Mentioned:

• Sebastian Mallaby, The Infinity Machine: Demis Hassabis, DeepMind, and the Quest for Superintelligence (2026)

Thumbnail: President Trump delivers remarks at the White House AI Summit in Washington, D.C., Wednesday, July 23, 2025. (Official White House Photo by Joyce N. Boghosian)

This is an edited transcript of an episode of “Executive Functions Chat.” You can listen to the full conversation by following or subscribing to the show on Substack, Apple, Spotify, or wherever you get your podcasts.

Jack Goldsmith: Today I’m chatting with Sebastian Mallaby, who’s a senior fellow at the Council on Foreign Relations and an acclaimed biographer and writer. And we’re going to be talking about his newest book, which is called The Infinity Machine. Sebastian, thanks for talking with me.

Sebastian Mallaby: Thank you, Jack. Nice to be with you.

So tell us what the book is about. Who is Demis Hassabis, and why did you write a book about him?

So the book is about artificial intelligence, and it’s centered on this character, Demis Hassabis, who is, in a way, the OG sort of AI lab leader, right? He starts DeepMind, this startup in London, back in 2010, before AI could even recognize the photograph of a cat—like nothing worked. It was full AI winter.

So this is five years before Sam Altman and Elon Musk start OpenAI. It’s fully 11 years before Anthropic gets started. So he was extremely early.

So if you wanted to tell the story of the making of modern AI through a personality, you know, Demis’s career and intellectual development maps perfectly onto that story.

So the thing that’s most interesting to me about him is that, as you emphasize in the book, his real interest in this, I think it’s fair to say, is scientific and not profit-making. And he, at least at the outset, and I think even today, has a rather idealistic—to me anyway, idealistic or optimistic—conception of the technology and how it can be used.

But the story I also see is someone who—and I don’t mean this uncharitably—but who has basically engaged in a series of compromises or fudges with regard to those values as he’s gotten deeper and deeper into the AI competition.

So is that fair? And can you talk about that arc?

Yes. I mean, he started DeepMind in 2010 with an absolute focus on AI safety. In fact, he met his scientific co-founder, Shane Legg, at a safety lecture in which Shane projected that by 2030 or so, AIs would be sophisticated enough—cleverer than humans—have their own sort of objective functions, and would maybe start to threaten humans.

And this was the lecture over which they bonded. And then in 2014, Demis Hassabis sells his company DeepMind to Google. And part of the sale condition was that AI would not be used for military purposes, that it would be safeguarded by a sort of ethics oversight committee that would be separate from the corporate leadership of Google.

So he took it very seriously. And then this continues for a while. Between 2016 and 2019, he wages a secret battle, a thing called Project Mario, where he’s trying to put pressure on Google’s leadership to have this independent safety oversight board, because Google kind of reneged on the deal at the point of sale in 2014.

And then after 2019, it kind of fades away. And, you know, by now you have Google being willing to provide AI to the national security establishment. In the US, there is no safety and ethics oversight board.

And Demis is left explaining to me, well, you know, I feel as if, you know, if I lean into Google and I’m part of the team there, and I, you know, understand the different pressures that a corporation is under, then I have a seat at the table. And so when push comes to shove, I can chime in in favor of safety. And so I’m a good person—trust me—is kind of the bottom line, which is a sort of flimsy scaffolding of reassurance for an alarmed world.

Especially since—I mean, this was also a time—a lot of this is happening at a time before ChatGPT amazes the world a few years ago with whatever model it was, I can’t remember. And suddenly there’s this massive competition among several frontier models that has been extremely fierce.

And now we’re in a massive competition among those labs and with Chinese firms, and the countries are in fierce competition. And he’s now leading—you talk in the book about how they combined, how Google combined its AI efforts—and he’s leading it.

So he’s really leading, in some sense, this fiercely competitive charge, which isn’t taking—doesn’t appear to be taking—safety all that seriously. Is that fair?

Yeah, it’s fair. And, you know, I think there’s a slight caveat in that his style is to pursue safety ideas secretly. I mean, he doesn’t talk about them.

And Dario Amodei, the leader of Anthropic, is extremely public when he picks a fight with the Pentagon, when he releases this new model called Mythos, where he’s publicly said, you know, this is too dangerous to release generally, so I’m going to release it to a sort of restricted list of people. He likes to be very out there in public with it.

Demis Hassabis, on the other hand, did two important things, to my knowledge, about safety. One was this secret battle I described before, which he was so unkeen to have sort of move into the public sphere that when I discovered it through leaks from other people, you know, I had to talk to his general counsel, who was trying to tell me I wasn’t allowed to publish that. So he really didn’t want that to be public.

And then secondly, he told Rishi Sunak in 2023, after ChatGPT came out, “Mr. Prime Minister, you know, I have an idea for you, which is you could have an international discussion on AI safety—invite the Chinese, invite everybody—start a process that might lead to some kind of understanding internationally on AI safety.”

Demis never told me that he told the Prime Minister that. I only know this because other people, like the Prime Minister’s advisers, told me. So he didn’t advertise what he was doing.

So I think he’s trying to do things now, but they’re not in the public view. So that’s a slight caveat. But basically, you’re right. I mean, he’s leading one of the major labs, Google DeepMind, in frontier AI, racing as fast as he can, even releasing, by the way, open-weight models, which by his own analysis are dangerous because you can’t control them once they’re out there.

And so there is this contradiction—you could call it even hypocrisy—between his stated beliefs about AI safety and what he’s actually doing. And so then the question is, well, how harshly does one judge him? And I’ve just floated the word hypocrisy.

But on the other hand, were he to quit his job and go off and become a professor somewhere and pursue research, which I think is the alternative path for him, it wouldn’t make the world safer, right? There’d still be this race dynamic.

To be clear, I wasn’t judging him. And he seems—I’m trying to understand—he seems like a thoroughly decent, honorable, brilliant guy. I’m just trying to understand the mindset of someone who, from a very young age, had these extraordinary scientific ambitions, which he’s been as important as anyone in making possible.

And—but safety and this kind of benign vision has always been part of it, and it just seems to have been overtaken by reality—mostly competitive, financial, and global competition reality. And I’m just wondering how he processes that. That’s what I’m getting at.

Absolutely. I mean, I was exactly trying to do the same thing—to kind of figure out how you process it and sort of portray that. And, you know, at the end of the book, he tells me, you know, I’m in a paradoxical situation.

On the one hand, Shane Legg and I projected back in 2009, 2010 that by around 2030, AI would be very powerful. And that’s kind of what’s going to happen. And we’ve been central to building it. So, you know, I’ve delivered on this vision in an amazingly gratifying way.

On the other hand, I had this hope that I could control the technology somehow and make it safe, and that hasn’t worked.

And, you know, when you want to ask, you know, why did it turn out so contrary to his expectations? You know, it’s the Oppenheimer syndrome. Oppenheimer led the Manhattan Project, built the amazing technology, and was an incredible scientific leader as well as a scientist, and thought he could sort of go and sell Truman not to use the bomb or to give the technology to the UN or whatever.

Truman just kicks him out of his office and says, “Don’t bring that guy in here again.” So scientists think that they can control their inventions, but often the inventions have their own will.

Okay, you’ve written a lot of interesting essays closer to the topics of this Substack in connection with the publication of the book. And I just want to talk about some of these policy and governance themes that are implicated—that are talked about in the book—but that you’ve talked about, I think, more in connection with the publication of the book.

First of all—and you’re, you know, the keenest of observers of these various relationships and where we are in these AI races—so I just want to get your temperature on, first, what is the state of the relations between the U.S. government and the frontier labs?

I mean, we know about the DOD confrontation with Anthropic and then with Mythos, the government trying to apparently get its hands on that technology, or at least try to reach some kind of accommodation with Anthropic.

How do you see—I want to go through different pieces of the relationships and the competitions at stake here—how do you see the relationship right now between the US government and the frontier labs?

Well, I think it’s just worth saying at the beginning that, you know, the background—if one goes back to 2023, 2024, the two years after ChatGPT came out, it made all this feel urgent—the background was that there was a lot of collaboration.

And if you talk to people, I think, you know, Ben Buchanan, who was at the National Security Council doing AI policy, is on the record about this. He says, you know, whenever I talked to the labs, there was no resistance to the idea of regulation. It was more the opposite—that they were telling me, “Hey, this is serious. This is powerful. This is scary. This is coming. You need to do something.”

And so when he wanted to, you know, set up the AI Safety Institute, which happened in the Biden years or so on, there wasn’t much pushback, contrary to sort of what one might suspect. There wasn’t much pushback from the objects of the regulatory impetus.

Now, the regulation at that point was extremely, you know, emergent. It wasn’t really biting very much. But still, it’s worth noting that the baseline here is collaboration.

Then you move into 2025. President Trump is in office. He essentially doesn’t want to talk about regulation. He’s more keen on competition, acceleration—just make sure the U.S. has the most powerful AI. And so that kind of goes off the agenda.

Now, in 2026, it comes back, both because of this fight over Anthropic’s models and how they are used by the national security establishment, and because of Mythos, the cybersecurity model.

And I’d say that right now you have this sort of, you know, weird and sort of unstable contradiction of impetuses, where on the one hand, you know, the administration has deemed Anthropic to be a supply chain risk, which is an extraordinary name to attach to a U.S. company. I think it’s the first time that’s ever happened.

And so it’s extremely antagonistic. And you had this Slack message that Dario Amodei, the head of Anthropic, wrote, and which got leaked. It was a Slack message to his own staff. And what he said was something like, you know, “The Trump administration wants dictator-level obeisance from me, and I’m not going to give it.” So that shows you the temperature of the relationship there.

But on the other hand, they are talking together about Mythos, and, you know, Amodei went to the White House to speak to people about it a couple of weeks ago.

So I think it’s a mixture of suspicion and sort of a distant relationship, but then at the same time, a need created by Mythos that—wow, you know, we have no choice but to talk to each other, and we’re going to have to do that.

So two questions following up. One is, I could never really tell how serious—I think some more serious than others, maybe Anthropic—how serious the labs were when they said, “Please regulate us.” Sam Altman testified to that: “Please regulate me.”

I don’t believe he really thought that. It’s clearly—but it’s a good thing to say when you’re developing this massively dangerous, consequential technology in the private sector. In case something goes wrong or, you know, it’s always nice to have said we asked to be regulated and you didn’t do it.

I’m just wondering how serious that request is, especially given in the last presidential election, a lot of the big money behind—or at least in connection with—the AI labs, the kind of Silicon Valley big money, went all in for Trump. And, you know, David Sacks gets installed in the White House and implements this kind of libertarian policy, hands-off policy.

So first question: how serious is that request to be regulated?

Well, I think when we talk about Silicon Valley in broad-brush terms, we need to actually break that down a bit. And you’ve got someone like Marc Andreessen, whose commercial interests as a venture capitalist are that he wants to back, you know, upstart challenger AI labs. He’s not going to back, you know, the big hyperscalers because they don’t need his money, right?

So he wants to back essentially startups that are probably going to use open-weight models. And so open-weight, by definition—you cannot be in favor of regulation and be in favor of open weight. You know, open weight is the least regulated type of AI distribution mechanism.

And so, yes, when you’re talking about Marc Andreessen, he’s extremely pro-Trump and extremely deregulatory and laissez-faire. And so there’s no seriousness whatsoever in any comment to the fact that you need more safety if it’s coming out of his mouth. I think, to be fair, it’s not coming out of his mouth.

True, correct.

So then you move to other people who are saying that they want to be regulated. And I think in some cases, someone like Sam Altman probably is straightforwardly, you know, insincere.

But I’d say that, you know, all of these characters who are running the frontier labs have said things at different times, and they’ve both wanted safety. And when Sam Altman created OpenAI in 2015, the rationale for creating it was safety and for the AI to be used for the public good. And who knows? I think he might even have meant it in 2015.

But as time has gone on and the race dynamic has become so white-hot, any instinct he might have had to be sincere about wanting to be slowed down has been overwhelmed by his desire to win the race.

But I think there’s both sides of that. Inside one human being, there could be two personalities. And I think, you know, there’s just a different balance in these different people.

So, you know, probably Sam Altman is the most prone to going for acceleration rather than regulation kind of when a stress point comes. I’d say Dario Amodei is the most prone to be safe when a stress point comes. Demis is somewhere in between. Elon—I don’t think he’s really been tested.

So last question in this vein—it just, how likely is it—let me put it this way—it seems inevitable, given the rapid changes in these technologies and the growing perceived dangers in the technology—Mythos being one example, but there are lots of examples.

And given the stakes of these technologies in private hands in the United States, so given the dangers they’re spewing and the stakes vis-à-vis the competition with China and national security more generally, it seems inevitable to me that—set aside legality for a second—it seems inevitable to me that the United States gets its hands on this.

And I don’t know what the mechanism looks like or what the institution looks like, but that the U.S. government cannot afford not to, in some sense, own this technology, both for its own purposes and for safety purposes. True or false?

Broadly true. I mean, by “own,” you would include, I think, control. I mean, it could be a regulatory model.

Yeah, let me be clear: I don’t mean technically own. I mean control. That’s a better word.

So then I think true, yes. I think, in fact, we’ve just run that experiment in the last few weeks, because you had a deregulatory, pro-accelerationist administration in power. And all of a sudden, an actual case of an AI that threatens stability emerges in the form of Mythos.

And it looks as if it could unravel all kinds of building blocks of the internet and cyberspace. Banks would have their bank accounts emptied, et cetera, et cetera. And they turn on a dime.

And all of a sudden, the Treasury Secretary is saying to the Fed chairman, “We need to call the heads of the banks and tell them to take this seriously. And, you know, we need to get our hands on this.” He’s—you know, Scott Bessent, the Treasury Secretary, reportedly has floated the idea of, you know, sort of—I forget what the term is—but you’ll know—essentially requisitioning or having power over the technology.

And Bessent just said, okay, this is so important in terms of statecraft that we’ve got to mention it, bring it up in the summit with Xi Jinping in the middle of May.

So they’ve flipped. They’ve done a 180, exactly as you predict. So I think you’re correct.

And—okay—and one reason they’re doing that is because of the larger China competition. So talk about the U.S.–China competition in AI, please. Where do you see it today? I know that’s a big question.

Well, maybe—I mean, I think actually that it’s—maybe I want to push back on your premise. I think that the reason they flipped is less about China and more about domestic chaos, right? They don’t want the internet to be hacked and, you know, the banking system to crash. It’s not—that’s not necessarily about China. That’s about domestic security.

Don’t you think it’s about both? I mean, if these technologies are—so fair enough, it’s about domestic security, but it’s also about China being able to defeat government systems, China threatening both to steal and to disrupt things in the cyber realm. I mean, I think technology has an advantage as a national security advantage.

Yes, but I think—okay, we’re going to—I think this conversation is going to go in a direction where we need to clarify one thing right now, which is that there are two kinds of worry about powerful AI in terms like—the big worries, I would say.

One is that—and essentially this is about bad guys getting it and doing bad stuff with it. And there’s two kinds of bad guy, right? There’s China. And in 2022, when the Biden team—before ChatGPT, by the way—saw this coming, they put the semiconductor export controls in place because they didn’t want China to have cutting-edge AI, because in their view, the bad guy to worry about was China.

And there’s a whole second category of bad guys, which is sort of rogue states, terrorists, criminals, et cetera, et cetera. And I think it’s just very important to clarify that there are two kinds of threat.

And I think it’s important to distinguish two ways of dealing with these threats, because in the Cold War analogy, the way to control the danger of nuclear war between the Soviet Union and the United States was mutual destruction and the balance of deterrence.

The way to control nukes from being loose and falling into the hands of terrorists and rogue states was a totally different mechanism, which was the Non-Proliferation Treaty, which wasn’t perfect, but it worked kind of for a while.

So, you know, that’s an important distinction because of, I think, where we’re going to go.

And so, going back to Mythos—Mythos is in the category of both, right? It could be that the Chinese government gets it, but more immediately, it could be just the criminals get it.

And the criminal threat or the terrorist threat is posed not merely by the prospect that China gets this technology and releases it on an open-weight basis, but also that Meta does, right? Or Mistral in France, or Cohere in Canada. There’s a bunch of other labs that could open-weight this stuff, or even just not open-weight—just release it on a proprietary basis, but much more expansively than Anthropic did.

Or have it stolen.

Yeah, right. And there was a hack of Anthropic’s Mythos.

Okay, fair—perfectly fair distinction. I want to come back to it, but can we talk about—you wrote a piece in The New York Times about U.S. policy toward China, chip restrictions, and why you thought that was a bad idea. And then I wanted to get to the nonproliferation treaty idea that you just mentioned.

So can you just tell us why you think that the Biden approach may not have been optimal?

Sure. So at the time I supported it because I thought it had a chance of actually preventing China from getting cutting-edge AI. It turns out it didn’t stop China from getting cutting-edge AI.

And the proponents of semiconductor export controls will say, well, it’s because the controls were not tight enough. But we’ve run this experiment since 2022, so it’s four years now. And we’ve done it under two different administrations.

And in neither case, if you look at the chart of the performance gap between the top model in America and the top model in China, that gap—America’s lead—has shrunk. And it shrunk under Biden, and it shrunk under Trump.

So I’m just skeptical of any regime in which we impose semiconductor export controls or some expanded version thereof that actually works in stopping China from getting powerful AI. They may be like two months behind us, six months behind us, something like that, but that’s not very long.

But just before we go further, I always understood the goal not to stop them but to slow them. And even with workarounds, the United States maintains its edge.

And are you saying that it would be better in maintaining that edge to get rid of the chip controls? That argument, I don’t understand.

Fair. Okay, that’s very fair. So yes, I agree that the controls slow them down. My point is simply that it’s not very much—it’s six months. Does six months really make us feel a lot more secure?

I’m open to—you know, this is a finely balanced argument, which is why I’m excited to have it with you. You could argue that a six-month lead is a lot, and that, you know, if you get Mythos six months before the Chinese do, that’s enough to harden your systems such that when they have it, they can’t use it to any devastating effect.

And also enough to compromise all of their systems.

Yeah, true.

Sorry, keep going.

So I guess my view—my view has been subject to revision—my view has been that a six-month lead over China isn’t enough to feel that’s a big geopolitical win, because, you know, I was thinking, I guess, more of, you know, the conventional integration of the AI into weapon systems, drone swarms, this kind of stuff.

And whether you’re six months ahead of the adversary in that case, you know, they just pick a time to invade Taiwan when they do have—they have caught up.

Now—and, you know, if you think about the Mythos example again—wouldn’t they wait until a moment of near parity? Because these things sort of—you know, you have a jump that the U.S. does, and then the Chinese are fast followers, and then they jump up so that the gap is reduced or minimized. And then they would have a strategic parity that they could use.

So I just observe that it’s not a very big lead. And then you have to ask, what was the cost of the policy? Because if there was zero cost to imposing the semiconductor equipment ban, and there was a marginal gain, then you’d probably take it.

But I contend that there is a cost, in the sense that if we go back to my distinction between two kinds of rogue—China is a rogue, but also terrorists and criminals are rogues—we need to keep our eye on that second category of risk.

And that involves getting China involved, because if they produce open-weight models, which is what they do at the moment, then we know that terrorists will get it.

So the deal is: We give you chips, and you shut down your open-weight models. And what does the form of cooperation look like? Because I’m not so sure that the third-party rogues aren’t going to get the bad stuff, even in the absence of open-weight models. That’s a large assumption.

Yeah.

And I’m skeptical—and I’ll get to this in a second—I’m skeptical that any agreement between the United States and China, about which I’m skeptical, can have powerful effects on third-party countries and private actors.

So what does the cooperation between the United States and China look like? And what is the virtuous story about what the deal is and what cooperation looks like?

I think the best analogy is the Cold War and nuclear nonproliferation. And so in that story, you have both the International Atomic Energy Agency, which kind of keeps track of nuclear material and tries to, you know, by accounting for it, prevent it from being used in secret nuclear weapons.

And then you have later—in fact, 12 years later—the negotiation of the Non-Proliferation Treaty in 1968, which kind of makes compliance with the IAEA mandatory for countries that want to have access to civilian nuclear power.

And I think that’s the sort of model where, in AI, you keep track of big clusters of compute, which can be used to train powerful AI models. And that’s one kind of safety provision, so that you know what’s being trained.

And then secondly, you don’t have open-weight models, which, once they are released, anyone can do anything with them. And that’s just too dangerous. You wouldn’t do that with nuclear material. I don’t think you should do that with AI.

And the deal is that, you know, countries which are going to want what they call sovereign AI—that’s already a bit of a catchword—if you’re France or you’re Germany or, you know, Kenya or whoever, yes, you want to be able to train your domestic AI models on, you know, the Swahili texts or whatever—the Swahili oral tradition—that, you know, you want to make it kind of culturally friendly to your own culture. And that’s all good.

But in return for getting access to models, which you can then post-train in a way that, you know, fits your sovereign AI objectives, you agree to these safety standards: no open weight, you know, we’re going to know where you’re training it, we know what the cluster is, we keep an eye on that so that if you’re trying to make cyber weapons, we’re going to see it.

And so I think it’s something like that.

So, I’m going to state why I’m skeptical and give you the last word on this, and then I have a final question.

So the nuclear Non-Proliferation Treaty—some people think—wasn’t a terribly wonderful success. Not every country joined it. The countries that didn’t join it developed nuclear weapons. It hasn’t prevented some countries from appearing to develop nuclear weapons.

But it’s a relatively simple treaty. It basically says no development of nuclear weapons. I mean, it’s a little more complicated than that, but—and we have a sense of what nuclear weapons are. There’s a problem about pre-development and how do you control that—that’s always been a problem with the treaty.

But let’s imagine that that treaty is a success, despite the adverse evidence. It just seems massively more complicated here because there are so many things you might regulate—chips, data, data centers, know-how, the models themselves.

So you’re going to be regulating all of these things. It’s much, much more difficult to have a verification regime in this context. Dual use is prevalent everywhere. It’s very hard—it seems to me, correct me if I’m wrong—to hide bad uses, bad developments, new models, and the like.

Verification regime in this context, which involves not just government-to-government, but deeply into the private sector of both countries, to private companies of both countries, and so on and so forth. Enforcement seems to me—not clear how that’s going to work.

Let’s just imagine that it works bilaterally between China—that we somehow work this out where there’s this virtuous regime where we’re both doing safety and none of us—or none of them—are doing bad uses or acceptable bad uses within the competition.

But I don’t understand, you talk about all the dangers from these third parties, both private and public actors, how this regime is going to handcuff them and all the verification and enforcement problems there, and then on top of that, there’s a massive need for speed.

This stuff is getting very, very dangerous very quickly. And it’s inconceivable that such a treaty or such an international organization would happen in the foreseeable future.

So that’s why I’m skeptical, in a nutshell.

That’s quite a long list.

You don’t have to go through the whole list, but is there a reason not to be skeptical?

Well, I think the first reason not to be too skeptical is that it’s kind of massively defeatist in the face of a serious threat, right? You know, the Mythos vision, where you have random criminals who can steal everybody’s money from their bank accounts, is not a good one.

And so I think trying to lean into the possibility of some sort of regulation that avoids that is important. And if you’re going to do that, it’s got to be international, because obviously you can commit those crimes from some offshore base on a Chinese model. And so if the Chinese have been excluded from any such deal, it’s pointless.

So I think, you know, “don’t give up” is the first message.

And then secondly, I think there are more kind of choke points on which the US, in particular, can get a grip. And if it had China with it, it could definitely get a grip.

And the choke points essentially are the making of the chips and the operation of the chips in clusters. I mean, at the moment, overwhelmingly, the majority of compute clusters are in the United States, and the ones which are not in the United States are using U.S. chips and U.S. equipment—or the equipment of, you know—even in the case of ASML, the Dutch company that builds the lithography machines to engrave the chips—that’s a Dutch company, but they have operations in the United States. They are fully going to comply with U.S. coercion.

So I think essentially the whole of the Western camp—and you and I have talked about this in the past in the context of other things, like the regulation of the internet—all of the Western AI, including—and I would include the Middle East and stuff like that—have touch points in America, which they care about, like the customers they want to serve, the way they’re going to raise money, the chips they’re going to use, other technology around the chips, the cooling material and all that.

If you brought the full might of the United States down on them, they don’t have an independent ecosystem, and they couldn’t do anything.

So I believe the U.S. has a lot of control. The only place it doesn’t control is China. That’s the only credible technological possibility of a totally separate ecosystem that rises up and, you know, provides an alternative source of AI compute power.

So if you had two countries on board—China and the US—I believe that some form of nonproliferation is possible.

Of course, I accept that in the past, nonproliferation for nuclear stuff was not perfect, and there would be people who didn’t sign and all that stuff—but better than nothing.

Fair. Okay, that’s a fair answer. Let me ask you one last question. And I haven’t heard you speak about this, so I don’t know if you have views on this.

At least in the United States, a lot of people are—we’ve been focused on—our earlier conversation was about the danger of these technologies being in private hands and the need, perhaps, since the federal government is the guarantor of the public interest and the supplier of public goods that we can’t rely on the private sector for, there’s a natural assumption that the government should perhaps, at least for dangerous uses—as for negative externalities and the like—have some access to and control over these technologies. That’s the basic argument.

But there’s another worry, which is the worry of the government possessing these technologies and the bad things the government could do with these technologies. And there’s an endless list. You know, it supercharges surveillance. One can imagine perfect control.

You know, we’re talking a lot in this country about the unitary executive. Well, you can imagine everyone being plugged into the presidential AI and taking immediate directions, and you can also imagine that taking nefarious uses.

There’s a real worry about government having these things. And so that’s just another worry in this whole horrible calculus of trade-offs that one has to think about. And I wonder if you have any thoughts about that.

Yeah, I mean, probably fewer thoughts than you do. But I would say that, first of all, I agree that AI is a centralizing technology. So the internet was a decentralizing technology in the sense that it became easier for less powerful actors with fewer resources to have as much information as the powerful ones.

And so the state, which might have had a good monopoly on some forms of important data, suddenly those data were democratized, and lots of people in the private sector could get them. You know, satellite surveillance of what’s going on on the ground, you know, was—I mean, it’s maybe not just the internet—but you have companies like Planet, which provide excellent images of what’s going on on the ground. That’s a private company, a startup in Northern California, which makes its data available to lots of customers.

And so you get the kind of democratization, privatization, diffusion of knowledge.

But with AI, you need to have as much data as you can. And then when you train on all that data, you sort of have this ability to centrally control, centrally understand things. And if the AI is held closely and not distributed, that’s an extremely centralizing phenomenon.

So, I mean, at the moment, you know, the way AI is being built is sort of in the hands of, you know, 10 or something companies. But they release it fairly openly, and so it’s pretty democratic.

But were that to change in the future—as we’ve been discussing—in a more closed way, it’s maybe government basically controlling domestic producers to the point where they could demand that they get one of the most powerful models only for the government or something like that.

Yes, that’s very centralizing and very disturbing in terms of civil liberties. And indeed, we’ve seen that argument come up explicitly in the fight between Anthropic and the Pentagon—not the Mythos one, the one before—where Dario Amodei made two objections. One was about autonomous lethal weapons. The other one was about the use of his AI for domestic mass surveillance.

And his point specifically was, you know, the law has not caught up to deal with the prospect of this very powerful AI in the hands of the government. And until it catches up, I don’t want my system to be used for this, because democracy lags behind where the tech is.

So I do think there’s a sort of whole set of concerns there, which you’re correct to raise. I’m not sure I’m pushing the argument beyond what you said in your question.

No, that’s great. I say I share his concern. The law has not caught up to the possible uses. And I think we’re going to be learning that over the next months and years.

Sebastian, thank you very much. It’s truly an extraordinary book. It’s not just about one person—it’s really a history of artificial intelligence and an exploration of all sorts of related issues. It’s a touching personal story, and it’s told accessibly and vividly and brilliantly, as usual. Thanks very much.

Thank you, Jack. It’s been a great pleasure.

Thumbnail: Classified intelligence material found during the FBI search of Mar-a-Lago on August 30, 2022. (Department of Justice.)

This is an edited transcript of an episode of “Executive Functions Chat.” You can listen to the full conversation by following or subscribing to the show on Substack, Apple, Spotify, or wherever you get your podcasts.

Jack Goldsmith: Today, I’m chatting with Professor Marty Lederman of Georgetown Law School, who is a former deputy assistant attorney general in the Office of Legal Counsel in the Biden administration. And we’re going to talk about the constitutionality of the Presidential Records Act.

The Presidential Records Act basically says it gives the United States ownership, possession, and control of presidential records. That’s what the statute says. And then it makes the president responsible for managing and preserving those records in office. It transfers custody and ownership—well, I guess just custody—to the National Archives once the president leaves office, and then it governs access to those presidential records thereafter.

And the reason we’re talking about this is because, on April 1st, the Office of Legal Counsel, where Marty and I both worked in different administrations, issued an opinion that said that the Presidential Records Act is unconstitutional for two independent but interlocking reasons—and that’s a quote from OLC. First, OLC said that it exceeds Congress’s enumerated and implied powers. And second, OLC said that the PRA, the Presidential Records Act, aggrandizes the legislative branch at the expense of the constitutional independence and autonomy of the executive.

Okay, Marty, we’re going to talk about these legal arguments, but let’s start off—if you could just—what’s at stake here? Why is this law important? And what’s at stake with whether it’s constitutional or not?

Marty Lederman: Yeah, so I’ll give a quick sort of historical background, which the OLC opinion tracks as well. So, from the beginning of the Republic until 1934, there weren’t a lot of records created by the president or the folks right around the president, but there were some. And Congress effectively allowed presidents to take those records with them when they left office.

And that, over time, was understood to be a congressional acquiescence in effectively transferring the property of those records to the president, to control the way he wished to do so after leaving office. And presidents treated it in various different ways. They made a lot of those records public, and some they didn’t.

In 1934, Franklin Roosevelt—so, there weren’t a lot of records, and there wasn’t really a repository within the federal government to hold them anyway, so it wasn’t like there was another option—in 1934, Roosevelt encourages and eventually signs the law creating the National Archives. And he also takes great pride in saying, I’m also going to donate all of my papers to the National Archives so that the people and the government can have access to them for historical reasons.

So, from 1934 until 1974, there is a place to store them. And every president between Roosevelt and LBJ, generally speaking, made all of their papers available and gave them to the archives, under certain terms and conditions, but generally speaking, to allow the public and the government to have access to them within certain parameters.

After President Nixon resigns, some of his tapes and other, you know, incriminating documents have obviously already been made public at that point—that’s why he resigns. Ford pardons him. And he says, well, the rest of these tapes in particular, and documents, belong to me, and I’m not going to follow in the footsteps of my predecessors. I want to take them and keep them and not give them to the archives.

You know, and this happens right in the wake of the greatest scandal of, you know, the post–Second World War era. And so that obviously engenders a lot of pushback from Congress and the executive branch. And, you know, President Ford encouraged and signed a bill that applied only to Nixon’s records. And it effectively—it did not transfer the property, interestingly enough. It actually gave him just compensation, but it required him to turn over those records to the archivists in the archives for the preservation of historical understanding and the public’s understanding, and the ability of future presidents to understand what Nixon had done, and Congresses and the courts.

And we’ll talk a little bit more about that—the various different actors who are able to get access and what the purposes are. So it’s a Nixon-specific statute that is then challenged by Nixon on about six or seven different grounds—constitutional grounds. And that gets up to the Supreme Court, and the Court rules seven to two against Nixon on every one of his claims. He had actually dropped one or two of them by the time the merits are there, but it rules against him on all of the claims, including, most importantly, a claim that this impermissibly impinges upon the ability of the President to perform his functions by creating a chilling effect going forward.

Or you write that presidents—if Congress were to do this to other presidents—they would know that their conversations are subject to eventual public disclosure or disclosure to their successors or to Congress, and that will inhibit presidents from doing the job the way it ought to be done. And the Court rejects that—what I’ll call the separation-of-powers argument—pretty decisively, and actually articulates a test, a very liberal test, pro-congressional test, for assessing that kind of constitutional claim: one that a federal statute impinges upon the President in a way that will unduly restrict the operation of his constitutional duties.

And so it’s a pretty determined—it’s a pretty strong opinion, supported by President Ford, who signed the bill, and by President Carter, who was the president at the time of the litigation.

Right after that decision, Jack, I think it’s important to note—and we’ll talk a little bit more about this coming forward—OLC nominally tries to distinguish this case, called Nixon versus Administrator, famously, tries to distinguish it on the ground that it was a Nixon-specific statute and it was in the wake of Watergate, and it was a sort of a one-off kind of thing.

The irony was many of Nixon’s claims—and the ones that caused some of the justices a little bit of angst—were that this was impermissible because it was limited to Nixon. It was a class-of-one statute that seemed—he raised a bill of attainder argument that got the most attention from the Court, an equal protection argument that he dropped before it got to the Court, and that sort of thing.

But Congress, after this decision, decides, well, you know what, that’s right—it shouldn’t be Nixon-specific. We don’t want this to happen again. And so we’re going to pass a bill that, with President Carter’s approval, starting in 1981, will make these basic rules applicable to every future president.

With one big change: this new bill—and the one at issue here, the Presidential Records Act—actually declares, as a matter of statutory law, we are now—this will now continue to be government property, not personal property. And then it subjects all of these presidential records to a series of regulations, which we can discuss.

I’ll stop here for a second. And that’s—and OLC at that time testified that this was plainly constitutional, and every president for almost a half a century has treated it as constitutional and complied with this statute.

I haven’t answered your question about what the implication is.

So you didn’t answer my question—what’s at stake here—but that’s okay. You raised a lot of things. So let’s just talk about some of those.

So I think that history is fair, except for one thing. I mean, wasn’t— you said that Congress, before the Nixon statute, you said that Congress had allowed presidents to do certain things. But really, wasn’t Congress mostly silent? And didn’t presidents actually claim—make various claims of ownership over their papers before Nixon?

I’m not sure there were ever any legal disputes about it. When Attorney General Saxbe actually issued an opinion when Nixon wanted—when Nixon made this assertion that he could just take these documents and never share them again. And Saxbe was asked, as Attorney General, is this government property or is it presidential property?

And his opinion basically said that Congress had silently—ordinarily, of course, you and I, Jack, the documents we created when we were government employees are government property, right? They were done—we were paid by the government, we used government facilities, it was done in our official capacity. All other actors within the government, of course, it’s government property.

And Saxbe basically argued that there had been an implicit acceptance by Congress over the years, without any specification, that it was okay for presidents to treat this as their own.

But I thought that the D.C. Circuit in the Nixon case held that they were his private records, and that the statute—well, it may have been valid—constituted a taking, and that he warranted compensation. And as I recall—I haven’t gone back to look at it—I think that case had a list of all of the prior presidents.

You’re talking about the Nixon case? The Nixon case in the ’70s.

Yeah. Sorry if I wasn’t clear about this.

Didn’t they assume or claim that it was Nixon’s property and that he was—yes, absolutely—the person that could regulate this, but that it was a taking, and that was because there had been no prior statute that had actually regulated this, and he had assumed?

Yes. So —

I don’t actually think this affects the legality of what happened.

No, no, it’s actually—Congress agreed. So with respect to the Nixon statute, Congress actually, for the first time, sort of said, we’re going to treat this as private property. The statute itself gave him a right to just compensation.

I think what you’re referring to—it’s been years since I read that decision—is that he ended up collecting, I think, something like $170 million—

Yes, exactly.

—from the government, because Congress actually came in and said, yeah, we’re going to continue this regime in which this is your property, but nevertheless, we’re going to require you to hold it with the archives.

So there was a statute for Nixon, and then there was a Presidential Records Act, and that was important because it prospectively claimed, for all presidents going forward, that this is public property.

And so—and then there was this regime which you just described that we were subject to, and that everybody in the executive branch is subject to, about certain papers being public property and subject to this regime.

So—and I don’t think that history is particularly dispositive one way or the other, but I think it is a little bit contested before the Nixon statute.

In any event, why is it important? We can come back to that. I don’t think it’s actually important, but why is this OLC/White House challenge to the Presidential Records Act, saying they’re not bound by it—why is it important? Is it because they’re going to destroy documents? Is it because they can use Signal now? I mean, why is this an important thing to worry about?

So, one of the first remarkable things about this OLC opinion is that it’s, I guess, kind of like some of the torture opinions in the Bush 43 administration—it’s OLC answering a question, a very broad question about the constitution, the facial constitutionality of a statute, without any indication on the front—on the page—about why they’re answering that question, right?

Ordinarily, OLC, you know, either the President or someone else in the administration comes to OLC and says, we would like to do X—is it lawful for us to do X? And OLC answers that question. Here, the question proposed to OLC by the White House Counsel was: Tell us whether this statute is facially constitutional.

Now, there may have been much more specificity in the internal discussions between the White House and OLC, but the opinion does not explain what turns on this or why the White House is interested in this question. So we’re left to speculate.

And I guess I’ll answer the question by—I guess—I’ll show what the statute does.

In the first instance, it requires both the President and the Vice President, and those working for the President in nine different components of the White House office—not OMB, not USTR, but most of the other White House components—it requires them to retain all government— all presidential records, which are broadly defined, but which exclude purely personal or partisan, you know, electoral records—campaign-like records.

And the President has a pretty wide berth at deciding what counts as that exclusion. But in any event, the President and those around him are required to basically maintain those records during the presidency itself.

And there’s an amendment in recent years that if you use non-governmental resources to create the records—like your personal phone or Signal or some of these new apps where they are not automatically retained—within 20 days, you have to transfer them to a government application or technology that will retain them.

So there’s an affirmative obligation to retain the records while you’re in office. And then, at the end of your term—of the President’s term—the records have to be transferred to NARA, to the archives, for preservation thereafter, subject to all sorts of different rules—which I think we’ll probably get into—about when they become available, to whom, and under what circumstances in the subsequent years after the President’s term.

So here, I think—here’s the answer to your question, what people are suspecting, but we don’t know.

The first concern is that it will allow the President and/or those around him not to abide by the statutory requirements of retaining certain information while he is in office—either tearing them up and throwing them out, or burning them, or putting them on, you know, Signal and letting them be deleted after a few days, or things like that. So that’s number one.

Number two would be not transferring them to NARA at the end of Trump’s term, but taking them to Mar-a-Lago, or doing like he did once before, and doing that in violation of the Presidential Records Act.

And number three would be what NARA does with the presidential records that it does have—both Trump’s and other presidents’ records. Will they turn them over as required by—will they engage in the processes that the statute requires and the disclosure that the statute requires in future years, or even right now, with respect to previous presidents’ records?

So it’s that whole series of different possible implications.

Okay, one more question on this. With regard to—just explain for people who aren’t expert on how the executive branch works—why is this important? Why is it important to retain presidential records? Why do we have this statute?

What is the danger of senior executive branch officials communicating on Signal or destroying records, putting them in a burn bag—which might be happening anyway, very hard to enforce in certain corners? Is it for history’s sake? Is it for accountability’s sake? What is the claim that it is important this statute exist and be abided by?

So I think the statute itself, and the legislative history behind it, and OLC’s testimony about it, and the Supreme Court’s decision in the Nixon case, all articulate what you could say are four different purposes for the retention and the ultimate disclosure.

Perhaps the most important one is, you say, just so the nation can understand its history and the way its government has worked, right? That the people—you know, it’s mentioned historians—but the idea is that people who study American government can explain to the world and to the governed how their government has acted, you know, for all the reasons we care about history, right?

It’s sort of obvious, and the Supreme Court states that right up front as the principal virtue of this statute, right, is to preserve the Republic’s history. It would be a little bit odd not to preserve them, you know—to hide what the government has done from the eyes of the public for centuries to come.

And then all three branches of the government itself have an interest in the preservation, which is reflected in the so-called special access provisions of the Act.

So, most importantly, the next president—and the one after him, and the one after her, and so on and so on—might have a great need to know what their predecessors have done in terms of negotiations with foreign countries, in terms of legislative strategy, all sorts of reasons.

I mean, Jack, we’ve worked in government. Of course, we’re constantly referring to records of what our offices have done in the past, right? That’s what you do every 15 minutes. That’s the only way you can do your job properly. So the executive branch has that.

Or, in particular cases, it could be that DOJ has a need for it if it’s investigating certain activities. And that’s what happened during the Biden administration, when DOJ was investigating Trump’s taking documents to Mar-a-Lago. The FBI had an investigation, and they asked NARA for those documents so that they could investigate what the former president had done. So there might be law enforcement reasons as well.

So the public—you know, public/history—and the executive, I think, are the two most important, but there are two others as well.

Congress might want to see some of the documents and have a need for them. And so, for instance, in this Congress, Representative Comer has requested from the archives—and the archives has given Comer, like my understanding is, enormous numbers of Biden presidential records—for his investigation of the autopen controversy and all manner of other things, for congressional oversight and consideration of future legislation and what have you.

And then the judiciary—like in the Nixon tapes case—you might have criminal or civil cases in which there’s a grand jury subpoena or a judicial subpoena for presidential records, and this statute also allows that.

So it’s the people and history, it’s the executive, it’s Congress, and it’s the judiciary—all of which have needs for these things.

And I think you also touched on probably one of the big motivations, which is the whole Mar-a-Lago documents matter, which is clearly still a thorn in the president’s side. He’s still upset about the Presidential Records Act—it was implicated in that—and he would like to unwind anything that could suggest that what he did was unlawful. He would like to unwind it and show it was illegitimate.

I think, at some level, they may want to be destroying documents now and manipulating the historical record, but I think that’s also—

It could be. There is this reference—I think you and I discussed briefly—in the OLC opinion that you could extrapolate from it and say, oh, and by the way, one implication of our holding is that this whole criminal investigation for violations of this Act was itself unconstitutional or invalid, because the statute is.

I mean, I think that was the driver of the whole thing.

Okay, let’s get to the constitutional arguments. OLC said, first, that the Presidential Records Act exceeds Congress’s enumerated and implied powers—i.e., there is no basis in Article I for Congress to enact this statute. What do you think about that?

So, Jack, I haven’t run into anyone from any political persuasion or any administration who thinks that these arguments—which take up the bulk of the OLC opinion—have any plausibility at all.

But explain why.

So, I will explain. It’s both what’s there and, most importantly, what’s not there. So, very briefly on what’s there—and we can go into this in greater detail, but I’m not sure it matters much other than as a reflection of OLC—OLC goes on for pages and pages about congressional oversight authority, for instance, which is really almost neither here nor there. It also takes extraordinarily—

Explain why it’s neither here nor there. OLC spends a lot of time talking about Congress’s powers of oversight, and this is important for the Supreme Court decision in Mazars and the like that it relies on. So why does it use that frame, and what’s wrong with using that?

You got me. It’s a weird frame to use.

But why? Why is it a weird frame to use?

Because those cases—Mazars and other oversight cases—are cases about when a congressional committee is seeking information with respect to its oversight authority or its impeachment authority from the executive, and it’s a particular request for particular documents.

And the Court has come up with various different doctrines related to when and under what circumstances presidents can resist, or executive branch officials—or private parties in Mazars—can resist those sorts of claims for congressional oversight. And that raises a whole slew of questions about: Is Congress’s interest in this particular investigation valid? Does it outweigh the needs of the president? Are there privilege questions at play?

It’s very case-specific, very document-specific, very specific to a particular congressional investigation. And those cases and those doctrines just have never been considered with respect to a facial challenge to a statute—which, yes, as I just mentioned, sometimes, like with Comer, Congress might use the PRA for oversight purposes—but that’s hardly the most important application of this statute.

Okay, so I agree with you. It’s not obvious why the constitutionality of a piece of legislation is viewed as an example of oversight, which we tend to think is more contextual and specific.

So what is the right lens for thinking about this? They say that Congress has no Article I authority to regulate. So what is the strongest basis of Article I authority?

So I actually think there are at least three fonts of obvious congressional power here, only one of which is in Article I, by the way.

And the one that’s in Article I is just the Necessary and Proper Clause. It’s necessary and proper for Congress to carry into execution the laws that it has created. Congress creates—except for the President and the Vice President—all of the positions within the Office of the White House, right, the Executive Office of the President.

