The Volokh Conspiracy

Mostly law professors | Sometimes contrarian | Often libertarian | Always independent | Est. 2002

Yesterday, the Supreme Court heard oral argument in Monsanto Co. v. Durnell. The facts of this case are likely not of concern to most constitutional law scholars, but the dispute is of serious concern to the business community. The plaintiff sued Monsanto (which is owned by Bayer) for harms caused by the Roundup herbicide. (Don't call it a pesticide, as Justice Thomas--a former Monsanto employee--reminded us.) The jury awarded $1.25 million in compensatory damages based on a failure to warn about possible harms. Bayer counters that the EPA did not require those additional warnings, so the state tort claims is preempted.

Going into the argument, I struggled to count to five votes for the plaintiff. Preemption is an area where the Court's conservatives do not line up neatly. Justice Thomas, the Court's most committed federalist, has often been skeptical of federal preemption. See his opinion in Hencely just last week. I though Justice Gorsuch would be in a similar spot. I was fairly certain Justice Kavanaugh would be troubled by the possible economic effects on companies like Monsanto, and would favor broad federal preemption. I suspected that Chief Justice Roberts would fall in a similar camp. Same for Justice Alito. I wasn't sure where Justice Barrett would be. It It was conceivable that Justice Kagan would favor broad federal preemption to promote uniformity, but I was doubtful.

Having now listened to the oral arguments, I will stand by my prediction. Everyone lined up about where I expected. Despite Paul Clement's best efforts, I'm not sure where Monsanto finds five votes. Indeed, the Justices seemed to have very few questions for Ashley Keller, counsel for the Durnell. Perhaps they were a bit tired, as the argument in Chartrie stretched nearly two hours, but Keller was able to talk uninterrupted for several minutes. And with about ten minutes remaining, and no further questions coming, Keller sat down. As a general rule, the side that gets the most questions is more likely to lose. Then again, the Justices may have been spent after the Fourth Amendment case.

Justice Alito was entirely silent during the argument except when he perked up during an exchange about Loper Bright. I flagged this issue in a prior post. Everyone agrees that Congress can preempt state law through a statute. But what about when an agency purports to preempt a state law through regulations? Under the Chevron regime, the agency would likely get deference when its regulation interprets a complex (ambiguous) statutory regime. But after Loper Bright, does that agency still receive deference? And if the agency does not receive deference, can a state court, in a tort suit, interpret the federal statute to determine if there is federal preemption? In other words, who gets to interpret the statute: the agency or the court. Who decides? Loper Bright would seem to suggest that courts decide legal questions, rather than agencies. Or does Loper Bright not envision a role for state courts to decide the preemptive effect of federal regulations?

Consider this colloquy:

In response to my two-part series (1, 2) on what to do with AI-generated scholarship, my good friend and former colleague Daniel Solove writes in with a question/comment:

What's the difference between you here and a judge? A judge directs legal opinions and puts their name on them, so aren't they doing the same thing, just with a human writer vs. AI?

Claude is just a law clerk.

Fair questions. I disagree, because I think the norms of authorship for legal opinions and scholarship are different.

Judicial opinions are exercises of formal government power, and the fact that one judge signs it is just a convention. Say there's a federal court of appeals case heard by a three-judge panel of Judge Ay, Judge Bee, and Judge Cee. If the panel hands down a published ruling, what makes the document important is that the formalities are met. It's a ruling in the case issued by a three-judge panel authorized to issue it that is binding authority in the circuit. It really does't matter who formally signed the ruling. It's equally important as precedent whether it is signed by Judge Ay, Bee, or Cee, or whether it is unsigned and issued per curiam.

Plus, we don't think it's really just the judge who signed the opinion whose view is reflected within it. We understand that an opinion signed by Judge Cee was really the collective view of all three judges. Perhaps it's more that of Judge Cee than the other judges, but it's something that Judges Ay and Bee could go along with, too.

In that setting, where individual authorship really doesn't matter and the document is important because of the formalities, it makes sense that we wouldn't have a law clerk's signature on an opinion they helped draft (or, in some cases, drafted entirely). The opinion is an institutional message, and it's the institution that matters. The names on the document don't matter much, but they understandably reflect those in the institutions who have the power over the message.

I see scholarly norms as different, at least when it comes to traditional law review articles. With scholarship, the scholar is saying, this is my view. I see the norms of scholarship as more like that of a soloist at a jazz concert. At a jazz concert, the solo is the musician's time to make a statement. If a tenor saxophonist gets up on the bandstand to begin their solo and instead pulls out a phone and hits "play," playing back a recorded solo performed by John Coltrane, we wouldn't say that the soloist is just as fantastic as John Coltrane. We wouldn't celebrate the soloist for expertly finding that Coltrane solo and skillfully hitting "play" at the right time. Instead, we would feel cheated. The soloist was supposed to make a statement, and instead he made no statement of his own at all.

