Judge Kavanaugh, the narrowest grounds rule, under which the Court's holding in a fractured opinion is deemed to be the view of the members who concurred in the judgment on the narrowest grounds, is really the middle grounds rule, under which the Court's holding in a fractured opinion is deemed to be the view of the members who occupy some middle ground between a broader opinion and the dissent.  

What does that mean in practice?  Surprisingly, Kavanaugh says it means that if the dissent "did not address the issue" on which the Justices concurring in the judgment disagree, there is no "Marks middle ground" and no binding precedent under Marks, even if one opinion concurring in the judgment is plainly narrower than the others.  That is to say, the dissent controls the meaning of a fragmented opinion.  Hypothetically, then, if two Justices say that a redistricting plan can never violate the Voting Rights Act and that the one before them therefore didn't, three Justices say that redistricting plans can sometimes violate the Voting Rights Act under a certain test and that the one before them didn't, and four dissent because they think the Court lacks jurisdiction, Judge Kavanaugh would say that under Marks, the three-Justice plurality isn't controlling* because their narrowest grounds aren't "middle grounds."  

That's an interesting idea, although there's no earthly reason to say that the plurality's opinion would become any more binding if the dissent had opined on the merits.  In saying it would, Judge Kavanaugh expressly trades on the fallacious view that lower courts can somehow be bound by Supreme Court "majorities" largely comprised of dissenters (the argument being that if the dissent had broadly read the VRA, there would be a controlling dissent/plurality majority for the proposition that redistricting plans that flunked the plurality's more forgiving test are illegal, and that because it didn't, there's not).  But however good an idea it may be, it mangles Marks,** which isn't just a rule about how to read precedent but itself a binding precedent of the Supreme Court.  However confusing Marks may be in difficult applications, every judge in America besides Judge Kavanaugh would agree that in my hypothetical, the plurality would be binding under Marks even if the dissent consisted of a blank sheet of paper.  To say it wouldn't because it wouldn't be a "Marks middle ground" transparently disobeys Marks.  People who wish Judge Kavanaugh had endorsed the government's rather less transparent invitation to disobey precedent in Garza, then, have every right to ask why he didn't.

To be clear, this isn't just a hypothetical, but a stylized illustration of something Judge Kavanaugh has done.  Specifically, Judge Kavanaugh refused to give stare decisis effect to a fragmented Supreme Court opinion, Shady Grove Orthopedic Associates v. Allstate Insurance Co., solely because the dissent didn't address the issues on which the Justices in the majority split, thereby preventing him from divining a "Marks middle ground."  In that case, Justice Stevens' grounds for concurring in the judgment have generally been deemed narrower than the plurality's and therefore treated as binding, though some courts don't find them narrower.  Whatever the application of the real Marks rule to Shady Grove, Judge Kavanaugh didn't decline to treat Justice Stevens' opinion as binding because it wasn't narrower, but because the dissent didn't address the issues it did.

** Not to mention en banc D.C. Circuit precedent, which flatly forbids Marks dissent-counting.