It appropriates funds for all of the materials—other than a private phone—to be making these sorts of records. It basically creates the structure of the executive branch, but for the president. And obviously, Congress enacts laws all the time that regulate the funding and operations and limitations of what folks in the Executive Office of the President, like the agencies, can do.

That’s just ordinary, everyday Necessary and Proper Clause legislation with respect to both congressional money and congressionally created positions and functions. That’s number one.

But two others are sort of even more prominent. One is the one the Supreme Court mentioned. Now, I will say here—I said Nixon made six or seven different constitutional arguments—it’s noteworthy that he did not argue—none of them was that Congress lacked the affirmative authority to enact his statute, right?

Like, this argument was so disfavored that even Nixon, who was throwing everything in the kitchen sink in the case, did not make it.

So one of them is the one the Court held was sort of the basis—and it’s kind of an inherent power of government to preserve the history of its operations, right? Imagine if you were writing a constitution for anything—for any government, a nation, a state, a city, a private institution—of course your lawmaking body would be—it would be assumed—you’d have the power to preserve the records of your own operations.

Like, who wouldn’t put that in the Constitution? So it’s that inherent power to preserve our nation’s history, which is what the Supreme Court referred to, among other things, in the Nixon case.

So there’s those two. And then the one that’s most prominently missing—which isn’t in Article I, but in Article IV—is Congress has the power to enact regulations with respect to government property—property of the United States—the Property Clause, which the OLC opinion doesn’t even mention.

It was mentioned in the lower court decisions in the Nixon case, and it was invoked by OLC when this statute was being contemplated. It is completely inexplicable why OLC—it didn’t—it either didn’t occur to anyone in OLC to think about the Property Clause; they don’t challenge that it’s government property, interestingly; they don’t mention it either.

Either someone did mention it to them, that they ought to address that question, or they didn’t look at the briefs in the old cases, and they didn’t talk to anyone who would have told them. I think it’s fair to say, Jack, that virtually anyone—any career lawyers in OLC or at the archives—would have told them in five minutes, what are you talking about? There’s the Property Clause.

I agree with you. They should have definitely addressed the Property Clause, especially since the government has relied on the Property Clause in very closely analogous contexts.

Does it matter—the Property Clause in Article IV—it comes in the context of talking about territories. It does say “property of the territories.” And what is the exact language? Congress shall have the power to dispose of and make all needful rules and regulations respecting the territory or other property belonging to the United States.

So do you think, in the Property Clause, they were—when they were focused on territories—they were just thinking about all government property, and that’s where they put it instead of Article I?

I don’t know about instead of Article I, but in addition to Article I. I think even if there weren’t a Property Clause, I think there’s a very strong argument.

I agree, but I’m just trying to figure out the relevance of the Property Clause. Obviously, the Necessary and Proper Clause in Article I applies to all government powers, so it doesn’t matter for that purpose that the Property Clause is in Article IV. I’m just wondering whether the Property Clause had this type of legislation in mind. I’m not even talking about it as an original matter. I’m just talking about whether they just focused on territories here—this is Article IV that is about horizontal sovereignty.

It clearly includes chattels and real property and buildings and materials and the stapler. You and I couldn’t take staplers home from the office because it was government property.

Here, I think it’s sufficient just to quote OLC—this is Larry Hammond of OLC testifying in 1978: “It is well established that the work product of government employees prepared at the direction of their employer or in the course of their duties is government property. Should Congress choose to extend this principle to cover records prepared or received by the president in the course of his duties, no substantial problems would, in our view, be raised.”

And then they did mention the Property Clause there. But I’m not sure exactly what your question is.

I’m just trying, for the general reader, to point out—I think that the Property Clause is probably deeply relevant here, and I think it has been treated as deeply relevant for a long time in the government. And it is remarkable that OLC didn’t even mention it.

I’m just pointing out that this is in Article IV, in a context that seems to be something else, and what you’re saying is basically this hasn’t mattered to the understanding of the Property Clause in the past.

I’ve never studied—quite honestly, I’ve never had any occasion to study—the full scope of the Property Clause. I do know that in the preliminary injunction motion that was recently filed in a case that we’ll talk about in a few minutes, the plaintiffs do have a little riff on the scope of the Property Clause. I haven’t really carefully looked at it.

Okay. What about the second argument, Marty? The second argument is that the Presidential Records Act aggrandizes the legislative branch at the expense of the constitutional independence and autonomy of the executive.

Yeah. So, just to put that in the context of this opinion—so what’s kind of shocking about the opinion: There are about 40 pages of analysis. The first 25 to 30 are on this—are just this sort of diversion about the lack of affirmative authority by Congress and oversight authorities and all this stuff, without talking about some of the most important authorities.

And then only then do you get to sort of this backup argument, they say it’s in conjunction with, which is effectively a separation-of-powers argument. It amounts to an argument that this statute unduly impinges upon or chills the conduct of the President’s activities in ways that will prevent him from performing his constitutional and statutory duties.

And that is the first argument that the Court addressed in the Nixon v. Administrator case, right? So perhaps the reason they relegated this to the back end of the opinion is they were going to have to deal with the Supreme Court case directly on point—and something they only mention in a footnote—with OLC testimony about this Act that addressed this question and rejected this constitutional objection, said that it was meritless, especially after the Supreme Court opinion.

So they use the word “aggrandizement.” I don’t think—it’s not necessary for your viewers—but that’s the wrong term. The aggrandizement question is just when Congress inserts itself into executive functions. There’s nothing like that in the statute here. So OLC is misusing that whole concept.

What they’re thinking about is—I think the argument is that the Nixon v. Administrator of General Services—that case engaged in a style of separation-of-powers analysis that the Supreme Court is not inclined to address now.

So the OLC seemed to be saying—I think they refer to it as the ancien régime of separation of powers—and there was a Brennan opinion, there was a kind of balancing test that the Court doesn’t really go in for now.

So, I mean, I have a response to that, but I’m wondering what your response to that is. Is that relevant here?

Well, I’d like to hear your response. You’re right about what their argument is. So they spend a page or so saying that Nixon’s distinguishable.

Actually, this is an easier case than Nixon in two ways. Here, the documents are government property—they weren’t in Nixon. And there, the way that OLC purports to distinguish Nixon is that that was only with respect to one president, not all presidents, and therefore it was more acceptable. Actually, that was the constitutional problem that was most acute to the justices there. This is better because it’s not a class of one.

So they don’t really make much of an effort to say it’s distinguishable. But what they say is basically—they use the word “wrong.” They say the Supreme Court was wrong. It’s the ancien régime, right? It basically—it’s a Bill Brennan opinion, therefore it must be wrong—is the not-so-subtle subtext there.

This is from a different era. We are just—it’s just a sort of—I don’t know if I’ve ever seen this in an OLC opinion, Jack. I’m wondering if you know of another example where OLC is saying even if there’s a Supreme Court precedent directly on point—and OLC has testified that it makes this an easy question—we are hereby not going to follow that because we think the Supreme Court got it wrong.

So I don’t think they quite said that. I agree —

They literally say it’s wrong.

I know they say it’s wrong, but they also distinguish it.

I’m saying the distinguishing is pathetic.

But they’re not—it’s a different statute, it was a different structure. It didn’t rule on the constitutional question of Congress’s power because they just assumed it.

No, on this one, though, on the intrusion into the president’s functioning—the Court did address that directly. That was the principal claim made there. And when OLC gets around to explaining why—

I agree—so here’s why I think, they didn’t really get into this, but what is the separation-of-powers analysis they would like? It’s a more formalistic one. But I don’t really understand how that helps them, because to me, if it’s a formalistic analysis, you start with Article I power.

And I think, for all the reasons you explained—especially the Property Clause, but the others as well, maybe not the inherent powers argument today, although that’s been an important argument in this context—I think Congress clearly has Article I power here to define and regulate government property.

Article IV. And Article I, yeah.

Article IV, but with the Necessary and Proper Clause on top of it.

And then the question is, well, you know, does the president have some exclusive power vis-à-vis this? And here’s what I don’t really understand. The president is claiming—it’s an Article II argument, something about the office of the president—but they’re relying on an Article II argument to say that the president has a private—in a way, kind of private—control and can do what the president wants with these documents.

I don’t see how you get an exclusive—I don’t even know where it would come from—the exclusive Article II power to dispose of this property as the president wishes. That’s the part that I think a formalistic analysis—

I’m going to make an analogy in a minute that’ll be a bit provocative, just because I just taught it to my separation-of-powers students, but let me hold that for a second.

I think I agree with you—I don’t see what’s there. I think it’s important for your viewers to understand that this statute bends over backwards in several ways to preserve presidential prerogatives and to ensure that there is not an undue burden on the president’s functioning.

It allows not only the current president at the time of a dispute, but even a former president, to raise an executive privilege claim, to have that adjudicated by the courts if it gets to that point, right? To get to the courts—it gives the former presidents a lot of control to make it so that, for 12 years, there’s basically not very much access to this. There’s no judicial review of archives decisions to withhold documents until the 12 years is up.

This is an extremely solicitous-of-the-president statute, which is not surprising because Carter was signing the statute, right? The president was approving this, so it’s got all these protections built in.

So then the obvious question is, well, how does this really make the president’s performance of his or her duties any more difficult? And the one concrete example they give—it’s like, I don’t know if I would have ever had the guts to write this—is it takes up too much time of the White House Counsel’s Office to advise the president on how to comply with the PRA.

I hope I don’t have to explain why that’s not a very good constitutional argument.

Explain why.

Well, for one thing, the Office of White House Counsel is created and appropriated by Congress for the purpose of advising the president how to comply with the law. To say that that’s diverting them from their functions is kind of a non sequitur. That is the purpose of that office.

And, you know, Congress doesn’t have to create a White House Counsel at all, right? There wasn’t much of a White House Counsel’s Office until after Watergate. And I just don’t even see how one would constitutionalize that argument.

And that can’t be what’s going on, right? They want to be able to destroy these documents. The broader claim is that even non-privileged communications—because remember, the privileged communications are going to remain secret—but non-privileged communications, the knowledge that they will come to light 12 years later or something like that, or be available to Congress or the FBI or a judicial subpoena in a shorter amount of time, will so chill the operations of the executive branch that they won’t be able to do their job properly.

And, you know, the plaintiffs’ main response to that is, well, it hasn’t chilled. The presidency has been working just fine for 50 years, thank you very much. The statute has not really had much of a detrimental impact at all.

Do you think—but do you agree with me that they don’t really explain how it is that Article II can create an exclusive power in the executive branch for the president to do what he wants with the document?

No, they don’t make either a legal argument to that effect or even one that sort of resonates in terms of the facts. I think, you know, if you press them with truth serum, they would say something like just the ordinary human thing, right?

Which is, a lot of people these days do things by phone or try not to create records, right? If you’re going to be more candid, more forthright, if you don’t have to write something down for, you know, history’s preservation, right? If you know no one will ever know what was going on, you will be more candid, even apart from executive privilege claims.

And I guess, in some respects, that’s true, right? That’s the way we are—when we have a pretty certain idea that someone we’re speaking to will never repeat it elsewhere, we tend to be a little bit more candid than if we know that it might come out 12 years later.

So I think that the strongest practical argument—it’s not really an argument, but a practical line in the opinion, and let me flesh this out before you comment—is when it says, well, if Congress can regulate presidential records, they could regulate Supreme Court records. And they’re going to say that district court records.

And I actually think that’s right. Congress could regulate records of the federal judiciary, including the Supreme Court. I think it would have to be done differently to take into account the very different structure of Article II and the ways in which administrations turn over, with judges retiring while their colleagues are still in the Court and the like.

But I have no doubt that Congress could enact a constitutional Federal Judiciary Records Act of some scope, but I do think this is going to give judges pause. What do you think about that?

I agree with that. It’s a common move in constitutional adjudication that might get up to the Supreme Court to say, well, what if this were you? Sam Alito’s out there publicly having said Congress can’t regulate the Supreme Court at all—that’s clearly not right. Congress has regulated the Supreme Court a good deal over the years.

Well, it can determine how many people are on the Court.

Yeah, when its term shall be, in a million different ways. Yes, it could require this. You might say there are questions about whether the judicial power—as you know, there’s a handful of cases that have construed those pretty vague words to impose some sort of baseline thing that Congress cannot regulate.

The judicial power includes the power to issue a final judgment that can’t be reopened by the legislature. I suppose someone might make an argument that having complete control over records infinitely into the future is part of what it means to be exercising the judicial power. I don’t know what that argument would look like.

I think it’s kind of tough. I mean, you’re really getting into whether the institution—the branch’s control over its deliberations—is something it is the master of. And once you acknowledge that the executive branch can deliberate however it wants, but it has to preserve a record of those deliberations when a record was created, I think that the same basic argument has to apply to Article III.

Although, again, I think there would have to be different kinds of accommodations. I don’t see how you could craft an argument—maybe just my lack of imagination—based on the judicial power to distinguish it from the executive power in terms of Congress’s ability to—

I agree, but you can imagine—and we’ll talk in a minute about whether DOJ will defend the merits of the OLC opinion—but you can imagine that, if they do, and it’s getting up to the Supreme Court, yeah, you start from really provocative hypotheticals.

Could Congress pass a law requiring that the justices videotape their Friday conferences so that, 40 years later, people can understand what their deliberations were? There are some who think maybe Congress can’t even require videotaping of oral arguments, that that’s something off-limits to Congress. I don’t think that’s a very good argument, but the Supreme Court would not take well to that statute on any of that.

This is why I mentioned it. I think this is—set aside the fancy doctrinal arguments—this is going to be the main intuitive argument that’s going to give many members of the judiciary, and I think members of the Supreme Court, pause in trying to understand how—well, you say that it doesn’t impact deliberations and it doesn’t affect the operation of Article II, but boy, we can imagine how it would impact us, so maybe we need to—

Obviously, the executive branch is just so very differently structured and organized, with a very different function from the judiciary, but I still think it’s going to be, if not on the surface, in the background.

No doubt, no doubt about it.

So let’s talk about—there’s a lawsuit, and it’s brought by scholars’ associations that are seeking a variety of injunctions and mandamus and declarations of unconstitutionality of various things related to the OLC opinion and the follow-up on it.

I mean, do you have any general thoughts about the lawsuit? The obvious questions are that, even if they have a powerful case on the merits—and I think that they do—the obvious questions are: do they have a right to sue, and especially, is there standing?

Yep, I think that’s exactly right. So just a little background for your readers—and maybe we’ll know more, maybe we can do a follow-up in the next couple of weeks as this goes on.

So the case was originally assigned to Judge Howell in the District of Columbia. It’s now been given to Judge John Bates, who’s a very, very serious—Judge Howell is as well—a very serious and scholarly and thoughtful judge who’s written on separation-of-powers cases before.

Interestingly enough, I don’t have any doubt that Judge Bates and the appellate courts would rule for the plaintiffs on the merits if the merits are eventually reached. I don’t think this is a hard case at all.

Okay, we just spent a long time talking about the merits. Let’s talk about getting to the merits.

Yes—and getting to the merits. So just to tee it up: there’s a motion for a preliminary injunction that is pending. The plaintiffs asked for expedited consideration; the government opposed that.

The plaintiffs said to the government, we’re willing to give you more time to write your brief if you will just agree that you won’t violate the Presidential Records Act while the PI is being litigated. The government refused to do that. So it sort of definitely leads to the suggestion that they are currently not complying with the statute in one manner or another—we don’t know how—or more than one. So Judge Bates ruled for the plaintiffs.

They pretty much said they weren’t complying with it, as I read. They sent instructions to have people continue to do things with records that were not as broad as the Presidential Records Act, which makes me—

It seems that way. But it’s going to be a little bit vague to Judge Bates exactly how—what are they doing with this OLC opinion, and how much are they doing that’s irreparable, that cannot be repaired?

I’ll ask that question.

So anyway, he’s expedited the briefing. The government’s brief is due Tuesday—so in four days. It’ll be very interesting to see whether they limit their arguments just to the standing/justiciability arguments or reach the merits. And the hearing is, I believe, on May 5th for the PI.

You said a second ago—and I still want to get your views, to the extent you can offer them, on justiciability—but you said a second ago you weren’t even sure the government was going to. I think the government—I don’t have any doubt that the government’s going to be all-in in defense of this position.

The position the president wants is the position the White House wants. There’s an OLC opinion on it. I don’t see why—I agree it’s going to be hard to square with some past representations, but I don’t see—But there is some slippage in some of—

All right. You may be right. I think this is a great challenge for the Civil Division and the Office of the Solicitor General, because I just think the arguments here are extraordinarily weak and would effectively require a district court judge or a court of appeals to disregard a governing Supreme Court precedent and OLC testimony from the ’70s.

Yeah, I don’t think OLC testimony won’t cut much weight.

No, but it’s reflective of how easy the question was, right? So one thing that’s interesting to me as a strategic matter is that the plaintiffs have asked for injunctive relief against several different defendants in the executive branch. One is the archives—that they should comply, and therefore they shouldn’t refuse to disclose something on the ground that this is unconstitutional.

Another is against all the sort of other officials within the Executive Office of the President—that they should comply with their retention requirements. Obviously, the requirements that take place on January 20th, 2029—those are three years away, right? So those requirements are three years away.

And they’ve made one claim—they’ve actually asked the court to set aside the OLC opinion, which I kind of think is not a thing.

Yeah, that is a—when I read that, I thought—

That was overreaching. A declaration that it’s wrong is all you need. There’s no such thing as setting aside an opinion.

And also, what is—they talked about final agency action here under the APA. What is that?

Well, that’s as to NARA. NARA is the only entity that’s—well, yes, there’s a claim against the attorney general about the OLC opinion. I don’t think that’s going to have any legs.

What is the final agency action by NARA? It would be not complying—systematically not complying with its obligation.

Anyway, one thing that’s interestingly not in the case is that they have not asked for any injunctive relief against President Trump in his official capacity. They are asking to enjoin him, once he becomes a private citizen again at the end of his term, from taking or destroying the documents.

But they are not seeking an injunction against the president’s own destruction of or failure to preserve documents that he’s involved in. Now, I know why they’re doing that, because, as you and I know, there’s this old Civil War–era case, Mississippi v. Johnson, that Justice Scalia was very fond of, that suggests that the judiciary doesn’t have the power to enjoin the president directly.

And I think there are a lot of justices on the Supreme Court—and there are justices—but I think it’s a very, very bad, incorrect reading of that case and of the Constitution. I think there are historical examples of presidents who have been enjoined, but I think that’s the reason the plaintiffs are not seeking an injunction against Trump.

I think that’s prudent, actually.

And it might be prudent. Okay, so what are the problems? The problems are potentially cause of action—although there are a bunch of cases where the Court has said, Armstrong, Free Enterprise Fund, that there’s an implied equitable cause of action to seek an injunction against executive branch officials for not complying with statutes.

The really hard question for Judge Bates, I think, is going to be standing. We haven’t seen the government’s brief yet. And I think it’ll be something like the following: yes, these organizations—the American Historical Association, American Oversight—they make requests under the PRA for enormous numbers of presidential records all the time. That’s what they’re in the business of doing.

And if all the records were being destroyed, you could be pretty sure that they would be injured at some point down the line. And the only way to prevent that injury is to act now, to not allow the destruction.

But right now, it’s a black box about which records aren’t being retained, which ones are being destroyed. Do we know for sure that anything they would ever seek from NARA would be destroyed? We don’t know for sure. And therefore, it’s kind of a speculation question.

It’s like a Clapper kind of thing—it sounds very Clapper-ish, but without the national security overhang of Clapper.

So on the one hand, this seems rather speculative, because it’s hard to identify particular documents that they will not have access to that they otherwise would have. But on the other hand, if you don’t issue an injunction now, anything that’s destroyed now—there’s no way of getting that back.

So it’s kind of a—I don’t know that it’s controlled by any particular—it’s got a weird mix of standing issues, I think.

That’s a problem—the Clapper problem, we’ll call it. It’s not exactly like Clapper, but it’s a cousin, maybe. Is that going to be a problem for any plaintiff, I think? Or is there—or can you imagine a plaintiff that would get around the Clapper argument?

It’s very easy to imagine a plaintiff, right? So if the following took place, right: If NARA has thousands of requests right now for presidential records, right—including from the first Trump administration, because five years just passed since the end of the first Trump administration—they’re processing them. They always take a long time. They’re always late on processing. They have a huge volume of things.

If they were to ever say, no, we would give you this document, but now that it’s unconstitutional, we don’t have to, that would obviously give someone standing. But I don’t think they’re going to do that anytime soon, if I had to predict.

Like, I’m not sure what they’re doing over at NARA, or if they’ve even been given instructions on what to do with pending requests, right? And so that would create standing, but that might not occur for six months or 18 months—who knows? And in the meantime, a lot of damage could have been done.

So that’s the problem. You get this case—the Court has said many times, including in Clapper, that even in a case where no one would have standing, that doesn’t mean you have to have adjudication. But the Court cares about whether the government would be able to violate a statute without anyone having standing.

And so I think this is a really tough question going forward. If I were Judge Bates, I sure would want to know—unlike in Clapper, where you couldn’t kind of publicly find out what the NSA was doing—here, there’s no national security reason you can’t find out.

Not yet.

Not yet. It’s all national security—

It will be raised. That’s coming, I bet.

Oh, even like, just tell us what you’re doing with presidential records? There’s no privilege claim or national security.

This is an administration that has made national security claims wildly, so I think it’s coming, but—

Okay, if I were Judge Bates, I would think he would be inclined to see whether there’s a way—I don’t, I haven’t thought about this very hard—to insist that the government reveal something about how it’s implementing the OLC opinion.

Do you think it can do that? Rules on standing as a condition of—

No, no—as a—because that will determine standing.

Yeah, so he has to connect those things.

Yeah, and I haven’t thought that through, Jack, quite honestly.

No, but I think that’s right.

What do you think? Do you think the district judge has that power, kind of?

I think if there’s a powerful standing argument, there will be a claim that they can’t get to the merits, and they would construe that question as a merits question. So I think it’s prudent, if he can, to tie that question into trying to understand whether there’s standing.

If I’m a district court judge, that’s what I would do.

Yeah, I think most of the action is going to be on this question, at least in the near term.

All right—anything else?

I don’t think so. Did we cover most of what you wanted to cover?

Everything on my list. Thanks so much, Marty.

Thank you, Jack. It’s always a pleasure and an honor to talk to you about these things. I hope it’s interesting to your viewers.

It will be.

The Presidential Records Act!

I know—If they stay for the whole hour. Thanks very much.

Take care.

Mentioned:

• Sarah Isgur, Last Branch Standing: A Potentially Surprising, Occasionally Witty Journey Inside Today’s Supreme Court (2026)

This is an edited transcript of an episode of “Executive Functions Chat.” You can listen to the full conversation by following or subscribing to the show on Substack, Apple, Spotify, or wherever you get your podcasts.

Jack Goldsmith: Today, I’m chatting with Sarah Isger about her new book, her first book, Last Branch Standing: A Potentially Surprising, Occasionally Witty Journey Inside Today’s Supreme Court. Sarah needs no introduction, but I’ll give her one.

She’s the editor of SCOTUSblog. She’s the host of the Advisory Opinions podcast. She’s a legal analyst for ABC News, and she’s arguably my most famous former student—and certainly the most famous student who went to so few classes of mine, that is, who skipped so many classes. Is that fair, Sarah?

Sarah Isgur: With your permission. That’s important.

Okay. All right. So, this is great. I love the book. Why don’t we start by just telling us what the title means?

So, the title came at the last possible hour for this book, as we were trying to figure out what would be fun and explain the thesis of the book in as few words as possible, like a hipster coffee shop. And Last Branch Standing is supposed to capture this separation-of-powers crisis that we’re in right now, where, if the founders were walking around today—this is what Scott, my husband, and I were talking about that night over a bottle of wine—if the founders were walking around today, I think they would be stunned.

They would look at Congress, and just their assumptions over how ambition would work would have failed. And they would be really sad about Congress. I think they would look at the presidency and be basically like, oh, James Madison wrote down the definition of tyranny for you guys, and you just thought of it as a checklist. How great. They’d be horrified by that.

But I think they’d look at the Supreme Court and say, yeah, that’s about right. It looks a little different. There are nine justices. They have their own building. But we wanted something removed from the majority, from the mobs that we were so afraid of—direct democracy. We wanted an independent branch, and we wanted it to be counter-majoritarian to the other two branches, to protect people’s rights against the government, against different parts of the government from one another.

And they’re the last branch actually doing their constitutional function, while the other two branches fail in opposite directions.

You don’t think—so I agree with that, and I think they would be most surprised about Congress and the presidency and the directions they went in. And I agree that the Court is probably more like what the framers imagined. But the Court—it’s the Court’s jurisdiction, broadly considered, in terms of the number and types of really controversial cases that it takes on, and the really extraordinary importance it’s had in American life really since the beginning, or near the beginning. You think they would have thought, yeah, that’s about what we expected?

I think so. I mean, they let each branch sort of develop itself. I think they knew there would be almost like a—I mean, they didn’t know who Darwin was—but a Darwinian process for each of these branches to some extent.

But part of the reason I think we feel that the Court is the final arbiter of so many of our culture-war, political, partisan fights right now is because the other two branches have failed. So take the mail-in ballot case or the remain-in-Mexico case just from this term. And that’s supposed to be just the Court saying—interpreting what a congressional statute means. Did they mean yes or did they mean no for this exact example that the state of Mississippi or the executive branch wants to do?

And if the Court says yes, but Congress is like, no, that’s not what we meant, Congress would just pass a statute the next day or amend the current statute and make it more clear, like they did, for instance, in the Voting Rights Act. Because nobody believes that Congress will do that, because nobody believes that the amendment process still works, is also part of this problem.

The Court becomes the final word on all of these questions. And so the fight for who these justices are, the fight for how these cases are decided—it has just turned the temperature up to boiling.

Yeah. So I want to come back—we’re going to talk mostly about executive power, since that’s what we do here—but I want to ask you a few more questions about the Court. I’m going to come back and ask you about the Court as the final word when we talk about Trump’s attitude towards the Court, because I think that’s another thing that the framers would be surprised about. But first, before we get to executive power, you’ve got a general theory in the book about the decisional dynamics on the Court, and it’s not as simple as conservatives versus liberals. Can you just basically sketch it?

Yeah. So I went into college as a math major and quickly dropped that, with future dreams of finding a mentor like Jack Goldsmith. But at that point, I had no idea. Anyway, so think back to high school algebra and your little grid paper of a vertical axis and a horizontal axis.

That horizontal axis is ideology. I don’t think it’s partisanship, but conservative to liberal, or something along those lines. But if you actually want to predict how the cases are going to come out, that’s not very helpful. Only 15 percent of the cases last term were decided with all of the liberals in dissent, and 15 percent of the cases, by the way, were decided with only conservatives in dissent as well. It was literally the exact same number of cases, and I sort of like that symmetry from last term.

So I talk about this vertical axis, and it’s this idea of something like institutionalism—how much you care about stare decisis. I like to compare Gorsuch and Kavanaugh. I mean, they’re twins in their résumés. They’re the same amount of conservative, right? They come up through the Federalist Society rock-tumbling system for justices, and they both end up on the Court nominated by the same president.

They only sided on the same side of the “V,” so to speak, 50 percent of the time last term. And it’s not because they’re so different on that ideological axis; it’s because they’re so far apart on that institutional axis. Justice Gorsuch thinks of himself as just Justice Gorsuch—“just Neil,” we could call it. And so his vote is all he can give you. He looks at the law and the facts inside the four corners, and what Neil thinks of that—that’s it. That’s his role.

I think Justice Kavanaugh thinks of himself as part of a choir, a middle school group project, if you will. The Court is speaking with one voice. He’s simply a member of that body. But it’s not about what he wants or what he thinks. And so if he can take out a paragraph and get another vote, they can narrow the aperture or not take that case—that is what he views as his role on the Court. They’re just very different views of that institutionalism part.

And the other institutionalists are Justice Barrett and the chief justice, and the other non-institutionalists—is that what you call them?

I have called them that. I mean, I could use some better branding if someone has ideas. So, in their high school lunch cafeteria—you know, who sits with who at lunch—you have the chief, Barrett, and Kavanaugh. They’re sort of our popular kids. They’re going to be in the majority the most—you know, jocks on the football team.

You have our freaks-and-geeks table of Alito, Thomas, and Gorsuch. They’re sitting together, but they don’t actually talk to each other at lunch. And then, you know, we have our high-academic-nerd crew—maybe Kagan, Sotomayor, and Jackson—although Kagan is a high institutionalist, whereas Jackson is a very low institutionalist.

And so you’ll see Kagan break off a lot more. A lot of 7–2 decisions from last term, and the interim docket dissents that Justice Jackson has written with our sort of most strident language—Justice Kagan joined none of those dissents. And I think, again, it’s not an ideological difference; it’s an institutionalism difference.

And Justice Kagan’s role as a strategy maker—you know, if she’s going to be in dissent, if she’s going to be more liberal, how do you get to the outcomes or prevent the outcomes that you don’t want? And Justice Jackson is like, my job, like Justice Gorsuch, is just to say, nope, that’s wrong. Nope, you’re wrong. Nope, I hate this. That’s not how Justice Kagan views her job.

Do you think institutionalists have—I don’t know how you’d measure this, because I can imagine both sides of the argument—do they have more power on the Court because they are more pragmatic about the details and more willing to find something, say, to achieve a majority? So therefore, they’re more—why don’t you give us the numbers on how often the institutionalists are in the majority? They must be in the majority much more often, I take it. Is that right?

Way, way more often. Justice Kavanaugh has been in the majority more than any justice in modern history. Over all of the terms he’s been on the Court, I think we’re at 93% on average. The chief, of course, is in the 90s over most of his tenure. Justice Barrett is the third, but she’s still at like 90–91%.

But it gets to this question of what it means to be influential. I think Justice Gorsuch and Justice Jackson, on opposite ends of that ideological spectrum but quite close together in terms of how they view their jobs, would say, no, no—influence is not being in the majority. It’s not being the fifth vote that’s deciding a case.

Influence is being John Marshall Harlan or being Justice Scalia. It is having a theory of the law and sticking to it and passing that on to posterity and future generations. My percentage in the majority is irrelevant to my influence on the American experiment.

And Justice Thomas has been on the Court for a long time, but he’s arguably an example of that, I take it.

I think so too. Although, as he’s handing off the baton of text, history, and tradition, that Rahimi case has to scare him a little bit. He seems to be deciding this in line with Bruen—his text, history, and tradition coming-out party—and all of his potential protégés on the Court are like, yeah, yeah, we’re definitely doing your test, but it’s coming out the opposite way you think it should.

Which, if you’re Justice Thomas on your way out the door—whether it’s this year or in 10 years—that’s not great.

Okay, I want to shift to executive power. So what is your basic take on the Court dealing with Trump 2.0? You know, there was a period in which the narrative was the Court is totally under the thumb of the Trump administration or totally caving to the Trump administration. That narrative, I think, is much more difficult to sustain now for a variety of reasons.

But what the Court has been dealing with—what I take it is the challenge of the Trump administration—is a very, very aggressive conception of executive power and a very aggressive, critical, harsh attitude towards the federal courts, at least the lower courts.

Well, I mean, the smartest person on this issue is my favorite former law professor, Jack Goldsmith, who has written extensively on it, and I associate myself with all of his remarks on this topic.

But what I think is helpful is to step back in sort of these grand projects that the Court has undertaken from time to time. In the Rehnquist Court, I think you start to see the beginning of a Commerce Clause project, right? This idea that Congress has exceeded its constitutional bounds of a limited government.

And so, of course, Morrison and Lopez and these other cases in the ’90s sort of don’t go anywhere, but that’s because the two branches change so dramatically. When Obama comes into office and does the “pen and phone” thing, the constitutional crisis that the Court has in its lap looks totally different than it did in the ’90s.

It’s not Congress with a Wickard problem—legislating every problem it sees in the world instead of leaving it to the states, the vertical federalism problem. It’s this horizontal separation of powers, where the president is just leaving Congress behind at the end of the hundred-year Progressive Era experiment of replacing Congress with the administrative state.

And so I think the Roberts Court has thought here now for two decades: this is our constitutional crisis to try to rebalance—not in any one case, but over the course of many cases and many presidents—because what the Court has that the other two branches don’t have is longevity. We’ve had so few chief justices—17—compared to nearly 50 presidents.

And so you have Roberts, over the course of Obama and Trump and Biden and now Trump again, limiting presidential power, giving the president more accountability over, for instance, personnel or money to spend within the presidency, but making it a weaker presidency.

Now, they can’t make Congress do their job, but they can let the problems pile up and hope that the incentives change such that voters wake up and say, I actually want people in Congress who are going to deal with this, because clapping at a Rose Garden ceremony and then actually having the Court strike it down—or having the next president repeal it the second he gets into office—ain’t working for immigration or climate change or name any other problem you want.

So I think that the Court has been quite consistent in that. I think that’s what Trump is upset about. The idea that they’re in Trump’s pocket—it’s literally the opposite, in my opinion, in looking at all of these cases.

Yes, what—he won 17 interim docket decisions in a row? Those were basically repeat cases. They were all about the same two issues. And in the meantime, every substantive policy thing by this administration: Alien Enemies, federalizing the National Guard, tariffs, and (I mean, does anyone think birthright citizenship is turning out any other way?) birthright citizenship—the four biggest policy things of his presidency to date—the Court is going to say or has said no. And it said it with his own nominees in the majority.

OK, so I basically agree with that. But just let me add a few things and clarify it. When you said that they are weakening the presidency, you’re talking about—you’re not talking about vertical control. You’re not talking about the unitary executive, the president’s control over especially senior executive branch officials in so-called independent agencies.

But it might go further than that—we don’t know. And the Court has, in a lot of interim orders, decisions—and probably going to happen in Slaughter—is siding with the Trump administration on that. And the Trump administration has run with that.

Now, I would say about that that the Court was already moving in that direction, and it was going to go in that direction no matter who was president. In fact, Collins was the last unitary executive decision before Trump 2.0—Collins during the Biden administration. The Biden administration was already engaging in fairly aggressive, at least from a historical baseline, removal strategies. So the Court was going to go in that direction in any event in terms of—because they believe it. They’ve been believing it since Free Enterprise Fund, if not earlier. They were going to empower the president to assert vertical control.

Justice Kavanaugh was writing about this in 2008, in 2013—long before he was a twinkle in any justice’s eye.

Yeah. So I’ll say two things about that, and I think that’s exactly right. One, Trump took that possibility—he ran with it. He’s been pushing it much, much more aggressively than any president, certainly more aggressively than a Democratic president.

And the only thing I find thus far—and we don’t know how Slaughter, the FTC case, Humphrey’s Executor case is going to turn out—the only thing I find surprising thus far is that the Court seems to be really sticking to its guns, even though we’ve got this hyper-aggressive, maybe even abusively aggressive unitary executive. Is that fair?

Yes. I think my hypothesis is that they try very hard not to think about any one person as president, and so they don’t want to be reflexively making rules for a president who is abusing the power. They want to come up with the sort of platonic rules for a presidency.

So they want a more politically accountable president in a law review article sense, but they want a weaker presidency. I think this, by the way, is potentially a flaw of the current model of the Supreme Court, where we have such narrow lanes for how to become a Supreme Court justice and the resume items you have to pick up along the way of your journey.

And it is professionalized—overly professionalized—the view of the justices of their jobs. And when they take cert, they answer questions, not cases. And you look at what this Court looks like compared to the Brown v. Board of Education Court, and you’re like, this must be a totally different group that we’re trying to build here.

You know, Justice Jackson hadn’t gone to law school. Eight of the nine had served in the military. Five or so had held elected office. You could come from anywhere and be a Supreme Court justice, especially if you were a political rival of the president.

And yet here we have this almost law professor who’s worked in the executive branch, who’s been a circuit judge for a long time. You’ve ideally clerked for the justice you’re replacing on the Court—the chief, Kavanaugh, and Jackson all replaced their justices. I mean, that’s getting a little bit much.

But I think that explains why they’re not too fixated on what Donald Trump is doing with these rules. They’re writing law review articles in the form of concurrences.

But—I agree. The reason it’s surprising—and it could be admirable that they are sticking to their guns in the face of this—is they really mean it. This shows that they really mean it, come whatever the consequences.

And we’re going to see how they write the Slaughter case and whether it’s a very, very broad conception of the unitary executive or whether it’s a narrow ruling. What do you think? I predict a narrowish ruling.

I think this could be quite broad. Oh my God, we disagree about something.

But I think they could—first of all, they could decide the case without even overruling Humphrey’s Executor.

That’s true, yes.

I think they’ll overrule Humphrey’s. But the oral argument was so concerned with all of the collateral implications of doing that—for non-Article III courts, for military courts, for administrative agencies, basically for all of the intersection between admin law and federal courts—that I think they’re going to try to write in a way that doesn’t deal with those implications. Do you think that’s right?

I think it’s certainly what the chief wants. I just don’t know if he gets to five. And, of course, it’s very different being the fifth vote versus the sixth vote, as I think we can sort of intuitively figure out.

I think the impeachment example is a fun one—like, let’s go to Congress and just how humans work. If you know that President Trump is not going to be removed from office and you’re a senator, you know the votes aren’t there—why would you vote to convict? You don’t come at the king and miss.

Well, same thing. If you’re the fifth vote, you decide which way the case comes out. That’s very different than if there already are five votes and you’re like, let’s jump in, see if I can narrow this—that’s the Kagan role, if you will. That’s why she is the seventh vote so often.

So in a case like Slaughter, we know there are five votes to overturn Humphrey’s. The question in this case is, interestingly, who’s the sixth and seventh potential vote, and how much does the chief want to have a larger majority instead of 6–3, and what’s he willing to trade for it?

And by the way, just to clarify for those—I don’t mean trade in some other case. They don’t trade across cases. I mean trade within the Slaughter case to narrow the ruling.

Right. And the tariff case was also consistent. So they’ll go along with a powerful presidency in terms of vertical power, but in terms of weakening the presidency, you mean in terms of the president being able to read vague or old statutes or statutes that arguably don’t give him the power. This is the major questions doctrine. This is the tariff ruling.

That case was hugely important because when the president started yelling and screaming and going crazy against the Court, my thought was, exactly what they wanted, because it shows that they’re willing to do this major questions thing—whether it’s coherent or not—even in a Republican presidency, when it matters the most to Donald Trump. Do you agree with that?

Hundred percent agree. Donald Trump’s attacks on the courts have been a gift to the chief justice at a time where the Court’s legitimacy and credibility has been a central focus, for all those reasons we talk about with the other two branches failing and the Court being the last word.

It is great for them to be attacked by a president reeling on losses and really bring those to the forefront of the media narrative. It’s arguably what’s built the court—Jefferson, Jackson, Lincoln, FDR—withstanding that type of political pressure and lasting through it. Because again, these chief justices, they last through three, four, five presidents. That’s what’s made this institution.

Right. Not everybody appreciates that point. I think it’s exactly right.

So let’s talk about Trump and the Court. Until the tariff case, the strategy of the administration seems to have been just be as difficult and abusive as possible towards lower courts. If you’re not defying the decisions, you’re coming very close to doing it—having competent lawyers, in many cases, don’t always follow judicial orders. Just totally abusive towards the lower courts, seemingly as part of the strategy.

At the same time, fawning over the Supreme Court. The SG is fawning over the Supreme Court, saying that you listen to their decisions, even if it’s not binding on him. The president also fawning towards the Supreme Court until the tariff case.

The president has not stopped being critical of the Court since then. The question is—there was a lot of worry last year. It seems to have dissipated, but now, with the birthright citizenship case probably going against the president, do you think there’s a danger of President Trump defying the Court?

No, because of the number of people it would take in that case. There’s a difference between implementing something in defiance of the Court and ignoring the Court to stop doing something. The other branches ignored Brown v. Board of Education for 10 years and then, of course, continued to quasi-ignore it even after that. That was fine in terms of the Court’s legitimacy and credibility.

You mean ignoring in terms of not enforcing it vigorously?

They didn’t enforce it at all. Schools didn’t desegregate.

That’s the states and the local governments. But the executive—

I just mean the political actors involved. At that time, it was far more about vertical federalism than now. Not doing something is much easier to do if you’re president. That’s the Jackson ignoring-the-Court example. FDR had prepared a speech, if the Court had come out the other way on the Gold Clause cases, saying, sorry, this is too important to the economy—we’re just going to ignore them.

Trump wouldn’t really be ignoring the Supreme Court in the birthright citizenship case. He would have to implement something and have a whole lot of other executive branch officials affirmatively do things in the face of all sorts of injunctions and mandamuses.

And even if he promised them pardons—no, it just gets too messy and too difficult. And look what he did with tariffs. He didn’t ignore it. He pulled down those tariffs immediately while railing against them.

But do you think if you lose the birthright citizenship case and he tells—and he orders the government to still not process citizenship in accordance with the statute or the Court’s ruling—do you think executive branch lawyers carry out those directives to defy the Court? Is there a point at which—do you have a sense—people in the Justice Department will draw the line at overt defiance of the Supreme Court? Would that be a problem for the president, or do you think he could pull it off?

There are a lot of DOJ lawyers who had prestigious jobs in the law before this. They will need jobs in the law after this. I don’t see it happening—not with those. The ones that make the headlines are sort of in the clown car. But I don’t know anyone in the SG’s office who isn’t a serious person, who would work at a major firm afterwards and isn’t going to be willing to blow that up. And clients aren’t going to hire you if that’s what you’re known for.

Interesting. Okay, two more questions. First, are we going to see retirements this summer? Do you have a prediction about that?

Boy, that Alito thing can go either way. CT—no, not happening. That I feel more confident about. Justice Alito is like a teetering for me. I think the fact that he’s writing a book means he thinks he’s nearing the end of his tenure, but I don’t find that the book coming out the first week of the term, when they’d be hearing oral arguments, to be some definitive statement either way.

His book sells much better if he’s still on the Court. On the other hand, it’s hard to do a book tour when you’re sitting in arguments. He’s got twin grandbabies he’d probably like to spend time with. His wife has a whole lot of flags that still need to be flown at various places. He loves solitude and animals.

But it’s also where he gets his socialization from—his clerks, whom he adores. His clerks—I talk about this in the book—describe him as maybe not talkative, but as the warmest boss and someone who cares deeply. I have this example where he argued all sorts of cases before the Supreme Court, and you get your white goose quill each time you argue. Oral advocates really prize those goose quills.

Some of them frame each one with the briefs. You can find out a lot about a person from where they keep their quills. Justice Alito’s are in one of those free vases that you might get with a flower arrangement on a shelf in his chambers. But you’ve got to dig through some other stuff, including a pink plushy axolotl that one clerk class gave him as an inside joke.

And it’s like this metaphor—his clerks literally come before his own memories, ego, career. He’s invested in them now, and I think he really enjoys that part of the job.

Okay, last question. You talk about Supreme Court reform in the book. I think that if the Democrats win the presidency in the next term, that they’re going to be under enormous pressure—at least, yeah, I would say enormous pressure—to reciprocate some of the Trump weaponization. Or they might not call it that—they might call it recalibrating, they might call it accountability.

But I think one thing that they’re sure to think about doing—and we already hear about this—is to try to do something aggressive with the Court. I predict that the easiest thing will be packing the Court, adding seats to the Court. You already hear talk about this among serious Democratic politicians.

And you already hear them saying that the justification is going to be that Trump v. United States is what led to Trump acting the way he’s acting, which I don’t think is right. So my question is: Do you think it’s likely, if there’s a Democratic president and they control the Congress, that we’ll see a serious Court-packing plan? And how do you think that will work out, if so?

This is sort of like that being-the-fifth-vote thing. I think there’s a lot of senators who are happy to talk a lot of game on the campaign trail about packing the Court, but it’s very different when you’re the one.

I mean, these aren’t stupid people. They know they’ll be destroying the Court and having an independent third branch of government—which sometimes is a real pain. You know, when they’re striking down Biden’s student loan forgiveness plan, when they’re striking down vaccine mandates, you shake your fist and scream at the clouds.

But it’s also what struck down Trump’s tariffs, birthright citizenship. You need a third branch that is counter-majoritarian, that has buy-in from both sides when they’re out of power, to stop the other party from doing things that are unconstitutional.