I get that such norms can be contingent. What kinds of expression are valuable for their individuality, and which kinds are not, is something to debate rather than to logically resolve. And at some point, the selection of others' works can become a kind of statement of its own. But when it comes to AI, or at least my own AI-generated scholarship, I see AI as doing more than just executing my formal directives. So I see the scholarly norms as different, and I don't see AI as "just a law clerk."

Or so it seems to me. Curious what others think on this.

From today's decision by Judge Gerald Pappert (E.D. Pa.) in EEOC v. Trustees of Univ. of Pa.:

In the wake of Hamas's October 2023 terrorist attack on Israel, the University of Pennsylvania's then-President Elizabeth Magill and others affiliated with the school stated publicly numerous times that Jewish faculty, employees and others at Penn had been subject to vile acts of antisemitism and harassment on campus…. Prompted by these statements, … [t]he EEOC … issued an administrative subpoena, with which Penn refused to comply, seeking contact information for Penn employees who may have been victims of, or witnesses to, such harassment…. The Court granted the EEOC's application [for] {judicial enforcement of the subpoena}, requiring Penn to comply with most of the subpoena by May 1.

Penn, and later the intervenors, moved to stay the Court's order pending appeal…. Penn does not have a strong chance of prevailing on appeal but makes, narrowly, a showing of irreparable harm. Staying the Court's order will not substantially injure the EEOC and a stay will allow the Third Circuit Court of Appeals to address in an orderly manner a matter of great public interest….

A request for a stay pending appeal prompts four questions: (1) whether the applicant has made a strong showing it is likely to succeed on the merits; (2) whether the applicant will be irreparably injured absent a stay; (3) whether a stay will substantially injure the other party in the litigation; and (4) whether the public interest favors a stay….

Penn does not have a strong chance of success on the merits, and its motion further exposes its vulnerabilities on appeal. The charge of discrimination is valid, the EEOC's subpoena seeks information relevant to the charge and the subpoena does not unduly burden Penn. The subpoena also does not violate substantive due process or the First Amendment. The Court explained its reasoning in its memorandum opinion [see this post], and Penn either ignores that reasoning, mischaracterizes it, or objects to it on superficial and conclusory grounds….

Show Comments (16)

The Supreme Court held oral argument this morning in Chatrie v. United States, the geofence warrant case. I live-tweeted the two-hour argument over at both X and Bluesky, and click the links there to see the play-by-play. For those wanting the big picture, here are my initial thoughts right after the argument:

(1) The most important takeaway is that the Justices seem likely to reject the broader arguments Chatrie made that geofence warrants are categorically unconstitutional or cannot be drafted in ways that could identify suspects. They seem likely to rule that geofence warrants can be drafted constitutionally, rejecting the 5th Circuit's view in Smith. I suspect they will likely hold that geofence warrants have to be limited in time and space, and leave the details to lower courts. There were some Justices who wanted to address the multi-step process of Google's warrant procedure and say new warrants were needed for those steps—that was the view, if I recall correctly, suggested by Justices Sotomayor and Jackson. But I don't know if the majority will get that far.

(2) I don't know if the Court will rule on whether there was a search. I would guess that they assume a search and say the warrant was constitutional, so they don't have to decide the search question. There was a lot of discussion of whether location history records were kept in a virtual private locker, and I heard the government conceding that this was a viable theory of protection for cloud-stored *content* records, like calendar entries and photos stored in the cloud, that the user directly controls. But I don't know if the Court will weigh in on that or just assume a search and focus on the constitutionality of the warrant.

(3) I was somewhat surprised by the government not pushing back more late in the argument, especially in response to arguments that its search arguments are maximalist, or that there were practical problems with getting warrants, or that warrants might be needed if the revealed location was a home. Late in the argument, the government seemed in a conciliatory mindset, perhaps sensing a victory on the warrant issue that it didn't want to place at risk. But Smith v. Maryland expressly holds that voluntary disclosure from the home that reveals presence in the home is still not a search. And more broadly, there's no practical problem with getting warrants for Location History because the technology doesn't exist anymore, and probable cause was not challenged back when there was. In any event, what matters going forward is all the other online records that exist online: IP logins, etc. (See my book for some of the other kinds of records that are important in investigations.) Those are records that lower courts have held don't require warrants, and it would be a massive shift in practice if the Court ruled on the search issue in ways that implicated those different records. I was surprised the government didn't flag that; that's what matters practically speaking. Anyway, maybe that counsels in favor of just assuming a search here, or maybe the Court will get into that. Stay tuned.

(4) There was some discussion of a bunch of the side issues: Do Terms of Service matter? Does privacy legislation matter? Can we recognize a mosaic theory? These are each massive questions on which lower courts are split. Although I think there are correct answers to each of the three questions— Full disclosure, I have articles on each of these topics arguing that the answers are, respectively, no, no, and no—I am not sure that the Court will get into them.