So regardless of what these guys say, in the end—maybe this is wishful thinking, but I don’t think so—I think these guys love their country more than they love the partisan politics or hate the other side.

Okay, I think we’ll be testing that proposition if the Democrats win in 2028. Sarah, thanks very much. Congratulations on your book, and thanks for chatting with me.

Thanks for having me.

Mentioned:

• Curtis A. Bradley, Historical Gloss and Foreign Affairs: Constitutional Authority in Practice (2024)

Thumbnail: President Trump participates in a press conference during the 2025 NATO Summit at the World Forum in The Hague, Netherlands. (White House Photo.)

This is an edited transcript of an episode of “Executive Functions Chat.” You can listen to the full conversation by following or subscribing to the show on Substack, Apple, Spotify, or wherever you get your podcasts.

Jack Goldsmith: Today I’m chatting with my friend and co-author Curt Bradley at the University of Chicago Law School. Curt’s most recent book, Historical Gloss and Foreign Affairs, is relevant to the topic we’re discussing today, which is whether President Trump has the authority to withdraw the United States from the North Atlantic Treaty of 1949, which established the NATO alliance. Everybody knows about Article 5 of the NATO treaty’s collective self-defense provision.

Less well known is that Article 13 of the treaty provides that any party may cease to be a party one year after its notice of denunciation. So the Senate in 1949 gave its consent, two-thirds consent, to that treaty. The president ratified the treaty.

It came into force, and it’s been in force ever since. Curt, in a second, we’re going to talk about the significance of a statute that proposed to restrict the president’s power to withdraw. But before we get to that, let’s just talk about the kind of naked constitutional issue.

Can the president, if there were no statute, could he withdraw the United States from this treaty?

Curt Bradley: Yeah, and thanks for talking with me about this, Jack. So the text of the Constitution does not give us much help on your question. All it tells us is how the U.S. government makes treaties, which it says requires two-thirds of the Senate to agree with the president to make treaties. And as you mentioned, that’s how the North Atlantic Treaty was concluded in the late 1940s. There’s nothing in the text of the Constitution that specifically talks about how the U.S. government can get out of treaties. Presumably it can get out of treaties like other nations—the other side’s breaching the treaty.

Or in the case of modern treaties like the NATO treaty, the treaties themselves say that the nations, usually with notice, can get out. But our Constitution doesn’t give us a lot of guidance about how that’s supposed to happen. In at least modern terms—the modern era—presidents have, for the most part, acted unilaterally in pulling the United States out of treaties when they think that’s warranted.

And that’s been true for probably a hundred years or so. There are a few counterexamples since the early 20th century, but for the most part, when the U.S. is pulled out of treaties, it has been through presidential action. And most of those withdrawals have not been controversial.

A lot of them are kind of low-level treaties, noncontroversial treaties where Congress doesn’t seem to have a strong view. A few times it’s been controversial, but the fact is the practice has been for presidents to act in pulling the U.S. out of treaties and not seek Congress’s permission to do it, particularly when there’s a clause like in the NATO treaty where it specifically says each party can withdraw. Presidents have used those sorts of clauses many times.

And so even though the text doesn’t give us much guidance, if you look to historical practice, it now is pretty heavily on the side of some presidential power here.

Okay. I have a couple of questions about that. First of all, you wrote a whole book about this, but why should we care about historical practice?

I mean, you basically said for a hundred years presidents have been doing this. Why should that matter? Why shouldn’t we be looking at the text of the Constitution and what the framers thought, et cetera, et cetera?

Yeah. I mean, I think if the text is clear, or maybe what the framers thought is clear, you might just use that. The problem for many foreign affairs issues, as you know, is the text is often not clear, or it doesn’t exist in this case, or the founding and current understanding is also often unclear or nonexistent.

Students have asked me, why isn’t there more in the text or the founding about pulling out of treaties? We don’t know for sure, but one possibility is they just didn’t think about it much. And the fact is we only had seven treaties at the founding.

We weren’t planning on pulling out of any of them at that particular time, and we now have many thousands of treaties. I don’t think they envisioned the numbers that would later come up in the 20th century. So it’s just, as you know, when they drafted the Constitution, they had a lot of priorities to deal with relating to experiences they’d had in the years before the Constitution.

Terminating treaties was just not high on the list of things to deal with. So that’s one theory, that they just literally didn’t think about it. So we may not know the answer to what the understanding is. And if you don’t, if you’re not sure about that, but we know the government has operated a certain way with some amount of acquiescence between the legislative and executive branches for a long time, there’s an argument for deferring to that in the absence of some better rule of decision.

And I will say courts often do that. They often, in foreign affairs, give a lot of weight to those kinds of practices. And not surprisingly, the executive lawyers who have to reason through these things on behalf of the president give a lot of weight, of course, when they advise presidents about what they can do. And I think, understandably, they look to what’s been done frequently in the past.

And so I don’t think that’s an unfair thing for executive lawyers to look to. And even Congress, surprisingly, maybe, Congress, when they have hearings on some of these issues, they actually sometimes just recite long bouts of practice as if that were important as well. So if all three branches are looking to it, then it does seem like it is a relevant part of our interpretive landscape.

So I just want to underscore something you said. Courts rely on this form of reasoning. It’s not just courts; it’s the Supreme Court.

The Supreme Court has, yeah.

Especially in separation-of-powers cases, especially in foreign affairs cases. Is that fair?

Absolutely fair. Just to give one example in foreign affairs, the case that many people know about—the Zivotofsky case about the president’s powers relating to recognition of foreign governments—a large part of that opinion is that the president has this power not because it’s specifically in the text, but because it has been the practice of government for a very long period of time.

And then there are other separation-of-powers cases from the Supreme Court that have the same style of reasoning. And again, if the texts were absolutely clear, there’d probably be less reason to do that. But many separation-of-powers issues don’t have a clear text for a variety of reasons.

Last constitutional question. Some people would say the text is clear and that the Vesting Clause speaks to this. And briefly—and you can fill in the arguments—the Vesting Clause is the idea that the executive power is, independently of other conferrals of power, given to the president, and that therefore all executive power, not otherwise allocated in Article II or Article I or somewhere else, resides in that Vesting Clause, and that the power to terminate treaties would flow from that. What about that argument?

Yeah, there’s some smart academic proponents of that idea, although I don’t think the majority of people writing about constitutional law agree with it. What we do know is the Supreme Court, at least so far, has not agreed with it. The only advocate on the Supreme Court of that theory really has been Justice Thomas. And even Scalia, when he was still on the Court, dismissed it out of hand as not an appropriate source of authority.

The reason to be skeptical of it—two reasons. One, it would smuggle into that first clause of Article II a large and unspecified batch of powers into the executive, and one might be wary of just allowing that much argumentation for the executive branch. And interestingly, I’ve looked—I’ve looked at all of the founding debates at one point many years ago when I was writing about this—you can’t find any discussion of this theory, the Vesting Clause theory, in the thousands of pages of state ratifying debate.

They were debating everything else. They were all worried about replicating King George and giving too much executive power. Nobody seemed to think—advocates or opponents of the Constitution—that that clause was giving some package of powers. And I don’t always want to put weight on what’s not in the record, but it’s surprising there’s no record of anybody thinking that out loud.

And then maybe the really telling point for treaty termination: no president thought the Vesting Clause gave them the treaty termination power throughout the first hundred years of American history. In fact, the practice was they weren’t sure, but they thought probably they had to go to Congress. And nobody argued they had this implied power under the Vesting Clause.

It’d be kind of weird for the executive not to know they have this big font of authority if it really were an understanding of that period. So, you know, if you believe in that, that’s obviously an easy solution. You could just say, it’s not in the text—it all goes to the president. But the Supreme Court has not gone that direction, and I don’t think there’s enough evidence in the founding period to embrace that idea.

Okay. So we’ve been talking about—we basically just discussed how the president could probably terminate the NATO treaty. Just one last question on this.

In his power to terminate the NATO treaty, is it your view that he has to comply with the one-year notification requirement? In other words, the practice is that the president can withdraw, but is the practice that the president has to withdraw within the terms of the treaty—i.e., I guess exercising a Take Care power or something like that? Or does he just have the power to blow it off altogether?

Yeah, I think it’s almost a logical point. And even aggressive presidents on termination have generally accepted that if they use one of these withdrawal clauses, they have to follow the notice period.

There’s a logical reason, which is they could say they’re out in one minute, but the rest of the world’s going to assume you’re in until the period has elapsed. So you can say a lot of things, but technically you’d have to wait until it’s understood that you’re out.

And interestingly, even this president and other presidents who have exercised the authority of withdrawal have almost always done it understanding they have to wait for the time period to elapse. And when President Trump—I’ll just give an example—his first term initiated withdrawal from the Paris Climate Change Treaty, that had a long period, the way it was structured, several years. He waited until it was over.

And he’s been aggressive about a lot of things, but seemed to accept that at least that’s one limit. It’s not obviously a major limit, but for NATO it would mean we’d have to wait a year before a withdrawal was effective.

Yeah. I’ll just note, before moving on to the statutory issue, that one oddity of the 1949 treaty, as you know, is that there’s a duty that the notice of the termination be given to the United States—not the United Nations—but to notify the United States. So in this case, the United States will have to notify itself of President Trump.

It just means nobody thought we would ever withdraw from this cornerstone of Western defense. And indeed, I would say until now, one of the points of very bipartisan agreement in the country was in the value of NATO. So this is a change in understandings in the country and also in the world.

That relates to one other thing, though, which is one reason why presidents assert that they can do these withdrawals is under international law, notices received by a head of state are just presumptively deemed to be valid ones—ones that trigger these periods of notice.

And if there were ever any doubt, obviously Trump is receiving his own notice, weirdly, if he initiated withdrawal here. But in any event, even another depository would deem the president’s notice to be a valid one just presumptively, because heads of state are just thought under international law to have this sort of authority.

Okay. Now we’re going to move on to the harder question, which is Senator Rubio, when he was a senator—now Secretary of State Rubio—joined with Senator Kaine in sponsoring a bill because he was very worried about President Trump’s first-term threats to withdraw from NATO.

And they sponsored a bill that ended up being a law. I think it was enacted in late 2023. I’m just going to read you the main provision of it: “The president shall not suspend, terminate, denounce, or withdraw the United States from the North Atlantic Treaty”—that’s a quote—and then I’m paraphrasing—except with two-thirds Senate consent or an act of Congress.

So basically, Congress is saying you can’t do this thing that you’ve been doing for a hundred years without getting our buy-in—either congressional buy-in or two-thirds Senate buy-in. How does that change the legality of the issue?

Yeah. And my own view is that really does have a significant effect on the legality. I looked at the voting margins, by the way, on that, and it passed, as I would have expected, with bipartisan majorities.

The Senate vote on that was 87 to 13, by the way, with Rubio and others joining it. And in Trump’s first term, when he started grousing about NATO, there was a lot of pushback by Republicans as well as Democrats. I don’t know whether we’d see as much this time.

But what we know is that we have a law on the books—a federal statute that was passed by two houses of Congress, signed into law by the past president. And the strong presumption in American law is that presidents have to follow the law, and duly enacted statutes are binding on the government.

So that puts Trump in a very different posture, I think, than just trying to pull out of a treaty where Congress is silent—where it hasn’t tried to legislate to the contrary. And it really puts at least the strong burden of proof on the president to explain why they don’t have to follow a duly enacted statute.

And I think that should become difficult for the president. I’m not saying the president could never disregard a statute—we know there are a few instances where Congress might overreach, and a president might have that ability—but it should be a pretty narrow set of instances where we would accept that a president can disregard a duly enacted law.

And I certainly think the presumption should be against the president when we get in that posture.

Okay. But before I ask you why—and I’ve got some thoughts coming too—I just want to point out, I think you agree with this: the Office of Legal Counsel in the Justice Department has issued opinions that take an extremely robust view prior to this event, going back to the first term.

And indeed, in a series of legal opinions going back decades that build on some related powers about the president’s power over diplomacy, the president’s power to negotiate, and the like—in an opinion, I think in 2020, about the Open Skies Treaty, which involved, I think, congressional notice before they pulled out—OLC said that that congressional statute is not binding on us.

And I just want to read one sentence from that. This is the key sentence that makes me think that this OLC certainly is going to say, “No problem, Mr. President.” The sentence from the—and I’m not going to go through the analysis from OLC—is: “Congress cannot regulate the president’s decision to exercise a right of the United States to withdraw from a treaty.”

And they talked about the president’s exclusive authority to execute treaties and to conduct diplomacy. They drew on the Vesting Clause and the like. So we don’t have to—you don’t have to answer that directly—but I’m just wondering: I have no doubt that Trump will get permission from the Justice Department, especially since they have to do whatever he says and his interpretation is binding on it.

But how would the court think about it differently? I mean, is it because there was this practice of concurrent authority going back to the beginning? Is it because the more recent hundred-year practice is just about presidential power to terminate in the absence of congressional restriction? How would you—or how would the court—do you think about it?

Yeah. So I think everybody can accept there probably are some examples where the president has some exclusive authority. That would be authority that Congress just can’t restrict.

And the one major foreign affairs case we have is that Zivotofsky case from 2015, where the Supreme Court did hold that Congress can’t restrict the president’s power to decide which governments to recognize and what territories they have. But one of the things the Court emphasized there a lot was the longstanding history not only of presidents doing recognition, but insisting that it was exclusive and pushing Congress back whenever Congress—even in the 19th century—started to try to intrude on that domain.

So we had a much longer history of certainly the executive maintaining that it was exclusive and really Congress, in a number of ways, backing off from efforts to regulate the executive.

The picture does look different to me for treaty termination for two reasons. One, so for recognition, nobody ever thinks the president has to start by going to Congress to decide on recognition. We’re already starting with unilateral presidential action and then asking another question about whether Congress could come in.

For treaties, we know that Congress has to come in at the beginning. So we’re not starting out in a kind of unilateral presidential space in the way we would say for recognition. And the history, to my mind, looks different.

The first hundred years of American practice, presidents worked with Congress to terminate treaties and appeared to believe they needed to do so and did not claim, unlike for recognition, that they had some exclusive domain here. And then you’ve adverted to the other part of the answer, which is even in the 20th century, when they started being more unilateral, they, for the most part, did not insist they could do it in the face of statutory commands.

To the contrary, we have seen that assertion pretty recently—you mentioned the Open Skies opinion from 2020—but interestingly, I looked at an opinion OLC gave two years before that. Trump was thinking about pulling out of the NAFTA trade agreement. At that point he didn’t, but he was thinking about it.

OLC decided he had the authority, but a large part of their opinion is he has that authority because Congress has not tried to restrict it. I looked back this morning—so a big part of their reasoning was he has this authority because Congress has not limited it. That would all be irrelevant if it’s true that Congress can’t regulate at all.

And within two years now, they’re saying in the Open Skies treaty that that’s all irrelevant. But that is not fully consistent with the way they approached it even two years earlier. So we don’t have a long tradition of some acceptance by the government institutions that Congress is just disabled from this space, particularly given that we know they have to be involved in the space to create these obligations like the NATO agreement.

So I just don’t think there’s enough history there to argue that somehow Congress is foreclosed from regulating at all. And since the presumption, in my mind—and this is like Justice Jackson’s views in the Youngstown case—the presumption should be very much against finding one of these exclusive rights of authority, because once you say something’s exclusive, there are no checks and balances on the topic anymore. The president has no checks at all legally.

So the burden ought to be very heavy against the president on this. I don’t see him carrying it for this issue, but you’re absolutely right about what his lawyers will likely argue. They’ll argue that because it concerns foreign affairs and diplomacy, Congress has no ability to limit his decision-making.

But I don’t think that’s true. And if you think about that, what can the United States do in foreign affairs? Oh—impose tariffs. Well, it turns out Congress can limit that, and the Supreme Court just upheld their limits on the tariff laws. Why? In part because they’re in statutes that Congress has passed, and we assume the president has to stick with those statutes.

So I think that may be a precedent that would hurt his chances as well.

Okay. Several follow-ups. First, I agree with everything you just said, but some qualifications. Jackson definitely, in his Category 3 in Youngstown, suggested there’s a heavy burden—it was very hard for the president to prevail in Category 3.

I’ll just point out that in both Zivotofsky and in Trump v. United States, where the president prevailed in that Category 3 arena, the Court didn’t pause over any idea that there was some huge burden for the president to overcome. They just didn’t seem to think that at all.

They just looked at whether there was an exclusive power or not. So I agree that’s a good description of Jackson, but it’s not even the way the Court proceeded in Zivotofsky. I don’t think you disagree with that, is it?

No, I don’t disagree with that. I view that as more specific to the kind of issue in Zivotofsky. But if I were in the executive branch, of course, I would argue probably more broadly than what I’ve just said. You could read the case more broadly if you wanted to, for sure.

Does your argument—in talking about the president—the fact that in the Zivotofsky case, where the president recognizes he didn’t even need to go to Congress, that’s just a unilateral presidential power that he can effectuate—does that mean you would think about this congressional restriction differently if this were a pure executive agreement that the president made under his own Article II power, which it wasn’t?

Yeah, certainly some of those. So if we think there might be some sole executive agreements—those are the kind the president makes on his or her own authority—there are probably some kinds of those that might actually fall into exclusive presidential power.

So one of the earliest sole executive agreements Madison made—for exchange of prisoners of war and humane treatment of the prisoners in the War of 1812—it may be that’s like in a domain where the president not only can make an agreement but really is the exclusive actor. And there might be some limits on what Congress could do in that commander-in-chief space.

But I wouldn’t define it as to whether the president happened to make it on his or her own, because they do that aggressively as well. I would want to know the subject matter. And since we know that—I don’t think anybody argues, as far as I know—that the president could have made the NATO agreement without the legislative branch. At least that’s my view anyway. And they certainly didn’t. At least there we know we’re in a space where the president thought they had to get the Senate involved.

And I would also add, I would say it was always pretty settled in American history that the president could act for recognition issues—they always had. It’s still debated whether the president can act on their own ever for pulling out of treaties. I think the better answer is they now can. But given that we’re debating that, it’s really pushing it to say not only can they do this contested thing—pull us out of a treaty on their own—but they can somehow disable the legislative branch across the board on that same topic.

I just think that’s a very aggressive kind of argumentation.

I can imagine someone arguing something like this: that, look, there’s a treaty—the Senate, they follow the treaty process—the treaty that they agreed to has a termination provision, and the president has a concurrent authority to exercise that. Maybe the Senate or the Congress could have withdrawn if they wanted to exercise that power.

But there haven’t been, I don’t think, a whole lot of statutes like this statute that purports to restrict the president from withdrawing. I agree. And you might think that the president—where does Congress get the power to kind of interfere with a treaty?

Which makes me think about the last-in-time rule, the rule that the last in time between a statute and a treaty prevails. Is that the reason why Congress can come in here and change the termination rule? Is it something like whatever its Necessary and Proper powers are, plus the last-in-time rule? Is that what’s going on here?

Well, I guess I’ll stop there.

I mean, I don’t think it’s exactly right, but I agree with you. We all agree—and it’s always been settled—that a majority of Congress can just nullify the domestic effects of a treaty the day after the president ratifies it, if they can get the statute through. Now, that’s not quite the same.

But this is the international effects.

Exactly. So it’s not quite the same because it wouldn’t change the international effects, although as a practical matter we might then absolutely need to withdraw if we’re in kind of permanent breach domestically. And in fact, some of the early presidential withdrawals in the early 20th century were because Congress had started to override the trade treaties, and the presidents felt like they had to get out at that point because why be in breach?

I don’t think it’s quite the same as that, although that’s probably adjacent to this. I think the Necessary and Proper Clause certainly gives Congress lots of powers to implement treaties—that’s always been settled—to provide standards for what we do under the treaties.

Imagine if, in the NATO treaty, the Senate had said, “We agree to this treaty, but you can’t use Article 13 to pull out unless you come back to us.” Well, that would have been maybe part of their advice and consent to the treaty. But if they can do that—and I think they probably could have done that—it’s not clear why Congress has less power to do that.

You’d have to have an explanation why the Senate could have done it, but somehow Congress cannot do it. I’m not sure what that explanation would be.

Okay. So it seems like there’d be strong arguments—in any event, very strong arguments—against the president’s power to terminate in light of the statute. But then the question comes: can anybody sue?

Because if no one can sue to vindicate Congress’s exercise of authority here, the president can, quote unquote, ‘get away with it.’ So what do you think about the plausibility of someone either having standing to sue or whether this will be viewed by courts as a political question?

Yeah. What we do know from the case law is that it’s unlikely the courts would allow Congress to sue. And you know some of these cases as well as I do. The Court has just really been strict in recent years about letting Congress sue the executive over things like this. So I’m thinking that’s not likely.

There’d be some senators or members of the House who would like to sue, but I doubt they’ll be allowed to do it. So the question is whether there are private parties who have enough of an injury, if we pull out of NATO, to justify what we would call standing.

There might be, and they would have to want to sue. But I was thinking to myself there may be contractors to NATO—U.S. contractors who would lose money if we pull out—long-term contracts—or even U.S. individuals who are employed to NATO. They’re not U.S. government employees, but they work for NATO. They would lose their jobs if the U.S. pulled out.

Those are plenty enough harm to generate what we call standing in terms of the economic harm. So if there are people like that or companies like that, then they might have standing.

And I actually think standing is the most difficult issue to get this to court. If you could find some parties—probably private parties—with an economic harm from the U.S., because remember their allegation would be: We will lose money because Trump is violating the law. That’s usually enough to get you into court, and a lot of cases against Trump are like that.

So if you can find somebody with an injury like that who has standing, then actually I think there’s a reasonable chance of getting to the merits.

Now, the big doctrine you mentioned that you might otherwise think about is the political question doctrine. And in the lower courts—I’ve looked at this—they do apply that doctrine with some regularity to dismiss difficult foreign affairs cases.

But one signal I think the Supreme Court has given is that doctrine does not have much life when there’s a direct conflict between a statute and presidential action. And in the first part of the Zivotofsky case—the initial one—that seemed to be the gist. Most of the justices, except for Breyer, just dismissed the political question doctrine. They said it is for the courts to decide whether the president has to follow a statute. That was how they defined the issue in Zivotofsky.

So even though it had lots of political repercussions—all about the Middle East and everything else—the Court thought when you have a direct conflict between a statute and presidential action, the only arbitral body available at that point is the court.

Because Congress has done what it can—it legislated in 2023 here. The president is saying no. And it looks like they might say that’s not a political question; they can decide the constitutionality of that statute.

So I actually think my inclination a few months ago—and I was thinking about this—would be really hard to get this to court. But I think if some private parties emerge who would lose money, we might actually see some real litigation over it.

And remember, we have a year. If Trump actually starts the notice process, that’s enough time for litigation to unfold and see what the courts are willing to do about it. So I think there’s a real chance.

Okay, last question. I want you to talk about Goldwater v. Carter. And this was about treaty withdrawal and the Supreme Court, in a complex decision, decided not to adjudicate.

And I’m wondering if you just explain briefly what happened in Goldwater, but also, do you think Zivotofsky changed Goldwater, or do you think Goldwater is distinguishable, or both?

Yeah, I’ll start with that. I think I’ll explain what Goldwater is, which is President Carter, when he was recognizing mainland China in the 1970s, decided as part of that he needed to pull the U.S. out of the mutual defense treaty we had had with Taiwan since the 1950s. And he did it unilaterally.

It probably would have been controversial in Congress because his China policy was controversial. That got litigated all the way up to the Supreme Court. And four of the justices said it was a political question—deciding whether he could terminate this treaty or not.

There was a fifth vote to dismiss the case by Justice Powell, who said it wasn’t really ripe from his perspective, because Congress really hadn’t pushed back. There was no resolution even in the Senate to oppose it. And Powell said he didn’t think it was a political question, but he thought the courts in these cases ought to wait until there’s really kind of an impasse between the two branches.

But I actually think both the plurality in Goldwater and, of course, Powell’s views are very differently implicated here. Because I don’t think the Goldwater plurality was talking about a situation when there’s a clear statute on point and the president is saying, “I just don’t have to follow that statute.”

And in any event, Zivotofsky is a supplement, because that’s a holding of the majority of the Court more recently that basically says if the question is whether the president has to follow a statute, even in foreign affairs, it’s not a political question—it’s a question for the courts.

And Goldwater was not faced with that. There was no statute in Goldwater—not even a resolution or anything. And then, of course, Powell’s view is even easier to show it would be met here. He said, “Congress, if you really feel strongly about it, pass a law, and then we’ll decide.” Well, that’s what Congress has done.

So I actually think Goldwater is pretty easy to distinguish, and it’s not a majority holding in any event. And I think the Court today, as far as I can read them, has a kind of narrow view of the political question doctrine, at least when Congress has actually taken some action.

All right. Great analysis, Kurt. Thanks very much.

I appreciate it. It’s been fun. Thank you.

Thumbnail: President Donald Trump attends the Dignified Transfer of remains of six U.S. soldiers killed in an Iranian drone strike in Kuwait. (Official White House by Daniel Torok.)

This is an edited transcript of an episode of “Executive Functions Chat.” You can listen to the full conversation by following or subscribing to the show on Substack, Apple, Spotify, or wherever you get your podcasts.

Jack Goldsmith: Good afternoon, Bob.

Bob Bauer: Good afternoon.

Today we’re going to discuss issues on the horizon in the war in Iran with respect to Congress, and I’m going to start off with some background, and then we’re going to go over three or four kind of fundamental issues on the horizon.

The operation began without congressional authorization on February 28th, about four weeks ago. The United States, apparently led by Vice President Vance, is reportedly in a negotiation with Iran through Pakistan to reach a ceasefire or a peace deal. In recent days, the Pentagon has ordered thousands of Marines and several more battleships to the Middle East. Both parties in Congress have complained that the administration has kept them in the dark in terms of objectives, costs, and timelines for combat operations in Iran. And a Pew Research poll from two days ago showed that 61 percent of Americans disapprove of President Trump’s handling of the conflict, while 37 approve.

So that’s the background. The first issue we should discuss are the efforts in Congress this month to stop the war. There have been three votes earlier this month, all failing by 47 to 53, and all pursuant to fast-track procedures to stop the conflict. And all three of them had language as follows: Congress hereby directs the president to remove the United States Armed Forces from hostilities within or against Iran. And as I said, the votes on that proposition failed.

Yesterday, The New York Times reported that Senator Murkowski of Alaska, Republican Senator Murkowski, is working with a group of senators for a formal authorization of force, and the Times story suggested that she’s focusing on an authorization to try to put some constraints around it, and I believe to try to get the president to engage Congress more, give Congress more information about what’s going on. So that’s what’s going on. What are your thoughts, first of all, Bob, about these votes in Congress and the politics slash law of this?

It’s always very difficult at the outset—and now we’re several weeks into it—but at the outset of a conflict like this, with the deployment of U.S. force in fulfillment, however described, however variously described, national security objectives, for Congress to vote to terminate the hostility. So three times Congress has voted on it; three times Congress has declined to cast that vote.

What will be interesting in the Murkowski case is how she’s going to craft a resolution that hits, if you will, the political sweet spot for a number of members. On the one hand, she’s asking them to take an unpopular step, which is to authorize a very unpopular war, and on the other hand, in some way setting parameters. She wants to give them the opportunity to say that they’ve expressed reservations and that there’s some limits or restrictions that they want to hold the administration to. And how she does that will be extremely interesting, because the core vote to authorize is a very tough vote at this point with a 37 approval rating for this war. And we don’t know yet because we haven’t seen the text.

And we can’t assume that the 53 members of the Senate who voted against the proposition that the war should stop would all be available to actually go on the record authorizing the war, right?

Correct.

And why is that? Just flesh that out. Why is that vote different? Why is it that we can’t expect that the 53 senators who voted to block the resolution to stop the war—it’s one thing to block a resolution to stop war; it’s another thing to affirmatively vote to authorize the war—and I’m just wondering what the political difference is?

The political—I think where she’s headed, and again this is based on very limited news reports—but I think what she wants to do is give them an opportunity to vote an authorization, but with exceptions, giving them an opportunity to say, well, I voted to authorize, I originally blocked an irresponsible Democratic resolution to stop the war, now I’m voting an authorization, but I have real concerns about the way the administration is proceeding here. So this authorization is in some way restrictive or conditional, and the question is, how does she establish that kind of cover?

Right. I mean, how unusual is it that in a war of this consequence, the president has basically stiff-armed Congress completely? I mean, I think there’s been a little bit of explanation about what’s going on in closed sessions, but even Republicans are now complaining that they don’t know what the objectives are. They’re worried about the cost. They’re worried about the timelines. They’re worried about the buildup.

How unusual is it for a president to engage in a conflict like this without making a greater effort to get Congress at least on board in terms of telling them what’s going on and what the expectation should be?

I think it’s unusual. I don’t know that I can call to mind immediately any cases like it. The risk that an administration runs proceeding the way this administration is going is that, as the public—and the polls certainly reflect this—begins to become increasingly restive about this war and oppose it, it cannot look even to its own natural allies for support, because those allies are hearing from voters, they’re seeing the public opinion polling, and yet there’s no background of building up trust, of presenting rationales, and appearing to take congressional questions seriously and answering them.

So they haven’t laid the foundation for the tough times. And one of the most significant steps that an administration can take to prepare itself for a deployment of force like this is to lay a foundation so that, as unpredictable as things can be, as unpredictable as these deployments can unfold, it has some foundation that it can fall back on. And this administration has none. It has just simply stiff-armed both parties. And that can come back to haunt them.

I think it very much can come back to haunt them. It means basically that Trump entirely owns it, no matter what happens. So if this somehow ended up in a spectacular victory over Iran, in which a new dawn rose over the Middle East and there was peace and prosperity, Trump would completely get credit for that. He owns that completely. But he also is going to own all of the downsides, including the military downsides, the security downsides, and the economic downsides—completely owns it.

No question about it. And things are not going to be tidied up that way. Whatever victory he ultimately tries to declare, however this all concludes, if conclusion is the right word, there will not be in the immediate prospect a resolution that is as tidy and satisfactory as the one that you hypothesize—he might be looking for a new dawn for the Middle East, a complete victory for the United States, a restoration of the economy. He said the other day he was confident that this war would end soon and immediately the economy would be restored, or at least the rate of inflation would drop rapidly. All of that is not going to happen. So he owns all of that. He owns all of the downside, with very limited opportunity to claim a major upside.

Wait is that a prediction that there can’t be a major upside here, or that if there is a major upside, he won’t get credit for it?

I think that there will be a very disputed upside, if there’s an upside. And I was just responding to your comment—well, if it all concludes and it appears that a new dawn is rising in the Middle East, I don’t think anything like that is going to be clear.

Already, the United States, having called for regime change at the very beginning, having said they were giving the Iranian people a chance to rebel and take their government back—already now we have the vice president of the United States commissioned to negotiate with that regime. I mean, the other day—I just have to mention this—the other day the president was asked about the Strait of Hormuz, and he said, well, you know, eventually it’ll be administered jointly. And the question was—I think the follow-up question was—well, how would it be administered jointly? Me and an ayatollah? That's a far, far cry from saying we're going to topple this government and the people will have a new day.

Okay, let’s talk about funding. The administration’s been hinting in the last week that it’s going to seek supplemental appropriations, perhaps as high as 200 billion dollars. And this has already been an extremely expensive war. I think they’ve put in requests for shifting around funding authorities. There’s talk about, I believe, about taking assets deployed in Ukraine and bringing them to the United States, either monetary or military assets.

How do you view—I mean, I have a legal point to make about this—but how do you view this aspect of things? Do you think he can get a $200 billion appropriation? Is that something likely?

The little I’ve seen in the news suggests that he may be able to do so, but everyone in Congress is aware, and I think the evidence is pretty strong, that this is extremely unpopular. It puts a dollar, if you will, exclamation mark on public disaffection with this war. The public was displeased enough, or restless enough, about the financial commitments made to Ukraine. Here’s 200 billion dollars, and Secretary of Defense Hegseth has said that’s not a fixed number, so he may need more. And I just think it adds to his political woes to a very significant degree.

Is that because that 200 billion dollars is not being directed to domestic matters, and it’s being directed, as you say, on an unpopular conflict. Is that the reason?

Yes. It puts a number on what people don’t care for in the first place, which is a war that they don’t approve of, with major consequences for the U.S. economy, which are being experienced when they thought he campaigned on the exact opposite set of objectives.

So the only other issue I want to make on this is that Congress—someone might write about this next week—Congress needs to be careful how it crafts any appropriation that it gives for this conflict, because an appropriation for a war can be deemed to be an authorization for the war, even though the War Powers Resolution says that appropriations should not be deemed as authorizations.

There’s an OLC opinion that says that, despite what the War Powers Resolution says, a subsequent appropriation, if it clearly authorizes the war, can do so. So there’s an issue about whether, in giving money for the war, Congress will be authorizing it, and it needs to be very careful how it words that appropriation.

Okay, on the War Powers Resolution, there’s a 60-day clock—the War Powers Resolution 60 to 90 days, or there’s a 30-day extension the president can give. The War Powers Resolution, after that clock runs, mandates withdrawal of troops that have been introduced into hostilities. We’re still probably two months away from that clock.

Would a normal administration be thinking about the War Powers clock this early?

Well, a normal administration would certainly do so in the same way that I mentioned earlier. A normal administration would be trying to lay a political foundation, establish more political support for this deployment of force, for this war. But this administration isn’t doing it, and I don’t think this administration is worrying about the War Powers Resolution at all.

It’s taken the explicit position that the statute is unconstitutional. And just the other day—I mean, in the last 24 hours—the president was essentially sneering at the whole question of whether it was a war or something else. He said, well, I won’t say war. He said something to the effect of, I won’t say war because I’m told I shouldn’t say war, so I’ll just call it a military operation.

He was taunting anybody who thought that the distinction was meaningful. And, by the way, I just want to take that really quickly back to the point you made about the care Congress ought to be taking in crafting support by appropriations. Here again, there’s maybe something at stake for Congress on that, but there’s nothing at stake for the administration. It just wants the money, and I don’t think it matters to them one way or the other whether, from the congressional point of view, it strengthens or weakens the case under the War Powers Resolution, because it doesn’t accept the authority of the War Powers Resolution anyway.

Yeah, so on the point that I think President Trump, Vice President Vance, and Secretary of State Rubio have all said in the last year that the War Powers Resolution is unconstitutional—and every administration has held—so that is a fraught and ambiguous statement.

There are many parts of the War Powers Resolution, and it’s not clear which part they say is unconstitutional. The legislative veto part of the original War Powers Resolution is not operational because of Chadha. They probably weren’t thinking about that. There’s discussion in the War Powers Resolution about what the president’s Article II powers are; presidents have disagreed with that articulation.

They might have been talking about the 60-day clock and saying that the War Powers Resolution was unconstitutional. There’s still an OLC opinion on the books, I believe going back to 1980, that says to the contrary—it is constitutional. I wouldn’t be surprised if they said the 60-day clock, or the 60/90-day clock, is unconstitutional and that Congress cannot, through that automatic procedure, terminate the president’s deployment.

It’s not clear, but they also have so many other options under the War Powers Resolution. They could conceivably say, depending on the state of the war—although it’s not moving in this direction now—that there’s not hostilities. There’s an old interpretive move going back to the Reagan administration that the Obama administration did for a while with the Islamic State: Every wave of a new use of force is a new introduction into hostilities and therefore restarts the clock. That seems preposterous, but it’s been done before.

I guess I don’t think, whether it’s through one of those interpretive moves or whether it’s through unconstitutionality, I don’t think that they’re going to be bothered legally by the clock. But politically, it’s a salient point to focus congressional deliberation. Do you think that’s right?

In what sense do you mean?

I mean the fact that—you know—you were there when there was a 60/90-day clock coming up on Libya, and I’m wondering whether that clock, whether it’s legally enforceable or not, is that kind of forcing mechanism to focus on the legality and politics of what’s going on?

Yes. The reason I asked a question about politically—you know, to what extent is the public focused on the legality question—and I suspect very little. They just simply don’t like the war for all sorts of reasons.

But take the Libya example. Yeah, the administration was very focused on that. It took a controversial legal position to avoid the problem with the 60-day clock, and then it sent the then-State Department Legal Adviser, Harold Koh, to Capitol Hill in June of 2011 to testify at length in defense of the legal position that they took. So they cared about it, and they wanted to take the position that their position was a lawful one and consistent with OLC precedent and also congressional intent behind the War Powers Resolution.

I don’t think we’re going to see this administration breaking a sweat on that topic. And I want to just add one other point: The president—you know, we get the presidents we have, with all their character and personality—this president is inclined to put on massive shows of machismo or muscularity. Think about the “bomb, bomb, bomb Iran” or the Truth Social videos that he ran after the bombings in Iran in June of last year, the video game montage that was recently put on, drawing on various movies and scenes of conflict and heroism, Hegseth’s martial rhetoric. . .

All of that just cuts completely against any concern whatsoever about constraint. It all depends on him and what he thinks he should be able to do, which is why you get that kind of language from him. And I think that plays into his general contempt for the constitutional issue.

Yeah. Okay, speaking of the constitutional issue—and so whatever the constitutional rules are about the use of force are contested—but there are a set of very permissive OLC opinions that we’ve discussed before, and that probably most of our listeners know about.

Those opinions, I say, are very permissive, but they do sometimes say that there is, or hint that there might be, one limit at least, and that is a prolonged and substantial military engagement involving exposure of U.S. military personnel to significant risk over a substantial period. If the conflict arises to that, and OLC says that ground troops is a good marker of that, then OLC has said that the president must, or might have to, go to Congress to get an authorization.

So there’s a question whether even out of the OLC opinions, if this conflict continues to ramp up—especially if the president introduces so-called boots on the ground—that the OLC opinions that say the president has to go to Congress might come into play.

My own view is that these opinions are very wishy-washy. There are a whole bunch of other opinions that OLC can and probably would rely on, including some that you and I, eight years ago, wrote to the Justice Department asking them to withdraw—opinions related to Iraq 20-some-odd years ago—and also opinions about self-defense.

So I don’t expect OLC to tell the president that he can’t continue with the war, but there is this issue about the language in the opinions being implicated.

Yes, and you know, you and I talked offline about one thing that I think this conflict really brings to the surface. This nature, scope, and duration test has been debated in recent years primarily against the background of whether the United States would have boots on the ground, and there was a significant risk of American casualties, even if the United States itself was using force and training fire on the enemy—whether it was encountering any high likelihood of return fire that could kill American servicemen and women. That’s the way this has been debated; that’s the context in which it has been debated.

But what we see currently, which I think requires really rethinking this old model, is a war that has caused the largest oil supply disruption in history—literally in history—in which the United States has also lost billions of dollars, according to Wall Street Journal reporting, and sensitive equipment, including drones and radar, and has had a convulsive effect on international relations.

So it’s a deployment of force, the significance of which—which in turn, of course, involves what we think of as Congress’s role—the significance of which cannot be measured only in terms of the anticipated loss of American servicemen or women or casualties.

So it’s a good point. I’ll point out that even on the kind of risk to troops, there have been 13 casualties thus far and many hundreds, I think, of injuries to U.S. servicemen and women growing out of the war—that’s the first point.

Your point is that this has been a massively consequential and costly war beyond loss of troops—troop death, and troop injury—and that the test should take that into account.

I mean, certainly, if we go back to the very beginning, what were the framers trying to do? It’s not clear how they did it—whether they did it with the Declare War Clause or with control over the standing army—that remains contested. It’s not clear exactly what the Declare War Clause meant, but it’s pretty clear that the framers wanted to ensure that the president did not use U.S. troops abroad in a consequential way that was going to be consequential for the country without getting the people, through Congress, on board.

And you’re pointing out that this war, even if there were no U.S. casualties, has been massively consequential financially and in terms of the economy, and probably will be for a very long time, and that a president should be required by law—and if not by law, by political prudence—to go to Congress and get Congress on board.

I’ll note that both President Bushes, in both Iraq wars—the first and second—President Bush in the first and second Iraq wars, they both claimed the authority to send the troops to Iraq and to use force there. They claimed the Article II authority without the need to get Congress on board, yet in both instances they both went to Congress in a tough situation, when it wasn’t clear in both cases that they were going to get the authorization, and they got Congress on board.

Whether that was a constitutional or a political imperative, they both sensed—they didn’t think it was a constitutional imperative, at least that’s what they said—but they both sensed that the stakes were so big that they needed to get Congress on board. And that thus far has not been, obviously, the impulse of the Trump administration.

That’s right, and I think the 37 percent approval rating reflects in part the absence of any effort to really build public support for this effort, or, as we said earlier, to line up political allies that would help build and reinforce that public support. And the price that a government pays for that is extremely high.

They’re making the point—this probably goes to their rejection of the constitutional issues—they’re making the point repeatedly, Trump has done so, Hegseth has done so, they’re not putting any limits on this war. Sometimes Trump says that it’s a matter of weeks—he may say that periodically because he wants to calm the markets—but both of them have made it clear that it ends when they say it will end.

Right, and this gets to the point you were making about President Trump saying, “I’m not supposed to say it’s a war.” I mean, I’m surprised he even knows that, because that communication came from his lawyers—but the idea in the OLC opinions is that if something is a war in the constitutional sense, Congress has to approve it, and that the conflicts that OLC has approved in recent decades have not been wars in the constitutional sense that require congressional authorization.

So that’s why he made that distinction, kind of dismissing the whole thing. But they’ve also, as you say, not taken anything off the table. They have not tried to say this is going to be limited along any dimension, and that is in serious tension with the OLC opinions.

Yes. You know, you can think about all sorts of things that could happen, right?

You could think about a cyberattack on the United States that the Iranians engineer that wipes out energy infrastructure in the United States, water desalination. . . You could think of a single drone attack in the United States, two or three terror attacks in the United States, bombs that are set off that completely tank the stock market. . . all of which is a result of this “military operation.”

And it’s just not part of the debate about when the president owes Congress and the public an explanation and some kind of legislative buy-in. It’s just not part of that debate.

I agree. First of all, he’s going to own it completely.

Yes, he is.

Whatever happens negatively, they’ll try to say it would have happened anyway—this is what we went to war to prevent, they were going to do this—but that’s going to be a tough argument to make.

But we’ve just really become deadened, on your second point, to the politics here. The idea that this is really Congress’s responsibility—and it really is Congress’s responsibility—the whole frame of understanding war powers has shifted to the one-man-decides model, and then Congress bickers from the sidelines. And, you know, if things get really bad, they may do something. That whole model is just out of culture. I mean, all of these things should be being debated; they should be being debated front and center, and they’re not.

Right, and Congress has not, and there are ways to do it, certainly beyond the one suggestion that I made in my posting, Congress isn’t even putting itself on the record on the legal issues. It’s just voting up or down to terminate the hostilities.

The findings that vary among the three, from Kaine to Murphy—the findings vary because there’s a little bit more detail built into the finding section about Trump having referred to war, some of the rhetoric that is contrary to any suggestion there’s anything other than a war—but Congress is not—there’s no—how would I call it—line in the sand that Congress is drawing legally.

It doesn’t have a mechanism for really doing so. It hasn’t attempted to do so. So when all of this is over with, we still won’t know a lot more, except that public pressure could cause this war to end earlier than it otherwise might, about when Congress believes the president owes that responsibility.

This is just a massive structural difference between Congress and the president.

Yes.

Presidents always have an incentive to justify their Article II power, and they can ensure that it can happen by the unitary nature of the executive, and basically the interpretive power flows from the president through the Justice Department down to OLC.

I mean, the reason that there aren’t legal findings in any of these resolutions is it would make it even harder to get the votes, because specifying in a list of congressional findings the nature of the president’s powers—it was hard enough to do, and they failed, in the War Powers Resolution. Doing that on the fly in findings in a law like this is—I agree with you—it would be great if Congress could do it, but they just can’t. They’re not structurally able to form that consensus.

Even the War Powers Resolution, when they were at the maximum consensus about the president’s powers and Congress’s powers—even there, there was vagueness in how they described the president’s Article II powers. It’s just a very hard thing for a decentralized body, divided by parties, to reach consensus on.

Yes.

All right. Thanks very much.

Thank you.

This is an edited transcript of an episode of “Executive Functions Chat.” You can listen to the full conversation by following or subscribing to the show on Substack, Apple, Spotify, or wherever you get your podcasts.