(5) Justice Alito expressed some frustration that the Court had granted cert to render what amounted to an advisory opinion, both because the good faith exception obviously applies and because Google no longer stores these records. Justice Alito suggested that he might rule for the government on the good faith exception even though the Court denied cert on that question presented. He also hinted that perhaps it was the minimum of four Justices among his colleagues who voted to grant in the case. Justice Sotomayor weighed in that guidance in this area was important. So presumably Justice Sotomayor voted to grant cert in the case, and Justice Alito did not.

(6) There was some discussion of the Stored Communications Act and how it applies. It wasn't a particularly strong discussion, I thought; it's a complicated question, and this was no context in which to delve into the glorious statutory details. I don't think the statute is relevant to the Fourth Amendment issue, for all the reasons I explained here. But if the Court does weigh in on the SCA, it might surprisingly lead to an important SCA precedent, albeit on a question not really briefed.

As always, stay tuned.

Today, the Supreme Court issued a very unusual order in a very unusual case.

In late November, a three-judge panel found that Texas's redistricting plan was motivated by race. Judge Jeffrey Brown, a Trump appointee in Galveston, joined (and likely wrote) the majority opinion. Judge Jerry Smith, a Reagan appointee in Houston, wrote a blistering dissent. It began:

"Fasten your seatbelts. It's going to be a bumpy night!"

Indeed, this night would be very bumpy.

Texas Solicitor General Will Peterson sought an emergency stay of the preliminary injunction with the Supreme Court.

On December 4, the Supreme Court granted an emergency stay by a 6-3 vote. There were two paragraphs of substantive analysis. The second paragraph focused on the timing, as the District Court issued its order after voting had already begun. The first paragraph offered two reasons why "Texas is likely to succeed on the merits of its claim that the District Court committed at least two serious errors." First, "the District Court failed to honor the presumption of legislative good faith by construing ambiguous direct and circumstantial evidence against the legislature." And second, "the District Court failed to draw a dispositive or near-dispositive adverse inference against respondents even though they did not produce a viable alternative map that met the State's avowedly partisan goals." The Supreme Court distilled Judge Smith's detailed dissent to a few words.

This stay order allowed the maps to go into effect for the 2026 midterm elections. But it did not end the case. Texas still had to challenge the preliminary injunction in the normal course.

On January 13, the Texas Solicitor General filed what is known as a jurisdictional statement. Unlike a usual discretionary case, which involves a petition for a writ of certiorari, Congress designated appeals from three-judge panels in redistricting cases as "mandatory" jurisdiction. The Appellants (not Petitioners) asked the Supreme Court to note "probable jurisdiction" in the case. The Appellees asked the Supreme Court instead to affirm.

Texas did not ask the Supreme Court to summarily reverse the case. Instead, the state wanted the Court to hear oral argument to clarify the Alexander issue:

Plaintiffs alternatively ask this Court to vacate the preliminary injunction (presumably without opinion), remand, and delay review until an appeal following final judgment. See, e.g., Brooks Mot. 37; Gonzales Mot. 35. Although the district court's errors are straightforward and may well warrant summary reversal, this Court should provide guidance for the district court in this case and future cases. The standard for issuance of a preliminary injunction in redistricting litigation is an important issue, and as this case illustrates, the erroneous grant of a preliminary injunction can create electoral chaos and require expedited action. To avoid such disruption in future cases, this Court should squarely hold that Alexander's alternative-map requirement and the presumption of legislative good faith apply with equal force at the preliminary injunction stage. And unless they are corrected by this Court, the additional errors discussed above may well recur at trial, leading to reversal and a waste of both judicial and party resources.

But the Supreme Court took a difference path. Today the Court issued an order in Abbott v. League of United Latin American Citizens (25-845):

For the reasons set forth in Abbott v. League of United Latin American Citizens, 607 U. S. ___ (2025) [25A608], we reverse the District Court's judgment. Justice Sotomayor, Justice Kagan, and Justice Jackson dissent from the Court's summary reversal.

Do you follow? The Supreme Court reversed the District Court's preliminary injunction in light of the emergency docket order staying the District Court's preliminary injunction. It is a bit recursive, but it makes sense. Again, Lulac is a mandatory jurisdiction case, not a discretionary certiorari case. I presume that the decision to summarily reverse required a threshold finding of probable jurisdiction, but the Court does not say so expressly.

I can't recall any instance where the Court summarily reversed in a mandatory jurisdiction case. It is common enough for summary affirmances. Bluman v. FEC (2012) was one such recent case. In my article on bilateral judicial reform, I proposed that all mandatory cases should be submitted for oral argument. At the time, I was responding to the Court issuing a summary affirmance in a mandatory jurisdiction case. The thought never crossed my mind that the Court would summarily reverse in a mandatory jurisdiction case. And I could have never fathomed the Court would summarily reverse in a mandatory jurisdiction case based on an emergency docket order. Yet, here we are.

Lulac was distributed for the conference on April 24, and decided on April 27. This does not seem to have been a hard one for the Court. They didn't want to decide the tough merits issue so just got rid of it. Justices Sotomayor, Kagan, and Jackson didn't even bother writing a dissent.

Just last week, the Supreme Court GVR'd a case based on an unargued per curiam decision. Now, the Supreme Court is summarily reversing a district court's preliminary injunction based on an emergency docket ruling that stayed the district court's preliminary injunction. It would have been impossible for the district court to have even considered that ruling when issuing the PI, as it did not exist yet.

Usually, a summary reversal is reserved for the rare cases where the lower court clearly violated a precedent and there is no purpose to hold more proceedings. But here, the district court is summarily reversed for not anticipating what the Supreme Court would later do to reverse that same district court. This is not like Judge Murphy and his colleagues who are reversed twice in the same case, consecutively. Here, Judge Brown was reversed twice, having only issued one opinion.

Going forward, this case is not yet over. There was only a preliminary injunction issued. In theory at least, the case can go through discovery, summary judgment, and perhaps trial. But what's the point? The Supreme Court has spoken twice. Does anyone think the outcome would be any different after discovery? Moreover, by the time this case actually gets to a final judgment, 2030 will be around the corner, and there will be new maps. The courts need to get out of the business of reviewing legislative maps. Let Texas be Texas and let Virginia be Virginia. The chips will fall where they may.

What comes on Wednesday? Maybe the Court was clearing the brush in advance of Callais.

Harvard magazine reports on recent comments by retired Supreme Court Justice Stephen Breyer on the "shadow docket." Although Justice Breyer would likely vote with the more liberal justices on many interim orders, he does not seem to share the concerns of Court commentators about how the Court has been handling things. From the story:

"Should we be worried about the Court's reliance on the shadow docket?" West asked Breyer, noting that many audience members had submitted questions about it.

No, was Breyer's short answer. "Every court has what you're saying is a shadow docket, which we call an emergency docket," he said, explaining that throughout most of the Supreme Court's history the docket had been used primarily to issue stays of execution in death penalty cases. "Or sometimes," he added, there would be a "very important case about an election or an election rule, and we might issue the stay."

Tracing the increasing use of the shadow docket in part to the rash of legal challenges that sprang up in the wake of vaccine mandates and other restrictions during the COVID-19 pandemic, Breyer—who now serves as Byrne professor of administrative law at Harvard Law School and as a visiting judge for the First Circuit Court of Appeals—rejected the notion that "there's some kind of plot involved within the Court to get this or that decided."

Instead, he argued, the nature of cases reaching the court on an emergency basis has changed: rather than death penalty and election matters, many cases more often involve constitutional disputes about "the nature of the constitutional relationship" between Congress and the president and the separation of powers."

"So it's hardly surprising to me, where you have a government that wants to bring those [issues] up, that you have more cases on the emergency docket," Breyer said.

Former Judge Michael McConnell does not share Justice Breyer's judicial philosophy, but has a similar take on whether the increasing use of the "shadow docket" should raise alarms. Writing in the Washington Post, he notes that this is an understandable and inevitable response to the increasing reliance upon aggressive assertions of executive power and resulting litigation. He writes:

Many shadow docket decisions are subject to legitimate criticism, but the wholesale condemnation of the practice is misguided. Such cases are driven by the practical reality that it takes months if not years for a case to wend its way through the judiciary. Whether a policy is implemented while the case is litigated is often the whole ballgame. By the time it gets to the Supreme Court, the harm is already done. . . .

The cases are hard because, very often, the harm is irreparable on both sides. . . . When a president is allowed to implement a new policy without authority from Congress — say, on tariffs or vaccines — it's often impossible to unring the bell. But if the president is forbidden from implementing a new policy where he genuinely has authority, the will of the voters can be thwarted for years.

The best the courts can do is to make an educated guess about the ultimate outcome on the legal merits and try to minimize serious, irreversible consequences. In a hierarchical system like our judiciary, it makes sense for the final decision-maker to become involved in a case when an interim order effectively resolves the dispute. I have little doubt that when a Democratic president is again elected, and district judges issue preliminary injunctions against controversial initiatives, many of today's critics will want the high court to intervene. . . .

Supreme Court decisions are often controversial and sometimes wrong. But the focus should be on the individual cases — the court's judgment on the probability of success on the merits and on which side stands to suffer the most serious and irreparable damage from an incorrect decision. That the court is forced by events to act quickly, and sometimes disagrees with lower court judges, is an inevitable result of the controversial use of executive power. The notion that this is an underhanded scheme to benefit one president is overly cynical. The practice began under Obama, has accelerated with each succeeding presidency and will continue when Trump is long gone.

To fully catch-up on worthwhile "shadow docket" commentary, I also recommend this NYT interview with Will Baude.

As before, you can find my prior posts on this subject here.

From Gov. Janet Mills' message Friday "vetoing L.D. 1911, An Act to Automatically Seal Criminal History Record Information/or Certain Crimes":

This bill would direct the Judicial Branch to review decades of criminal docket files—much of this work by hand—to identify those that are to be removed from the public record. The sealing of these records would occur without regard for whether the subjects of the files have requested that they be sealed, whether victims have objected, or whether a compelling public interest exists in the records remaining accessible.