Jack Goldsmith: Today, I’m going to talk to my colleague and friend, Cass Sunstein, about his new book, Separation of Powers: How to Preserve Liberty in Troubled Times. Cass, thanks for chatting.

Cass Sunstein: Thanks so much for having me.

So, your book is about all aspects of separation of powers, and you focus on the judicial branch and the legislative branch, as well as the executive branch. But since this is for Executive Functions, we’re going to talk about executive power mostly.

The subtitle of your book, which is about separation of powers, is How to Preserve Liberty in Troubled Times. And Chapter Six is entitled, “The Executive is the Most Dangerous Branch.” So how do these things relate together? And why is the executive branch the most dangerous branch? And what does that have to do with liberty?

What I saw, particularly in my White House job, was the massive knowledge advantage the executive branch has, and the massive advantage it has by virtue of the power of initiative. So courts, of course, are the glamorous entities for law students, often, but they’re after the fact. Sometimes they’re really after the fact. Sometimes they don’t get involved at all. And sometimes they don’t know what they’re doing.

And when they don’t know what they’re doing, part of the time they know they don’t know what they’re doing. And that creates a kind of humility and modesty, which is both salutary and sometimes dangerous. So the courts are in a weaker position.

Congress—I was flabbergasted to see in my second stint, that is, in the Office of Information and Regulatory Affairs—is often flailing. And it’s not really their fault. They have to get re-elected, the members. Their staffs aren’t nearly as numerous or as expert as the executive branch’s staffs.

So Congress is frequently thinking, you know, this is wrong, what you did. And Congress might be right on that, but really doesn’t have anything like the understanding that the executive branch does. And it is able to hold a hearing and maybe annoy. And sometimes it’s able to write a letter, which can cause a stream of events to follow, including potentially a subpoena.

But the executive branch really is front and center of the apparatus of governance. And when I was there, of course, in the Obama administration, I didn’t particularly lament that. But I did notice it and thought that the most beautiful and wonderful things that happen in our country are frequently a product of executive branch creativity. And some of the least beautiful things are either potentially, or in fact, a product of—let’s call it—executive branch creativity.

But most people, when they think about the dangers of the executive branch, don’t think about its information advantages. They think about the fact that it exercises the sword of the community, that it has coercive power, and that it can take the initiative to exercise those powers. So how does that factor in, and how does that relate to this knowledge or information point?

Okay, they’re closely related. So suppose the executive branch wants to issue a bunch of regulations that are, let’s say, costly or harmful, or motivated by politics rather than by really figuring out what’s best. It can do that. And then the very fact that it’s issuing regulations often causes change on the ground.

And the people who try to stop it from happening might make arguments to the public or to courts that are in the direction of, let’s just say, inadequately informed. And the executive branch will have a machine of response. And that can, on the regulatory side—which is a keen interest of mine in two of my three stints—on the regulatory side, just the knowledge and the authority to initiate can make all the difference.

So you might even say you’re going to issue a regulation. It might involve the environment, it might involve immigration, it might involve safety in the air. And then the private sector is just going to do stuff, knowing that the executive branch might act, and thinking it doesn’t want to be behind the game as the game starts.

And because the executive branch knows a ton, the private sector knows that it’s going to have an uphill battle in winning in court, even though it might win. And even if it wins in court, it might win late. And that might mean there’s all sorts of economic consequences.

Of course, when people think of the prosecutorial authority of the executive branch, which I think is what you have in mind, I saw that in the Department of Justice—not that close, but pretty close, because it was the same building. And I saw it at the Department of Homeland Security; ICE—I work closely with ICE—the executive branch can do all sorts of things to bring enforcement power to bear.

And it knows things. So in a contest with people outside the executive branch, those people might lose, not because they’re wrong, but just because they don’t know enough. Now, it is often the case—and we’ve seen this under both the Biden administration and Trump administration—that the executive branch is going to lose in court a lot. But there’s a lot that doesn’t go to court at all. And if it doesn’t go to court, then the executive branch typically wins.

Yeah. So those are the two points I was going to make. The second one first—so much, I mean, you say in the book that the executive may—I’m sorry—you talk about how courts can constrain the executive branch. And I’m going to talk about an important example that you mentioned in a second.

But so many of the things that the executive branch does, and it can take the initiative on, whether it has superior information or not, it can do—and there’s no judicial review. So many things, when I was at OLC, very few things that I worked on in OLC ever appeared in court, either because it was secret, or because of just justiciability, standing, and related things, or because no one had a cause of action.

So it seems that the executive branch has a massive advantage, not just because of information and because of its coercive power, but also because there’s, in many circumstances, no checking branch, or potentially Congress.

That’s true. Let me tell you two stories, shall I?

In one discussion we had in the executive branch, the Department of Homeland Security was in a contest with another department, which I won’t name for reasons that will become clear. The other department wanted to do something that was thought by it to be in the public interest, very much so, and left of center in a way that that department really liked.

The Department of Homeland Security thought that that was—it was unlawful—that the preferred option by the other department violated governing statutes. And we said that, and the other department said, look, no one left standing to sue us. So we’ll completely succeed; no one is standing to sue.

And the Department of Homeland Security said, you know, we took an oath to follow the law, the Constitution, the Take Care Clause, took an oath—we have to follow the law. And the other department said—I remember this as if it was yesterday—said no court has ever struck down an act of the executive branch solely on the ground that it violated the oath, which suggested they missed the point, that even if courts wouldn’t be involved, it was still inconsistent with the law.

And in that particular discussion, the Department of Homeland Security prevailed, because people agreed you shouldn’t do something that violates the law, even if there’s no court involvement. So that happens. I’m sure you know that. It happens a lot.

Does it happen all the time? Not. It doesn’t happen all the time.

So that’s one story of the role of legal constraints when things are working well, with no judicial involvement. I think it was just today announced that the Department of Homeland Security, my old department, is going to return to the BRIC program, which is basically a resilience program in the face of weather-related harm, like flooding or wildfire.

And the Trump administration had abandoned this program, the BRIC program, something I had some work to do on when I was there. And the court said, you can’t just abandon the program. A court did get involved in a suit brought by 22 states.

And the Trump administration just said, okay, we’ll go forward with it, which suggests, though you’re right, of course, there’s a serious category of cases in which there’s no court involvement. Still, there are a lot of cases in which there is. And that’s a beautiful thing from the standpoint of the separation of powers.

I just want to underscore something you said, which is, though in those cases where there’s no judicial review, it really depends on the integrity of the lawyers inside the executive branch. And, you know, executive branch lawyers have their own interpretive principles, their own institutional constraints, their own attitudes toward law.

But there’s such a large percentage of cases where the executive acts without effective judicial review that that form of internal review is just vitally important. And that’s not really separation of powers, formally.

You’re right. It has an integration of powers, where it’s the executive authority and interpretive authority. And if the executive has a strong policy preference—meaning the political leadership—maybe the lawyers will recede, or that the lawyers will try to figure out a path, even if it’s really not a legitimate path.

One thing that I observed, and that we’ve been circling around, I think does have a separation of powers component, which is the morality of legality. And let’s just name it, where an executive branch lawyer thinks that there’s a moral obligation to follow the law, which is independent of anything else and a trump card.

And the reason that it has a separation of powers component is that typically what legality amounts to is fidelity to congressional instructions.

Right.

For the Constitution, which is not really a separation of powers issue.

But so you describe—you say this is one of the most important—I’m going to move on to an area where the executive is effectively checked, and it’s vitally important. You say—I’m going to read from the book—that this is one of the most important paragraphs in the book, you think, and I’m quoting from it now:

It is also true and fundamental that the executive power and the judicial power are separated in the sense that the executive initiates cases, including prosecutions, and judges decide on both facts and law. If the Department of Justice thinks that you have committed a crime, it cannot put you in jail just because it wants to do that. It has to prove its case before an independent tribunal, and often with a jury.

And you might have added, with a grand jury intervening. That point has been vitally important in the last year and a half, and has, I think, is a more important separation of powers point, as a fundamental constraint protection of liberty, than people appreciated before recently.

I agree completely. So the words that you kindly read were written kind of low temperature, in the background, where it was, you know, a recognition. But if you write them now, there’s a lot more heat in the sentences.

And it is the case that, you know, if you love the current president or not so much, the idea that any president can put people in criminal jeopardy without the buffer, the independence of court—that’s Kafka. It’s kind of literally Kafka. And it’s a nightmare from the standpoint of liberty.

Yeah. And the framers—I mean, as you say, you made this point outside of—you wrote this point outside of the current heat—but it’s kind of a foundational point. It goes back to the framers. They designed this system with this protection of liberty in mind.

Completely. So Hamilton, in an obscure passage in the Federalist, celebrated constitutional review, but said it’s not with a view to infractions of the Constitution alone that the independent judiciary is important. And he talked about a civilizing effect.

And, you know, we could write a book about that. But part of the civilizing effect is making sure that the facts of the law are supportive of the charge, and that, you know, for a large number of people, some of whom were in the newspapers, most of whom are not, that is a lifeline.

Okay. I want to move on to what you call in the book the “grand narrative.” And I think it’s fair to say this is a grand narrative about administrative law, but it might be a grand narrative about executive power. What is the grand narrative, and why is it important?

When I was in law school the day before yesterday, the majority of us thought Article One gives legislative power to Congress. That’s true, which means that Congress can’t give open-ended authority to the executive branch by telling the executive branch to do what’s reasonable or feasible—that that is a violation of Article One, Section One.

Article Two gives executive power to the president, which means there can’t be independent agencies like the Federal Communications Commission and the Federal Trade Commission that excise executive power from the president and give it to other people. That’s very, very bad—or so we thought.

And Article Three gives judicial power to judges protected by independence, life tenure, salary guarantees, and that means that adjudication can’t be exercised outside of those independent people. And yet the Social Security Administration and many other executive officials are engaged in adjudication.

So what we thought back in the day was that there’s sequential violations of Article One, Article Two, and Article Three. And that’s the grand narrative—that the Constitution created these entities, and then the New Deal or the Great Society, or maybe Woodrow Wilson, did something horrible, which was to take a red pen to Article One, Section One; Article Two, Section One; Article Three.

And that’s terrible. And the grand narrative suggests we have to go back to the original constitutional settlement, which was, you know, very focused on self-government and why our country became independent. That’s not a trivial matter.

And also very focused on liberty. And so the obligation of the legal culture now is to obey the lesson of the grand narrative, which is to restore the beauty and pristineness of what the Constitution plainly says. That’s the grand narrative.

And I’m going to flesh this out, and I want to talk about the Slaughter case, which has raised these issues exquisitely since you wrote the book. But what is your basic take on it? I mean, what is your take on the grand narrative?

It’s a little false. And it’s a little made up. And it’s a little bit simple. And it’s too cool for school. That’s actually the phrase—too cool for school was made for this. It’s very cool, but in school, it’s just not going to work.

And the reason is that, as simple and seductive as it sounds, it has a very tendentious understanding of Article One, Article Two, and Article Three. I say not, you know, baseless or ridiculous understanding of Article One, Article Two, and Article Three, but it has a tendentious understanding.

So we have to investigate each of the three. The idea that Congress isn’t permitted to grant discretionary authority to the executive branch—that’s the first part of the grand narrative, or the one that tends to get people’s juices flowing most.

This is the non-delegation doctrine, is that correct?

Yeah, yeah, yeah. So let’s take one idea, which is Congress can grant as much discretion to the executive as it wants. Then it’s exercising lawmaking, and the executive is implementing pursuant to the admittedly open-ended grant of discretion.

So one view is that’s fine. If Congress says, do what you think best to handle immigration or to handle climate change, no problem—that’s how Congress has made the law. That’s one view, and a lot of people hold that view. It has some historical support.

I think that view isn’t justified by the history. So let’s have a more modest view, which is that broad grants of discretion are fine, and the founding generation understood them to be fine. If it’s completely open-ended, it’s not fine.

So that’s not inconsistent with the founding settlement. But if you say, do it to the extent feasible, or do it to the extent practicable, or do it to the extent reasonable, those things are consistent with what the founding generation itself did, with very little in the way of constitutional objection.

So that part of the grand narrative—let’s just say—is historically reckless.

So you mean false from an originalist perspective? Yeah, completely, yeah. And because, as you’ve written, there may be reasons today, given the change in the nature of government, where we might view these things differently. I agree with you. I think that the originalist debate on that has cut pretty strongly, based on what the early Congress did and how the early courts understood what the early Congress did, to allow pretty broad, broad delegations.

So I concur in that. What’s next?

Okay. So let’s say—and I want to underline your point—that we could think, on originalist grounds, an aggressive restriction on Congress’s power to give discretion to the executive branch is very hard to defend. But we could say that some kind of non-delegation doctrine is essential now, given the nature of executive power. And I think that’s an interesting view, and not at all preposterous.

And I want to come back to that idea later.

Okay. So on the second part of the three-part grand narrative—we actually have three grand narratives—the second part is that Congress can’t authorize officials to exercise executive authority without the president’s removal power untrammeled.

And that would mean that little me, as a 20-something in the Office of Legal Counsel, would be an at-will employee of the president, notwithstanding the Civil Service Act. I wasn’t that. As a staffer, I had for-cause protection.

Under the second part of the grand narrative, that would be constitutionally very suspicious. Worse than that, from the standpoint of the second part of the grand narrative, is the Federal Communications Commission, the Federal Trade Commission, the Nuclear Regulatory Commission, the Federal Reserve Board, et cetera.

— explain why? It’s because of the restrictions on the president’s removal power.

Because the executive power is being taken out of the president’s domain—and what part of Article II, Section 1, do you not understand? Executive power is vested in a President of the United States.

That view—the view that the strongly unitary president is constitutionally mandatory—it hasn’t taken the kind of battering that the non-delegation doctrine has taken, but it’s suffered some very serious blows.

So, if you’re a boxing fan, it’s Muhammad Ali—jab, jab, jab, jab, jab, jab—rather than George Foreman knocking you out. And the jab, jab, jab, jab, jab consists of findings that the early Congress probably thought it had the authority to limit the president’s removal authority over principal officers. It just decided not to exercise that authority.

So there wasn’t a consensus in the founding period that the president had unlimited removal authority over the people who execute the laws, or at least this is a very hard historical question.

But on that one, I mean, you started off by saying you weren’t removable. We’re talking about the grand narrative about the unitary executive here and the president’s removal power. And a couple of points—and I’m wondering if you agree.

One is it’s not at all clear what the pure theory entails, that the president can control every person in the executive branch on every decision, no matter what it looks like. The court has never gone nearly that far in its case law.

Even Myers acknowledged that there could be constraints on the president’s removal power of certain officials, right? And the originalist point doesn’t really get at whether, you know, these multi-member independent agencies with massive combined—whatever kind of powers you want to call rulemaking and adjudication—it doesn’t really speak to that, does it?

Well, I agree. But the people who believe in a strongly unitary president, I think, should agree—and many do—and, as you say, Myers did, that the civil service law is okay, because the lower-level officials are, through the chain of command, subject to policy control by the president.

But if you have a principal officer, like people who are running the Securities and Exchange Commission, the view is—the unitary executive view that I think Chief Justice Roberts holds, and that many law professors hold—the constitutional settlement doesn’t allow that.

So the multi-member commissions, whose heads are for-cause protected against the president—that I thought, when I worked in the Justice Department as a kid lawyer, I thought that was a constitutional atrocity. And I think a lot of people, on originalist grounds, think that.

It’s not clear that, on originalist grounds, they’re wrong. That’s my view. It’s a really hard question. I think it’s 60–40, 40–60.

Yeah. Okay. So a lot of this issue has come up in the slaughter case, which is about the constitutionality of the independent commissioners on the FTC. And everybody thinks, based on the oral argument—and, frankly, based on the last three or four unitary executive cases concerning removal, anyway—that the Court is going to say that the president does have the authority to remove the members of the FTC, and that his removal power extends that far, at least.

I just want to assume that’s true and try to think through the implications. And I think this goes to the heart of your book. And I want to do so by listening to two sets of questions by Justice Kagan at the oral argument in Slaughter, where I think she raised, as I say, issues that go to the heart of your book, and I want to see what you think about them.

So here’s the first clip from Justice Kagan in the Slaughter argument.

The idea is that the president was supposed to do the executing, but he wasn’t supposed to do the legislating, and he wasn’t supposed to do the judging. And here’s my next proposition, which I think you have to agree with, because we just look around the government and it’s obviously true.

Some people think it’s a real distortion from what the founders thought, but what you think of as executive branch agencies, including independent agencies, they do a lot of legislating and they do a lot of judging. And you listed it a bunch of times. You said this is obviously executive power.

Why is it obviously executive power? Because they’re doing a lot of rulemaking, and they’re doing a lot of adjudications leading to enforcement. And those are—although we’ve said that this is executive power in some sense—but they’re legislative functions. That’s what rulemaking is. They’re adjudicative functions.

And isn’t it problematic, given what we know about the founder’s vision, that what this is going to amount to, at the end of the day, is putting not only all executive power in the president, but an incredible amount of legislative slash rulemaking power and judging in the president’s hands?

Okay. So I'm very keen about what you think about that. Is Justice Kagan embracing or rejecting the grand narrative? Is she embracing a different narrative? What do you make of that set of questions?

Okay. I think she is kind of embracing it, thinking that the exercise of rulemaking authority by the executive is a constitutional violation, and the exercise of adjudicative authority by the executive is a constitutional violation. I don’t think it’s right, though.

So rulemaking—suppose there’s a statute that says, on a diagnosis of cancer, the Occupational Safety and Health Administration has to require some reporting. And then the Occupational Health Safety Administration defines what a diagnosis is through rule.

Now, there’s not a lot of room to say what’s a diagnosis. There’s not a lot of discretion there. It’s a pretty mechanical—maybe that’s an overstatement—but it’s not a big exercise in policymaking, open-ended policymaking authority.

So this is a way of saying rulemaking can be just a form of execution unless we insist it isn’t, but it might not involve much in the way of discretion at all. It might be that you decide whether the permissible level of some pollutant is 9.11111 or 9.11112, and you do that through a rule.

That isn’t an exercise of legislative authority unless we insist that any rulemaking is—and why would we do that? Rulemaking has been around a long, long time. The idea that rulemaking, because it’s rulemaking, is legislation—which Justice Kagan, my former dean, is saying—I think, on reflection, she wouldn’t accept her own proposition.

But I think there’s a way of interpreting it, and this is going to bring me to the second clip. I don’t think—I’m fairly confident—Justice Kagan is not saying that rulemaking is an exercise of Article I legislative power, and that adjudication is an exercise of Article III judicial power.

I think what she’s saying is that there’s a massive concentration of various powers that we call executive, but that, as you talk about in your book, have adjudicative elements and have discretionary lawmaking elements, whether it’s the exercise of legislative power or not.

And I think she’s just pointing out the extraordinary concentration of power in the executive that now the court is on the road to saying the president can do with what he wants. I think—so let me run the second clip, which I think enlightens the first one, and then you can tell me what you think.

Here’s been the bargain over the last century, and I think it has been a bargain. Congress has given these agencies a lot—a lot of work to do that is not traditionally executive work—that is more along the lines of: make rules when we issue broad delegations, and do lots of adjudications that set the rules for industries and entire bodies of governance, right?

And they’ve given all of that power to these agencies, largely with it in mind that the agencies are not under the control of a single person, of the president, but that indeed Congress has a great deal of influence over them too.

And if you take away a half of this bargain, you end up with just massive, uncontrolled, unchecked power in the hands of the president. And it’s really hard to affect both sides of this bargain, because it’s already been done.

So the result of what you want is that the president is going to have massive, unchecked, uncontrolled power, not only to do traditional execution, but to make law through legislative and adjudicative frameworks.

So Cass, I want you to comment on that. I just want to point out, because I thought this was remarkable, that in the same oral argument, Justice Gorsuch and Justice Barrett make similar claims about this bargain, and wondering whether the elimination of the removal power leaves the president in a commanding position that maybe Congress might not have wanted.

I think that’s a great comment by Justice Kagan, and it shouldn’t be surprising because she taught administrative law for many years, and she’s an expert—a field expert.

So let me tell you a story that exemplifies what she’s worrying over. When the Consumer Bureau was created, I was in the government and spent a lot of time working on the legislation that led to it. And we discussed a lot whether the Consumer Bureau should be an executive agency or an independent agency.

And whatever the right call as a matter of policy, it was completely clear at the time that to get that legislation through Congress with the Consumer Bureau being an executive agency would have been really difficult and possibly impossible. And pause over that, if you would.

It’s the deal that Justice Kagan is discussing in action, where Congress thought, okay, we’ll create this Consumer Bureau, and it’s going to have a lot of power, but we’re not going to make it subject to the president, because if we did, then the president would have a lot of power.

And you can see this through agency after agency. With respect to the Federal Communications Commission, whatever exactly happened at the time, the idea of Congress creating an FCC that controls the communications sector and that is run by employees of the president—it doesn’t matter what your political party is—that should make people’s blood run cold.

So the deal that she’s describing—that’s real. Now, what to do about that is a very nice question. If you say all of these agencies exercising rulemaking and adjudicatory authority are now under the president’s thumb, that does unsettle the congressional understanding that led to the creation of those agencies.

Does that argue—I don't think you talked about this in your book, and I actually don't know if you've written about this, but you probably have—what severability regime does that argue for? Does that argue for getting rid of the removal restrictions, or does it argue for taking down the whole thing, because Congress would not have given the one without the other?

This is really hard. So if you believe that the strongly unitary vision is constitutionally correct, then you just have to insist on that, and the severability questions become very reasonably answered in the direction of throwing the whole statutes into the garbage. But that itself is a cataclysm.

And that’s something that’s a little like a French Revolution, which a tradition-loving, occasionally Burkean court should be really nervous about. So we throw the SEC statute down, we make the NLRB not exist anymore, because of a contentious but plausible, maybe more likely than not, right—we’d have to investigate that severability argument.

So my own view is that the constitutional attack on the independent agencies should be resisted on the ground that the constitutional question either is an originalist matter or as a matter of what do we do, given where we are, which is what Justice Kagan is asking.

It’s a hard question. And to throw out decades of precedent on which our government’s institutions rely on the basis of a contentious, maybe right theory—that we shouldn’t do.

So I would be in favor of amping up a little bit the president's policy control over the independent agencies through interpretation of the existing statutes, which give him power to fire for inefficiency, neglect of duty, malfeasance in office. I'd ramp those up a little bit, but basically not throw Washington on these issues into a tailspin where it's going like this.

Okay. Here’s my last question. It’s clear that the court’s moving in the direction of unitariness and a broader removal power and more presidential vertical control.

Some people, including some conservatives on the court, think that the complement to that—the important complement to that—is to narrow the president’s ability to rely on vague statements to do novel things. And this comes under the heading of the major questions doctrine. And there are some people that see those things as going hand in hand.

I think Justice Gorsuch is the loudest articulator of this view. So you talked about the major questions doctrine in the book. Just what do you think of the major questions doctrine, especially given the more unitary executive?

And how do you read the Learning Resources tariffs case and the major questions doctrine after that? This case came down after your book.

Yeah. Okay. So let me state two views and then try to figure out which I like better.

So view one is the major questions doctrine is a made up thing by people who don’t like administrative authority. And as Justice Barrett has said repeatedly, to create made up substantive canons isn’t a very judicial thing to do. It sounds more like policy stuff.

And, you know, if the president lacks authority under the statute as rightly interpreted, which I think is correct in the school loans case, then the executive should lose. But to have a doctrine that puts a special thumb on the scales against the executive when it’s doing something big—that’s a little like the old canon that statutes in derogation of the common law shall be narrowly construed, which seems, by current lights, a little too policy-generated, and we shouldn’t do it.

So that, I think, has a lot of force behind it. And it says don’t develop new canons of construction that suit your policy preferences.

The other idea is we should be a bit more upbeat on the major questions doctrine, partly because of the breadth of current executive power, partly because of the misallocation, by constitutional lights, of discretionary action—let’s just call it—from the legislature to the executive.

And on that view, the major questions doctrine isn’t a terrible idea. It says if, you know, the president is going to try to forgive school loans, or the president is going to try to shift from coal to solar and wind, or the president is going to have a tariff regime, Congress must explicitly authorize it.

I’m kind of talking myself into it as a plausible idea that isn’t bizarrely adapted to the situation in which we find ourselves. So the second view, which is a little bullish on the major questions doctrine, I like better than I did seven years ago.

So the second view—you can disagree if you disagree—strikes me as closer to Justice Barrett’s conception of the major questions doctrine, which is a little softer, more contextual, as she puts it. She views it as just an instantiation of textualism, although it’s a very—it seems to me—I don’t want to say squishy, but soft version of textualism.

I mean, is that—is her—and, by the way, she and Justice Gorsuch, who have sharply opposing views, substantive canon versus contextual interpretation of the major questions doctrine, both joined the formulation by the Chief Justice, which was shrewdly ambiguous on this debate.

I mean, is that the more attractive way to see the major questions doctrine?

It is, it’s more humble. So what Barrett’s doing—there’s a passage by Wittgenstein in Philosophical Investigations, where he says, I tell you, teach the children a game. You teach them gambling.

Is that consistent with the instruction—simplifying Wittgenstein—but the basic idea is it isn’t—that if you say teach the children a game, you didn’t mean gambling.

And Barrett says, basically, that if you have a reaction to what the interpretation is, you need Congress squarely to have authorized it. And that’s because the what—“what?”—the skeptical “what” reflects how communication works.

And so if there’s some agency doing something completely new, and it seems, you know, well beyond what anyone had ever thought, Congress must squarely authorize it. That is humbler.

I think in the end, the distance between Gorsuch and Barrett is less substantively large than it seems. There’s clearly a big difference in temperament, where the separation of powers—heavy artillery—Gorsuch loves, and Barrett is very cautious about.

But Barrett needs some claims about Article 1, Section 1, I think, to get her more cautious, maybe softly textualist wannabe argument going. So it’s—I like it better. I even think that the Barrett view isn’t so far from Kagan’s view in the end.

But it’s—it is singing a slow, soft, quiet, short version of Gorsuch’s song.

It also seems more on the road to—or resonant of, or in the vicinity of—some kind of purposivism, because you can only say “what?” if you have a sense of what the statute was for.

Yeah, I think so.

You have a sense that some executive action is just not what they meant. So it seems—it seems like a modified, strongly modified to me version of what I understood textualism to be.

Yeah, I think so. I think, you know, I really admire her effort to be within textualist boundaries and not wanting judges making things up. That’s kind of her thing.

And it does have a disciplining effect. And it is part and parcel of her caution and humility. You’re right that in I teach the children a game, I teach them gambling, you need a sense of the purpose of the instruction to get that gambling answer to be the wrong answer.

Okay, Cass, thanks very much. Terrific book. It’s about much more than executive power, which is what we focused on, and well worth reading.

Thank you.

Thank you so much.

Thumbnail: FBI Joint Terrorism Task Force photo. (U.S. Government Work.)

This is an edited transcript of an episode of “Executive Functions Chat.” You can listen to the full conversation by following or subscribing to the show on Substack, Apple, Spotify, or wherever you get your podcasts.

Jack Goldsmith: In light of the U.S. attacks on Iran over the weekend, the Department of Homeland Security and the FBI announced a heightened alert for domestic attacks, including cyberattacks, and U.S. Northern Command, which covers the United States, basically sent out a signal of heightened security on U.S. military and related bases. I think it’s important to understand how the U.S. government infrastructure would normally operate to deal with such threats and what the main concerns and issues this government faces are. I’m very grateful that Elizabeth Sherwood-Randall has agreed to talk about these issues.

Liz has a long career in U.S. national security and most recently was the White House Homeland Security Advisor from January 2021 through January 2025, a position now held by Stephen Miller. Liz is a senior fellow at the Harvard Kennedy School’s Belfer Center. Liz, thank you for coming on to talk about this today.

Elizabeth Sherwood-Randall: Thanks for inviting me today, Jack.

I would like for you to walk us through how, if you were sitting in your old chair in the White House, you would be thinking about these issues. What would you be doing in the current situation if there were a military conflict abroad that implicated domestic threats? What would you be doing?

So, Jack, the Homeland Security Advisor role at the White House was born right after 9/11. I lived for four years in a basement SCIF — a secure compartmented information facility, a room with no windows in the West Wing — where I worked basically 24/7 to ensure that we kept Americans safe. The job responsibilities include keeping Americans safe at home and also Americans around the world.

So wherever our people are — our diplomats, our military service members, our intelligence community — we watch out for them from that post. If this had been anticipated, which it clearly was for some time internally as the military preparations got underway, I would have been working for weeks to get us ready for the homeland implications of the attack on Iran that began over the weekend.

And we would have worked across the federal enterprise with every relevant department and agency to ensure that we were prepared for the possibility that there could be homeland consequences of these attacks, as well as for the possibility that there could be retaliation against our personnel in the region of the attacks and around the world.

In addition, we would have worked to ensure that we were aligned with our state and local counterparts. For example, the Department of Justice and its Federal Bureau of Investigation would be working anticipatorily through the FBI’s field offices — more than 50 around the country — to ensure that local law enforcement was ready for what might happen and was monitoring carefully for the possibility that there could be an increase in threats to American citizens.

Finally, importantly, we would have been working with allies and partners to prepare them as well. We’ve already seen significant retaliation against our partners in the region. In that case, one of the things we look at is whether we need to prepare to remove American personnel and representatives in the region or elsewhere, and to develop plans for evacuating American citizens — which is a quite complex operation, especially in a situation of live conflict.

So all of that work would have been done in advance to ensure that we were ready, and that in the event of adverse incidents at home or around the world, we could respond quickly to help our citizens.

Okay, I want to unpack that, and I want to focus on the at-home part of it, just so we can — otherwise I think it’ll be too much. You talked about what you called working “across the federal enterprise.” Can you give us the high points of what that looks like? You mentioned FBI and DOJ dealing with state and local officials, and I want to get to state and local in more depth in a second. But what is FBI and DOJ doing in this context? And how did it look from the White House?

So, the federal enterprise is large and has many different capabilities. We certainly would have involved in the work I described the FBI and DOJ, the Department of Homeland Security, the Department of Defense, and the intelligence community. Those are the key departments and agencies you would want to have at the table to begin planning for what you need to do.

We would be meeting in the Situation Room — a classified setting, obviously — where you can have confidential discussions. The objective would be for the leader of each department and agency to come forward with: here are the things I think we need to worry about if this happens, and here’s what I propose to do about it. Then you develop a coordinated plan.

And basically there’s an operational tempo in a situation like this in which you’re probably meeting every day — whether in person or on screen — to ensure that everyone is working together like the many instruments of an orchestra playing well together to prepare for what may be coming our way.

The Department of Homeland Security, for example — what would they be worrying about, in the time in which I served? Certainly they’d be looking at whether there would be cyberattacks. That would be a very important element of this. Right off the bat, we had an entity that has now been basically stripped of much of its capability at DHS called the Cybersecurity and Infrastructure Security Agency.

It would have been the tip of the spear in doing that work, alongside the FBI, which would be doing the investigative work to identify what’s happening on networks and to ensure that every possible capability was activated in support of those on the front lines — our government networks, but also the whole nation’s networks: cyber effects, telecommunications, financial systems, and health systems, as you well know. So that would be an example of the kind of work we’d be doing in advance to make sure we were, as I would describe it, “shields up” — that you get ready in advance to prevent and to be ready to mitigate any consequences of anything that might happen.

Okay, let me follow up on that. Two related questions. One, how would what DHS is doing in a heightened alert situation differ from what it’s normally doing? What would that look like? And DHS isn’t the only entity focused on cybersecurity, I’m sure, I mean, the FBI has a role there. So how does the government get activated when there’s a heightened cybersecurity threat?

In a heightened environment, what you’re doing is you are previewing to your partners, whether they are the state and local partners I mentioned, whether it’s the homeland security representatives in each state, the governor’s teams, the local law enforcement across the nation — you’re working with counterparts across the nation. Oh, and I should have mentioned, I’m sorry, the private sector, importantly. So, for example, on cyber, all of the providers of capabilities in this country, you’re working collaboratively to say, these are the kinds of tactics, techniques, and procedures that we have seen this adversary use in the past. Here is what you should be watching for.

That’s a capability the federal government can provide. And we provide that based on our intelligence capabilities, and we provide it based on our experience in doing this work. And so what you say is: Here’s what we anticipate may happen. And here’s what you need to be doing. And here are the resources we can make available to you if you need help.

Now, it’s possible that that would be very difficult to do if you’re trying to keep operational security around a military attack. The thing is, the president had been previewing that there could well be military action for some time. So that would have given a pretext for doing this because it was already out in the public domain that there was some consideration of undertaking an operation. And you could therefore do this work and say, should anything happen, here are the things we want to be sure we’re ready for.

Okay, what else would the FBI be doing? I mean, I take it that it might — I don’t know — but I take it that there are confidential sources that it would be working in a more aggressive way. I take it that it had, again, I’m speculating, that there are probably a list of suspects or people that it might be worried about that it would give extra attention to. Is that right? And what else would it be doing?

Jack, that is right. And I think what we see in a system in which the government is working effectively to bring to bear all the tools in the toolkit is that we bring together foreign intelligence information that gives us insight into what bad actors may be trying to do. And we merge that with domestic information that is available through the investigative work that is done on an ongoing basis by the FBI, where they are tracking numerous people over time — criminal networks, drug traffickers, potential terrorist actors.

Of course, you have to have an actual legal basis for bringing somebody in if you decide that they are trending toward violence and you need to take action. Much of the time there’s wait and watch, to gather more information, to ascertain whether there are networks of people working together. In the effort to be as effective as possible, not to just blow an investigation early, but rather to gather as much as you can, to do the greatest work that you can to break up anything that may be threatening to Americans.

So we saw this in a variety of contexts in the homeland, and there’s press reporting on some of the work we did in the years in which I was in this Homeland Security Advisor role with some ISIS members who were in the homeland. And that work was painstaking law enforcement work, took a great deal of time and effort. And ultimately, when it became evident that there was a possibility that these individuals would cause harm to Americans, we worked with the Department of Homeland Security to use our immigration authorities to remove them from the country. And that was a very good example, again, of collaboration between two entities that work in the homeland security space.

You mentioned the integration of foreign and domestic intelligence, which, I guess, typically tends to be collected by different entities. Can you talk about the challenge of that, how the coordination is done and where?

Of course, we’re very, very careful about any collection on Americans. We have very strict laws around the civil liberties and privacy of American citizens. So what we’re working to do is gather information about foreign actors and, to the extent necessary, bring it together with information that is gathered by state and local law enforcement and the FBI about developments in the homeland.

It’s all done very rigorously and generally without any involvement of the White House whatsoever, because that work is privileged information and the White House does not get involved in it. But when there is an action that will be taken — for example, there might be a case brought against an individual — then that becomes potentially public knowledge and you may find out about it in the news. It is not necessarily the case that the White House will know about that in advance because of how carefully protected those processes are of investigation and prosecution under traditional norms of conduct of the Department of Justice and the FBI.

We haven’t mentioned, I don’t think, the Office of the Director of National Counterterrorism and the National Counterterrorism Center. Could you speak to the roles that they play?

So after 9/11, the National Counterterrorism Center was created because, as you’ll recall, there was a diagnosis that one of the reasons we experienced what we experienced on 9/11 was that the dots were not being connected across all the different points of information that had been gathered, and where we might have known that this was happening and been able to prevent it had we connected those dots. So NCTC was established as a clearinghouse for all the intelligence information that is related to terrorism.

And experts are based in NCTC, coming from all the different intelligence agencies, in order to do the painstaking analysis that leads to that connection of the dots and which shows that there is a risk here of this individual or this group in this location or these locations for which we need to be prepared. We want to try to disrupt what they’re doing, if at all possible. And so we want to get ahead of the action that they might take.

That function was one that I relied on very, very heavily. And indeed, the head of the National Counterterrorism Center was often in these meetings that I convened regularly to ensure that we were seeing all of the developments in the international sphere and aligning them with what we knew was happening in the homeland. And so we would have in meetings together the head of NCTC, the appropriate representative from the Department of Justice and the FBI, similarly from DHS, similarly from other departments and agencies, so that there was cross-referencing and people would know that this was happening, this may affect your personnel, you may be able to contribute something to doing something about this. Let’s all come together to ensure that we prepare in every way, we prevent wherever possible, and if we can’t prevent, then we’re ready to respond effectively.

Okay. Can you talk about the distinction between a foreign agent in the United States or a suspected foreign agent in the United States versus someone in the United States who’s not an agent of a foreign power but who might be a threat of some sort through inspiration?

Yeah, so when I think about possible terrorism in the homeland, I think about three categories, Jack. One is a terrorist who is directed by one of the terrorist groups overseas to conduct an attack in the homeland, like 9/11, for example, with Al-Qaeda. You could see Lebanese Hezbollah or Hamas. You could see the IRGC today try to conduct an attack in the homeland where it’s funded and organized by a foreign terrorist group or by a state sponsor of terrorism — in that case, Iran for the IRGC or Hamas.

Second is a terrorist who is seeded into the country, who would be potentially in what might be called a sleeper cell. That could be an individual who is more of a mercenary than an actual believer in the ideology of the group that is directing his or her action, and they could be here for a long period of time, and they may or may not ever take action. But that’s also a challenge for our law enforcement agencies to suss out and to track and to discern whether somebody is trending toward action or whether they are here potentially with some motivation that is supported by a foreign entity but not necessarily intending to do something immediately.

And the third, and quite importantly, especially at this moment, is a homegrown terrorist who is inspired by but not directed by one of the foreign terrorist groups. So a good example of that would be the individual who was a former military service member who conducted a terrorist attack on January 1st, 2025, in New Orleans. And in that case, this individual was inspired by ISIS but was not directed by ISIS, to the best of our knowledge. And he committed mass murder on New Year’s Day, of course, in New Orleans.

So those are three different kinds of actors who we have to be monitoring for and trying to disrupt. The lone wolf actors in the homeland are particularly difficult. And it’s possible — we don’t know the full story yet about what happened in Austin — but it is conceivable that that individual falls into that category.

And that’s something that right now law enforcement will be working very, very hard to ascertain: whether there’s any intelligence about that individual; indications of conversations of direction or support to that individual from a foreign actor; what his connections are in the homeland. Was he acting in any group? Are there others who might do similar things with whom he is in contact? And how quickly can that be disrupted if that is the case?

And so that would be work that would be happening in real time, with the FBI as the tip of the spear.

I don’t know if you can speak to this, but do you have a view on either the impact on all of these considerations of the funding lapse in the Department of Homeland Security or the firing of career officials by the Trump administration across the intelligence and law enforcement terrain, including in the FBI, which does both of those things? Do you have a view about how either one of those things might affect protecting the homeland at the moment?

I do. So I have relied very heavily on the deep expert competence of people across the federal enterprise who do very unglamorous work every day to keep Americans safe. And that includes the people in the Department of Justice and the FBI who do this work that we were discussing, the people in the Department of Homeland Security at the Cybersecurity and Infrastructure Security Agency, at the Transportation Safety Administration, at FEMA, at the Secret Service, in the intelligence community — the professionals who are connecting those dots.

The fact that so much of the federal homeland enterprise has been diverted to focusing on immigration enforcement concerns me greatly. And if coupled with the mass departures and firings, in some cases, of very senior experienced leadership in these departments and agencies, we have a real risk of not having the human capacity we need to do the investigative work to prevent and disrupt attacks in the homeland. It’s hard enough when you have everybody in place and working closely together, as I described it.

When you’ve had such dislocation, reorientation, and loss of human capacity, it really does raise the risk that we could suffer experiences that we do not need to suffer in the homeland. I hope very much that that is not the case. And I hope that we’re able to withstand what may be coming our way.

We also need to count on our foreign partners in this regard, Jack. And as I mentioned earlier, that would have been an early element of the equation for me in terms of preparation. I can think of a number of instances in which my foreign counterparts called me to warn me about something and said, this is what we’re seeing in our intelligence. We are worried this is coming your way. And without that tip, we would not have been shields up at the time.

And so the relationships we have, the trusting relationships we have with our foreign counterparts, and between the — again — the departments and agencies that work with their foreign counterparts, FBI counterparts, intelligence community counterparts, all of that really matters right now to ensure that we’re getting the cooperation we need to get early warning and to prevent bad things from happening.

You spoke about the diversion of FBI resources from counterterrorism to immigration-related matters. I believe I read that last year, during last year’s attack on Iran, that those resources came back to the FBI to work on counterterrorism. I haven’t yet read anything on that, but it might well have happened. Do you have any thoughts on the going back and forth, if that’s what’s happening?

It’s hard to know without more detail. I saw the same reporting. I would say that this work is so detailed. It’s so continuous. It’s not something you do episodically and then drop it to follow the leads that were given, to be in communities with your counterparts, staying on top of emerging potential threats, to track people who are potentially trending toward violence. You want to have continuity, not episodic engagement.

And so I would say my strong view would be it would be better to have people in place continuously. That said, if people have been brought back and if there’s capability that’s being put on this challenge, that’s better. And it is also, I’d say, an important moment for the administration to consider a rebalance, because what we know is that, in general, these attacks don’t happen immediately. It may take weeks to months, especially if you’re trying to position somebody to conduct something at scale. And so we have some time to prevent if we get organized to do so.

In addition, some of these things are years in the making, and the radicalization of human beings can be several years in the making, too. And so one of the things we’ve known, tragically, is that the Hamas attack on Israel in October of 2023 and the subsequent consequences of that in terms of the ongoing crisis in Gaza have led to radicalization of individuals who are more likely now, after 18 months or two years, to be trending toward violence. That is the trajectory, in general, we have seen of people who have been radicalized over time.

That does not mean that somebody wouldn’t be motivated today. And an interesting additional and worrisome point is that we’ve seen the confluence of groups come together who previously were divided by an ideological divide in the Islamic world between Sunni and Shia. And so though Hamas is a Sunni group, quite quickly many of the Shia groups became involved in a parallel activity in support of Hamas, whether it’s Lebanese Hezbollah, which just attacked Israel overnight (this morning, Israel time), or the Houthis, or we have seen other groups as well, ISIS and Al Qaeda, in the wake of the crisis in the Middle East, call for solidarity with Hamas against the United States, Israel, and their allies.

So we could see additional actions by groups that are not directly involved in the conflict but want to express solidarity with Iran in this moment. We don’t know that that will happen. We have seen Lebanese Hezbollah, which is an Iranian-supported terrorist group, act. We’ll need to see what others decide to do. But again, we should be shields up for all of this, to make no assumptions.

Can I ask you — I’ve got two more questions. The first one is to ask you to go back. You mentioned at the beginning the importance of state and local authorities. I think people might be surprised to learn how absolutely crucial they are in these types of threats. Could you say more about that and how the federal government intersects?

I can’t say enough about that, actually. This is the way our country is supposed to work. We have the capabilities residing in communities where our law enforcement is supposed to be out in the community, feeling the pulse of the community, understanding what is trending and identifying it for possible action if need be.

Now, that can also be abused in certain circumstances, and we have wrestled with that as well. But in cases related to terrorism, optimally what is happening is that our local law enforcement is working closely with the FBI team that is in the region of that community to receive information and provide information. So there’s a two-way street that is full of activity all the time, in exchanging information about what is happening so that we are not blind, so that we know we’re seeing something happen here, and the locals may ask for help — may ask for help in immediate response should something be emerging, or may ask for help in trying to investigate and prosecute and disrupt something.

And that has to be, as I said, ongoing, and there has to be a trusted relationship. One of the things that’s happened in the last 14 months is that we’ve seen many departures of individuals in the FBI offices around the country who were in leadership roles — the special agents in charge, their deputies, and others — who have left for various reasons in frustration over the rule of law not being implemented. And so that also degrades our capacity to work in communities because we don’t have those trusted relationships that we’re building on.

Last question: do you have a view about — you know, we’ve talked about certain resources going from national security/counterterrorism to maybe national security immigration enforcement. And it seems in the administration’s mind, anyway, that there’s been a trade-off and a perceived need to rebalance and then maybe readjust that balance. Do you have any views about how the heightened threat now and possible threats — and again, this is speculative, from the Iran conflict, which seems to be, seems as we’re speaking on Monday afternoon like it’s going to be going on for at least many weeks — how that is going to affect other items, domestic agenda items of the administration, especially immigration enforcement?