There are several significant problems with the legislation. First, as drafted, L.D. 1911 would mandate the sealing of records for Class D domestic violence assault—a result that is plainly contrary to the public interest. Second, a ruling by the U.S. Court of Appeals for the First Circuit strongly suggests that categorically sealing criminal records without conducting a case-by-case review of the circumstances violates the First Amendment.

Free Speech Right of Access

Show Comments (7)

From Manhattan trial judge Judy Kim in Rodney's Comedy Club v. Omari, decided April 17 but just posted on Westlaw a few days ago:

[P]laintiff's motion for an order enjoining defendant from "any further social media posts naming, inferring or addressing Plaintiff or Plaintiff's employees in any way or posts related to Plaintiff" and ordering her to "delete any and all posts on social media naming, inferring or addressing Plaintiff or Plaintiff's employees in any way" is denied.

"Prior restraints on speech are the most serious and the least tolerable infringement on First Amendment rights, and any imposition of prior restraint, whatever the form, bears a heavy presumption against its constitutional validity." Accordingly, "a party seeking to obtain such a restraint bears a correspondingly heavy burden of demonstrating justification for its imposition and, to do so, must show that the speech sought to be restrained is likely to produce a clear and present danger of a serious substantive evil that rises far above public, annoyance or unrest."

Plaintiff has not carried its burden here. While the law permits "the restraint of speech that communicate[s] a serious expression of an intent to commit an act of unlawful violence to a particular individual or group of individuals," the allegedly libelous speech plaintiff seeks to restrain "does not meet this exacting constitutional standard."

Looks correct to me; though most courts generally allow injunctions barring repetition of material found defamatory at trial, pretrial anti-libel injunctions are generally seen as unconstitutional prior restraints (see my Anti-Libel Injunctions article). The court cites Brummer v. Wey (N.Y. App. Div. 2018) for these propositions; for more on that case (in which I filed an amicus brief, on behalf of Profs. Steve Shiffrin and Martin Redish and myself), see this post.

Back, to Rodney's Comedy Club, here is an excerpt from plaintiff's argument in favor of an injunction; remember that the substantive defamation claim is still pending and that this decision just dealt with the requested injunction, without determining whether the statements were true or false:

Free Speech Libel

Show Comments (1)

Plaintiff Bear and Defendant Bear, from plaintiff's TRO filings.

From Judge Arun Subramanian (S.D.N.Y.) Wednesday in Those Characters from Cleveland, LLC v. Schedule A Defendants:

[P]laintiff has failed to sufficiently address the potential fair use and First Amendment claims related to the products in question that use "Don't Care Bears" (or a variation) along with marijuana imagery. Plaintiff relies heavily on the Supreme Court's decision in Jack Daniel's Props., Inc. v. VIP Prods. LLC (2023). There, the Supreme Court rejected a First Amendment defense to a dog toy that had the same shape and design as a bottle of Jack Daniel's whiskey. But key to the Court's holding was the fact that "the accused infringer ha[d] used a trademark to designate the source of its own goods—in other words, ha[d] used a trademark as a trademark." The Court made clear that its opinion was "narrow" and its infringement holding only covered cases "when the challenged use of a mark is as a mark."

That's not the case here. The "Don't Care Bears" defendants are not using the Care Bears marks as marks; unlike in Jack Daniels, for the most part, the products don't look like the products that plaintiff sells. They don't even use bears resembling Care Bears—some have teddy bears with bows, others have bears a-la the main character of Corduroy, still others have non-Care-Bears-looking bears that appear to have eaten too many Cheetos. For these, defendants seem to just be using "Care" and "Bear" as part of a phrase. So Jack Daniels is inapposite.

Free Speech Trademarks Copyright Due Process

Show Comments (3)

4/27/1822: President Ulysses S. Grant's birthday. He would appoint four Justices to the Supreme Court: Chief Justice Waite, Justice Strong, Justice Bradley, and Justice Hunt.

Show Comments (306)

As I explained in my previous post, I recently tasked AI with comparing two transcripts of the 1807 treason trial of Aaron Burr. My ultimate question is, what do I do with the document that resulted? And that breaks down into two sub-parts. First, do I publish this, either just online informally or with some kind of journal? And second, how do I describe what my relationship is to it? Am I a co-author? The author? Just a prompter?

Let me start by explaining how the memo was created, and then turn to the questions I have.

I. How the AI Memo Was Created

I need to start with what I did to help create the memo, as that might be relevant to my questions. The transcripts that needed to be compared were .pdfs of two-volume books from 1807 and 1808 that go for hundreds of pages, although the only parts I cared about were the parts on the privilege against self-incrimination. I used Claude (Opus 4.6 extended), and I tasked it with comparing the discussions of the legal arguments about the privilege against self-incriminagtion to get a better sense of whether my 2021 article on those arguments based on the Robertson transcript was accurate in light of the Carpenter transcript.

To say that "I tasked Claude" covers up a lot of detail, though. I went through around 30 rounds of prompting with Claude, over the course of a few hours. As I went along, I learned about what Claude could and couldn't do and pushed it to do a better job when it was resistant to do more. For example, when I first asked Claude to compare the documents, it declined, saying it was just too big a task to take the two long pdfs, to make them readable, and then to compare them. So I started with an easier task: Take my 2021 article, read it, and understand what it claims about the Burr trial, and then read the Carpenter transcript and write an article presenting a comparison. The first draft reply was a start, and made me think that the enterprise might be ultimately useful. But it left a lot to be desired.

Over time, I came to realize that there was an art to getting Claude to make the comparisons I needed. Ultimately it agreed to do a direct comparison of the two transcripts based on the claims I had made in my 2021 article. And I quickly realized that the comparison really needed lots of direct quotes and page references for everything, so that took a lot of extra time: Among other things, Claude had surprising difficulty with page numbers, in part because one of the two .pdfs had two volumes back to back and Claude could not figure out the pagination. I had to spot check, and I kept finding errors, which Claude had to keep correcting.

A big breakthrough came when I realized through trial and error that Claude could do comparison screenshots. That is, instead of just telling me what the two transcripts said, Claude could take screenshots of the relevant discussions and show them side-by-side. That way I could be more confident that I was getting a real comparison. Even then, the screenshots needed a lot of correction: Claude started off giving me only about 1/3 of the comparisons correctly, and I had to keep telling it to go back and make sure it was screenshotting the exactly equivalent sections.

As I neared the end, I also added more tasks. For example, I asked Claude to read the Carpenter transcript and tell me if my 2021 article had accurately summarized the arguments, as well as whether there were any parts of the argument from the Robertson transcript I had missed. I also asked Claude to say if there was any legal source Robertson had reported than Carpenter hadn't and vice versa, and any legal argument that one had reported that the other habn't.

After about 30 rounds of prompting, I ended up with a 22-page comparison memo. A typical page of the memo looks like this:

The good news, at least accordiing to Claude, is that all of the substantive points matched really nicely. The variations between the Carpenter transcript are apparently very minor, the kinds of things you might expect with two independent human beings trying to write down hours of court proceedings and hearing some small things differently. The parts I cared about were a match.

II. Should I Publish This, At Least in Some Way?

This brings me finally to the big queston I have, what do I do with this 22-page AI-generated memo? On one hand, if you're interested in my 2021 article, I think the AI comparison is of scholarly interest. The comparison is pretty noteworthy, at least for the small number of nerds who care about the substantive topic. On the other hand, the AI-generated memo doesn't slot into any traditional understandings I have of either scholarship or non-scholarship. So I don't know what to do next.

My uncertainty breaks down into two questions. First, do I publish this? And second, if I publish it in some sort of way, what should I state as my relationship to it? Let me explain my thinking as to both questions.

First, on the question of whether to publish this, I am conflicted. Read More

Among other things, Otis responds to my post from yesterday; an excerpt:

The main criterion in a democratic system is not whether a given prosecution is common, but whether in this particular case fairly evaluated, the facts could be viewed by a reasonable jury as establishing the prospective defendant's guilt beyond a reasonable doubt. For one thing, adopting the "common prosecution" criterion leaves open many of the problems it's supposed to solve. How "common" is common enough to be confident the case isn't merely political? Will that get decided by the line prosecutors — careerists (or, less generously, bureaucrats) or their more accountable (but also more political) superiors?

More importantly, to focus on commonality system-wide is to risk losing focus on why we have a criminal justice system to begin with, namely, to hold wrongdoers to account and give justice to their victims. Contrary to some of my liberal and libertarian friends, I do not see "the system" as being perpetually on trial. Its balance and fair-mindedness are, to be sure, "on trial" before the legislature, which properly has the power to address systemic problems, such as they may appear to be. But they are not on trial in deciding whether Mr. Smith or Mr. Jones from the SPLC fleeced any given contributor by giving him a song-and-dance rendition of what his money would be used for.

The flaw in Prof. Volokh's second criterion (whether the SPLC's fundraising actually was fraudulent) is that this is simply a question of fact for the jury to decide. It goes to the strength of the case, not its legitimacy….

The SPLC indictment does raise non-trivial questions about weaponization of law and the boundaries of prosecutorial discretion, but in my view, having been a federal prosecutor under administrations of both parties, falls inside those boundaries.

As I've mentioned before, Bill and his coauthor Paul Mirengoff are my go-to people for hardheaded, pragmatic, but principled conservative views. They tend to be somewhat more conservative than I am, but I always find their work interesting (and well-written).

Free Speech

Show Comments (21)

I have a question about how to present the results of legal scholarship generated in part with AI. I pose it as "an April 2026 question" because what AI can do is changing quickly. I would guess that how we think about AI assistance in legal scholarship will change over time, too. But I wanted to explain why I ask, and then open it up for feedback. I'm very interested in your thoughts.