Yeah. I mean, Jack, there’s a scenario that I worry about, which is that if there are terror attacks in the homeland and the military is fully engaged overseas in an escalating conflict, as you’ve just suggested, it could well be — we hope that’s not the case, but we don’t know where this is going. That could lead to a presidential decision to activate more military service members from the National Guard in the homeland to support, in his view, the suppression of violence or of civic unrest, or to manage terrorism if he perceives that there’s a rise in terrorist threats or activities.

That would be a development that will cause us to question whether it is necessary. There seems to have been an interest in doing this, obviously, in a number of communities over the last year. And if there is a heightened terror threat in the homeland, I could foresee that there would be an inclination to use that as a justification for putting more service members — potentially National Guard members — in our communities to provide security when the military is occupied overseas.

So we don’t know whether that will happen, but it’s something to watch because of the trends that we’ve already seen.

Okay, Liz, thank you very much. I appreciate it.

Thank you so much for giving me this opportunity to talk with you.

Thumbnail: President Donald Trump announces the administration’s plans to invoke tariffs under the International Emergency Economic Powers Act on April 2, 2025. (White House Photo.)

This is an edited transcript of an episode of “Executive Functions Chat.” You can listen to the full conversation by following or subscribing to the show on Substack, Apple, Spotify, or wherever you get your podcasts.

Jack Goldsmith: As everybody knows, on Friday the Supreme Court ruled in Learning Resources that the International Emergency Economic Powers Act, or IEEPA, does not authorize the tariffs that President Trump has imposed under that statute. The president has begun to assert substitute tariffs under other statutes, which raise a whole host of legal issues. But today we’re going to discuss a follow-on from last week’s case — an issue that the Court did not address and did not say a word about — namely, what is the remedy, if any, including refunds, for the tariffs that were collected unlawfully under IEEPA?

These tariffs have been estimated to be in the range of $130 to $175 billion, or maybe even more.

To discuss this with us, we’ve got Matthew Seligman, who’s a fellow at the Constitutional Law Center at Stanford. He filed an amicus brief on behalf of former government officials and legal scholars in Learning Resources opposing the government’s tariffs, and he also represents importers who may seek refunds following Learning Resources.

Good afternoon, Matthew.

Matthew Seligman: Thanks for having me, Jack.

So I would like this to be just a primer on how to understand the issue of refunds — or whatever other remedies may be available. I think we should start very basic. Why don’t you just tell us how tariff collection works and how it worked for these IEEPA tariffs?

Sure. So we’ll start with the basic legal and logistical framework for how tariff duties are assessed.

When an importer brings goods into the United States at a port of entry — whether that’s a physical port, an airport, or a land crossing — the Customs and Border Protection agency assesses a duty based on the customs declaration. There will be a provisional amount of money that’s paid.

Then sometime after that — the legal limit is up to a year afterward — CBP will liquidate, or finalize, the customs assessment, the duty assessment. So that’s how the collection process works.

Importantly, since the IEEPA tariffs were imposed in early 2025, leading up to so-called Liberation Day in April 2025, CBP has been collecting these tariff duties all along.

Now, importantly, even though the Court of International Trade — which is the specialized federal court that handles these types of trade disputes — held that the government, the president, didn’t have the authority to impose these tariffs, it stayed its decision. And that decision has been stayed throughout this litigation.

As a result, the government has continued to collect these billions and billions of dollars of tariffs, even though a federal court had already ruled that they were unlawful.

Right. So if I understand you correctly, there’s a provisional assessment and then there’s a kind of final assessment, which is called the liquidation. Is that correct?

Yeah, that’s right.

And do we have a sense — maybe we don’t — what percentage of the tariffs have been liquidated or not, i.e., subject to a final assessment?

The best public data that we have on this comes from a filing that DOJ made in a Court of International Trade case in early December. It attached a declaration from a CBP official.

As of December 10, there had been approximately 34 million entries. Each time, say, a container comes in for an importer, it creates an entry. That’s the legal object for which the duty is assessed.

So there have been 34 million of these entries as of December 10.

Thirty-four million or thirty-four thousand?

Thirty-four million. And those 34 million entries were for over 300,000 separate importers.

As of Dec. 10, DOJ represented that 19.2 million of them had not yet been liquidated, which means that about 50 percent had been. And that was in early December. We’ve had two and a half months since then.

OK, so we’ve got a giant mix of liquidated and unliquidated assessments. Is that right? Is that the right word?

Yeah — entries, because each time a container comes in, there’s an entry.

OK. So the Supreme Court ruled that the government collected these funds unlawfully. I think that’s the right reading of Learning Resources. Do you agree with that?

Absolutely.

OK. So then, in a normal case — and this has happened before — when customs collects duties that turn out to be unlawful, what does the normal refund process look like? Is there a normal refund process?

There is, but it involves an agency process.

Once an entry is liquidated, the importer — the importer of record, who actually paid the duty — has 180 days, and only 180 days, to file a protest within CBP.

Now ordinarily, the types of disputes involved are things like whether the category of goods was correct — maybe the goods were categorized incorrectly and should have come in under a different classification with a different percentage rate. Occasionally there are instances where there’s a question about the legality of the tariff duty at all.

This is the process through which refunds are supposed to be sought.

OK. And I take it that the refund is not going to be automatic. The government is not going to start, after this decision, cutting checks to the importers who paid the IEEPA tariff. So, I take it — is it true that to recover, the importer who paid has to file some type of process?

For most cases, that’s going to be correct.

So I want to be precise here. For the entries that aren’t liquidated yet, the government could just choose to liquidate them at a level lower than the initial assessment. It could say, “OK, the Supreme Court has ruled; we’re going to follow the ruling, and we’re going to liquidate these entries at the lower level, subtracting out the unlawful IEEPA tariff duties.”

It has the power to do that. Whether it will exercise that power is an open question. We can watch what the president and some of his advisers said over the weekend — it doesn’t seem likely to me — but it’s at least possible.

But why would that only apply to unliquidated entries and not liquidated ones?

Because the process always works this way: the entry is initially made and then liquidated. CBP has unilateral and discretionary authority to make adjustments to the amount of the duty that was initially assessed at the time of liquidation.

After liquidation, there’s actually a 90-day period when CBP can voluntarily reliquidate. But after that, there’s no longer any kind of discretionary authority on the part of CBP — or the federal government more generally — to just cut checks.

So someone has to seek to recover the funds.

That’s right. And the big question then is going to be whether they have to file protests or whether they can sue directly in court.

Filing protests before the—

Before CBP. So it’s an interagency process.

So it’s an administrative process before CBP, or going directly to the Court of International Trade.

That’s correct. And there’s been some disagreement among trade lawyers about which of those two approaches is viable.

OK, and I take it this is all that question — that foundational question. Assuming that the government is not going to start cutting checks, which I think we can assume — the president, as you alluded, said that this is going to be in litigation for five years. So I guess that suggests that the president, anyway, thinks that there’s going to be a lot of contestation here.

So that’s a foundational question: whether you go to CBP first in an administrative process versus going straight to the Court of International Trade. And does that turn on the liquidated–unliquidated distinction?

It turns on something else. And so the question about whether you go to court first, or whether you have to protest and then — if the protest is denied — then go to court, has to do with whether the administrative exhaustion requirement — that you have to go through this agency process first and give the agency a chance — is mandatory or not.

And there is some uncertainty in the precedent of the Court of International Trade about whether it’s mandatory. I think that ultimately there’s a significant likelihood that the Federal Circuit — which is the appellate court for the Court of International Trade — or eventually the Supreme Court again, is going to say that the administrative exhaustion requirement — that you have to protest and only then can you go to court if the protest is denied — is mandatory. I think there’s a pretty significant chance that the appellate courts are ultimately going to hold that that’s mandatory.

OK, and I take it just to be clear, because I don’t understand this area of law very well, that it’s the importers who paid that can bring the claims. They’re the only ones that can bring the claims.

That’s right. And this is something that’s been discussed more broadly, because as the economic impact of the tariffs has been discussed over the last year, there’s been this question about whether, and how much, the cost of the tariffs has been passed on from the importer to a retailer and then to consumers. And the answer to that is: some. And it’s hard to say exactly how much.

But the only party that has standing to challenge the assessment of the duty that was paid to the federal government is the importer of record. Whether there are other legal remedies down the line is a separate question. You know, whether consumers can sue companies is a completely separate question. But as far as seeking a refund of the tariff duty that was paid from the federal government, only importers can do that.

OK, before we get to what the government might argue, it just seems like this process is going to be a nightmare because of the scale of the issue. I mean, is CBP and/or the Court of International Trade set up to process all of these claims at this scale?

The short answer is, it seems that — like — no. And in fact, in that December filing that DOJ made —

What case was that again? Tell us about that case.

So that case was called AGS. And so there were about 1,500 — maybe 1,700 — companies, importers, that had filed lawsuits in the Court of International Trade back in November and December, even before the court had ruled. And the relief that those importers sought from the court was an injunction telling CBP not to liquidate any more entries.

And the hope there was that if CBP stopped liquidating entries, then once the Supreme Court ruled that the tariffs are unlawful, CBP could just liquidate at the lower levels. I think that was an optimistic request in a variety of ways. And ultimately the Court of International Trade denied the motion for an injunction.

And part of the reason why it did — the principal reason why it did — is because DOJ represented, and CBP represented, that it would be a complete logistical nightmare for them to suddenly stop liquidating. And it would create this rush of bureaucratic process all at once once the court rules one way or another. And so the government claimed then that it was going to be too much of a problem for them to put a stop on the process.

But now they’re in a situation where they are going to have to do it — and even more process — because they’re going to have to handle these protests and/or lawsuits, which is even more than they would have had to deal with if they had just stopped liquidation and voluntarily given the money back.

OK, and what — yeah, that’s good. So what has the U.S. government, in its legal filings — has it represented anything that it’s going to stop from re-litigating? What has it represented about — I mean — legal defenses, qualifications on a duty to refund, anything about the proper process? What representations has the government made?

In that AGS case, DOJ represented — along with its representations in maybe two dozen other similar cases, which were all consolidated together — that it would not oppose an eventual court order ordering CBP to re-liquidate previously liquidated entries for the plaintiffs. And that last phrase — DOJ was absolutely consistent in limiting its concession to relief for the plaintiffs in that case.

And that ends up being really important, because what it means is that what I think a lot of importers and general observers of this process may think — well, OK, look, all of these cases are basically the same. Why don’t we just have one case, and then the Court of International Trade says, OK, the Supreme Court ruled — we’re just going to order refunds for everybody?

And what DOJ has been saying — and this connects with a case that the Supreme Court decided last year about universal injunctions, Trump v. CASA — DOJ was saying that it was only conceding that the court had the power to issue that refund relief for the plaintiffs who have actually sued, not for anybody else.

The plaintiffs who had sued before Learning Resources was decided to allege that the tariffs were unlawful.

That the tariffs were unlawful and that they sought this relief about an injunction.

So are you suggesting that the government might argue that anyone who didn’t sue before Learning Resources can’t now bring suit and recover?

No, I don’t think it’s going to say that. I think what it’s going to say — and I think there’s actually a Supreme Court precedent that we can talk about that supports this view — is that they will grant refund relief, if they stick to this concession, to importers who sue later, but not to importers who don’t sue.

Yeah. In other words, there’s not going to be universal relief. That’s why you mentioned CASA.

That’s exactly correct. Yeah.

So you have to sue to get the refund. And the government is suggesting that you won’t get a refund unless you sue. Although my read of CASA was that the Supreme Court, if it issues relief, that decision would be binding on the government universally. So maybe — and again, we’re getting ahead of ourselves a little bit — a fast track to the Supreme Court might be one way to get universal relief.

Yeah, that’s absolutely right. I mean, my read of CASA is the same as yours. And I think that the Supreme Court did — you know, we can talk or not about whether this is a principled distinction that the Supreme Court made between its own remedial powers and the remedial powers of lower federal courts — but it did seem to hold for itself the authority to eventually order universal injunctions in a way that the lower courts cannot.

But the best-case scenario here is that we have cases that go to the Court of International Trade, and then the Federal Circuit, and the Supreme Court. And so maybe the Supreme Court ends up stepping in there and ordering a universal injunction. But we don’t know that for certain.

But are there no other — I mean, the Court obviously in CASA left open the possibility of many other forms of aggregate universal relief, just not universal injunctions — the class action process, complete relief, etc. Are there not other mechanisms that could, in effect, be like a class action here? It seems like there are a handful of jurisdictional issues that need to be resolved that are common to all of these cases — or at least maybe they could be clumped into two or three categories. Is that a possibility?

So the Court of International Trade does have its own class action rule, Rule 23. Whether that’s viable or not depends on a couple of other sort of antecedent jurisdictional issues. What I can say is that there has never once in history, as far as I can tell, been a class that was certified in the Court of International Trade seeking refunds of tariff duties paid.

Now, it’s certainly true that this is the largest tariff refund set of cases that has ever existed in American history, but it’s not the first. And so the reason why — and we can talk a little bit more about this jurisdictional question later that goes to whether you have to protest first or not — is that the jurisdictional provision, or venue provision, for challenging a protest says that only the importer of record can sue. And so it seems that that precludes the possibility of class action aggregation.

And there’s some precedent in the Court of International Trade to say that. And even if Rule 23 is viable for challenges to protests, there are going to be individualized issues. OK — did you exhaust your administrative remedies? Did you protest in a timely way? Was your protest properly filed in all of the technical ways that it needs to be? So even if Rule 23 is available, just as a general matter there’s going to be a serious question about whether Rule 23 is satisfied, because the government will likely argue that those individual issues predominate over the common ones.

Are there no common issues that could be lopped off and put into some kind of aggregate litigation, the way a claims process normally does these things? Is there nothing within the current set of rules or statutory jurisdictions that allows that?

I mean, I think the main common issue has already been decided — whether the tariffs were unlawful. The Supreme Court has already rendered its decision about that.

Now, there may be other arguments the government makes about why it doesn’t have to pay the money back. I tend to think those are quite weak arguments, but it may make some arguments about, well, equitably speaking it’s too much money for the government to have to pay back. There have been some gestures about that from the administration. So that could be a common issue as well.

But as is often the case in class action law, the liability issues are common — but the question of predominance really comes down to the relief.

Right. So that suggests to me one of two things. Either — and again I’m speculating — a lot of potential plaintiffs aren’t going to sue, because the money they might get back, discounted by the probability of recovery, will be dwarfed by litigation costs. That would mean the firms that paid a whole lot of money are going to bring the suits.

Or, if there can be a fast-track representative case that produces universal relief in the Supreme Court, that might be a way of getting broader refund relief.

A third option seems unlikely to me: that Congress could impose some process to expedite all this stuff.

Yeah. I’ll address your third possibility first. That’s clearly the best outcome — and I think there is approximately zero chance of it happening as long as President Trump is president. Any legislation Congress passed, even if it were willing to do it, you have to think the president would veto. So I think that the best solution here — that Congress creates a streamlined administrative process — is probably off the table for political reasons.

Then we get to whether smaller importers — those who don’t already have trade counsel at outside law firms — will have effective access to relief. And even within that, I think there are two categories that are worth breaking out. The first category is, you know, if it turns out that filing a lawsuit in the Court of International Trade is necessary, there's a $400 filing fee.

And so there are going to be some importers where, even if they had pro bono legal representation, you know, there are going to be tens of thousands of — you know — small, small, small importers. You know, somebody who has an Etsy shop or something like that, where it literally is not going to be worth paying the filing fee to get their money back. And that seems like a remedial failing.

But even if you're above that, then you're going to have the range of importers where it does make sense for them to pay the filing fee. And the question is going to be the structure of the legal profession's ability to provide that representation in a cost-effective way. And that's a very serious problem, because trade law is a specialized practice. And, you know, the big firms are really expensive.

So do you have — is there any way for you to guesstimate? The answer is probably no, but I’m going to ask you if you can do it anyway. I mean, what percentage is it? Again, this is maybe impossible to assess, but what percentage might we be talking about of potential plaintiffs who were assessed these tariffs that just is not going to — probably not going to — be cost-effective to try to get refunds? Is there any way to know that number or percentage?

The best information that we have on this — the best data — is that the thousand importers that had the highest volume of tariff payments account for approximately 80 percent of the total dollar value. And so if we do the math and just pick the midpoint of the estimates — let’s say there’s been $150 billion dollars’ worth of tariffs collected since the beginning of 2025 — that means that there’s $125 dollars that goes to these big companies that presumably have their trade counsel and they’re going to be fine. And then there’s another $30 billion dollars that is going to everybody else.

Now, some of those at the top end — you know, the thousand-and-first-largest importer — maybe they can hire a Wall Street law firm. But there’s going to be — you know — I think it’s fair to say tens of billions of dollars that the government has taken illegally. And there’s a real question about whether the importers are going to be able to, as a practical matter, get that money back.

And are the thousand big firms that you just talked about — are they the ones largely who have already brought suits last fall?

Interestingly, no. And this goes to some of the issues that you and Bob have talked about more generally. So there are some big companies like Costco, for example, who brought suit. But a lot of them are smaller companies, and interestingly ones that are owned by private equity firms.

The big companies — the biggest, biggest payers — Walmart, Amazon, and so on, have not filed, too. Now, the reason why is they didn't need to. It was not yet time where they needed to take a public stand. And notice that the plaintiffs in Learning Resources — it was Learning Resources, you know — and then one of the other cases was brought by a small clothing company called Princess Awesome that makes sort of little girls' clothes — that's sort of positivity about Wonder Girl.

And these are small businesses. And the best guess that I can give about why that is is that large corporations like Walmart, like Amazon, have been concerned about the possibility of retribution from the administration.

Yeah. And I take it — that’s what I would have assumed before you said it. And that’s happened in other contexts. I wonder — do you — is there any sense in the news about how these big companies that could get potentially large refunds are thinking about the financial gains versus the political and retributive costs?

I think the best indication that we have is that the trade associations are already issuing press releases calling on the federal government to give immediate refunds — you know, holding aside the fact whether that was even possible. But, you know, the fact that the trade associations are out there now speaking clearly — you know, we don't yet have press releases from Walmart. But, you know, I think that, you know, before they thought it was maybe unnecessary for them to stick out their nose. But now, you know, this could be $10 billion for Walmart. I think that's just — yeah — it's real money. And so I don't think that they're going to just let it slide.

Yeah. OK, what else — what else haven’t we discussed that’s important in understanding this issue? Is there anything else?

I mean, the only other thing that I think is really important here is — well, I’ll say two things. The first is that these administrative exhaustion requirements — the Court of International Trade has been a little bit wishy-washy. And actually, in the AGS case that we talked about before — which was the Costco, etc., those other companies had sued in December.

The court, when it denied the motion for an injunction, said essentially that there was an avenue for judicial review that didn’t require protesting. And that draws on a line of cases going back about 25 years in the Court of International Trade and the Federal Circuit that suggested that this protest administrative exhaustion requirement was at least sometimes optional if it was futile. Now — and the argument here is that protesting would be futile because CBP was just following orders from the executive order.

Now, I think that futility argument is weaker now than it was before, because the executive order has been determined by the Supreme Court to be unlawful. But even beyond that, there’s a lot of Supreme Court precedent that says that when you have a statutory exhaustion requirement, it’s mandatory. And courts can’t carve out exceptions to that. So I think that there’s a significant risk that what the Court of International Trade has already said about these cases is actually going to get reversed on appeal.

So let me ask you one final question. Politico said in the story in the last couple of days — I'm just going to read from it — the Justice Department and parties in the tariff litigation have already asked the Court of International Trade to establish a steering committee to coordinate the more than 1,000 refund-related cases now pending. A common step in large, complex trade disputes. Was that the Rule 23 process you were talking about earlier, or is that something else?

No, that’s something else. And this is something that is borrowed from the multidistrict litigation process for mass torts. And so these are cases where — so the most famous recent one is there’s a military base, Camp Lejeune, where there were toxic chemicals, and some tens of thousands — maybe even hundreds of thousands — of service members and their families were exposed to these toxic chemicals and got a variety of forms of cancer.

Now, this can’t be a class action, because every case has — you know — you have to have the medical records for every case, et cetera. So they’re individual cases, but they get grouped together in one court that then handles all of these cases and comes up with a judicial administrative process to make it more efficient. And it often — so the process often involves picking what are called bellwether cases to go to trial first.

Now, in these tariff cases, there’s not going to be a trial, but it makes sense that they would litigate one or just a handful of cases first to see what the government’s defenses are, and then sort of come to a judicial determination about whether those defenses are valid, whether it has to go to the Supreme Court again or not. And then whatever the legal resolution of those cases, those bellwether cases would just apply to all of the other cases that had been filed in court. But that still requires everybody to file a lawsuit.

Right. So last question — and it's a bit of a wrap-up, but it may be repetitive. So one question is: who can take the initiative? Which institution can take the initiative to make this more efficient and more fair — this process of refunding the illegally collected, the illegally collected moneys?

Congress could do so, but it's not going to. The Trump administration could maximally refund unilaterally within whatever the law permits it to. Probably unlikely, I'm guessing. We've just been through a couple of different alternatives that the Court of International Trade might pursue. We've talked about a fast-track case in the Supreme Court, which might be able to give a type of universal relief. Is there anybody else in the system anywhere that could improve this process, or are those the options?

Those are the options, as best I can tell. You know, I think it's worth underscoring again — Congress could fix this problem, but Congress and the president won't do it. CBP could, as you say, maximally within the procedures and legal framework that exists right now. You know, it's remarkable — CBP is, as of today, still assessing IEEPA tariffs because they haven't issued new guidance. And it's not until they issue new guidance that they'll stop assessing IEEPA tariffs. And so, you know, the level of administrative — let's just say slow walking — is remarkable.

Yeah. OK, thank you very much, Matthew. That was elucidating.

Thanks, Jack.

Mentioned:

• “The Bad News in the D.C. Grand Jury’s Refusal to Indict Six Members of Congress” by Bob Bauer (Executive Functions, Feb. 12, 2026)

Thumbnail: President Donald Trump on Nov. 20, 2025 calls for the prosecution of lawmakers who urged service members to refuse unlawful orders. (Truth Social.)

This is an edited transcript of an episode of “Executive Functions Chat.” You can listen to the full conversation by following or subscribing to the show on Substack, Apple, Spotify, or wherever you get your podcasts.

Jack Goldsmith: Good morning, Bob.

Bob Bauer: Good morning, Jack.

This morning, we’re going to discuss actions taken by the Trump administration against six Democratic members of Congress, all former military or intelligence officials who are serving in Congress, and who produced a video in which they counseled that members of the U.S. military can or must refuse legal orders, which is a true statement of law.

The president had an allergic reaction to that and said that they should be arrested and put on trial. Subsequently, Secretary of Defense Hegseth criticized in particular Senator Mark Kelly, who is a retired naval captain subject to retirement benefits, saying that he acted illegally in trying to influence the chain of command and to influence discipline in the military, and suggesting that he could be subject to criminal actions, and also suggesting he could be subject to his retirement benefits being reviewed and perhaps reduced.

And then subsequently, DOD took that action against Senator Kelly to try to reduce his retirement benefits. Senator Kelly sued the Department of Defense, and yesterday Judge Leon issued a preliminary injunction enjoining those proceedings against him.

And also earlier this week, we learned that a grand jury in the District of Columbia refused to indict the six members of Congress who produced the video under — I believe — the statute that the U.S. Attorney in the District of Columbia had sought indictment for was 18 U.S. Code 2387, which talks about interference with the loyalty, morale, or the discipline of the military or naval forces.

So these were members of Congress speaking out in their capacities as members of Congress. The Trump administration tried to bring legal proceedings against them and failed on both counts.

Let’s start with Judge Leon’s decision yesterday, the injunction enjoining this Hegseth–DOD action. What do you think about it?

It’s a very strong opinion, strongly expressed. He doesn’t, by the way, review all of the defenses that Senator Kelly put forward. He didn’t really address the separation of powers questions or speech and debate questions.

He layered the fact that Senator Kelly is a sitting member of Congress in an opinion that was largely focused on his First Amendment rights as a retired serviceman. But it’s very difficult to see the opinion as anything but strong. By the way, it’s, as I said, strongly expressed. I lost count of the number of exclamation marks. He literally showers the government with contempt and does not see the issue as a close one.

Yeah, I didn’t think the exclamation marks were terribly judicious, but I agree the opinion was largely persuasive. There are a couple of difficult legal issues. The whole question of the scope of judicial review and the timing of judicial review over discretionary military actions like this is a little murky to me, and his analysis of that was pretty compressed.

There’s also an open question about — I have no doubt at the end of the day that this was retaliation for the exercise of First Amendment rights and that Senator Kelly will prevail at some point — but there’s also an open legal question about whether, let me just say that members of the military have significantly diminished First Amendment rights. So one question is whether, as a retired member of the military, still subject to military discipline, receiving military benefits, subject to being called up again, Senator Kelly — how much, if at all, are his First Amendment rights diminished?

And how is that impacted by the fact that he sits on the Armed Services Committee and the Intelligence Committee and is a member of Congress? These are all novel issues that will ultimately need to be worked out. But it’s an amazing rebuke, especially coming in the context of the grand jury refusal to indict — which almost never happens.

It’s so rare that a grand jury fails to indict. We’ve seen it a lot in this administration, but this was an especially noteworthy one.

It certainly was. There’s a report out that the U.S. attorney brought somebody onto the case who hasn’t practiced law for a number of years. I mean, we may be seeing again the effects — I’ll have to look into this a little bit more closely — but the effects again in this case of a Department of Justice that is thin on talent and having difficulty recruiting people who you would want to bring the cases that they care the most about.

But I think it’s troubling that they sought to bring the case at all. I mean, I want to go back to something the judge, as I said, layers into the opinion, but doesn’t put as much stress on it as I would have liked to see. And that is that Senator Kelly is a sitting United States Senator, and the administration threatened him and pursued him with legal action for fundamentally disagreeing with the administration — and disagreeing with the administration by articulating, as you point out, a point of law that is absolutely correct, as stated was absolutely correct.

That may not make any difference in some respects because perhaps they could have disagreed with each other about whether that point of law was correct. In this case, it clearly was. But I just think the attack on sitting members of Congress — the weaponization of the law against them over what is flatly a fundamental policy or legal difference — is an extraordinary expansion in the weaponization program of this administration.

Right. I mean, the civil case by the Defense Department over Kelly’s retirement benefits raised a lot of technical questions. I’m confident Kelly will prevail ultimately. But the criminal action against members of Congress, some of whom weren’t in the military, for stating an opinion about the legality of the Defense Department’s actions is — I’m not sure — I don’t think it’s actually going to fall within the speech and debate clause — but it can’t be the case that the administration can retaliate with criminal process against a member of Congress for offering opinions about the legality or even the prudence or whatever of the Defense Department’s actions.

And the fact that a grand jury wouldn’t indict is pretty good evidence of how thin the legal case was.

I could not agree more. It’s astonishing, as I said before, just remarkable that the case was brought. Remarkable that — and admirable that — the grand jury rejected it.

Let me ask you this, Bob. You said this is an extraordinary case of weaponization, and I agree. I mean, the president gave marching orders when he had this allergic reaction, and the U.S. attorney in the District of Columbia and the secretary of defense tried to execute those marching orders, and they both failed, at least as of today.

But one — someone might say, well, this is an example of failed weaponization. It didn’t work. The courts or the legal system rebuked both efforts, and therefore this is a victory against weaponization. I don’t think that’s your view, right?

No, I don’t. I think there’s — as I wrote in a piece yesterday — it’s good news and the bad news. The good news is the grand jury rejected it. The bad news is that the administration — and there had to have been lawyers in the administration who knew the case was weak — brought it anyway. And that suggests that they’re hell bent on pursuing these paths.

And we should never discount the costs of these kinds of cases being brought, even if they’re brought ultimately unsuccessfully. They target people. They put them to the expense of defending themselves. They have to deal with leaks in the press and the entire communications problem around looming criminal action against them. These are not costless.

And the message that the administration is sending is we are prepared to do this to our political enemies. I think that — that — that is extremely dangerous. There’s no reason to think that because they failed in this instance, they won’t attempt to do it again.

Right. And I want to emphasize that point. I mean, the impact of these cases, even if they fail, is not limited to the six Democratic lawmakers. What the administration is basically saying is we will use every tool at our disposal, and we don’t care if we win or lose in court. We’ll use every tool at our disposal to do harm to people who disagree with us.

And you know, Senator Kelly is retired and he’s got another salary and he’s going to be fine no matter what happens. But the impact of this goes beyond Congress. It goes throughout DOD. I think that they’re — I’m confident that this has a chilling effect on every action within DOD in terms of whether people are afraid to disagree over some matter. The lawyers who are offering legal advice, I’m sure, are impacted about this.

The deterrent effect is much broader, even if they lose, than these six, because it’s costly. It’s reputationally costly. It’s stressful, and people don’t like to go through this. So I think even though these cases failed, they’re like many things in the administration: they lose in court, but they might still win — and within their worldview of what counts as winning.

Absolutely. We’ve seen this in other areas — unrelated areas. I mean, they’re related in theme, but they involve not a conflict between Congress and the executive branch, but the executive branch and private organizations. The attack on the law firms — these executive orders were issued. They’re no longer being issued. There are no new ones after the wave, the first wave that the administration issued, but they were able to create a kind of interim environment in which it appears from the reporting large law firms are thinking hard about taking steps that could conceivably provoke the administration.

Now, maybe that period of time has passed because the administration appears to have moved on from these attacks on the law firms. But there’s no question there’s an intimidation factor. The administration is perfectly happy to generate, to put into play, even if it knows that ultimately it’s not going to win.

It's not going to win in court, but it can still win within the executive branch and even vis-à-vis Congress. You can imagine some members of Congress being intimidated at the margins by this.

Absolutely. Now, it is interesting, speaking of the grand jury rejecting the charges, it is interesting that in this instance we have heard Republican members — and not the ones from whom we hear dissent from time to time, but Republican members generally — expressing grave reservations about the steps that were taken against the six Democratic lawmakers. So that’s interesting. They recognize that this is a problem for everybody — Democrats and Republicans — and for the institution of the Congress as a whole.

So that may have some effect on the administration — that in this instance the president may not be able to rely on his own party to support him in this course of action.

But those steps — I mean, they didn't do anything about it. There was no pushback from Congress as an institution. I agree there were a few members on the Republican side who said this wasn't a good idea. But one would have imagined a much more robust reaction across the aisle from Congress at this pretty aggressive intrusion on congressional prerogatives.

Yes, but it's also possible that if there's a next round, the response will be more robust, especially in light of the grand jury. It’s so embarrassing to the Department of Justice, just turning those charges away, just rebuffing the government in the submission of those charges. And it may strengthen the hand of Republicans who say that, you know, they cannot support this and pressuring the administration to desist. Again, I'm speculating — maybe not.

So let’s talk about — in your piece yesterday, you proposed a reform. You talked about how bad this is from a separation of powers and free speech perspective in terms of congressional prerogatives. And you proposed a reform, not for now, but for later. After this administration, you proposed a reform to dampen the executive branch’s ability to bring a criminal action like they did — tried to bring — in the District of Columbia. Why don’t you just outline that?

Certainly. It seems to me that this is a category of criminal prosecution that the executive branch should not be able to proceed with in an unfettered fashion. So I proposed that the Attorney General of the United States be required to apply to the United States Court of Appeals — a division of the United States Court of Appeals — for the appointment of a special counsel in any case where they are targeting a sitting member of Congress.

Now, we know from United States that the Attorney General cannot — the executive, the president — cannot be dispossessed entirely of the power to exercise prosecutorial discretion, proceed with the prosecution. So the Attorney General would have the ultimate say. But at a minimum, the Attorney General would not be able to pick the lawyers who evaluate and ultimately decide whether to bring the case.

The court would appoint such a lawyer, presumably to appoint a lawyer that had the reputation and background that would suggest an independence of judgment — that would confront the administration over the merits or lack of merits of a particular prosecution.

Just to make this short, the special counsel would have to report to the Attorney General, but any steps that the Attorney General made to overrule a prosecutorial choice by the special counsel would have to be reported to the Congress. And in the end, if the special counsel recommends against prosecution, the Attorney General decides to proceed anyway, that too — that decision to overrule the special counsel — would have to be made known to the Congress.

And at that point, there’s nothing more that can be done by law. The ordinary operation of politics would have to do the work, and the administration would have to defend a prosecution that was brought in the face of a recommendation of the contrary by the special counsel.

But yes, it’s a little bit of a — it’s an echo of the independent counsel statute of old, but that statute was directed toward investigation of wrongdoing in the executive branch to try to introduce some independence in the executive branch prosecution, if you will, of itself. And this is different. This is an attempt to shore up the separation of powers and to protect the Congress from what we see the Trump administration doing right here.

It would seem to me that members of both sides of the aisle would have an interest in a reform of this nature.

So I disagree, not surprisingly. I don’t think it’s going to be fruitful to discuss much whether this proposal would be constitutional. I tend to doubt it. I would have doubted it before Trump v. United States. I especially doubt it after Trump v.United States, given what the Court said about the president in robust terms and in an extreme context about the president’s exclusive power over investigation and prosecution.

But setting the constitutional issue aside, I just don’t think it’s a good idea. And the reason I don’t think it’s a good idea is I just don’t think it’s — Congress has plenty of tools to protect itself that don’t require it micromanaging how the executive branch chooses lawyers and brings prosecutions, which I think is a bad idea. It could change the criminal laws. It could just fix the criminal laws. It could exclude Congress from whatever criminal laws it wanted. It could put conditions on the criminal laws for when they’re implicated. Congress also has a constitutional speech and debate clause that it can — to protect itself.

And moreover, one of the big problems with the independent counsel statute was that it politicized the courts. The courts got in the business. It was very controversial when the courts appointed Ken Starr to pursue Bill Clinton. And one of the things that were mentioned in the 1990s, when there was unanimous agreement that the independent counsel statute was terrible, was it had the impact of politicizing the courts.

So why wouldn’t it be better for Congress to stay within its lane of its Article I powers? You’re going to tell me this is an Article I power — but why wouldn’t it be better for them to simply change the criminal law that can apply to them?

It won’t do it for a couple of reasons. There are major collective action problems we’re all familiar with. We also know that we live in a system — and it doesn’t look like it’s going to change anytime soon — where we have a separation of parties rather than a separation of powers that seems to significantly diminish Congress’s interest in or ability to defend its institutional prerogatives. And I don’t think that we should leave it to Congress in this era of our politics to give up that kind of ground.

I think we should take steps to shore up Congress’s ability to protect itself and to provide it with incentives to protect itself. So I think that on paper everything you say is correct about Congress having the tools to respond, but it cannot be counted upon to use those tools. We probably should doubt that it will use those tools.

Also, the courts here have a much more limited role under my proposal than they did under the independent counsel statute. Here they’re limited to counsel selection. They don’t have any ongoing supervisory role. They were much more deeply enmeshed in the operation of the independent counsel statute. So I’m taking into account that experience. And I think in that respect, I think that the factor of politicized judicial involvement — or the risk of politicizing judiciary — is diminished.

And let me just close. I’m sure I won’t persuade you on this. I just set out in the piece the reasons why I thought there was constitutional space for affirming a reform of this kind. But we are in a completely different era now where, contrary to the concerns, for example, that Justice Scalia entered in his famous dissent in Morrison v. Olson upholding the independent counsel law, the risk of weaponization here is of a completely different order. And this is an anti-weaponization statute or reform. The one that I’m proposing is an anti-weaponization reform. And Justice Scalia was concerned that the independent counsel statute was a weaponization vehicle. So it’s the exact opposite. My reform would flip it into a completely different function. That’s how I see it.

I’ll say a couple of things, and I’ll give you the last word.

One, I think you can call it anti-weaponization, and I understand the differences with the independent counsel statute. Every one of these things — experience since Watergate has shown — that every one of these mechanisms to try to put a prosecutor with some distance from the attorney general has ultimately become very, very politicized and weaponized. And I think that would happen here too.

Second, the court’s decision to choose a lawyer is going to be massively controversial, and that will become politicized in the courts — even if they have no more ongoing role — that would become part of the political debate in ways that I think were terrible for courts back then, and I think would be now.

And third of all, I don’t understand a world in which — yes, separation of parties, not powers, which means it’s very hard for Congress to pass legislation — but I don’t understand a world in which Congress could pass legislation to do this statute that pretty clearly gets into the Article II business, would have the support in Congress — the bipartisan support in Congress — to do a reform of that type, but not do a reform of substantive criminal law.

I mean, basically, those are two separate options for Congress enacting statutes, and it seems to me that if Congress is worried about how the president might enforce criminal law against it, it should change the criminal law and not try to change the Article II process.

Well, before I respond, can I ask you one question? What kind of reform of the criminal law — if something had other tools they have to punish the executive for doing this — what other reform of the criminal law do you have in mind?

I mean, we don’t know quite what statute, but the statute that apparently — if it’s 18 U.S. Code 2347 — the statute could say this statute does not apply to members of Congress. So members of Congress are just not subject to the criminal prohibition about interfering with, impairing, or influencing the loyalty, morale, or discipline of the military. And any statute in which they thought that there was a chance of that happening, they could go through and act within their prerogatives to control the content of federal criminal law, to give themselves the space they think they need to be able to exercise their First Amendment rights.

And moreover — I mean, your proposal applies not just in weaponization cases, but it would apply in genuine cases of congressional corruption. Basically, you’re saying that in a real case of congressional corruption, that the Justice Department is going to have to be stuck with whomever the court says should prosecute that case. I just don’t think that is a good solution when Congress has more direct solutions.

Well, let me just — a few closing points about that. First of all, I’m totally open to statute-by-statute amendment to protect the Congress in circumstances like this. I’d have no objection to that. And if it turns out it’s easier for Congress to do that — which is a pretty thoroughgoing law reform project — than what I propose, then at least, you know, you and I share the same objective, and there’s a law reform that maybe we can agree on. So I’m totally fine with that.

Even public corruption cases in the past have been highly controversial. The administration can — any administration can — gin up a public corruption case under the very elastic theories of criminal liability that we have against a member of Congress that is 100 percent powered by political animus. And that has been a claim that, you know, maybe rightly or partially or wrongly, defendant members of Congress have made in the past — that they’ve been picked out for a particular charge because the administration is trying to take them out politically. And so I don’t see any problem with, as you put it, sticking the administration with a court-appointed lawyer in those circumstances.

I just think that what we are seeing now with this administration is a commitment to weaponization, as you point out, that is indifferent to the merits and is looking to achieve an intimidation factor that requires a radical rethinking of constraints on the executive branch, where it is trying to, in effect, dominate the United States Congress, where it’s trying to basically undermine the ability of the United States Congress to perform a critical constitutional checking function.

And we could say, well, when Donald Trump leaves, that’ll end. I don’t know that that’s the case. We see it, and we should respond to it in some way. But, by the way, I do like your suggestion of going statute by statute. I just wonder whether that — as I said — sort of thorough law reform project is more likely or less likely than what I propose. But I think it’s a perfectly good idea to address the problem that way, too.

Well, let's end on that note of semi-agreement.

Yes, it's good.

Thank you, Bob.

Thank you, Jack.

Since this chat was recorded, President Trump today signaled support for the Chagos sovereignty transfer, or at least a rollback of his criticism.  He also said: “If the lease deal, sometime in the future, ever falls apart, or anyone threatens or endangers US operations and forces at our Base, I retain the right to Militarily secure and reinforce the American presence in Diego Garcia.”

Mentioned:

• “Britain’s Chagos Deal Can’t Go Forward Without U.S. Consent” by Richard Ekin (National Review, Jan. 30, 2026.)

Thumbnail: Diego Garcia photographed by the Sentinel 2 satellite in 2021. (European Space Agency / CNES).

This is an edited transcript of an episode of “Executive Functions Chat.” You can listen to the full conversation by following or subscribing to the show on Substack, Apple, Spotify, or wherever you get your podcasts.

Jack Goldsmith: Today’s chat is about Great Britain’s plan to hand control of the Chagos Islands in the Indian Ocean to the Republic of Mauritius. This event may seem irrelevant to executive functions, but it is quite relevant for two reasons. The Chagos Islands are home to the vitally important U.S.–U.K. military base, Diego Garcia.

And until yesterday, anyway—and the news reports are still unclear on this—President Trump has been a strident critic of the impending deal. To help sort this out, I’m going to chat with Professor Richard Ekins, who is a professor of law and constitutional government at Oxford, a distinguished global professor of law at Notre Dame, and head of the Judicial Power Project at Policy Exchange, a leading U.K. think tank. Professor Ekins is an expert on this matter, and he’s been very critical, in publications both in the United Kingdom and the United States, of the British government’s plan to return sovereignty over the islands to Mauritius.

Richard, good morning. Can you explain, for listeners who don’t know much—if anything—about this, the background to this episode?

Richard Ekins: Sure. So, the Chagos Islands are an archipelago of about 60 islands in the Indian Ocean, 300 miles south of the Maldives, 1,300 miles northeast of Mauritius. As you say, the Diego Garcia base is located therein.

Diego Garcia is far and away the biggest island in the archipelago and exceedingly well located, sometimes referred to as the Malta of the Indian Ocean—a single aircraft carrier, as your term is familiar, I’m sure. So, a very, very significant base. The Chagos Islands became part of the British Empire in 1814, as part of the Treaty of Paris.

They were, before that, French, and ceded to the U.K., to Britain, in 1814, along with Mauritius and the Seychelles. And in 1965, the Chagos Islands were excised from the colonial territory of Mauritius. They were nominally administered as part of the colony of Mauritius, despite being a very long way away and having no meaningful connection other than that sort of accident of colonial boundaries.

So, they were excised from the colony of Mauritius, renamed the British Indian Ocean Territory, and preparations were made for a U.S.–U.K. military facility. The Mauritian government was compensated for this. This was in advance of independence.

Mauritius became independent in 1968. The people who lived on the Chagos Islands, the Chagossians, were removed from the islands, and the base was built. In 1982, Mauritius initiated a claim to the Chagos Islands, having otherwise said, well, they’re British.

Mauritius played quite an effective hand through the international institutions across the subsequent decades. I think it did a very effective job of leveraging the injustice done to the Chagossians, who were removed in a pretty high-handed fashion, without adequate compensation being paid.

In 2017, Mauritius persuaded the General Assembly to refer the question of the Chagos Islands to the International Court of Justice for an advisory opinion. That was over the head of the U.K. and the U.S., who said this was an inappropriate matter for an advisory opinion—there’s a live dispute between Mauritius and the U.K.

In 2019, the International Court of Justice issued an advisory opinion saying the decolonization process in the Chagos Islands had been incomplete. The U.K. government said, well, it’s an advisory opinion; we’re not going to do anything about it, thank you very much.

In 2021, a special chamber of the International Tribunal for the Law of the Sea, in a dispute between the Maldives and Mauritius, took for granted that the 2019 advisory opinion had established that Mauritius was sovereign.

In 2022, the then Conservative government of the U.K. entered into negotiations with Mauritius about the future of the Chagos Islands. I published my first paper for Policy Exchange on this matter in 2023, saying it was a very bad idea to hand the islands over.

The then government didn’t. But in October 2024, shortly after they came into office, the new Labour government of the U.K. reached an agreement with Mauritius to transfer sovereignty and lease back Diego Garcia for 99 years, with substantial annual payments to be made each year for continued use of the base—or the island for the base, I should say.

The agreement has yet to be ratified. It was officially signed in May 2025, but the parliamentary process for ratification hasn’t been completed, and it’s not clear, obviously, whether the U.S. administration supports it or not. That’s where we’re at.

Okay, great. So, I want to talk about legal issues first, and then we can talk about policy issues, and they may blend together. So, there’s a 1966 treaty between the United Kingdom and the United States that says the territory shall remain under United Kingdom sovereignty—referring to the Chagos, I believe, or Diego Garcia?

No, referring to the whole Chagos Islands, and actually an island chain off to the west of the Indian Ocean, too. That was handed over to the Seychelles when they became independent in 1976.