I'm going to present the question in two posts. In this post, I'm going to explain why I turned to AI for help with a scholarly problem I had. In my next post, I will explain what AI was able to do and present my question about what I should do with what AI produced.

Here's the context. A few years ago, I wrote a law review article, Decryption Originalism: The Lessons of Burr, 134 Harv. L. Rev. 905 (2021). The article sought to understand the original public meaning of the Fifth Amendment privilege against self-incrimination and its possible application to unlocking cell phones. It was based on a fascinating historical coincidence: In 1807, in the treason trial of Aaron Burr, there had been an extensive oral argument and then subsequent opinion by Chief Justice Marshall on how the privilege applied to obtaining testimony from Burr's private secretary about an letter in cipher that Burr was thought to have sent.

I wrote my 2021 article based in large part on a transcript of the proceedings made in shorthand by a lawyer in the courtroom. The lawyer, Mr. Robertson, had written everything down: Every argument, every legal source, even all the pincites, in what he claimed was a verbatim reconstruction of the proceedings. The idea of the article was that, given the prominence and experience of the lawyers in the case, the details of the 1807 arguments would likely reflect the Founding-era understanding of the privilege. So my article presented a very detailed reconstruction of what the lawyers relied on, what sources they looked to, and what arguments they made, all based on the Robertson transcript.

That article came out in 2021, and I moved on to other projects.

Just last year, however, I became aware that there is a second and independent transcript. Another lawyer, one Mr. Carpenter, claimed to have done the exact same thing that Robertson claimed to have done. Like Robertson, Carpenter claimed to have written down the whole trial in shorthand, including the legal sources and pincites. Both Carpenter and Robertson had published their transcripts as books shortly after the trial ended. The Robertson transcript is much better known. It is the one referenced in histories of the Burr case, and it was the one that was cited as the report of the trial in 19th Century caselaw. Those references had pointed me to the Robertson transcript, and I had studied it in great detail. I hadn't known the Carpenter transcript even existed.

This created a problem. The premise of my 2021 article is that the Robertson transcript accurately presented the arguments made in the Burr case about the privilege against self-incrimination. But a quick skim of a few spots in the Carpenter transcript suggested that they were not identical. There were things that appeared in one or not the other, or arguments presented somewhat differently, or parts summarized in different ways. If Robertson and Carpenter independently reported the same things, I could be pretty confident that it happened that way. But what if they reported key moments and arguments differently? In that case, I couldn't be confident that my 2021 reconstruction of the privilege arguments in the 1807 Burr trial was accurate.

My scholarly obligation, it seemed to me, was to conduct some sort of comparison of the two transcripts to alert readers to any meaningful discrepancies between them that might relate to my 2021 article. But this would also take a lot of time, as I would first have to go back and re-familiarize myself with the very long Robertson transcript, and then go through all of it and compare everything relevant from my 2021 article with the Carpenter transcript. It's certainly doable, but also pretty time-consuming. It's been on my list of scholarly things-to-do since last year.

And then in March 2026, I wondered: Hmmm, is this something that AI can do for me? These days, AI is really good at going through large documents and summarizing them, comparing them, and the like. And it just gets better and better as the weeks pass. Maybe, instead of going through the two transcripts myself, I can save time by asking an AI service to go through the two transcripts and compare them. Maybe AI can tell me quickly if there are substantive disparities between what Robertson says the lawyers argued and what Carpenter says the lawyers argued.

At least, I figured, it's worth a try. In my next post, I'll say how it went, and ask what I should do with the document AI produced.

On April 15, Justice Clarence Thomas delivered a lecture at the University of Texas at Austin in honor of the 250th Anniversary of the Declaration of Independence.

Here's a taste:

The second paragraph of the Declaration proclaims: "We hold these truths to be self-evident, that all men are created equal, that they are endowed by their Creator with certain unalienable Rights…." Throughout my youth, these truths were articles of faith that were impervious to bigotry or discrimination. The American Heritage Dictionary of the English Language defines "self-evident" as "obviously true, and requiring no proof, argument or explanation." Whether they had a divine source, or a worldly one, they were never questioned. They were the Holy Grail, the North Star, the rock – immovable and unquestioned.

Despite the multiplicity of laws and customs that reeked of bigotry, it was universally believed among those blacks with whom I lived and who had very little or no formal education, that "in God's eyes and under our Constitution we are equal." This was also the case with my nuns, most of whom were Irish immigrants. At home, at school, and at Church, we were taught that we are inherently equal; that equality came from God; and that it could not be diminished by man. We were made in the image and likeness of God. That proposition was not debatable and was beyond the power of man to alter. Others, with power and animus, could treat us as unequal but they lacked the divine power to make us so.

Somehow, without formal education, the older people knew that these God-given or natural rights preceded and transcended governmental power or authority. When you lived in a segregated world with palpable discrimination and the governments nearest to you enforced laws and customs that promoted unequal treatment, it was obvious that you did not get your rights or your dignity from those governments, but from God. Though not a literate man, my grandfather often spoke of our rights and obligations coming from God, not from the architects of segregation and discrimination. Men were not angels. They were subject to the constraints of antecedent rights. And, we were not subject to them even as we were subjected to their whims. We knew that life, liberty, and property were sacrosanct. These truths were self-evident to the adults in our lives and were taught to us as undeniable truths. Those around us could endure with dignity the insults of segregation because they knew that, in God's eyes, they were equal.

All too often, there is an unfortunate tendency, when discussing the Declaration, to make these self-evident truths and first principles of government obscure. Intellectuals want you to believe that our founding principles are matters of esoteric philosophy or sophisticated debate. Even those who support them too often talk about them as if they were academic playthings. They overcomplicate them, take the spirit out of them, and discuss them in a manner that puts us to sleep.

Justice Thomas' speech has received significant attention, and stoked a fair amount of controversy, largely over Justice Thomas' criticism of progressivism. In the New York Times, prominent Supreme Court critic Jesse Wegman even suggested Justice Thomas should apologize for his remarks.

Were Justice Thomas' remarks really so outrageous? So readers can make up their own minds, here is a transcript, courtesy of Civitas Outlook, and the video is below.

4/26/1995: U.S. v. Lopez decided.

Show Comments (225)

This evening, immediately after I turned my phone on, I saw headlines about shots fired at the Washington Hilton where President Trump was attending the White House Correspondents Dinner. For the second week in a row, I do not get a leisurely return to the cloud. Shavua Tov.

For the past two years, the Federalist Society has held its national lawyers convention at the Washington Hilton, better known as the Hinckley Hilton. This is the place where John Hinckley, Jr. shot President Reagan and Press Secretary James Brady in 1981. You may better know Brady as the namesake of the Brady Gun Control Act that was challenged in Printz v. United States. Randy and I feature a photograph of the hotel in our discussion of Printz.

In 2023, I bid farewell to the Mayflower, and in 2024, I wrote about the new experience at the Hilton. The experience was, on balance, negative. I am beyond grateful that the Convention will be returning to the Mayflower in 2026 (thanks Sheldon).

But having spent some time at the Hilton, I can see why it poses unique security threats. To start, it is a functional hotel. Even if there is a "hardened" event in the ballroom (located in the basement), there are still thousands of guests walking around inside and outside the building.

When I traveled to the convention last November, Vice President Vance was speaking at the 250th U.S. Marine Corps Birthday Ball. Because I was a hotel guest, I was able to clear the perimeter security. Later that evening, I decided to go to the gym, which was accessible through a different elevator. The gym was also in the basement. After I finished working out, just of curiosity, I walked towards the ballroom. By that point, the event was over, Vance had left, and the magnetometers were gone, but I was able to walk right to the the ballroom, even though access was still being restricted. The thought crossed my mind of how easy it would be to sneak into the venue.

There are many unknowns now.

President Trump is about to give remarks in the Brady press room.

Update: It appears the alleged shooter was a guest at the hotel. The video shows he was taken down near the Terrace Foyer on the Terrace Level. You can see from this map how close the Health Club was.

The Volokh Conspiracy

Loper Bright and Preemption

Why AI Isn't Like a Law Clerk

Judge Reaffirms: EEOC May Subpoena Penn's Records as to "Jewish-Related Organizations" (and Others) in Investigation of Anti-Semitic Harassment at Penn

SCOTUS Summarily Reverses Three-Judge Panel In Mandatory Jurisdiction Case Based On Earlier Shadow Docket Ruling In Same Case

Justice Breyer Says Not to Worry about the Shadow Docket

Maine Governor Vetoes Broad Criminal Records Sealing Bill

Comedy Club Can't Get Injunction Blocking Claims of Sexual Assault, Racism, Anti-Semitism, and Sexism

Don't Care Bears and Intellectual Property Law

Today in Supreme Court History: April 27, 1822

Open Thread

Bill Otis (Ringside at the Reckoning) on the SPLC Indictment

What Do You Do With AI-Generated Legal Scholarship?: An April 2026 Question

Today in Supreme Court History: April 26, 1995

Open Thread

The Hinckley Hilton

Recent Volokh Posts

Volokh Conspiracy Daily E-mail

I WANT FREE MINDS AND FREE MARKETS!

I WANT TO FUND FREE MINDS AND FREE MARKETS

SUPPORT HONEST JOURNALISM

PUSH BACK

HELP KEEP MEDIA FREE & FEARLESS

STAND FOR FREE MINDS

PUSH BACK AGAINST SOCIALIST IDEAS

FIGHT BAD IDEAS WITH FACTS

BAD ECONOMIC IDEAS ARE EVERYWHERE. LET’S FIGHT BACK.

JOIN THE FIGHT FOR FREEDOM

BACK JOURNALISM THAT PUSHES BACK AGAINST SOCIALISM

FIGHT BACK AGAINST BAD ECONOMICS.