Okay, and you’ve argued—and I think the U.K. government has acknowledged—that United States consent under that treaty, the 1966 treaty, is a prerequisite to the transfer of sovereignty to Mauritius. So, did I get that right, and can you explain that?

Exactly. I mean, this came sort of to light in some public discussion only a few weeks ago, when President Trump came out and denounced the agreement on the 20th of January. I think the U.K. government had understandably thought American consent was in the bag, but it was made clear—by me and by others—that we have a treaty commitment, the U.K. to the U.S., that the Chagos Islands will remain under U.K. sovereignty. And unless that’s changed, if we go ahead and transfer sovereignty to Mauritius, we’ll be in breach of the 1966 treaty.

So, that treaty has to change—that agreement has to change—before ratification can go ahead, at least can go ahead without the U.K. immediately being in breach of the treaty with the U.S., and being in breach of its treaty with Mauritius, because we agree in that treaty that we’re not breaching any other obligations when we hand the islands over.

So a couple of questions here. First of all, I guess we should first talk about what is the U.S. position as of Thursday morning? I mean, President Trump several weeks ago said this was a stupid idea; the impending transfer represented, “total weakness,” and was great stupidity.

It seemed then that he was going to veto this or wasn’t going to give consent. In the last 36 to 48 hours, I’ve read news reports in Great Britain—but less certain in the U.S. news reports—that President Trump is indeed going to consent. There’s been some agreement reached.

Can you tell us where things stand as of Thursday?

Well, it is unclear, as you say. So, there were news reports last night in the U.K. that, surprisingly, in a way, President Trump was going to sign off, and that Number 10 was awaiting a phone call from President Trump to sign off, so to speak—which was a surprise.

And the news reports suggest that somehow Prime Minister Starmer had persuaded him that all was well. It’s a suggestion that further security assurances were going to be given by the U.K. to the U.S. Quite unclear what that means. Quite unclear, actually, how the U.K. would be in a position to give further assurances, unless that means reopening the treaty with Mauritius—but there’s been no suggestion of that yet.

So, we have to wait and see if it’s true, in fact, if that is President Trump’s position, and if he is, in fact, going to let this go forward.

Okay. Let’s talk about what’s been in the background—the security assurances. Just explain why Diego Garcia is so important and what the concerns are from a military, national security, defense perspective. Just list the concerns as you see them from transferring sovereignty from Great Britain to Mauritius.

Well, so it’s an exceedingly significant air and naval base because of its location—very good for projecting power into all sorts of utterly critical locations. I think that’s acknowledged by—

Flesh that out, I mean, a little bit.

Oh, so, well, I’m not a military expert, but I think given the—

The standard understanding of the importance of Diego Garcia.

Sure. So, it was used throughout the Iraq wars. It’s a base that is within striking range of the Persian Gulf and the Straits of Malacca. And if you’re looking at the map, it’s sort of right in an utterly critical position for, sort of, dominating the Indian Ocean—maybe not dominating, but projecting power from.

So, there are very good reasons why the U.K. has wanted to retain control for defense purposes, and why the U.S. has viewed it as very significant too. And I think there are a host of important U.S. Navy facilities there, and Air Force, and so on.

So, it’s, I think, very widely acknowledged to be extremely significant. The question is, in the U.K. government’s argument for, sort of, handing the islands over to Mauritius is: yes, it’s a very significant base; they’re not attempting to diminish it. It’s because of its significance that we have to reach an agreement with Mauritius. That’s their argument—or at least their headline argument. There’s another argument going on beneath, I think.

I don’t buy the argument, because I think exchanging U.K. sovereignty, subject to Mauritian challenge, for Mauritian sovereignty, subject to the terms of the treaty, is a bad trade. I think it’s a bad trade for a host of reasons—but to begin with—the term of the lease agreement is 99 years. There is an option of renewal, but the option is not practically enforceable, I think, and one could certainly foresee it not being capable of being actually exercised.

And it’s perfectly foreseeable that in 99 years this will remain an utterly critical strategic asset, which would be foolish to yield—not to be able to continue to exercise, or worse, to be taken up by a hostile state, the obvious one being China, but who knows what things are like in 100 years.

So, I think it’s short-sighted. But even within the 99-year period, I think exchanging U.K. sovereignty—again, subject to Mauritian challenge—for Mauritian sovereignty introduces dangers that are not otherwise there.

And the dangers being that, yes, the U.K. and the U.S., according to the terms of the treaty, will remain in sort of exclusive control of Diego Garcia, and there are some related legal risks we’ll probably come to, but they’ll stay there on the island.

But there are a lot of other islands in the archipelago, and their use—or their misuse—would be a security risk for Diego Garcia. Now, the terms of the treaty provide for the exclusion of foreign military forces from there, but there are options open, including Chinese fishing vessels, or nominal scientific installations, or so on and so forth, where, if projects go ahead there—which the U.K., as it stands, could simply deny and evict—things could go ahead that pose a threat to the base.

So, what are the—two questions, actually. One: what is the reason beneath the stated reason why the British government is doing this?

Yeah, so I said the reason beneath. I think there’s a mix of two. So, one is post-colonial guilt, which is related to—it’s maybe part and parcel of the same thing—but the Chagossians, as I said, were badly treated in the late ’60s, early ’70s.

Now, they’ve been compensated, and in a sense more recently compensated. Very many of them are now British citizens, but there’s, I think, understandably, a kind of guilty conscience about what went on. And Mauritius has done a very good job at leveraging that injustice and that guilty conscience into a “you can remedy it by transferring sovereignty to us.”

That’s certainly the view of many international human rights lawyers and similar. Now, I think it’s nonsense in a way. I think actually going ahead with this treaty will compound the injustice rather than address it, because Mauritius has a very poor record for how it deals with the Chagossians. Most Chagossian groups are opposed to the treaty.

So, it’s a misleading ground for action, but I think it’s part of what’s in the mix.

The other ground, which maybe is even more interesting, is I think there’s a certain vision of international law that’s driving the present government. The present government was led—Keir Starmer was a distinguished human rights lawyer; his attorney general was a leading human rights lawyer as well.

They have a certain understanding of international law, human rights, that I think drives their approach to governing. And this was a decision made, you know, to hand the Chagos Islands over—a decision made early on in their government, before, frankly, things started to go wrong and they had to think more prosaically about the project of governing.

And it’s an early speech given by our attorney general where he refers to the succession of the Chagos Islands to Mauritius as, “this showcases our sort of total commitment to the rule of law.” Now, again, I think this is nonsense.

I don’t think international law requires the U.K. to hand the islands over. I think, in fact, we’re surrendering to the abuse of international adjudication—effectively letting an International Court of Justice advisory opinion function like a binding adjudication—when the U.K. didn’t consent to adjudication, doesn’t have to consent to adjudication, and yet is being—having that sovereign right subverted.

But I think that that’s definitely part of what’s going on. And you can partly tell that not least by public statements, but also what the Labour Party was saying when they were in opposition, where there were many appeals to the international rule of law as just making it obvious we have to hand the islands over.

Okay, I want to come back to that vision of international law in a moment, but let’s move to, I think, what you’re referring to when you talk about legal risk for the United States. So, explain what potential restrictions there are under the transfer of sovereignty treaty between Great Britain and Mauritius—what possible restrictions there could be on the United States’ use of the base.

I’m thinking about—and it’s actually a different treaty; you can explain this to us—the possible storing and usage of nuclear weapons from there, and the use of force generally. Can you explain what the legal regimes are that might impact that?

Sure. So, I mean, the terms of the treaty: the Chagos Islands, including Diego Garcia, become sovereign Mauritian territory, but the U.K. will have effectively exclusive control over Diego Garcia. So, it will be a U.K.-based—well, U.K. facility—which is jointly operated with the United States.

And the U.K. will undertake various commitments to Mauritius in relation to this, one of which is that all operations that take place out of Diego Garcia will be compliant with international law, which obviously includes prohibitions on the use of force.

And U.K. consent is required before Diego Garcia is used for U.S. operations. That’s the case now; it’s the case after this treaty goes ahead. And so, the U.K. really will be undertaking to Mauritius that it won’t let the U.S. use the base as a base for military operations unless it’s compliant with international law prohibitions on the use of force.

So, it’s perfectly foreseeable if, say, the agreement were in place now and President Trump decided to tamper with his reaction against Iran in support of the protesters—serious argument, of course; not saying compelling, but serious argument—that that would not be compliant with international law, and that would place pressure on the U.K. to deny use of the base to the United States for that purpose.

And now, those arguments would loom large with the U.K. in any case, but Mauritius would have standing to insist that we were in breach of the agreement if it took the view that the actions that were going on there were non-compliant.

So, I think you’re adding in not just, you know, what does the U.K. think about the use of force? The U.K. may take a different view to Mauritius, especially if Mauritius comes under pressure—as is entirely foreseeable, I think, in the medium term.

But—and what about the Treaty of—is it Pelindaba?

Pelindaba, yeah—nuclear-weapon-free zone. So, Mauritius, along with a lot of other African states, has undertaken to prohibit the stationing of nuclear weapons on its territory. If this treaty goes ahead, Diego Garcia will be Mauritian territory, and the Treaty of Pelindaba will apply.

And so, Mauritius will be required, as a matter of that treaty, to prohibit the stationing of nuclear weapons on Diego Garcia. Now, the U.K. government seems to have either persuaded itself, or will think, there’s some way around this—that this just won’t be a problem.

And I don’t really know what they were thinking, other than maybe Mauritius just won’t be that interested; it’ll look away, which doesn’t seem a sustainable strategy—especially given that any signatory to the treaty can raise this as a concern, and there can be forced inspections to determine if nuclear weapons are being stationed.

There are arguments about what “stationing” means, but it seems there’s a pretty serious risk that the U.K. and the U.S. would be placing Mauritius in breach of the treaty.

And then, of course, Mauritius could say, well, you can’t place us in breach, and thus you’re in breach of the agreement, and we’re entitled to countermeasures—we can let the Chinese set up a base on the other islands without being in breach of the treaty ourselves, et cetera. It’s quite hard to know how this would exactly unfold.

It looks like there’s a pretty serious legal risk to me there.

So, this leads to the thing I really don’t understand about all of this. I understand the current U.K. government’s vision of international law, I understand that perspective. I understand the guilt argument clearly. I can understand why the U.K. might want to transfer sovereignty here.

What I don’t understand at all is why the U.S. is even thinking about giving consent. That vision of international law is the antithesis of the vision of international law that the Trump administration holds.

Given—just to go sideways a little bit—given the rhetoric and reality of what they’re trying to do in Greenland, and the importance that they place on control of strategically important landmasses for national security, and given the Trump administration’s vision of international law, and given the risks—whatever the risks actually are—we’re talking about international law, legal entanglements, and the politics related to those, and how that might influence decision-making in Washington.

But all of these things are things that the Trump administration is not remotely interested in, doesn’t care about. So, I’m having a very hard time understanding, in light of all that, why the Trump administration is even considering consent to this.

Is there a good answer to that?

Well, I share your bemusement as part of the answer. I mean, when Trump was elected, I think the British government were deeply worried that he would veto the agreement for sure. They put considerable effort into persuading him and the administration that it was the right thing to do.

My impression—and it might be, you know, it’s not based on anything inside—is that Trump was willing to go along with it, maybe sentimental attachment to Britain, or maybe Keir Starmer is somehow awfully persuasive when he’s one-on-one with the president. I don’t know.

He was willing to go along with it without looking too closely at the detail, perhaps. Surprising, because I think, like you, that he places a premium on control, and it’s pretty surprising to think that a—what really is a campaign of lawfare, or harassment by legal means, and it is abusive international law, I think, rather than international law as it probably is conceived—it’s strange to think you’d be complicit in the surrender to that.

Now, it looks, two weeks ago, as though President Trump had, in a sense, turned his mind back to this and was not liking what he saw. Maybe that’s still the case today—unclear.

But it certainly does seem surprising to prefer Mauritian sovereignty, when Mauritius is clearly going to be a target of Chinese subversion. I think it is a risk to U.K. sovereignty.

So then—this is my last question. It’s still a puzzle to me. It sounds like it’s a bit of a puzzle to you, too, about the U.S. position.

What possible security guarantees could the British have given the Americans? This is what’s being reported in the news—that…

Yeah, I find this very mysterious.

Just, is there anything that they could have said or done that actually would give the U.S. assurance on these questions, or is that—

Well, I mean, we’re just going off the reporting, of course. So, maybe the assurances—I mean, this wouldn’t be giving new assurances, would it? It would just be trying to explain why you think what you’ve already agreed is good enough, which, for the reasons we’ve been through, I don’t find persuasive.

I don’t see how the U.K. is in a position to give any further assurances unless it reopens the treaty with Mauritius and somehow reaches new terms. And if they do that, they could come up with a more watertight deal. But that’s going back to the beginning again, and there’s been no suggestion of that.

So, I don’t think the U.K. is in a position to offer further security assurances. And, you know, the kind of fundamental position—you hand over sovereignty; Mauritius will be sovereign, and it will have capacity to decide what happens in important respects. I mean, it’s not going to forcibly evict the U.S. Navy from Diego Garcia, obviously. But it can make life awfully difficult, and it can do so under a color of legal right, because it will be sovereign.

As it stands, it can try and make life difficult, which is what the British government fears, but it can only do so by abusing international institutions and by directly threatening U.S.–U.K. defense interests.

At that point, if Mauritius starts to threaten those defense interests, you should think, well, maybe you can take some measures in response. So, it’s a very strange sort of chain of reasoning.

All right. Well, we’ll see what happens. That was all very elucidating, Richard. Thank you very much.

You’re very welcome.

Mentioned:

• “Is the Administration Prepared to Deploy ICE to Police This Fall’s Elections?” by Bob Bauer (Executive Functions, Jan. 27, 2026)

Thumbnail: President Donald Trump speaks with members of the media before boarding Marine One on the South Lawn of the White House. (White House Photo.)

This is an edited transcript of an episode of “Executive Functions Chat.” You can listen to the full conversation by following or subscribing to the show on Substack, Apple, Spotify, or wherever you get your podcasts.

Jack Goldsmith: Good morning, Bob.

Bob Bauer: Good morning, Jack.

This morning, we’re going to talk about two more examples of President Trump’s grievances from his first term, continuing to drive his conduct during his second term.

I’m speaking about the $10 billion lawsuit that he brought against the IRS and the Treasury Department for the leak of his income tax returns and information, and also the search warrant that was executed at an election center in Fulton County, Georgia for, quote, all physical ballots from the 2020 general election. I propose that we start with the Georgia matter first. Why don’t you tell us, just summarize what happened there?

As you say, there was a search warrant executed, a large number of boxes, I’ve seen press report reference to 700 boxes removed. This investigation, by the way, I think it’s important to note, was not conducted under the supervision of the U.S. attorney in the Northern District of Georgia. It was conducted by a U.S. attorney appointed to coordinate all civil and criminal cases of this nature, who holds an office, holds his office in the Eastern District of Missouri. I believe his name is Thomas Arbus. So what we have is, in effect, a 2020 election fraud special counsel. He can do this anywhere in the country.

He can supervise this anywhere in the country, and that’s what he’s doing in Georgia. Now, we have to keep in mind, I think this is all well known, but I think I’ll just emphasize it for the record here. The president has been focused on Georgia, among other states, where he believes fraud denied him the election in 2020.

He was confronted there with a claim completely to the contrary by the Republican Secretary of State, Brad Raffensperger. He famously, on a telephone call, asked Raffensperger to find the 11,000 plus votes that he needed to make up the margin against President Biden. And subsequently, the state of Georgia conducted a risk-limiting audit, which included a hand count that confirmed that President Biden had won the state by more than 12,000 votes.

There is no credible expert anywhere who has looked at this issue, much less with reference to the outcome of that hand count audit, who agrees with President Trump that there’s any evidence whatsoever of fraud in elections. He’s focused on Fulton County. That’s where these records were seized.

But I think we need to talk about the fundamental question as we go along of what this means for the use of law enforcement consistent with his claims of authority to overseas elections, the use of law enforcement generally for this purpose, and whether this is only a matter of grievance on his part that he’s continuing to pursue, or whether he’s laying the groundwork or testing the possibility of deploying these tactics in the future for future elections and in particular the midterms.

Okay, before we get to that hugely important question, can I just ask you one more question about the action that the government took? What could they, I mean, what is the point, do you think? I mean, we’re going to get to whether, you know, what the implications are for this use of federal executive power.

But before we get to that, is there any point to this particular raid or request for physical ballots from the 2020 general election? What is the ostensible purpose here?

It could fulfill a number of purposes. One is that it’s scratching his itch. He wants an investigation, he’s launched the investigation, he’s been thwarted at every turn in raising questions about it, and now he has struck out at Fulton County and he’s showing that he means it, that he doesn’t believe the election was honestly held and he wants an investigation.

That’s one thing. Secondly, he’s now raising new questions, presumably within his base in particular, I don’t know where else, about the election. After all, there’s now an active criminal investigation.

So in some sense, he’s taken an action that he can allege proves his point. A judge signed off on this warrant. There was some basis, presumably at law, for the seizure of this material.

So he’s creating facts on the ground, if you will, in support of his claim about a rigged election. And then thirdly, the last purpose is the one I know we’re going to get to, which is road testing the use of this kind of law enforcement authority as a means of addressing the control he’d like to exercise with respect to the midterms.

Okay. Why don’t you explain that? Explain what your concerns are, maybe tie it into the executive order that was issued last year.

Yes. So in the executive order that he issued in March, in which he took any number of actions to assert presidential authority over rulemaking in federal elections, he addressed mail-in voting, machinery, of course, both of which he believes is the basis or are the basis for fraud and rigged elections, strong views on the policies that states ought to be pursuing, for example, in providing their citizens with the option of mail voting.

And at the time he said — and I think it’s important for me to quote from what he said in a Truth Social post that accompanied the executive order. He wrote that the states must deal on these issues, on these fundamental rules issues with the federal government as represented by the president of the United States. The basic quote is they must also deal with the federal government. What the federal government is represented by the POTUS tells them for the good of our country to do.

So he’s taking a position that has no support in the constitutional text. The time, place, and manner rules confer the authority to make rules for the conduct of elections on states and localities, it resides with the states, and the Congress can intervene with respect to federal elections if it chooses to do so. There’s no mention of the president of the United States there.

Except, and I agree with you one hundred percent, but their theory is, well, that may be true, but there’s lots of federal law, civil and criminal law that relates to elections and the president is charged with the duty to faithfully execute the law. So doesn’t the president’s power come in that way, or is that the wrong way to think about it?

No, this is, I think, what is most worrying about what’s going on. Or he has strong views about not just what happened, but how the election system ought to be reformed. That implicates presidential authority, if you will, over rulemaking.

Rules that should be implemented, rules that should be appealed. Now, quite apart from the executive order, and there may be another one in the works, the use of the criminal justice system as a further wedge to seize control of or influence the course of the way election administrators perform their responsibilities around the country is something we need to be concerned about. It’s a claim that there are criminal issues, but the criminal issues are a pretext for inserting the federal government into the electoral process in the fashion that we may see now unfolding in Georgia.

So it has both a backward-looking, his grievance about 2020, and a forward-looking purpose, which is yet another means by which he can attempt to force federal control into the operation of states and the way they administer their elections.

But my question is — that goes to his purposes and his illicit or illegitimate purposes. My question is, isn’t it true that for a benign president, there is federal law enforcement authority related to criminal and civil law related to federal elections?

There absolutely is. And the question is —

That’s what he’s leveraging.

That’s what he’s leveraging, correct. And let me just add to that, there’s also the threat of it, the weaponization of the law for this purpose is potentially also something that he hopes will be a warning to election officials, even if he doesn’t necessarily deploy that force in the first instance.

And the last point on that I want to make real quickly is this, we haven’t talked about this piece of it. There are steps that he can take if he believes that the people have speculated about, if he believes that he needs to intervene in this election with the use of force in some fashion.

There are steps he can take, like the much discussed potential invocation of the Insurrection Act, as an exception to the Posse Comitatus statute, to deploy troops on the ground. And of course, he has already confronted the issues around his federalization of the National Guard. There are significant issues presented for him in both cases, but what about the use of law enforcement, just the use of law enforcement to bring force into these communities?

That’s what’s happened, of course, in Minnesota, and where it was specifically linked up to election security. In a letter that the Attorney General sent to Governor Walz, and we’ve written about this on our site, she linked cooperation with law enforcement in immigration matters with the state’s willingness to turn their voter rolls over for federal government inspection, which is a very hot issue and a point of contention between the states and the federal government at this time.

So let me ask you, can you just give a concrete example of how President Trump in the midterm elections might plausibly, illegitimately use his law enforcement power to influence, interrupt, or shape the outcome of an election? I assume we’re talking about congressional elections.

Yeah, for the moment we are, although, of course, you know, it’s a template that could carry over into presidential elections.

Just for this fall, what is a plausible use of this, plausible thing that he might do?

Well, so to sketch out a worst-case scenario here, or next-to-worst-case scenario, he alleges that fraud is afoot, that the only way, and he said this repeatedly, that Democrats can win elections is by cheating. All right, so right away he’s raised the presumption that there may be criminal legal violations afoot. So, he threatens on a number of fronts that he’s going to stop that fraud and he’s going to hold those who are responsible for it accountable.

That could mean he does invoke the Insurrection Act, there are some legal issues there we can discuss, but long and short of it, you know, he does that. Or alternatively, and I raised this issue in the post that we had up just the other day, he decides that he’s going to deploy ICE and he moves them either by threat, prepositions them or starts to move them into states in a way that, depending on timing, has a significant impact on voter confidence, voter concerns about their safety. It has a voter-suppressive effect.

And let’s assume he launches that only in the jurisdictions where, you know, he thinks it’s going to have an effect on Democratic turnout. Again, I’m sketching out the worst-case scenario. I’m not trying to, you know, put together here some fantastical ...

I want to be realistic about what could happen without at the same time saying that it will happen. But those are things that, looking at what he has said and done to this point, are not easily discounted as possibilities. And the point I want to stress here is the mere threat can have an effect even if he decides not to, you know, bring full force to bear down in a voting jurisdiction.

OK, given that what we’re talking about here are discretionary law enforcement decisions, basically, that as a matter of fact and as the Supreme Court has said in doctrine, the president has exclusive authority over, what, if anything, can be done to prepare for, redress, block, regulate these uses of federal law enforcement? It seems like it’s something that’s hard to stop.

I do think that the states can be prepared to move immediately into court. And for all the discretion that he can claim to exercise, I think the administration has significantly drawn down on its capital, if you will, on its credibility with the course on these issues. And so I think there are issues that you might not imagine a dissenting state could have prevailed upon in some other administration, that it might have a very good chance of getting a serious hearing about here, particularly in relation to the operation of the electoral process.

His positions are so aggressive, the position he took in the executive order is so completely without constitutional foundation. I think all of this suggests the states were in a position to potentially respond by taking emergency legal action. So I think that’s one thing that could make a very, very significant difference here that you might not otherwise think is possible.

So you’re talking about states going to federal court seeking injunctions against the president enforcing federal election law?

Absolutely. Yes, yes. And the other thing I’m going to say, and it’s significant in many, I think this needs to happen earlier rather than later.

It is hard to underestimate the degree to which civil society pressures can make an enormous difference. I mean, I think we have seen the effects of staunch resistance within all the relevant civil society sectors in the state of Minnesota and the effect that it’s had on public opinion generally. And I think drawing a line now and soon within these sectors.

How do you draw the lines? I mean, I don’t see how a federal court can issue an injunction or anyone has standing to seek the injunction until the president takes action or threatens in a way that’s very close to taking action.

I’m not well, I’m not referring to a legal action by civil society. I’m talking about pressure from the business community, faith communities. Lines are drawn into the sand.

You saw, for example, I would have liked to have seen it worded a little bit with a little bit more muscularity of the businesses in Minnesota, major businesses in Minnesota and a good number of them that signed a letter basically speaking to the escalation of this ice conflict in in Minnesota. I think that kind of pressure with the effect on public opinion that will have could make a significant difference in how the administration calculates his chances of success here and the decisions that it makes about what it’s prepared to do. And last but not least, let me say this election officials around the country, Democrats and Republicans really resent what he’s doing.

I mean, they resent the attack on professional election administration, the suggestion that there were parties to fraud in 2020, that they don’t know how to do their jobs. They can’t administer mail in voting. And then there’s a widespread frustration with that.

And they’re going to be speaking to their members of Congress and they’re going to be speaking to their communities. And I think that intersects with that larger civil society response that I think could make a difference here.

OK, let’s talk briefly about the lawsuit that President Trump brought, I think, yesterday, 10 billion dollars against the IRS and the Treasury Department in connection with the leak of his tax return information, his tax returns and tax return information during his first term. This is not a frivolous lawsuit. There are federal laws that allow someone whose tax returns are unlawfully disclosed to sue the federal government in federal court for damages.

And he’s got a non-frivolous Privacy Act claim as well. There are problems with it. I’m not saying he’s going to succeed.

There are statute of limitations issues. There’s a question whether the person who leaked the tax information was, in fact, an employee of the IRS, which is required under the statute. He was actually a contractor for Booz Allen.

So it’s not clear whether that counts as being an employee or officer of the United States as statute requires. And then there’s some problems with the Privacy Act claim as well. But setting those aside, I mean, it’s not a frivolous lawsuit on the merits, but it’s a strange one because of President Trump’s conception of his control over the executive branch, because he’s suing the federal government.

The Justice Department defends these lawsuits. They defend them very vigorously. Very analogous lawsuits brought in the past, even in connection with the same leak of information, have been defended robustly by the Justice Department.

Will President Trump allow his Justice Department to defend the lawsuit that he brought? Will he allow the Secretary of the Treasury, who is also the commissioner, the head of the IRS right now, will he allow him to be on the other side of the lawsuit against him? Will he demand that Pam Bondi settle with him for billions of dollars?

These are things that other presidents wouldn’t try. So this type of concern, for example, I don’t think arose to nearly the same degree, if at all, when Hunter Biden sued the IRS and the Justice Department defended. But they clearly arise here.

What thoughts do you have about that?

Well, I entirely agree that the questions that you raise about what it means for him to exercise, as he’s made it very clear he intends to do, full control of the executive branch, you know, those claims bear directly on the integrity of this legal action. Who really believes that the Department of Justice is in a position, even if not directly, in order to do so, to put up any kind of defense on the merits or to address this case in any way on the merits? He feels very strongly about this.

He’s made it very clear he believes he’s suffered not just speculative, but actual damages in the billions of dollars. He wants punitive damages, as I read the complaints, I recall the complaint as well. So what kind of integrity or guardrail is built into the operation of his version of the unitary executive in these circumstances?

Just one other quick point. Let’s assume, and this goes to how he conceives of the exercise of presidential power and the very personalized view he has of the president, that he believes he was wronged and that the Trump organization and his sons were wronged by the leak of his tax information and that he has some basis to believe that. And of course, the individual in question, who was, I think, a full-time contractor, has been sentenced for and held accountable under the criminal laws for this conduct.

The normal approach to all of the failures of the IRS security and screening procedures, the normal response would be to direct the IRS commissioner to tighten them up, to tighten them up so all taxpayers are protected. And he may be a particular example that he’s concerned with. But all taxpayers are protected by an improvement in these procedures.

And instead, he wants $10 billion because he’s singling out, in this instance, the wrong done to him. Instead of proceeding as president, it’s a matter of policy and legal authority. He’s proceeding through a private lawsuit to collect billions of dollars for the wrong done to him.

And I think that conception of the role that he has as president and sort of how he would respond in a circumstance like this of policy and administrative failure is extraordinarily telling. The only other thing I would say real quickly, the other thing I think about this, and this may be more of a political point than anything else, but it is quite remarkable. The lawsuit alleges that harm has been done not just to his reputation, but to his business, to his financial prospects, to the success of his business operations.

This against the background of the amount of money that Donald Trump and his family made in the presidency really strikes you. And it’s just remarkable that there was no second thought about making that kind of claim in that context.

So I’m not sure I agree with all of that. I don’t. I mean, it’s a very, it’s completely, it’s an unusual, it’s an unprecedented situation.

He does have plausible federal law claims against the government under both statutes that he invokes. They’re not frivolous claims. I’m not saying he’s going to win, but they’re not frivolous claims — unlike a lot of other lawsuits that he’s brought, for example, against the press, which I do believe to be frivolous. This is not a frivolous claim at all. You know, there already have been steps taken, as the IRS has argued in these lawsuits that have been brought out of these leaks to make sure they don’t happen in the future.

I’m not sure there’s much more for Trump to do. So for me, it’s, you know, I think it’s, I agree, extremely unusual. And it’s almost bizarre for the president to be suing the government for damages.

But he has a legal basis to do so. For me, the interesting question is, what is the Justice Department going to do in response? Are they even allowed to show up and defend?

What is the IRS going to do in response? Are they allowed to show up and defend? Are they allowed to make all the arguments they would normally make?

If they don’t make those arguments and Trump wins, what does that mean? What does that look like? Those are the types of questions that I think this case raises, as opposed to the legitimacy of him suing the government.

He has a legal claim, a plausible legal claim, it seems to me.

Well, let me let me just respond quickly to that, because it’s an interesting point of, I think, apparent disagreement between the two of us. When someone’s elected president of the United States, they surrender all sorts of private rights and privileges, if you will. They just agree to do that.

There’s certain things they cannot do. They cannot, for example, shake their security detail and go take a walk along the Potomac so that they can sort of reflect on things. There’s certain things they’re not expected to do, not and should not do, because they have a constitutional role, a constitutional role that puts them in a position of attending to the public interest at some considerable sacrifice, presumably, to private rights that they would otherwise want to exercise.

That’s what we want in a president. We don’t want a president, for example, speaking of the first topic we just discussed, to decide to use the criminal laws to strike back at his political opponents. That’s also a private grievance.

He ran for president as a candidate. He didn’t run for president. You know, he was an incumbent president in 2020, but he ran as a candidate.

And in his personal capacity, using the government to settle scores with political opposition is something that maybe as a candidate who lost, he might find various ways to do various claims to make. But we don’t expect him to do that as president with the full power of the U.S. government behind him. And I think the same thinking applies to his use of government power to settle scores with the government over the release of this tax information.

I accept all of that, except one thing, and I would not—I don’t think that it’s prudent for a president to do this. But when you say we don’t expect this to happen, I’m not sure who the we is. I mean, he ran on these grievances and he has a legitimate legal claim here.

I’m not saying it’s a good idea for him to sue or that it’s moral for him to sue. I’m just saying here, unlike in many of the lawsuits he’s brought, it seems to me he has a colorable legal claim and, you know, he’s exercising his legal rights. I agree with you that presidents have to give up all sorts of things.

This is completely one off. President has to give up a claim to a billion dollars in damages or ten billion dollars in damages if it’s that much. But I don’t know about that.

That is a judgment call about, you know, the norms of a president who has a plausible legal claim, who is motivated by grievance. And I don’t love this lawsuit. I just think it stands on a much different footing from a lot of the other lawsuits he’s brought in terms of its legal validity.

And it raises all sorts of completely very difficult questions about how the executive branch is going to function here.

How do you see it in relation, for example, to the lawsuits he’s filed against media organizations?

I think those are frivolous. Totally frivolous.

Well, ome of them resulted in multimillion dollars.

Totally frivolous and abuses of the government. They’re totally frivolous and abuses of government power. I don’t think — Let me put it this way — And I think you may agree with this. If he were a private citizen bringing this lawsuit, there would be no doubt about it that it would be fine to bring. And but I don’t I don’t think that the media suits, I think those are frivolous lawsuits.

This is not a frivolous lawsuit on the merits, is what I’m saying.

Well, just to just to add this, I mean, this is an interesting question, again, of what we what we expect presidents to give up so that we’re confident they’re attending not to their personal —

Bob, you keep saying we. Who is we?

Well, that’s a fair question. I believe that if you were to put the question and maybe I’m wrong, if it were polled, do you think that the president’s paramount concern as president should be the public interest and should be in those circumstances prepared to defer the pursuit of private interests? And by the way, this applies, of course, to his business activities.

I think the answer to the general question would be overwhelmingly, yes, that’s what we expect when we elect the president of the United States. So I think the we you’re right, I shouldn’t say we because there may be some dissent there, but I think there would be a general public view that his overriding responsibility here, even if it means some sacrifice of private rights, is to the public as a whole and to the administration of his office in the interest of everybody and not in the interest of his family or the Trump organization or himself and the billions of dollars like in this lawsuit.

I agree with that. I completely agree with that. I think it’s much more complicated here because this was an egregious violation of federal law, unprecedented in its scale and context in the run up to a campaign.

And I think, you know, again, I don’t like you know, I don’t like the tit for tat. I don’t like the president making money off the presidency. I would not have advised him to bring this lawsuit.

I just think it’s it’s different from the other ones. And it’s because it has a valid legal claim or not a valid legal claim, a plausible legal claim at the basis of it. I’m not sure we’re disagreeing.

I mean, I don’t disagree that this is not something I would want the president doing. I think there are a lot of people that are going to cheer it on. A lot of certainly a lot of his supporters will and even people who I think won’t necessarily be supportive of his lawsuit strategy in other contexts, like the media might think he was genuinely wronged here in violation of federal law in a way that’s entirely unprecedented and terrible for the country to have the IRS leaking presidential candidates’ and sitting presidents’ tax returns is really bad thing.

It wasn’t an action by the IRS as an institution. It was a failure on their part to stop this rogue contractor from accessing their —

I know, but there are rules about the way that Congress set up the system to have when this happened, it was so egregious that they contemplated that the individual whose tax returns were disclosed could sue the federal government for damages, which is an unusual thing. And there’s a separate claim for the fact that the IRS did not for the under the Privacy Act, for the notion that it didn’t take precautions, adequate precautions to prevent this type of thing, which they have admitted since then that their precautions and their systems were inadequate. So I just don’t think you can that one can discount what this is a response to.

I mean, I agree with you that this is you know, we’re out in Trump lalaland where he does things other presidents don’t do and they go beyond the traditional expectations of the office. This one just seems a little different to me.

Well, and so just and so where I hear it, I think what many probably a majority hope is that a president steps into that situation and says, look, I was wronged and maybe the Trump organization sues and he doesn’t, you know, maybe he says to his son, to the Trump organization, you bring the lawsuit. I can’t do it. I’m president of the United States.

Yeah. It’s the wrong thing for me to do.

Speaker 2

Yeah, I agree. I mean, this is this is not what I want my president doing. I agree with you.

I agree that it’s not what my president doing. I’m just saying that he was really and the Trump family and organization were seriously, seriously wronged here. It’s a serious violation of public trust and a serious violation of the law, whether they can recover or not.

And and, you know, it was it was a huge deal. So I just think the the fact that this was a huge deal and a serious breach of public trust, you know, way up there in terms of seriousness and the fact that there’s a plausible, though not airtight legal claim, makes this one different to me. I agree with you much better if just the Trump organization or his children sue.

Certainly, you know, again, President Biden’s son sued and that case was dismissed. And Biden Justice Department acted at arm’s length from the White House that, you know, that’s not going to happen here. But that would have been much better.

I agree. And that’s what a that’s what a normal president would have done in this situation. A normal president at most would have stayed out of it, would have shut himself off from the Justice Department and the IRS.

You know, if the Trump organization or the children want to bring the lawsuit, that’s their business. But there would be a firewall.

Correct. Exactly.

It would have been a much better way to do it. I completely agree.

And then the people who were responsible for the IRS for, you know, failing to enforce these safeguards are fired. There’s a huge reform project that’s put together. I mean, there’s so many things a president could do.

Well, I agree. OK, I’ll give you the last word on that — Thank you.

Thank you.

Mentioned:

• Saikrishna Bangalore Prakash, The Presidential Pardon: The Short Clause with a Long, Troubled History (2026)

• “Trump’s Circumvention of the Justice Department Clemency Process” by Jack Goldsmith and Matt Gluck (Lawfare, Dec. 29, 2020)

Thumbnail: President Gerald R. Ford Signing a Proclamation Granting Pardon to Former President Richard Nixon in the Oval Office. (Public Domain.)

This is an edited transcript of an episode of “Executive Functions Chat.” You can listen to the full conversation by following or subscribing to the show on Substack, Apple, Spotify, or wherever you get your podcasts.

Jack Goldsmith: I’m chatting today with Professor Sai Prakash of the University of Virginia Law School about his new book called The Presidential Pardon, The Short Clause with a Long Troubled History. I read the book in galleys and blurbed it as the best book written on the pardon power, which I think it is. And so, Sai, I’m thrilled to talk to you about your book today.

Sai Prakash: Well, I’m thrilled to be here with you today, Jack.

So, the pardon clause in the Constitution in Article Two has become hugely consequential in modern times. Let’s just start off with what it says and unpack that, and then we’ll go from there. The pardon clause says that the president “shall have power to grant reprieves and pardons for offenses against the United States, except in cases of impeachment.”

As you say in the title of your book, a short clause, but it’s hugely significant. Can you just kind of unpack the clause and then we’ll go from there?

Sure, sure. So, the pardon and reprieves were preexisting concepts. They existed in England, and the idea is they want the president to be able to prevent the imposition of a punishment, that’s a reprieve, and usually temporary, and then also be able to fully forgive the offense against the United States. And so, that’s the pardon.

There are other kinds of clemency that are thought to be included, even though they’re not separately listed, like a commutation or an amnesty. So, even though they’re not listed, they’re just sort of thought to be there. And then the further thought is it’s got to be against the United States, and therefore, it doesn’t cover state offenses of various sorts.

So, can you just explain briefly what a commutation is and what a clemency is and how that relates to pardons and why it’s thought to be included in the clause?

Yeah. So, I mean, clemency is just an umbrella term for any kind of mercy. And there’s a sense in which the clause is about clemency as a whole, precisely because it goes beyond mere pardons and reprieve.

So, reprieve, we talked about, it’s just a respite in the punishment. It could be, you know, we’re going to temporarily suspend your prison time to let you free for a couple of weeks or months, or it could be we’re not going to execute you right away. We want to think about whether we should execute you a little more carefully and thoroughly, and maybe we’ll eventually pardon you.

So, that’s reprieve. And the other stuff, you know, is a commutation is basically saying we’re not going to say that you didn’t commit the offense. We’re not going to act as if you didn’t do anything wrong.

We’re going to shorten your sentence. You know, you had a 10-year prison term. We’re going to make it five years.

You had a three-year prison term. We’re going to make it time served. And so, the idea is it doesn’t erase the fact that you were found guilty of something or that you’ve done something wrong.

It just means that we think you should suffer less for it. And so, typically, it just means that we’re going to let you out early, earlier than you otherwise would be let out. And of course, I guess it could be issued before you actually serve time.

Theoretically, someone could be convicted, you know, they could be in the courtroom, and the punishment could be five years. And then, as they’re being taken to the jailhouse, the president might commute that to three years. That’s certainly a possibility.

But typically, a commutation is issued to someone who’s already in jail. An amnesty is a situation where the president decides that many people at once who are found guilty of particular crimes should get some form of clemency, either a commutation or a full pardon. And they’re often issued in conjunction with rebellions.

The idea is, if you lay your arms down, we promise you that we won’t prosecute you. And the way we memorialize that is through a pardon that says, if you lay down your arms, you get the pardon. Which brings us to another type of pardon, which is a conditional pardon.

And the conditional pardon says, you only get this if you do X first, or you only get this if you continue not to do X going forward. And so, it’s like a contract, right? You have to do something or refrain from doing something in order to get the pardon.

And the pardon no longer is effective if you violate the condition, right? And so, that’s why even though it doesn’t talk about commutations or conditional pardons or conditional amnesties, it’s thought to have all those components in part because talking about a pardon power would have been understood to include all those subsidiary authorities, even though not expressly mentioned.

So, that’s an extraordinary array of ways that a president has discretion to alleviate the consequences of commission or possible commission of a federal crime. And just to underscore what you said earlier, the pardon power is limited to, it only applies to federal crimes, not state crimes, right?

That’s right.

And just briefly, before we move on, what is the meaning of the exception for the pardon power does not apply, it says except in cases of impeachment. What is the purpose of that?

So, I think it’s twofold. They wanted to make it clear that the president couldn’t stop an impeachment by pardoning the person, but they also wanted to make it clear that the president couldn’t pardon a conviction by the Senate. So, it turns out in England that the crown for a time could pardon people after they were convicted by the House of Lords and thereby prevent whatever punishment that the House of Lords had attached to the conviction.

And they want to make it clear, you know, impeachment is really left to the House and conviction is left to the Senate and the punishment as limited as by the constitution is left to the Senate. And the president can’t just say, well, you know, I think you should continue to serve notwithstanding the conviction by the Senate, or I think I’m going to lift the bar on future office holding that the Senate can apply to convictions, impeachment convictions. He can’t do any of that, right?

The Senate gets to decide what the punishment is without any presidential interference and the House gets to conduct its impeachment trial without the possibility that the president will cut that short.

Okay. So, this is an extraordinary grant of discretion to the president. I mean, it’s an astoundingly broad grant of discretion.

In the same article of the constitution that says that the president has a duty to take care that the law be faithfully executed. So, and the framers were somewhat skeptical, at least many of them were about this new entity, the presidency they were creating and about whether this power might be too broad or get out of control. So, why?

Why include the pardon power in Article II?

So, I think it reflects several features of that age, several strands of thought of that age. I think criminal law was quite harsh in the 18th century. And there was a thought that someone somewhere in the government had to have authority to mitigate the harshness of the criminal law.

Many crimes were subject to capital punishment, far more than today. And there was a thought that we needed to have someone who could dispense mercy, maybe better fit the punishment of — the circumstances of the crime. Because laws are written in general terms, but you might think more factors are relevant.

Maybe we don’t want the courts to wade through a seven factor test, but maybe we think the president or someone else who can issue the pardon can or should do that to possibly mitigate the sentence. They also thought that it was important to have a pardon power in part because nations suffered from rebellions from time to time. And they thought a well-timed offer of pardon could nip the rebellion in the bud, basically lead people to lay down their arms.

So, the president would say, I’m going to pardon you if you lay down your arms. The people lay down their arms and then the pardon vests. And they will do that.

They will only lay down their arms if they’re certain that they’re actually going to be pardoned. If they’re fearful that the president will renege, well, then they’re not going to lay down their arms. It’s an all out struggle because if anything less than that is, their heads are on the chopping block, so to speak.

So, I think they think that’s going to be a feature of our politics. And there are sort of rebellions of various sorts in the early years. And of course, we had a very big rebellion in the 19th century.

But their experience was there was the rebellion up in Massachusetts, Shay’s Rebellion, and they thought this might be a regular feature of politics. And so, it was necessary to have it. Other reasons for a pardon power would be to create some measure of allegiance to the state.

The state is both punishing you, but also showing some forgiveness. And people are often quite grateful for their forgiveness. And this is certainly an aspect of English policy.

And so, you document in one of your early chapters, despite those reasons, why the Anti-Federalists in particular were very worried about this provision. To me, like so many other elements of the Federalist, Anti-Federalist debate, the Anti-Federalists were more prescient in understanding how the Constitution was going to work. Am I right about that?

Tell us what the critiques and worries were about the pardon power.

So, the pardon power looks very much like the British Crown’s pardon power in the 18th century, but it looks rather dissimilar with the pardon powers that existed in the states. So, at the state level, the governors or the plural executive councils don’t have this sort of broad authority. What they instead have is some executives have the power to reprieve, with the legislature having the power to pardon.

Other state executives have the power to pardon, but not in cases of treason or murder. There, they have a power to reprieve, and presumably, the state legislature can pardon. And then, of course, sometimes what you have is a plural executive that’s wielding the pardon, which is itself a built-in check on its exercise, because there’s going to be some deliberation and thought to whether to issue the pardon of the sort that you can’t necessarily guarantee if one person is the one wielding the pardon pen.

So, they see an array of pardon wielders with lots of checks on the pardon. And when they look at the U.S. Constitution, the proposed U.S. Constitution, they don’t see any checking function there. They don’t see any requirement of getting the Senate’s consent.

They don’t see any offenses that are accepted from the pardon power, like, say, treason. And then, they fear that the president will sometimes act contrary to the Constitution, perhaps try to overthrow the Constitution, install himself as a monarch. And they think, oh, it’s weird that the president can basically pardon the people that might help him try to overthrow the government.

And they say, this will prevent us from finding out that the president’s involved in the conspiracy, and it will prevent us from understanding the full extent of the conspiracy. And this is what the anti-federalists say, but it’s hardly some fantasy. I mean, I think Hamilton himself, in the Constitutional Convention, proposed a pardon power that didn’t cover treason.

He doesn’t tell you that, you know, as Publius, because he’s not revealing himself to be Hamilton. And, of course, the convention records are sealed at that point, and no one knows, you know, what people proposed. But the point I’m trying to make is that even strong advocates of executive power had some trepidations about giving the president as broad a power as was ultimately found in the Constitution.

And so then why, I have to say, I’m a little surprised that it prevailed. Why did it stay in there? What was the clinching argument?

Well, I think there’s the clinching argument with respect to not having a treason exception is, well, I mean, are you not going to be able to pardon rebels, right, during rebellion? Then they won’t lay down their arms. You need to have that authority.

And we can’t rely upon Congress, because, you know, today Congress kind of meets all year round, but Congress is not always in session. And if they’re not in session, are you going to wait, you know, are the rebels going to lay down their arms for six months while they’re waiting to figure out what Congress is going to do? But I actually think the bigger reason is everybody is looking at George Washington and thinking he’s going to be the first president, and they think this is a man of virtue and restraint, and he won’t abuse it.

And there are several people who say after the Constitution was drafted, this presidency would look rather different if people didn’t imagine Washington serving as the first president. And they remarked that the presidency was created for a man. And then they lamented this fact, they tended to be anti-federalists, they lamented this fact, and they said, you know, the presidency is far stronger than it should have been, in part because it’s George Washington who’s going to be the first president.

And they say, of course, not every president is going to be a Washington, and so we should have perhaps, you know, imposed more constraints.

Yeah, let me just ask you about that, because that argument about Washington being the first president was made, as I recall, with respect to lots of different elements of Article II, and everyone was sure Washington was going to be the first president, was sure he would exercise power responsibly, but as you just said, it was obvious that there were going to be presidents after Washington. So was the idea that Washington would set the precedents and kind of fill in the presidency and the implicit constraints on the presidency, and that would be the model going forward, or what was it?

Well, you know, I should mention a third thing. There are lots of able advocates of executive power in the convention, James Wilson, James Madison, to a certain extent, Gouverneur Morris. They have an experience with weak executives at the state level and at the national level.

At the state level, the executives are described as being ciphers and not serving as a counterweight to the state legislatures, and they don’t want to repeat that. And at the national level, you know, it’s a plural executive that’s divided, distracted, and lacks competence and expertise because delegates, you know, it’s the Continental Congress that’s the executive, and people come and go in the Continental Congress and don’t stay very long and don’t develop expertise and aren’t always around to superintend, you know, the officers who do the business on the ground.

So there’s a concerted effort to have a stronger executive. The losers in that battle, I think, are still right to say that the executive was far stronger than it would have been absent Washington. I think there’s just a general tendency, Jack, to focus on the present and what were the problems we’re trying to avoid and, you know, think in the short term, even for something as important as a constitution.

And, you know, it’s speculation on their part. I think they’re right to think that this would have affected the outcome. You know, I tell my class, if we wrote a constitution after Richard Nixon resigned, it would look rather different than Article II, even though you might say Richard Nixon, you know, is an aberration and most of our presidents weren’t like him.

I still think, you know, the presidency would look much weaker if the constitution were rewritten after Richard Nixon’s departure from office.

Fair point. Okay. The meat of the book, the middle of the book, is a really wonderful, colorful history of the uses and abuses of the pardon power.

I don’t want to march through every one because I want to get to current controversies, but can you just give us some of the highlights, say between Washington and up through the Trump era, some of the highlights of the abuses and maybe especially how they started to increase starting in the late 20th century? Sure.

Sure. Yeah. Well, I have a chapter on Washington’s wielding of the pardon pen because I just wanted to see what he did and think about, you know, the precedents he established.

And Washington was very careful and very thorough. He would seek out the views of the prosecuting attorney and the judge about whether the person merited clemency. And sometimes they would say yes, right?

Because they thought the law was X, but maybe this person shouldn’t have been punished. That actually happened from time to time. And then another thing that was interesting is Washington would always list multiple reasons in the pardon itself so that he could justify it to the world that he was pardoning this person because pardons, you know, even though they benefit one or two people, they are actually documents written to the world.

They’re they’re written as if, you know, they’re public documents to be read by by people like you or me. So I was just interested to see what he did and to see if there were any controversies. Some people said there were, but I don’t think so.

I don’t think there were any major controversies during the Washington administration. That changes with John Adams. There’s a rebellion called the Fries’s Rebellion in Pennsylvania four or five years after the Whiskey Rebellion during the Washington administration.

And they’re opposed to a house tax, a windows tax, a tax on windows, a direct tax. And, you know, there’s a lot of praising of the French Revolution and praising of guillotines. And several people are tried for treason and they’re convicted.

And John Adams pardons them all. His reason, he says, is that they shouldn’t have been they weren’t guilty of treason. They were just guilty of riot.

And so they were overcharged. But he doesn’t commute their convictions. He actually pardons them.

And federalists are up in arms about this. Now, federalists are, you know, they’re part of the same party as Adams, but there there’s a sort of an arch-federalist faction within the federalist party, the high federalists. And they don’t like Adams.

They think he’s soft on France and they think he’s, you know, soft in various other ways. And they claim that Adams did this for political reasons. They think that Adams did this in order to secure more electoral votes in Pennsylvania and to possibly run with Jefferson on some sort of unity ticket.

Obviously, the second thing never happens. But Adams does get more votes, more electoral votes in Pennsylvania in the 1800 election than he does in the previous election. And, you know, people do speculate that that’s why he did this.

Certainly he would have known that many Pennsylvanians wanted these folks to be pardoned in part because they know these people and they’re related to them. And, you know, maybe it couldn’t have been a bad thing that this would make him more popular in Pennsylvania. Whether that was his motivation or not, it’s really hard to fathom.

It’s always going to be hard to figure out whether the president is doing so for political reasons. The next controversy is Jefferson and his pardon of some allies. Abigail Adams excoriates Jefferson for pardoning people who had libeled or slandered John Adams.

And it’s, you know, it’s all kind of a tawdry affair. During the Civil War, post-Civil War, Johnson wants to run as the Democratic nominee for president. He’s the accidental president because Lincoln is assassinated.

And he’s told by several people, if you issue an amnesty while the Democrats are meeting in New York, it will help your candidacy. And he issues an amnesty on the first day of the convention. And he’s told this is very helpful to him.

He doesn’t end up getting the nomination, but people speculate that he issued the amnesty precisely to influence them. He was unsuccessful, but there was the speculation. And then we will do, you know, what you said, we’ll flash forward to more controversial stuff at the tail end of the 20th century.

We’ve got the Iran-Contra pardons where Lawrence Walsh is still investigating the Iran-Contra affair. George H. W. Bush was the vice president of Ronald Reagan during that controversy. And George H. W. Bush issues pardons of people that the, you know, Iran-Contra prosecutor Lawrence Walsh is continuing to investigate.

And Lawrence Walsh says this is the continuation, the completion of the coverup, because we won’t know everything that we need to know about Iran-Contra. So it’s very controversial at the time, certainly in those quarters who thought that President Bush would be implicated by the special prosecutor. The next kind of major controversy, as you know, is the Clinton pardons.

There’s the Mark Rich pardon, but there’s also pardons of members of the FALN, a terrorist group related to Puerto Rico, and some pardons of Hasidim Jews in New York. The latter two pardons are thought to help Hillary Clinton in her Senate campaign. And there’s speculation that, you know, she’s going to do better with the Puerto Rican vote and the Hasidim vote by virtue of these pardons.

And it’s probably true that she benefited from them in some way. And then the Mark Rich pardon was, you know, a situation where his former wife donates, I think, to the presidential library for Bill Clinton, and then she asked for a pardon for her husband, and her husband gets one. It doesn’t go through the ordinary process in the Justice Department.

They don’t, you know, approve of this pardon. That’s normally how pardon applications get to the president, at least in years past. And so there’s a firestorm about it, and both chambers of Congress investigate, and there’s a U.S. attorney investigation up in New York about it, but nothing comes of it. I mean, I guess I should mention the Nixon pardon by Ford, right? That’s pretty important, and that was quite controversial, and there was a claim that there was a corrupt bargain. That’s not quite late 20th century, but it does tar Gerald Ford for a long time.

So if we flash forward to the Biden and Trump administrations, you know, Joe Biden issues a bunch of pardons at the end of his term. He pardons his son after insisting for years that he won’t do that. He pardons members of his family because he fears that they will be unjustly prosecuted by Trump.

It turns out that his brother’s actually being investigated by his own Justice Department, and that’s cut short. Those pardons are controversial because they extend for 10 years. They cover a 10-year period, and they don’t list the offenses.

The Hunter Biden pardon covers all offenses, criminal or non-criminal. The pardon of his brothers and, you know, sisters and in-laws is only non-violent offenses, and so it’s kind of weird that the pardon of the son includes violent offenses, but this one doesn’t. And then, of course, there’s these, you know, what I call sort of, you know, pardoning people that Trump might prosecute category of pardons, including the January 6th committee.

I think Mark Milley, and the former head of the CDC, what’s his name? Famous scientist.

Dr. Fauci, Dr. Fauci.

Yes, Dr. Fauci. There’s all these people that are pardoned at the end who Biden says will be targeted by Trump, and there’s obviously some truth to that possibility. We’ve seen targeting going on now, but it’s controversial nonetheless because it suggests, you know, some people look at that and say, you’re just pardoning people that are your ideological allies who went after the president in some way, and it raises the possibility that Trump might do the same.

For Trump, you know, the most controversial pardon was the January 20th pardon of the January 6th-ers. He had campaigned on this, but he had suggested he wasn’t going to pardon the violent offenders, but he comes into office and doesn’t want to figure out, I think, who’s violent and who’s not, and he pardons almost all of them. A few of them get commutations, like 20 main people get commutations, but everybody else gets a full pardon, and it’s quite controversial.

Republicans actually criticize the president. Democrats obviously do so as well. As your listeners or your viewers are well aware, Republicans are very loath to criticize this president, so it’s unusual that they’re criticizing him.

He had campaigned on this, and so it wasn’t a secret that he was going to do something. The breadth of it was a bit of a surprise, but it portends the possibility of presidents running on campaign promises to pardon one or more groups and then hoping to secure their votes, their votes and the votes of friends and family. Obviously, if you knew someone who had stormed the Capitol on January 6th, you would feel great gratitude to someone who pardoned your loved one.

Both Biden and Trump made promises of pardons when they were running for office. Biden’s related to marijuana. He was clearly trying to get the youth vote.

Trump’s related to January 6th. Maybe he thought he would get votes, but it was also partly a view that the January Sixers were political prisoners of some sort, which would have played to his base and perhaps got him votes as well. We can imagine presidents running on promises to pardon tax offenders and environmental polluters or whoever’s part of their coalition.

I think we’re going to see more of this. Since the January 20th pardon of the January Sixers, Trump’s pardoned contributors to his campaign, people who owe potentially hundreds of millions of dollars in fines. That taboo about pardoning someone who contributed to your campaign no longer exists, at least during this presidency.

One has to wonder whether that taboo has just been utterly broken.

More generally, both in his first term and this term, Trump has made it practice of giving pardons to friends or friends of friends or people that he thinks are ideologically consonant with him. It’s not just seeming quid pro quos in some instances. It’s just kind of ideological pardons, it seems, and hundreds of them in the first term and now in the second term as well.

You have a line, and you just alluded to this, you have a line in the book where you talk about where we are with the pardon power today that’s pretty arresting. I’m going to read it. It says, as they wield the pen, presidents advance their ideological and personal agendas, shore up their electoral bases, inflict psychic wounds on their opponents, and wreak real injuries upon the rule of law.

That’s kind of where we are today with the pardon power. I expect Donald Trump to issue many, many hundreds, if not thousands of pardons at the end of his term, kind of picking up on what Biden did to preclude investigations by the next administration into his administration. Not to mention who knows how many hundreds or thousands of pardons he’ll issue in the interim, because basically, as you’ve described, the president can just wield the pen and it happens.

It’s the perfect power for Trump, because there’s no intermediary. He can just bark the order and it happens. Just talk about that.

Is that where we are today? Is this now a feature of the American constitutional landscape?

It’s certainly going to be a feature of this administration. I think it’s going to be a feature of subsequent administrations as well. It’s just far too tempting to be able to hand out goodies to your allies and friends and potential voters.

Let’s focus for a moment on Trump. I think you’re right that the president likes to be able to exercise a power that Congress can’t constrain and that the courts are unlikely to police. The courts may interpret a pardon more narrowly than the president wants, but the solution to that is just to issue another pardon.

There have been cases out of the January 6th pardons where the administration wanted to read the pardon broadly, the courts refused to do so, and then the president just issued a broader pardon. The only way that the courts can constrain the executive is if the executive doesn’t like the breadth of a pardon and the executive is trying to read it narrowly and the courts say, you’re wrong. That’s the only way in which the president can lose.

If the president wants a broader pardon to someone, the president can unilaterally do that. I think this is a very tempting tool for the president. He can see the advantages of it.

You alluded to the ideological basis for some of these pardons. The president apparently loves to hear that the prosecutors who prosecuted someone were driven by politics or driven by personal agendas of various sorts because everyone, it seems, who’s applying for a pardon is saying, I was railroaded by a corrupt prosecutor, basically mirroring the language the president uses. They just say this, I assume, without regard to whether the person was corrupt and whether they were railroaded or not.

I highly doubt that the system is working so poorly that people are being routinely prosecuted and found guilty on thin or flimsy evidence. They’re playing to the idea that the president feels that he was a victim of a process, and they’re just using the same language. That’s ideological.

It’s biographical, but it’s certainly ideological. It says there’s something wrong with the system. It’s interesting.

He’s pardoned people that aren’t Republicans, either as a means of trying to lure them over or some sort of sympathy with them. I think he was mad at Eric Adams for not running as a Republican. There was no condition on the pardon that Eric Adams become a Republican.

The same thing with the Texas House member that he pardoned, and then he was angry when he decided he was going to run again as a Democrat in the House election.

Right. President Trump was a businessman. He thinks in transactional terms.

Everything he thinks about is in transactional terms, really, when you think about it. He thinks there’s a transaction here, but of course, there’s nothing in the pardon suggesting there’s any condition on it that you have to run as a or become a Republican. They don’t view it that way.

They may have sought a pardon, but they didn’t agree to any conditions. He’s going to often find that his expectations are going to be thwarted if he thinks that he’s going to buy loyalty just by issuing pardons.

It seems to me, and you suggest this in the book, that we’re on the verge of the pardon power getting out of control. Whatever norm-based constraints were on it have been dissipating starting in the late 20th century, certainly accelerating under Trump one and then Biden and now under Trump two. What is to be done?

Do you think that reform of the pardon power is appropriate? If so, how should we think about reform?

Jack, you wrote a paper with, I think, a student or a lawyer about how Trump had bypassed the ordinary process for issuing a pardon. Ordinarily, you apply for a pardon with the Department of Justice. They vet it for years.

It goes through one vetting process in the office of pardon attorney, and then it goes to some higher level, I think it’s an assistant or deputy attorney general, who then makes a recommendation to the White House, who then conducts its own kind of review. And what happens certainly during the Trump administration, but the Biden administration as well, is that more people just apply directly with the White House, because why would you go through this long process if you have connections inside the White House? And how do you get connections inside the White House?

Either you know someone in the White House, or you pay someone who knows someone in the White House. And so that’s introduced more cronyism into the process and less vetting into the process. I think you’re absolutely right that we’ve entered into a new era.

I don’t think this will change with, say, a Democratic president. It’ll be too tempting for them to issue pardons to members of their coalition. Biden ran on a marijuana pardon, and he issued the pardon mere months before the midterm.

That is to say, he waited until mere months before the midterm to issue the pardon for some inexplicable reason. I mean, it’s not inexplicable. That’s sarcasm.

He was trying to get votes. He issued the marijuana pardon on the same day he issued the student debt relief. Both were designed to gin up the student vote, the youth vote for Biden, for the Democrats in the midterms.

Whether that worked or not is hard to say. But I just think it’s going to be too tempting. But you could have internal reform where the president commits to not pardoning someone unless they’re recommended by the Justice Department.

That’s what Jimmy Carter said he did in the wake of the Mark Rich pardon. He said, I never pardoned anybody that the DOJ hadn’t approved of. And so you can imagine someone running on that.

And then the question is, are they actually going to follow through on that? Because presidents have all kinds of promises that they make, that they are sincere about when they make them. And then the reality hits a little differently once they’re in the White House.

The other alternatives to have some sort of constitutional reform, the Supreme Court as of now has cut off statutory reform by saying that Congress can’t constrain the effect of a pardon. And I think they would have likewise suggested that Congress can’t limit the offenses that would be pardoned. And so that just leaves a constitutional route.

And of course, you could imagine any number of reforms, maybe saying that certain offenses couldn’t be pardoned, maybe saying that Congress has to approve of pardons, or maybe saying that one or two chambers could veto a pardon. I also discussed the possibility that we don’t have enough pardons, that the system is sort of set up to make it more difficult to get a pardon, far more difficult to get a pardon now, I think, than it would have been in the 18th century, in part because there’s far more people now and more offenders. And so it’s just hard to get to the president’s desk for him to make that decision.

And so maybe, you know, there certainly are people, progressive folks who think that there should be more pardons, not less. And they would kind of look askance at reforms that only made it more difficult to issue a pardon and perhaps didn’t create another route for perhaps securing one.

Well, Sai, it’s an extremely timely book. And I’m sure it’s going to be widely read because I think there’s going to be more and more agitation about the pardon power and what to do about it over the next three years, especially in the last few months of the Trump administration, I predict. So thanks very, very much.

Thanks so much, Jack, for having me. This has been a fun conversation.

Thank you.

Further reading:

• “Trump Threatens to Invoke the Insurrection Act” by Bob Bauer and Jack Goldsmith (Executive Functions, Jan. 15, 2026)

• “Thoughts on the Interim Order in Trump v. Illinois” by Jack Goldsmith (Executive Functions, Dec. 24, 2025)

• “Trolling About Habeas Corpus” by Jack Goldsmith (Executive Functions, May 12, 2025)

Thumbnail: Department of Homeland Security operations in Minneapolis, Jan. 6, 2026. (DHS photo by Tia Dufour)

This is an edited transcript of an episode of “Executive Functions Chat.” You can listen to the full conversation by following or subscribing to the show on Substack, Apple, Spotify, or wherever you get your podcasts.

Jack Goldsmith: Good morning, Bob.

Bob Bauer: Good morning, Jack.

President Trump, a few days ago, threatened to invoke the Insurrection Act in response to standoffs with ICE in Minnesota.

Today, we’re going to talk about why he’s making this threat now and what it might entail.

Let’s begin then with what our viewers are very familiar with, and that is the Court’s interim order in Trump v. Illinois. And there the court had something to say about the authorities under which the president could deploy the National Guard. That was directly before the court, but also there was discussion of the Insurrection Act.

Can you talk a little bit about that as background for how we might expect things to unfold from here?

Sure. Without getting into the very complicated details of that order, that was a statute, Section 12406, under which the Trump administration had federalized the National Guard to perform a protective function in Illinois, protecting ICE. The Supreme Court, in a very brief but complicated decision, essentially said that the president couldn’t invoke that statute to federalize the National Guard unless he could show that the regular military was authorized to enforce the law domestically and unable to do so.

And since the president had not made that showing, the Court basically said that the president could not rely on that other authority. Now, as Justice Kavanaugh said in his concurrence in that case, and I’m reading from it now, one apparent ramification of the Court’s opinion is that it could cause the president to use the U.S. military, as opposed to the National Guard, more than the National Guard to protect federal personnel and property in the United States. I wrote about this also.

You can see that interim order as inviting the president, or at least giving the president an opening to invoke something like the Insurrection Act, which is a statute that overtly authorizes the use of U.S. military forces to enforce the law domestically.

Address, if you could, just one thing right at the outset, which is it’s an opportunity that’s always been available to the president. In other words, Supreme Court didn’t create an opening that didn’t exist before. However, by closing off this other statute, 12406, it may have, some argue, driven the president in the direction of the Insurrection Act because this other source of authority was not available.

How do you assess that?

Yeah. As you say, the Insurrection Act has always been available. I think the administration perceives, probably correctly, that the political stakes are much higher if he invokes the Insurrection Act, because that’s really a statute that’s meant for an extreme circumstance. But the Insurrection Act was available and has been available for the president since the day he became president.

So it’s not really true. And again, I tried to make this point clear in what I wrote about the interim order. It’s not really true to say that the Supreme Court is incentivizing or inviting the president to use the Insurrection Act.

And that’s not what Justice Kavanaugh was saying. Nonetheless, I can imagine the administration, if it invokes the Insurrection Act, trying to make an argument like that, they’ll certainly say that, well, the court said that we hadn’t shown that we had a statutory basis to use the military to enforce the law. And here we are using the military to enforce the law under the Insurrection Act.

Let’s talk then about how the courts might respond. So we’ve written, as others have, that this is an antiquated statute. The terms are antiquated and indecipherable.

That is to say, the so-called triggers that the president could use to invoke the statute and to deploy the military. No role for Congress, no time limits on deployment. And there’s a general sense that, well, this is once he’s invoked it, it’s not at all clear, given the broad discretion of the statute seems to confer.

It’s not clear what limits might apply in any challenge. Can you talk for a second about how you see this playing out? Because one might argue one of two things.

One might say that’s correct. The Insurrection Act seems quite open-ended and the president will receive great deference. On the other hand, as the Illinois case suggests, these cases may rest entirely on the facts and the courts may be suspicious of the grounds of invocation that the president asserts, particularly against the background of these extraordinary controversies over the deployment of the National Guard and now the operations of ICE.

So I’d be interested to get your sense of what might we expect from the courts, how how wide a playing field has been afforded the president?

So I’ll say a couple of things, as you said, I mean, the Insurrection Act, and as we’ve been emphasizing for years now, is an extraordinarily broadly worded statute that that gives the president really many predicates to call out the National Guard and the regular military forces to enforce the law. And there are some 19th century precedents, Martin v. Mott and Luther v. Borden that can be read and have been read to give the president extraordinary discretion in making the determination that the Insurrection Act predicates are met. So in that sense, just on the face of it, the president, as we’ve been arguing, has extraordinary discretion.

But a couple of caveats—several caveats, actually. One is, as you suggested, also the Trump administration has shown itself capable of overstepping, even with broad authorities given to it, of not developing predicate facts needed to implicate a statute of suggesting bad faith in the way that they’re using authorities. And then it actually also depends on what the actual facts on the ground are. And when the president is invoking the Insurrection Act, how they justify it in the executive order and the like.

So that’s an uncertainty. A second consideration is that, as we saw in the interim order case in Illinois, the court was there interpreting a statute it really hadn’t interpreted before. It was looking at questions freshly and asking foundational questions about the statute. I think that every word of the Insurrection Act is going to be poured over and the Court is going to be forced to give it — the Court really hasn’t done a deep dive into the Insurrection Act in a very, very long time.

And so I think that there are going to be some foundational issues that come up. The whole question of the deference under Martin v. Mott and Luther v. Borden is going to be looked at very carefully. There are ways to distinguish those cases when you get down into the weeds.

So that’ll be a question. And then as Justice Gorsuch sort of hinted at in his dissent in the Illinois case, there are some foundational constitutional questions that invocation of the Insurrection Act will address. I mean, do you want me to talk about those?

Yes, please.

So Jay Bybee, senior state’s Ninth Circuit judge, issued an opinion in a statement in one of the en banc Insurrection Act cases in the Ninth Circuit where he made the point that, and I’m going to be talking in both of these foundational constitutional issues, they both have to do with the Militia Clause of the Constitution. And this is the Article I clause that says Congress shall have the power to provide for calling forth the militia to execute the laws of the Union, suppress insurrections and repel invasions. Now, that is the constitutional basis, the primary constitutional basis for the Insurrection Act.

And as we’ve discussed before, the Insurrection Act, as it’s called, is actually a series of statutes going back to the 18th century that have been enacted over time, changed and now kind of put together in one part of the U.S. Code. So Bybee raises the question about, you know, the militia clauses say that Congress has the power to call forth the militia to execute the laws of the Union. But there’s another clause of the Constitution that he points to, the Domestic Violence Clause of Article IV, which says the United States shall guarantee to every state in this union a Republican form of government and shall protect each of them against invasion and on the application of the legislature or of the executive when the legislature cannot be convened against domestic violence.

In other words, Article IV says that the federal government shall guarantee to the states of the union protection against domestic violence, but only when the states request it. So Bybee is basically suggesting that the Militia Clause has to be understood in light of the Domestic Violence Clause, basically suggesting that the president can’t or shouldn’t receive deference on the question of using the regular military to enforce the laws domestically unless there’s been an invitation by the state. That has not been the understanding of the Insurrection Act, but I think that that’s a kind of foundational question that will be asked.

A second foundational question that will be asked about the Insurrection Act is— and again, this has been long settled, but I think it will be questioned — The militia clause says that the president can call out the militia to enforce the law when needed. But the Insurrection Act has authorized the use of the regular military, not just the militia.

This was an innovation by Congress way back in 1807. The constitutional basis for saying that the regular military, as opposed to the militia, can enforce the laws domestically is not clear. It certainly isn’t the militia clause.

It’s probably the war power. So there’s a foundational question there about whether the president has the constitutional authority to send in the regular military to enforce the law. Both of these questions, I think, have been long settled by practice, but they’ve never been, to my knowledge, thoroughly vetted in the Supreme Court.

So these these are the kind of foundational questions that the court is going to have to encounter. I don’t think it’s going to be a walk in the park for the administration.

On its face, the president has extraordinary authority. That is the legal claim, and it’s a true legal claim. But the reality of how that thing gets litigated, of how the statute gets litigated and of how the administration goes about doing it, how it defends it, what the kind of narrative context is. And then all of these, we’re going to see just as we saw after 9/11 with the president’s military authorities asserted there with respect to detention, military commissions and the like, there were a lot of there was a lot of presidential support for President Bush’s unilateral actions soon after 9/11.

And three years later, four years later, the Supreme Court had blessed some of them and pushed back on a lot of others. And it turned out that precedents from World War II that looked rock solid and from the World War II context didn’t operate the same way after 9/11. And so we’re going to see that dynamic playing out.

There’s no doubt about it, that the old precedents are going to be subjected to scrutiny in a new and different context. And how that works out, I think, is just uncertain. So the real bottom line point is that the court will decide the modern relevance of the old precedents.

And while the old precedents are very strong in the president’s favor, they’re not airtight and it’s open to the court to rethink them and reapply them in a different way, in a different context. I just think it’s it’s just too early to tell how that might play out. We don’t even know if the president’s going to invoke the Insurrection Act and we don’t know how he’s going to invoke it.

We don’t know what the political context will be. So it’s hard to say how it will play out. But the administration can’t be sanguine about the Insurrection Act arguments.

And I think maybe that’s one reason why they’ve been hesitating. I don’t know.

This seems to be an extremely important point that the court may not be writing exactly on a blank slate, but it’s it’s a page on which there’s plenty of room to write. And, you know, for example, on these foundational questions, understandings that we haven’t had for a long time that could be revisited. I’m reminded of the sort of course correction or change in course that they took in the Illinois case where the understanding of the term regular forces suddenly received much closer attention ultimately was pivotal to the decision in the case.

And so I think it’s important for generally people to recognize that not only will every term in the statute be closely vetted and scrutinized against the facts of the case, but there’s some pretty deep constitutional questions that could wind up being revived in the analysis of the court and lead to some surprising results. Am I overstating the case?

No, I think that’s right. That’s what I was trying to suggest. And this may be one reason why the administration has hesitated.

I mean, on its face, the Insurrection Act is an extraordinarily broad statute. Why haven’t they invoked it yet? They could have invoked it on the same predicates that they used the other statute, which was a narrower and more obscure basis to federalize the National Guard.

They could have invoked it to actually use the regular military to enforce the law under the terms of the statute. They’ve hesitated, I think, because the politics are more fraught. I think they are — I’m not sure. I think the politics are more fraught of invoking the Insurrection Act and certainly bringing in the regular military is probably politically more fraught. But there are all of these foundational questions and that that are going to be addressed and especially in the Supreme Court.

I mean, the lower courts, as we’ve seen, are going to feel bound by Martin v. Mott and Luther v. Borden and some of these other precedents. Certainly the Supreme Court will as well. But there are ways to read those opinions that might not lead to full support for what the Trump administration is trying to do.

So, yes, there’s plenty of questions to be answered that the court will ultimately have to answer.

Yes. And there’s something else, I think, that’s going to play into their thoughts about whether to invoke it. And that is, there are polls and there are polls, and they vary significantly in sample size methodology, frankly, even credibility.

But the public is turning against the overall picture of these ICE operations. They continue to support robust enforcement against illegal immigration, but they don’t like the way this is being done.

Yeah.

And for the president in the middle of all of this to invoke the Insurrection Act with the uproar that that is going to create, I think is certainly not going to be helpful to his political standing at a time when the White House is, you know, as far as I can tell, rightly and keenly concerned about the president’s political standing in advance of the midterms.

That’s another way of saying just that the politics seem to be moving against the president on this issue, the way that immigration law is being enforced and that the politics might preclude implication of the Insurrection Act. And let me just say, they’ve threatened a lot of things in the last year. Many of them they’ve done, but many of them they haven’t done.

I mean, again, Miller six months ago came out and said, we’re thinking about suspending the writ of habeas corpus. Everybody’s head exploded. I was doubtful about that at the time for reasons that I wrote about.

But they’ve threatened all sorts of things that they haven’t followed through on, in part, who knows why, in part to gaslight, in part to play to the base, in part to discombobulate people, in part to put up a trial balloon. But the president says a lot of things in his Truth Social posts that never come to reality. And the Insurrection Act may be one of them we’re going to find out.

Yeah, I should add, by the way, I’ve always thought one other possible reason for hesitation, although that may be dissipating with time, is that the public heard a good bit about the Insurrection Act maybe for the first time in a long time in connection with January 6, when the administration, there were administration officials discussing potential use of the Insurrection Act in support of the president’s challenge to the 2020 election. So I thought that was also a sense of potentially a sensitive point.

Again, I’m not sure how sensitive the president remains to that since his his rhetoric and actions on that subject have definitely escalated. Let me ask you your reaction to other thoughts that I’ve had about the difficult posture that the administration may be in when it comes before the court if the president invokes the Insurrection Act, say, in relation to what’s taking place on the ground in Minneapolis. We’ve talked about the potential they’ll make claims that don’t hold up against the record, question of bad faith.

There are two other elements here that stand out for me. One of them is, the Court has to be sensitive to all of the attention that’s being paid to whether, in fact, the outbreak of lawlessness is being exacerbated by the conduct of ICE, whether the administration itself is creating the very conditions that it may later claim require the invocation of the Insurrection Act. And I can’t imagine that there isn’t some way that advocates in writing or in argument won’t point to that.

So let me start with that. Could that be a background concern of the Court as it thinks about the implications of the use of the Insurrection Act in these circumstances?

I’m certain it will be a background factor for the Court, but there’s another side to it. And the Court has talked about this in its interim orders, and that is the president has lawful authority to use the law enforcement machinery of the federal government to remove unlawful immigrants, you know, as long as it’s complying with the law. So they’re going to be competing narratives there for sure: that on the one side, the president ran on a platform of reversing what the Biden policy on immigration and and correcting the Biden policy on immigration.

And that’s what he’s doing. And the other side is going to say the president is using ICE in a way that is extralegal or on the borderline of legality and is creating the very situation that is being taken advantage of to invoke the Insurrection Act. I think we’re going to see competing rhetoric both in the briefs, probably, and certainly in the in the public discourse.

I don’t know how. It’s just hard to know how that’s going to play out for the Court, I think.

Yeah. And on that second point, I think one of the arguments about ICE’s conduct will be that it hasn’t simply encompassed communities where there was good reason to believe there might be illegal immigrants who needed to be brought out and faced legal process, that it has touched on the lives of American citizens. There have been videos of citizens who have been attempting to establish their citizenship but were effectively not given the opportunity to do so before force was applied.

So I think, again, there’s so much in the background here to color the understanding of what’s taking place in Minneapolis, not to mention the very sort of heated rhetoric that the president puts out on Truth Social. Another background element that I wanted to ask you about that may come up, the administration after the Renee Good killing took the position that it was not going to cooperate with Minnesota law enforcement in investigating the the death of Ms. Good and one question is whether or not someone’s going to take the position. Listen, you can’t government allege that local law enforcement, state and local law enforcement is unable to execute the law because you’re, in effect, shouldering them out of the picture.

There’s such hostility between the federal government and the state. Some of it, of course, based on the president’s allegation, it’s kind of a runaway rogue blue state. And so, again, you’re in a situation where the administration is claiming that the state can’t do something that the federal government itself is making it hard for the state to do.

Do you think that has any potential background force in the arguments in the case?

I have basically the same reaction to that as the other one, which is there’s another side to that rhetoric, which is that the federal government is claiming in all of these cases that state and local law enforcement are not being cooperative, are not being helpful, are resisting the force of federal law and federal law enforcement. I can’t, I mean, there’s so many different contexts in which these arguments come up, and I can’t sort them out. And I think it’s extremely difficult to predict how those competing narratives are going to play into the I’m not denying that they will play into the Court’s decision making.

I mean, the court decides on the law, but it’s not irrelevant. It’s not indifferent, obviously, to the reality of what’s going on in the world. I just don’t know which which narrative is going to win out and how they’re going to play out among the nine justices.

Can I ask you a question?

Yes.

We’ve written about the elections this year, in November, and we’ve written about the possibility that the president might be — and indeed, it seems clear that the president is — contemplating using federal law enforcement in novel ways to perhaps impact the election or bring about favorable results in the election or, to the president’s view, enforce election law the way the administration thinks it needs to be enforced.

Can you just talk about that a little bit and talk about how the Insurrection Act and law enforcement by the military may play into that and why we might be worried about that?

Certainly. Well, I mean, the president recently said, for example, and this doesn’t necessarily require the involvement of the military, but recently talked about sort of forcible intervention in the election process by saying he regretted that voting machines — I believe it’s voting machines, if not ballots — had not been seized in the wake of the 2020 election. And as you know, there was an executive order in draft prepared to that effect in the last Trump administration, in the waning days, as part of that that challenge to the election that was mapped out in various directions.

He has taken the position in an executive order that he issued some time ago now, promising that he would issue a second that contrary to the way the election of this country is structured constitutionally, contrary to the way fundamentally the entire legal order is set up for that purpose, that for the good of the country as president, he can, in effect, federalize the electoral process and oversee functions in the context of the election that have typically been the province of the states. So you have bellicose rhetoric and a broad assertion of authority. And then, of course, there’s the question of whether or not the president would use the military to carry out some of these interventions.

And let me just say one thing about the formal use and then the informal use and its potential effect. I mean, the formal use would be using the military to intervene on a claim that the election is being rigged, the election laws are being criminally violated. In effect, there’s been a subversion of the constitutional order and he has to call out the troops.

There are some very clear legal issues, to say the least, being raised by this, including federal statutory provisions that appear on their face to prohibit the deployment of troops to polling places. Then there’s the question of whether the president would ever recognize those kinds of restrictions. And rather, and I’m going to come back to his Article II authority question, I want to ask you still whether he’ll just invoke his constitutional authority in the circumstances, essentially issuing a constitutional emergency decree.

So there are whole sorts of issues about those kinds of direct interventions. But there’s also the question of what these kinds of threats and the mini-execution of these threats could mean for voting.

Let’s assume, for example, that the president starts to send out fevered, true social messages the day before the election, three days before the election, early in the morning of the day of the election, saying crimes are about to be committed. The constitutional order is about to be subverted. The rigging that I’ve always claimed would take place is currently occurring. I am calling out the National Guard. I’m operating, I’m invoking the Insurrection Act. I’m acting pursuant to my constitutional authority. And he puts out images of airplanes taking off and troops being mobilized.

That’s obviously going to have in some jurisdictions an extraordinarily chaotic and disruptive effect. Whether he actually acts upon it to the degree that he claims or not, of course, that matters enormously. And there will be an immediate response to this in the courts.

But the damage may have been done. And so a lot of effort has to be put in, in my view, before the election to put up every single guardrail against an attempt to issue these kinds of threats. If you take the darkest view of what he might be prepared to do to stave off the loss of the House of Representatives and maybe even the Senate in this midterm election year, I’ll just close by saying the stakes for him, he has claimed, are super high.

He’s not on the ballot, but he has suggested that quite apart from it being obviously very bad for the last few terms of his administration, if the Democrats regain control of the House, he has posited that he will be impeached. They will impeach him.

So there’s a direct personal stake in the outcome of these midterms.

I have a couple of questions about that. So it seems like a, it seems like a legal nightmare to have novel questions of federal executive authority concerning congressional elections combined with novel questions of the Insurrection Act being played out in litigation up to, during, and beyond the election, which takes place on kind of an emergency basis. Those cases always go very quickly with all those novel questions, with all those high stakes claims with it happening on a fast track.

It just sounds it sounds concerning. But you said that we should try to have guardrails in advance. But how can we have guardrails in advance to stop the president from engaging in certain threatening rhetoric, or even guardrails in advance to prevent him from trying to make these executive power gambits?

I mean, what kind of guardrails can possibly be put in place before November?

Well, of course, a lot depends on what happens between now and then. And I posit it’s sort of a last minute intervention. But some of these steps may become visible, may even be taken well before the election.

He has said he’s going to issue another executive order to supplement the first. The first was intended to extend intended to expand his authority over mail-in voting, over voting machinery that he continues to claim is unreliable over the monitoring of potentially illegal foreign national intervention in the elections. And that’s being litigated.

A second executive order or steps pursuant to the first will be litigated. So there are some steps that certainly could legally be taken and likely will have to be legally taken before Election Day itself. And obviously, you know, the earlier, the better.

I’m going to say something I’ve said before, and I think it does matter. A whole civil society response to these threats is going to be critically important. There are actors in the political process and in the broader civil society who are in a position to influence the course of events.

I mean, I’ll give you an example and how it would apply in these circumstances. I really can’t predict, but an example would be the unwillingness of the Indiana legislature to agree to the president’s attempt to have more seats added to the Republican column through a sort of mid-decade election year redistricting. These kinds of pressures are critically important.

And by the way, you and I have talked about it. Others have written about it. The response of the military to these orders is going to wind up being critically important.

And right now, the administration is trying to send a message about this more generally with its threat. First of all, it’s acted to enforce discipline against Senator Mark Kelly for advising members of the military that they are duty bound not to follow illegal orders. Those same threats have been directed and I think investigations have been undertaken of the other senators who have issued similar, completely correct legal guidance on the obligation of the military.

And so there are points in this system where those with huge responsibility and choices to make have to choose wisely. But those choices, if made wisely, can make an enormous difference.

You said you wanted to ask me about Article II.

Yeah, so let’s assume the president says, and by the way, whatever you say about the Insurrection Act and whatever, I have Article II authority that transcends all of it. It’s I can for the good of the country to restore domestic tranquility, protect against rebellion, against state and federal authority. I can deploy the military.

I’m commander in chief. And so, yes, I’m going to invoke this statute. But any responses, legal responses, you know, if they’re wound up to be in that setting deemed to be effective, do not nullify my capacity to turn to this inherent authority to accomplish the same purpose.

What are your thoughts about the —

Just so I understand, military authority to accomplish what purpose exactly?

Well, for example, to send troops into Minneapolis on the ground that the city, to quote what he said about Portland, is burning to the ground. You know, it’s chaos. We have basically civil war in Minneapolis, complete insurrection against the federal government.

This statute is Congress’s last word on the subject. But Article II is the final word on the subject.

So I’ll say there’s a lot to say about that. I don’t I hope I don’t think that scenario is going to happen. I hope it doesn’t happen.

But here’s what I would here’s what I would say about it. One is this is kind of the powers that Lincoln claimed in the Civil War, which was obviously a completely different situation than anything we’ve seen to date in this country. So, you know, there are a bunch of Lincoln Civil War precedents in about presidential action, but the circumstances there were so completely different that I don’t think they would be pertinent.

So but setting that aside, basically, here’s how I see the law. I mean, the president’s Article II power to use the military and the domestic sphere such as it is is the protective power. And this is the power, as we’ve discussed, to use the military in the domestic sphere to protect the property personnel and operations of the government in a protective short of law enforcement capacity.

As we’ve discussed, this is a longstanding executive branch practice and understanding. It has not ever been it has not ever been blessed and blessed in those terms by the Supreme Court. So it seems to me that even using the troops in a protective function is something that would be subject to full judicial review.

And maybe, you know, we’ll see. I mean, it’s a novel interpretation. It’s not a novel interpretation.

It’s a stretchy interpretation of a case called In re Neagle that the court has never passed on. But beyond that, I think you’re talking about something even more aggressive than that.

You’re talking about something like martial law, which is basically the replacement of military rule for civilian rule. And I don’t believe I mean, just staying within the four corners of the law, I don’t believe the president has the authority to unilaterally impose martial law in the sense I just described it. There’s no canonical definition of that.

But in the sense of replacing military for civilian rule, basically that would entail going further and spinning the red of habeas corpus, whicah, again, Lincoln claimed he had authority, the authority under Article two that the president had the authority to suspend the writ. I think the better understanding and the clearer understanding now is that that’s a power under Article one, section nine that clearly lies with Congress, even though it’s written in the passive voice. So I don’t believe that the president could successfully suspend the writ of habeas corpus, although I’ve written about this.

They threatened it. I mean, Stephen Miller came out six, eight months ago and said, we’re thinking about spending the writ. He was ambiguous.

I thought he was just gaslighting. It seems like he was at the time. So I don’t think that gamut can work.

I don’t think there’s any legal route for the president to be able to do that under anything like current conditions or really at all absent something like the Civil War situation, which doesn’t mean they won’t try it. I just think that they’ll have serious resistance in the courts. But that just begs the question about whether, you know, if we’re in that kind of extreme scenario.

And again, I don’t think we’re going to be there. If we’re in that kind of extreme scenario, it’s not clear what role the courts will be playing.

Well, I think we we probably conclude by saying the in this latest saga with the threat to invoke the Insurrection Act, it turns out that, you know, as bad as the statute is, as much discretion as it seems to afford, there are a whole host of issues facing the administration attempting to invoke it and the circumstances as we currently think about it. And we’re about to we’re about to learn a lot about the constitutional guardrails against various presidential assertions of authority in a number of contexts, including these elections that are coming up. Let’s hope not. But that may be the case.

It may be the case. We’ll see. Thanks, Bob.

Thank you, Jack.

Thumbnail: Federal Reserve Chairman Jerome Powell delivers a statement regarding a Department of Justice subpoena of the Federal Reserve, January 11, 2026. (Federal Reserve.)

This is an edited transcript of an episode of “Executive Functions Chat.” You can listen to the full conversation by following or subscribing to the show on Substack, Apple, Spotify, or wherever you get your podcasts.

Jack Goldsmith: Good morning, Bob.

Bob Bauer: Good morning, Jack.

It’s Monday morning and we’re going to discuss the news about the subpoena that was issued by the U.S. Attorney, the District of Columbia, to the Federal Reserve apparently late last week in connection with the testimony of the Chairman of the Federal Reserve, Jerome Powell, last summer concerning the renovations of the Federal Reserve. Obviously, there’s been lots of acrimony between the president and Powell. Trump has been pressuring and complaining about Powell for a long time. The subpoena is an extraordinary escalation of that.

President Trump claimed that he knew nothing about it, even though two weeks ago he said that the government was thinking about suing Powell in connection with the renovation. He didn’t say more. President Trump also said yesterday, I wouldn’t even think about doing it this way, whatever that means, but it doesn’t strike me as a credible statement given that he’s presided over an overtly weaponized Justice Department.

Bloomberg reported that Bill Pulte, the head of the Federal Housing Finance Agency, the person who has been agitating for various forms of weaponization by the Justice Department and who was behind the Lisa Cook for cause firing, was behind this in some way. Chairman Powell yesterday issued an extraordinary statement in which he said this had nothing to do with the renovations and it had everything that it was a pretext. I’m going to quote now:

He said that this whole matter about renovations or pretexts, and quote, “the threat of criminal charges is a consequence of the Federal Reserve setting interest rates based on our best assessment of what will serve the public, rather than following the preferences of the president.”

Yesterday also, Senator Thom Tillis, Republican on the Banking Committee, said he would oppose any nominee of the president to the Fed. There’s an opening on the Federal Reserve now.

One of the seven governors, there’s an opening and Powell’s term is up in May as the chairman. So it had looked like Trump was in a position by May to nominate and confirm someone as a new governor and to replace Powell. Not clear what impact this has on that.

The last thing I’ll say is even though Powell’s tenure is up as the chairman in May, he still has two more years to serve as governor. It was, it’s typical, I think, for the chairman of the Fed to resign from the board once the term is up, if not renewed, but it’s possible that Powell could stay on for two years and therefore deny Trump a second opening. So I think that’s a catch up on the news as of this morning.

What should we talk about?

Let me start by asking you this question. Will Trump now, pursuant to the for cause provision in law, fire Powell?

This could be a possible setup for that. I should, in that connection, mention one other fact. As we know, another governor, Lisa Cook, was last, late last summer, I believe, fired by the president for cause in connection with a housing matter.

That firing has been unsuccessful to date. The matter was enjoined by the lower courts. The Supreme Court refused to stay it and has oral argument on that question, I think, on the 21st of this month, so very soon.

So there’s a legal question there. It’s notable that the court did not rush to support the president there, unlike in other contexts of firings that it has. It’s possible that Trump is setting up Powell for a for-cause firing.

I don’t think it would be successful in light of what’s going on in the Cook case. Moreover, as best I understand it, I think this is right. The president technically has the legal authority at any moment to remove Powell as chairman of the Federal Reserve.

There’s no express statutory for cause protection as chair. He has statutory protection as one of the seven governors, but there’s no statutory protection as chair. Certainly, that could be an area that the president could push, so he could use this as an excuse to, although I don’t think he needs a legal excuse, to remove Powell as the chairman, but not take him off the board several months early.

I have to say, I don’t understand why any of this makes sense. Powell’s term is up in four months. I don’t see how this helps the administration on the economy.

We’ll see what the markets do today. We’ll see what the impact of it is, but I don’t understand it strategically. It’s not obvious that this was a concerted, coordinated in the White House, so it’s all a bit of a puzzle to me.

I would note that Powell is represented by the law firm of Williams & Conley, which is a powerhouse law firm in Washington, D.C., and he, presumably on the advice of lawyers, decided to take a pretty aggressive stand last night. He wasn’t required to put out that statement. He wasn’t required to write it the way he did, and so he’s digging in, which is quite remarkable.

Now, one question for you, I mean, the power to remove the chair without having to allege cause could lead, of course, he could exercise that to remove Powell. That wouldn’t address Tillis’s complaint in the Senate, and that wouldn’t solve his problem with moving the nomination of a successor to the floor, would it? It wouldn’t.

It’s still an independent issue.

Yeah. Let me emphasize, I’m pretty sure I’m right about this removal issue as chairman of the Fed, but keeping him as governor, I’m confident that the statute doesn’t provide anything, and the OLC has interpreted that type of situation to allow removal if the position atop is not statutorily protected. Certainly, it would be open to the administration to make a gambit to try that.

They would have a powerful legal argument, but as you said, it wouldn’t impact Tillis’s point because Tillis is concerned about the attack on Fed independence by the president. So I just don’t see how this helps the administration at all. It hurts.

We’ll see what happens, what the impact is on markets—on treasuries and the like— but politically in getting his person on, whoever that will be, whoever’s nominee will be getting that person on the Fed and as chairman strikes me as much harder this morning than it was 24 hours ago. Do you agree with that?

I completely agree with that also. I’m going to approach this as someone who had a stint in government as a lawyer and thinking about the lawyers who were trying to defend the administration’s legal position before the Supreme Court in the removal case that, as you point out, is going to be argued in January. I can’t imagine any one of them is going to be happy that this step was taken while the court is deliberating on this case, on the Cook case.

It just seems to me that it, tell me if I’m exaggerating, we will never know, but it is hardly the foot that they wanted to put forward as they’re attempting to put a credible and they hope persuasive position to the court in the Cook case.

Yes. I mean, it just deepens the notion that the president is weaponizing law enforcement, weaponizing the firing power and anything at the margins, I think he’ll be hurt on this. The president has extraordinary prerogatives.

The for cause standard, whatever it means, and we’re going to find out is not a super high standard. I think there’s a legitimate question about the independence of the Fed. Other presidents have chafed at an independent Fed.

There are all sorts of legitimate issues here, but the way that the administration has gone about it, I don’t think it’s going to help their legal case. It could conceivably impact, although I doubt it, the Humphrey’s Executor matter and how the court is thinking about structuring the unitary executive that it’s clearly in favor of, but I don’t think it’s going to help in the four cause case at all. I wouldn’t imagine the Solicitor General’s office is terribly happy this morning.

No, it occurs to me that there’s an overlay here where they’re taking their unitary executive case and they’re damaging it by introducing the unitary weaponizing executive into this debate in a way that it seems to me is hardly helpful to their position. On the question of reasons, you say, why in the world would they have done this? Here’s one argument: Because it is a remarkable thing to do.

They’ve been thinking of doing it for some time and of taking aggressive action against Powell. As you mentioned, Trump at one point said he was going to, they were thinking of suing him. It wasn’t clear what he meant by that, but they didn’t.

And now they have served these subpoenas and so the battle is on. The only political, I guess I’ll put it that way, the only political reason that I can think of, the strategic reason why they would do that, is that they’re beginning to become extremely agitated about the impact of the economy on the midterm elections. So we’re into this new year, within less than a year, the midterms are going to take place.

He’s made it very clear he’s extremely worried about losing the House. Most presidents would be anyway. In his case, he believes that the stakes are even higher.

He said, they’re going to impeach me if the Democrats regain control of the House. And it may be that they think, and I don’t know whether it’s a matter of economic reality or economic theory, this is accurate. They may think they need the Fed to move much more aggressively and raise the rates—excuse me—lower the rates and juice the economy to stand any chance of protecting themselves against the damage that his ratings on the economy are currently doing.

And the polling data on this has to be very discouraging to the administration. Is that a plausible theory? It’s the only one I can think of.

I certainly don’t know, but it strikes me as not obviously rational because I don’t see how this gambit is going to lead the Fed, is going to pressure the Fed into lowering interest rates. It would be, I mean, who knows? It’s very hard to assess the Fed’s motivations when they do these things, how much they are subject to pressure or not.

But Powell’s statement yesterday was basically drawing a line in the sand saying, I won’t be bullied. That’s the first point. Second point is, as we just noted, it makes it harder, not easier for Trump to get his person on the Fed.

I just don’t see how—and also, you know, time’s running short for anything the Fed would do to have an impact on the economy that’s going to impact the election. So I just, again, I’m not an expert on the politics of this, but I just don’t see that angle and how it would be part of a rational plan. But it’s not obvious that this was . . . clearly the president has been agitating for this, clearly he sends signals throughout the administration, clearly there’s a decentralized, weaponized structure to the administration.

It’s not clear how centrally planned this was and it’s not clear to me that it’s rational for that end. But maybe there’s something I’m missing.

Well, I don’t think you are. I think the reasoning may be very skewed, but politicians sometimes don’t act very rationally when their deepest interests, like, for example, holding onto office or holding onto their best political fortunes, may be at stake. The other question I have is, is Tillis going to find allies in the Congress on this?

He’s not running for re-election. So he’s found new space in which to operate and speak freely on these issues. And the question is, other members of Congress are going to have to make a decision here.

The five Republicans who bolted on the War Powers vote the other day and chose to support Democrats in bringing the question of the War Powers resolution on Venezuela to the floor were immediately threatened by the president, who said they shouldn’t be, I don’t have this quote in front of me, but effectively they’re done for as far as he’s concerned. He doesn’t think they should be elected to anything. So there’s always a fear of crossing the president.

But one question in my mind, I’d be interested in your views on this, is this degree of disorder, if you will, constant conflict, with the potential that we’re going to start seeing signs, and there are early signs, that this isn’t sitting well with the markets, is going to have an effect on constituent responses to what’s taking place in Washington and how it’s affecting the economy. And that in turn could mobilize some opposition to Congress, even in the face of that fear factor. Now, it’s hard to tell.

It’s hard to tell how long those effects would have to work their way through. What do you think about that?

Again, hard to assess. We’ve been saying for a long time that the closer the midterms come and the more that the member, the congressional member reelection imperative, that interest departs from the demands of the president, the more we’re going to see this type of plea. But then we started to see it more and more in the last few months.

But I just, I don’t, I don’t know, you know, the down in the weeds politics of particular members, for example, on the banking committee. I don’t think the banking, I think if Tillis votes for the Democrats, I think it’s, I think it’s hard for a nominee to get out of that committee. So I’m not sure how many more, how much more Republican support the president could bear to lose.

But let me just say a couple of things broader, more broadly about that. This type of, it seems, political use of the law enforcement power and the criminal law enforcement power. There are basically two checks on it.

I mean, there’s the check of the grand jury and the trial, but that is kind of down the road. There’s also the market check. And we’re, you know, to date, the market check has, the market has not reacted, at least not in a violent way to other agitations against Powell by Trump.

So we’ll see what happens this morning. And then as you’re suggesting, there’s the congressional piece. And we’re going to see whether we’re starting to see Congress members in the Republican party stand up to the president a little more.

He’s raising the stakes. We’ll see. It’s going to be interesting to see today what, how the other Republicans in the Congress, especially in the Senate react.

Let me just say one more thing that I forgot to include. That is the inspector general of the Federal Reserve is Michael Horowitz, who was the long time, very credible, well-respected inspector general in the Justice Department. He moved over to the Fed, I think last year.

And he’s now conducting an investigation of the renovation. Powell asked him, I believe it was last summer, to conduct a full investigation of the integrity of the renovation. So that is going on in parallel.

To the subpoena, just to mention, because it’s not clear how far along he is or whether or when he might issue a report. But that’s another factor in this.

Well, all the more reason to question the motives. I mean, I don’t know. I don’t know anybody’s going to really contest the motives behind this move to proceed criminally against Powell.

And that is, there is already a mechanism of accountability and a record that’s presumably being established about how the renovation was being conducted and whether there was any kind of misconduct in these cost overruns that the president alleges that he’s concerned about. And yet, nonetheless, they forge forward with this attack in the service of these subpoenas. And let me ask you, let me ask you another question.

And again, I hope this isn’t in the realm of, you know, completely rank speculation or drawing connections or trying to tie two points together that probably should not, can’t be rigorously tied together. He just completed what he absolutely believes was a triumph in Venezuela. And just a few hours ago, I guess yesterday, he posted a modified Wikipedia page, sort of a mock Wikipedia page in which he described himself as the acting president of Venezuela.

He’s also been issuing threats in various directions against Greenland and Colombia briefly until apparently he reached some kind of a peace with the president of Colombia for the time being. And of course, he’s made threats about Iran and intervention there, depending on how the government proceeds in dealing with these protests on the streets. We could be dealing, could we not, with somebody who’s just in this immediate space has concluded that he can do whatever he wants, that the use of force here, if you will, and this isn’t a military use of force, but the exercise of raw power is going to achieve his ends.

And so it’s hard for me not to draw, you know, a sort of some kind of connection as we think about what mentality has gripped the White House between, you know, his ability to impose his will in the Caribbean and south of the border and his opposing his will domestically. Now, again, is that entering into a realm of reasoning that just is sort of beyond what we ought to be talking about here?

I mean, clearly. So here’s what I think. The president has much more leeway in the foreign affairs, military war context to bark an order and see the order carried out. And this is why Trump has, I think, been gravitating towards foreign affairs, military intervention in part because it’s an area where he can realize his monarchical dreams of basically being able to run the world, to do what he wants and the like.

And that’s not an exaggeration to say that’s what his aim is, because they’ve proclaimed that he’s proclaimed that nothing checks him in the international realm other than his own morality, as I think he put it. I just think it’s and that mentality might be carrying over into the Fed. Again, it’s not clear how centralized and rationed and thought out this was, but it’s just a different situation in the domestic realm because of the reasons we’ve discussed.

Not clear what the markets are going to do. Not clear also, though, that he’s going to be able this is going to help him with the Fed because Powell’s back is up now. Powell might not leave in May, even if he retires from the chairmanship.

It’s not clear Trump’s going to be able to get his replacement on there. This action, I think, unless we’re missing something, makes things pretty clearly worse for him in his effort to control the Fed. At least we may be missing something, but it just seems to me that this is going to be self-defeating in terms of coercing the Fed to bow to his will.

So I just think on many domestic matters, not all, the president has a harder time immediately implementing his will compared to foreign affairs. And that’s why threats against Greenland, Venezuela, running Venezuela, maybe invading Colombia, maybe bombing Iran. These are all things that at least if the DoD culture can continue to be shaped as they’ve been doing, that the president can basically execute when it’s much harder to do so in the domestic realm, I think, especially on the Fed issues.

And yet it may be increasingly hard for him to accept limits. I mean, this is a president who notably said Article 2 permits him to do whatever he wants, that he has, quote unquote, total authority as president. And as he sees success in one realm, he may find it hard to believe why he can’t achieve the same success in the other.

And speaking of the use of military force, of course, he has attempted, and here the Supreme Court has thrown up obstacles, he’s attempted deployments of force in American cities and says he’s going to, even though he’s now withdrawn from Los Angeles and Portland and Chicago, he said he’s ready to go back in as necessary. So it’s just hard not to see a total environment of willpower that he believes that he can bring to bear in all these circumstances, just the raw exercise of power to affect his own will. But I’m not contesting your point.

Obviously, he has much, much more operating room as a matter of law practice and precedent in the international realm than he does domestically.

He basically lost that round on the National Guard. I mean, really extraordinarily lost that round. And there’s judicial review there.

Again, as I’ve written many times, the president has other legal cards for domestic deployment of the military. But there’s going to be judicial review there, and the political stakes are much different. And there’s not going to be judicial review of a bombing of Iran, or even probably, I don’t think there’s going to be an invasion of Greenland, but there’s just not going to be a judicial check about foreign military intervention the way there is about even domestic military deployment or other domestic matters.

It’s just, again, it’s too sharp to distinguish between foreign and domestic affairs because many times they blur, but the president is more constrained in the domestic realm than he is in the foreign affairs realm.

Certainly, but I’m glad you mentioned the other cards he has to play and that he hasn’t played. He said just in the last, and I’m speaking roughly here, 10 days to two weeks, that he may need to turn to the Insurrection Act, which so far they haven’t attempted to put into action. And there, you have a statute that you and I have written, and we’ve worked on this topic that just seems to confer on the president extraordinary authority, ill-defined terms, no congressional consultative provision, no limits on troop deployment if needed to address conspiracies, combinations, domestic violence, insurrection, all these terms that appear in the statute.

So he hasn’t finished testing the limits of his ability to work his will in the United States with potentially the use of armed force. And obviously, it’s beyond our remit right now or beyond our topic right now to talk about what’s taking place and how the public’s reacting to the use of ice in American cities and the questions that have been raised about the use of force in that context. But this is quite the period in the next 24 hours, at least in the Fed situation, with the reporting I suspect we’ll still see is going to be quite revealing and maybe faithful.

No shortage of topics for us to talk about. Never.

Thank you, Bob.

Thank you, Jack.

Mentioned:

• “On the Legality of the Venezuela Invasion” by Jack Goldsmith (Executive Functions, Jan. 3, 2026)

• Curtis A. Bradley, Historical Gloss and Foreign Affairs: Constitutional Authority in Practice (2024)

Thumbnail: President Trump delivers remarks at a press conference at Mar-a-Lago in Palm Beach, Florida, following Operation Absolute Resolve in Venezuela. (White House Photo.)

This is an edited transcript of an episode of “Executive Functions Chat.” You can listen to the full conversation by following or subscribing to the show on Substack, Apple, Spotify, or wherever you get your podcasts.

Bob Bauer: Good morning, Jack.

Jack Goldsmith: Morning, Bob.

Well, since the last time we spoke, much has happened.

We’re going to focus on what you most recently wrote about—and that is the deployment of force by the United States in Venezuela—and the different explanations the administration has provided. But in the course of that, of course, they seized the president and first lady of the country, brought them out of the country to the United States to be tried. Let’s break down some of the points in your most recent posting, the one that you put up over the weekend, and talk about that also somewhat in light of questions that have been raised just generally on this topic since then.

Does that work for you?

That’s great. We should also talk a bit about the trial going forward and some of the legal issues there.

Absolutely. So as you begin in your post, you say there are few, if any, effective legal constraints on presidential deployments of force.

So the question I have for you, and you walk through that, explain that at length, the question that I have for you is, should somebody take from the way you have described the evolution of presidential authority in this area, opinions produced by the executive branch and so forth, practice. Should somebody conclude that the whole question of the legality of this kind of deployment of force is really largely a pointless question, that what we’re going to look at now is a political struggle between the branches, if there is one, but there really isn’t any law to be discussed in any meaningful sense?

So this is kind of a hard, I basically agree with that, but it’s kind of a hard question to answer, and let me explain why. I mean, first of all, we don’t have any Supreme Court precedents. So the court really has not engaged with the question ever about the circumstances under which the president can use unilateral force.

The only context really was in the prize cases in the Civil War when it basically upheld Lincoln’s exercise of essentially self-defense there. So when we’re talking about law, one major source of law is just not there. So what are the sources of law?

Well, we’ve got the constitutional text and we’ve got 250 years of practice. Now, there’s a great debate over the original understanding of the War Powers and the Declare War Clause, the Commander-in-Chief Clause, and the like. There’s been 250 years of practice since then.

It’s all been in favor of the president. The president has exerted over 250 years greater and greater and more extravagant and extravagant uses of force without congressional authorization. Congress has basically acquiesced.

The War Powers Resolution in 1973 was Swiss cheese and has not proven to be much of a check on the president. And Congress in these uses of force (and there have been lots of them) tends not to rise up and check the president. So when we’re talking about law here, we can either look to the founding documents, which are contested and have been overtaken by practice in 250 years, or we can look at the executive branch precedents.

My point in the piece was that under the executive branch precedents, which I’m not saying are the definitive statements of the law, that this was fairly easy to justify. And the final point I’ll make is, and it’s basically where you started, I’ve been emphasizing this for a while and we’ve talked about it for a while. I just don’t think there are effective legal constraints on the president, i.e. a situation where an executive branch lawyer is going to tell the president in a situation where he thinks the use of force is necessary, that there’s a powerful argument against it. The only possible legal argument in the OLC opinions against unilateral uses of force in any situation where the president might want to, is in an extreme boots-on-the-ground situation where there’s a persistent US presence and a serious threat to US troops on the ground. And in that situation, OLC has held out the possibility that congressional authorization may be necessary. But right now, anyway, we’re not near that.

Let’s go to practice just for one more second here. What is the legal significance of that? So let’s assume presidents—and I’m going to put this in deliberately sort of blunt terms—they get away with it because the political process just is simply not structured to respond effectively to presidents who overreach their authority, give reasons of national interest, rally public opinion behind them.

Congress just is weak in that circumstance. What does it mean that the president has repeatedly gotten away with it as a legal matter? How does that bear on how we should think about the law?

I don’t know how to answer that beyond what I just said. It’s not clear what one would refer to as the relevant legal principles. They’re contested.

One can make an argument from founding sources, and even this argument is contested, but one can make an argument from founding sources that where the president is in his claimed authority to use force now is unconstitutional. One could look at the Declare War Clause and have a certain interpretation of that. One could have a certain interpretation of the Commander-in-Chief Clause.

So there are legal materials one could draw on, but they’re contested. All of these principles are contested. So I’m not sure what it means to say that the president got away with it.

I mean, this is an area where presidential legal interpretation has been dominant, and the only force that can potentially check it is Congress. Congress did assert a theory of Article II in the War Powers Resolution. It stated its theory of what the president’s powers were.

Congress has plenty of tools to assert its own legal prerogatives, its own view of the Constitution, and to check the president. There are many ways Congress can push back through funding, through conditions, through oversight, and it has at times. I mean, in the Somalia matter, in the Lebanon matter in the 1980s, Somalia in the 1990s, Congress rose up, pushed back against the president.

But this is an area where we can talk about law, and the executive branch has all these opinions. But the truth is, it’s really law in the hands of the political branches, and it’s up to the political branches to sort it out. That’s the way I view it.

So let me just restate the question to make sure I understand, again, the legal significance of practice. Presidents do something. Congress is unable as a political matter to respond, and the absence of that response means presidents have added to the law in their favor.

We’ve done it before, so we clearly can do it again. You have acquiesced, and therefore that has legal significance that you didn’t respond in that situation to the claim of authority that I, the president, made.

So that’s not quite the argument I was making, but let me clarify. So there are two ways to think about it. Let’s imagine that this case actually came before the Supreme Court, that the Supreme Court set aside the political question doctrine, set aside standing concerns, and that there was a justiciable case that could brought before the Supreme Court.

And the question then would be, well, what are the president’s powers to use force without congressional authorization? The president would make the argument you just made, that there’s been a long course of practice that has been acquiesced in by Congress, and that this practice liquidates the meaning of the Constitution in separation of powers, and that therefore that long practice of congressional irrigation of power has liquidated into an interpretation of articles one and two that allow the president to use this force. That would be the constitutional argument. It would be a tough argument to make.

Curt Bradley’s got a great book on this, on the role of practice in foreign affairs and political branch practice and understanding the meaning of separation of powers. It would all hinge on whether Congress could be deemed to have acquiesced, and Congress would claim, or the parties representing Congress would claim, no, they haven’t actually acquiesced. They had the War Powers Resolution, and you can’t view silence as acquiescence because there hasn’t been affirmative acquiescence.

That would be the terms of the debate if we’re trying to figure out the significance of practice for the meaning of the Constitution. It would be, how do we think about liquidation as informing separation of powers? I think that answers your question.

That’s, I think, the legal or legalistic answer. As a practical matter, however, the courts aren’t going to adjudicate this. At least, they haven’t yet, and I don’t believe that they will.

As a practical matter, separate from what I just said, the president has lawyers who, over the course of time, have written dozens and dozens of these opinions that now constitute a body of precedent that drives executive branch decision-making. There’s no judicial force to stop that. Only Congress exercising its political prerogatives, perhaps making legal arguments, can check that.

Does that make sense?

Yes, it absolutely makes sense. But turning now to the precedents that the executive branch itself generates, again, somebody would say, well, what can we make of that? Normally, presidents get out of their lawyers what they want.

I’ll say that maybe uncharitably because that’s not always true. But by and large, the president, and certainly this president, has been very selective about the lawyers chosen for the task. What does it mean that presidents order up these legal opinions, and that in turn shapes the legal debate when they’re entirely self-serving?

The branch is basically not constraining itself in any way.

Yeah, I definitely do not think that that has historically been true as a general matter. There are lots of instances of executive branch lawyers saying no to the White House or advising the White House that some course of action is unlawful, and the White House acquiescing in that. There are lots of examples of that.

Yes, of course, there are lots of examples of executive branch lawyers finding ways to allow the president to do what he wants to do. In the war powers context, especially, I believe the lawyers have three, because these are the highest possible stakes. When the president decides that he thinks he needs to use force, and I’m talking about a whole run of presidents, not just Donald Trump.

Given the long, long line of practice of presidents using unilateral force, going back to the 18th and 19th century, at least, early 19th century, it’s very hard for lawyers in light of all those precedents and all those opinions. I’m not saying it’s impossible. I’m just saying it’s very hard for them, and it obviously depends on the circumstances, to say no.

This is the point I’ve been making for years now. The precedents are so overwhelmingly in favor of the president, of the president, both on the self-defense front, hugely capacious precedents and opinions, and on the offensive use of force front, that I just don’t think, as I’ve said this many times before, and people don’t like it when I say it, but I believe it’s true. I just don’t think that even within the four corners of the OLC opinions, that there’s much constraint on the president.

I’m not sure if that answers your question.

No, it does. I think you’re right about, certainly in my experience, such as it was, there are certainly cases where OLC does not concur in a request that the White House makes or the president makes for a legal blessing of a particular initiative. But as you point out, the situation changes dramatically when it’s really important to the president, particularly in the national security and foreign policy context.

And there, it isn’t easy to find many cases where the lawyers in the administration, wherever they may be found, who ultimately are relied upon to produce the opinion, stand up to the president and say, no, you can’t do it in a way that effectively constrains the president.

I’ve got a few scars myself from my time in office in dealing with that, but I agree it doesn’t happen very often. Okay.

Let’s talk about some of the doctrines or legal justifications that appear in executive branch precedent. National interest. Secretary Rubio took to the networks yesterday.

I don’t know that he swept all the Sunday TV shows, but he was on many of them. And he kept on bringing up the subject of national interest. It was our national interest.

And he interpreted it or he applied it very broadly. Here’s a quote, and I’d like to have your comment on this. It’s just a matter of its fidelity to executive branch treatment of this particular term, national interest.

Quote, and the first steps are securing what’s in the national interest and also beneficial for the people of Venezuela. And those are the things that we’re focused on right now. Again, here are the interests, no more drug trafficking, no more Iran Hezbollah presence there, no more using the oil industry to enrich all our adversaries around the world and not benefiting the people of Venezuela, or frankly, benefiting the United States and the region, unquote.

So there’s a whole range of things there. And I could mention a few others that I think by implication, he included in the term national interest. Is there any boundary to the president’s self-proclaimed legal authority to declare something to be in the national interest and therefore support the deployment of force?

There’s no declared boundary in the OLC opinions. The OLC opinions have invoked the national interest as a basis for the president’s offensive use of force. It’s come up with a long list of national interests.

Some of them were hinted at there, self-defense is one, Hezbollah matter, humanitarian concerns have been one, regional security is often invoked as a national interest. These are very capacious national interests as a basis for the president to use unilateral force abroad. It’s one of the reasons this national interest test is one of the reasons why I think the opinions are not meaningfully constraining on the president, because even after you have that long list, the list doesn’t purport to be exclusive.

And they’ve been adding to them in recent decades. So this has the patina of legal analysis, but there are no stated limitations on the national interest test in the OLC opinions, other than that point I made earlier about significant boots on the ground with significant presence that poses a significant threat to US troops. OLC has recognized that as a check on the national interest, but for operations like this one, where it’s in and out quickly, it’s mostly force at a distance, although not completely, there are no stated constraints on what can be the national interest.

Some of the ones he mentioned haven’t been mentioned in prior OLC opinions, but I have no doubt that he used the word national interest to kind of key off of what the lawyers were telling him was the important word. And another important word that they used a lot during the press conference was that this was an apprehension operation and that this was an FBI arrest supported by the US military. That ties into another legal justification, the bar memo that I referred to in my piece, that basically says that the executive branch can enforce arrests extraterritorially.

So in other words, on the national interest point, just to bring it sort of to a conclusion, a president could plausibly, and I think this administration has come close to saying this, maintain that it is in the national interest of the United States, and therefore the president can deploy forces necessary to ensure that no regimes take hold south of our border in Central or South America that we deem hostile.

I mean, there’s nothing in the precedents that would rule that out. You would have to bring it into the language of the OLC opinions and talk about regional stability and the like. But look, Bob, as I mentioned in my piece, it’s not like there’s not a long history of United States intervention in South and Central America violating sovereignty and running things.

I mean, and this is another problem for the OLC lawyers, there’s a very long history of just this kind of action or something very similar to it. So, and I’m sorry to sound so skeptical about legal constraints here, but I just don’t believe that, and I haven’t for a while believed that there are significant ones. And let me just say, I think this is a very important point to underscore.

Immediately after these operations happen, every time this happens, Libya, Kosovo, Iran, all of these unilateral uses of force without congressional authorization, we immediately jump to the law and commentators immediately say this is illegal depending on whether they like the war or not, or they defend it as being lawful. And we have this debate about whether it’s lawful or not. And I frankly think it’s kind of a meaningless debate in almost every circumstance.

The issue is why has Congress given the president this massive military force without constraints? Why does it continue to acquiesce in the president’s use of force? Why isn’t Congress exercising its prerogatives to its constitutional prerogatives and constitutional responsibilities to check these things?

The lawyers tend to flee to the legal arguments. I think the legal arguments in this context are not terribly meaningful and that the focus should be on the politics of this. And the politics are that Congress has let the president get away with it knowingly across administrations, left and right, Democrat and Republican.

And the Democrats tend to complain about Republican uses of force and vice versa. But all of this stuff takes place in the rhetoric of law that I think is largely meaningless.

Before we move on to the trial, now the Maduro’s in the United States and I think is being arraigned today, I assume also the first lady, let me just ask you one last question about the sort of debate that you correctly say erupts immediately on episodes like this. Is it important that we even indulge those kinds of debates? Isn’t it important that anybody actually cares enough to raise the question of whether something is legal?

Because if it passes completely from the dialogue in the aftermath of something like this, then we’re really sort of completely outside the zone of any concern for law. Or is it healthy that we don’t bother with it so that we can stay focused on the political struggle, which you think is really the material concern?

I’m not saying people shouldn’t make legal arguments. I mean, law is the language through which we criticize presidential uses of war powers. That’s just the natural language that we use.

And, you know, and people tend not to get down into the weeds the way I do in terms of and the way others do. I’m not saying I’m the only one down into the weeds of what these opinions say, what the historical practice is, what the contestation about the founding was. I’m not saying people shouldn’t make the argument.

I’m just saying that I’m not sure. I’m just not sure what purpose it serves. There’s not going to be judicial review.

And there’s the president has his precedents. The legal rhetoric, if it’s relevant, it’s relevant for influencing the politics, I think. And there are occasions when Congress rises up.

I mean, they came very close. It was by a split vote. Congress almost denied Clinton.

It was a tie vote, the authority to continue the war in Kosovo. And as I say, they’ve pushed back in other contexts. So I’m not saying that people should not argue in the language of law.

They can argue in whatever means they want. And sometimes I suppose that this legal rhetoric can be useful to the politics. The point is that, frankly, it’s the politics that are going to matter in checking the president.

It’s the politics of what the president thinks he can get away with, the politics of how this is going to play out, owning Venezuela, the politics of whether there’s going to be pushback in Congress, the politics of whether Congress can get his act together to do anything.

So let’s turn to an area where I think we’re in a much more clearly defined zone of law, and that is the trial that will take place, or at least legal proceedings will take place, who goes to trial, doesn’t go to trial, I don’t know, in the Maduro case. Talk for a second about what the impediments may be for a court hearing the case, or if a court hears the case and dispenses with any argument that it can’t, what legal defenses, based on the extraordinary circumstances of his arrest and rendition to the United States, what legal defenses are available to Maduro, and how would you assess their strength?

Sure. So I think we’ve seen this play out already. And I think that the issues in the Noriega trial that took place in the 90s, I believe it was, a lot of the issues that are going to be fleshed out in the Maduro trial, assuming the trial goes forward, were anticipated and resolved in the lower courts in the Southern District of Florida and the 11th Circuit in the and brought back to stand trial in the United States for violating U.S. federal criminal law. And the first issue is going to, and I’ll just tick through them, and I might not cover them all, but these are the ones I know that will come up. Did the abduction, or whatever you want to call it, the snatch or the extraterritorial arrest, does that by itself constitute a due process violation such that Maduro can’t be trialed?

The answer to that question will almost certainly be no, under a doctrine called the Care Frisbee Doctrine. It goes back to the 19th century, some Supreme Court cases that basically said the circumstances in which the person is brought to trial do not impact due process. That turns entirely on the process given in trial.

He’ll also maybe try to invoke the UN Charter. First of all, he’ll probably try to invoke head of state immunity. He’ll claim immunity from prosecution.

It’s pretty clear to me he won’t get head of state immunity. The executive branch’s views on this will have a lot of weight, if not dispositive weight with the courts, and the executive branch will claim that he wasn’t a I don’t think there’s a case called Alvarez-Machain that involved another US extraterritorial arrest where the Supreme Court said that the extradition treaty, unless it expressly precluded extraterritorial kidnapping, did not prevent the trial from going forward.

I doubt that the UN Charter will work. It didn’t work in the Noriega trial because the treaty wasn’t self-executing and didn’t confer individual rights. I doubt that he’ll be able to claim POW status since there wasn’t, unlike in the Noriega context where Noriega had declared war against the United States, there was no war here.

Those are some of the standard defenses that I’m sure will come up. They’ve all been fleshed out in the 11th Circuit and in the Southern District of Florida. I’m not saying, I mean, I basically think the courts there got it right, but it’ll all be relitigated.

Certainly those decisions aren’t going to be binding. So all of these issues and others will be relitigated. There’ll be a question about the extraterritorial application of the federal criminal laws.

I think the government will win that. I want to emphasize that I’m giving thumbnail sketches of these legal issues and thumbnail sketches of how I think they’ll come out. They’re all more complicated, but those are some of the main legal issues that will come up.

Let me press further on one question. Noriega was the de facto head of state. Maduro, who we’d recognized as the president of Venezuela for some time until an election that was widely understood to be fraudulent, the United States recognized him as a head of state.

He was, as of the time of his capture, he was the quote-unquote president of Venezuela. Does that make any difference here that Noriega was in a different position than Maduro?

I don’t actually know the factual answer to this question about whether Maduro was the recognized head of state by the United States at the time of capture. I don’t believe that’s true, but I actually don’t know the answer to that question. That will inform the answer, but also I believe, although this is not set of law, that the United States would be able to prevail on a head of state immunity defense if it said he doesn’t warrant it, even if he were the recognized head of state.

That’s a more complicated issue. If he was the recognized head of state, obviously, recognition is an exclusive presidential power. He can be de-recognized or recognition can be removed by the president unilaterally.

The executive branch has been given extraordinary deference by the courts in the determination of who warrants head of state immunity, but I don’t actually know the status of recognition at the time of capture.

An extraordinary set of circumstances. But as you point out in your piece, this is not only where we are, but where we have been for a long time.

I think it is, unfortunately. The president gets to decide how to use military force abroad, and that’s where we are.

Thanks, Jack.

Thank you, Bob.

Further reading:

Chris Mirasola, Sovereignty, Article II, and the Military During Domestic Unrest, 15 Harv. Nat’l Sec. J. 199 (2023)

Jack Goldsmith, Thoughts on the Interim Order in Trump v. Illinois, Executive Functions (December 24, 2025)

Chris Mirasola, Unpacking the Protective Power, Lawfare (June 12, 2025)

Thumbnail: President Donald Trump in the White House, April 7, 2025. (© noamgalai, Shutterstock.)

This is an edited transcript of an episode of “Executive Functions Chat.” You can listen to the full conversation by following or subscribing to the show on Substack, Apple, Spotify, or wherever you get your podcasts.

Jack Goldsmith: Good morning, Bob.

Bob Bauer: Good morning, Jack.

We haven’t chatted in a while, and there’s so much going on in the news.

We’re going to talk about three things today that have been simmering in the background persistently. One, we’re going to talk about the latest on pardons. Then we’re going to talk about some transparency issues for the executive branch. And then we’re going to talk about the law firm cases and how the Trump administration has been handling those.

So let’s start off with pardons. Just to remind everyone, the president has been on a pardon spree this year. Early on, he pardoned 1,500 individuals related to January 6th. He later pardoned 77 people in connection with challenges to the 2020 election. He has, by some counts, pardoned 30 people for financial crimes, including the crypto Binance founder and Representative Henry Cuellar.

And then, recently, he pardoned the Honduran president, Juan Hernández, who was serving a 45-year sentence for conspiracy to distribute more than 400 tons of cocaine. This comes at the same time that we’re engaged in this boat-strike war over illegal drugs entering the United States.

So in some sense, this is an acceleration even over Trump’s extraordinary first term. And in some sense, I’m not really surprised by the Hernández or the Cuellar or the Binance pardons. What do you think?

I want to add one more to the list, which is that he has issued a federal pardon for a state conviction. This involves a county election official who was convicted on several counts of assisting a breach in security protocols to stop the Steele inquiry—an unlawful effort to stop the Steele inquiry. This particular election official is now in prison and has been seeking release.

State prison, just to be clear.

Yes, state prison. And he has issued what he calls a full pardon.

Now, that pardon applies to federal offenses. However, it’s interesting to note that he has been pressing very, very hard for a state pardon as well. He would like to see a state pardon issued. The lawyer for this individual takes the position that the president’s pardon power should be read to apply to state convictions too, and he sent a letter to the president making that case.

Now, where Donald Trump wants to take this beyond that is unclear. Right now, it seems to be part of a campaign to put pressure on the state of Colorado—a pressure that, so far, the state of Colorado has shown no interest in giving in to. Whether there’s some other step he wants to take to broaden the understanding of the president’s pardon power is unclear. But I just wanted to note that.

But on that score—before you get to the broader points—it’s not his call. First of all, the pardon power doesn’t extend there. And second, it’s not his call as to whether it does. He can issue whatever pardon he wants, and then it’s a question of what executive officials in the state and what courts examining it will do.

So, you know, I guess the Justice Department could file an amicus brief, but that argument is going nowhere, I think.

I can’t imagine it would go anywhere, but it will be interesting to see whether they take it to the next level and incorporate, as the constitutional policy of this administration, the position that the lawyer for this incarcerated election official has articulated.

So my overall reaction is, yes, it’s a pardon spree. There is no filter, as far as anybody can tell. I want to stress this in the case of the pardon of the former president of Honduras, who, as you point out, was convicted for massive drug trafficking and, by the way, related weapons charges for providing armed support for drug trafficking activities.

He explained in a Nov. 28 Truth Social posting that he had heard from people he greatly respected that this particular drug trafficker—the former president of Honduras—had been treated very unfairly and harshly. Who these people are who told him about this unfair treatment is not entirely clear, although it is known, at least according to press reports, that Roger Stone—himself a pardonee and a former aide and political supporter of President Trump—had provided some avenue for the former president of Honduras to make a plea by letter to the president for this pardon.

And so the doors are open for political influence, with very little in the way of deliberation, explanation, or elaboration to the public about why the pardon power was exercised in this particular case. He heard from Roger Stone, and potentially others, that Hernández had been treated unfairly, and he went ahead and issued a pardon.

Yeah. I mean, one thing that’s important to point out is that there has traditionally been a process that runs through the Justice Department that vets these pardons. Presidents have not always followed that process, of course. In recent years, many presidents have departed from it—President Biden, I think, at the very end with regard to his family. President Trump, in his first term, broke that process completely. Other presidents at times have not followed that process.

Right now, I believe Ed Martin is in charge of the pardon process at the Justice Department. He is someone who is kind of leading the weaponization campaign there. So you’re right—there’s no process. It sounds like it’s whoever gets to the president. The president has said more than a few times, “I don’t even know that person.” I think he said that about the Binance person he pardoned. That sounds doubtful to me.

But it does seem like if you can get access to him on a good day, he’ll sign the paper. I guess I don’t understand. I get some of the pardons—the ones that are retributive or that he viewed as political vis-à-vis himself. But I don’t understand the pardon for the former president of Honduras. Politically, it seems not great for him to be pardoning a former Honduran president under these circumstances.

Some of the financial crime pardons don’t seem to have any political angle at all, other than that the convictions occurred during the Biden administration. I mean, what is going on here? Is it just random? Is there any political play behind that part of the pardons?

Well, in the Hernández case—the former president of Honduras—I don’t think we have a full explanation. It may simply be that this particular former president, guided by people Trump trusts—people he said he greatly respects—managed once again to tap into President Trump’s openness to the argument that some people have been treated as unfairly as he has been by a weaponized Department of Justice, just as he believes he has been.

He said something similar about Congressman Cuellar. He said that Cuellar had been victimized by the Biden administration for speaking out against its border policies.

I want to mention one thing about Cuellar. The irony—if you want to call it that—of the president, in this instance, apparently displaying some bipartisanship by pardoning a Democratic member of Congress should not be lost. When Cuellar then decided to continue running as a Democrat, President Trump lashed out at him and said he had exhibited disloyalty, and that he should have shown more loyalty to the president after receiving a pardon.

So it turns out that this was likely an entirely political pardon. As you know, Trump is very concerned about losing the House, and we are in the middle of this major controversy over mid-decade redistricting. And we’re in the middle of this huge controversy over mid-decade redistricting. He seemed to have had some thought he could pick up a Republican seat, if you will, or convert a Democratic seat to a Republican seat by pardoning Cuellar and Cuellar exhibiting the requisite loyalty. So this nonpartisan or bipartisan pardon was probably very much a partisan pardon.

Although, just to show why I think this doesn’t quite make sense, he also pardoned Rod Blagojevich, who wasn’t going to run for anything—a former Democrat. I think he also pardoned the former Democratic mayor of Detroit. So there seems to be something larger going on. I agree that Trump was disappointed and angry and had different expectations about Cuellar, but it does seem to go beyond that, given that he has pardoned some Democrats.

I should mention—by the way—go ahead.

No, go ahead.

As I mentioned in the case of Cuellar, the question is how he developed this expectation in the first place. And because there’s so little transparency around these pardons, he may have had some reason to believe that Cuellar was going to act differently.

That’s interesting. Yeah. Cuellar could have said, “Give me a pardon and I won’t run.” He got the pardon and then decided to run. There’s nothing Trump can do about that.

Okay, let’s talk about the transparency issues. There are lots of transparency issues floating around Washington right now, but two in particular stand out.

One is the Epstein files. There was a statute passed a few weeks ago—the Epstein Files Transparency Act—requires the Justice Department to release most of the Epstein-related files. I think they’re due next Friday, Dec. 19. There’s an exception, of sorts, for ongoing investigations.

The other issue is whether the public—or even more members of Congress—will be able to see the videos and legal analysis related to the September 2 boat strike, including the follow-on strike. The House appears to have closed its investigation, but the Senate Armed Services Committee remains open. There’s also a provision in the House-passed National Defense Authorization Act that cuts into the Secretary of Defense’s travel budget if he’s not more transparent.

Those are two pretty important transparency issues. How do you think these issues will play out?

At this point, I don’t know. I’ve been very skeptical in the Epstein case, and I’m also skeptical—I’ve written about both of these issues on our site—that we’re going to see anything the administration doesn’t want us to see.

In the Epstein case—let’s take these in order—we recently had Kash Patel saying that he didn’t know whether there was anything more to be released. He also raised concerns about compromising ongoing investigations, and that is protected under the Epstein statute that Congress passed.

There’s word circulating that expectations should not be too high, and that there will be redactions and withholdings. Already, Epstein survivors and Democrats have called for the inspector general to conduct an audit of how the administration has managed these files.

In the meantime, you’ll recall that in March there were reports that 1,000 agents had been put on 24-hour shifts to look for any reference to Donald Trump in the Epstein materials. So there’s a real question about what we’re actually going to see. Are we going to see what the administration chooses to put out, while withholding other material under one justification or another?

I can’t say for certain what will be withheld or why, but I don’t think we’re going to see all of the Epstein files.

Before we move on, let me say something about that. Isn’t this a disaster for him politically? If they’re not fully transparent—and if the exceptions to transparency aren’t completely credible—doesn’t that make the situation even worse?

Haven’t expectations been ratcheted up yet again, with repeated pledges of disclosure? It seems to me that they’re in a terrible position if they don’t produce a maximal release.

Part of it probably depends on whether there are disclosures elsewhere that raise questions about the completeness or integrity of the administration’s release. In other words, if they’re caught out—if something emerges from another source, such as the Epstein estate or a leak, that casts doubt on the administration’s account—then of course that would be catastrophic for the administration.

I do think it’s interesting that in both cases we’re discussing—the Epstein files and the boat strike case—we’re seeing legislative activity, including support from Republicans, that’s putting pressure on the administration on transparency. This is something he hasn’t had to confront before.

They’re under enormous pressure, but how much they give in to that pressure depends on what they’re trying to protect against. And we just don’t know.

That leads to the boat strikes. My sense, from reading the stories—including Republican claims that there’s “no there there”—is that there actually is a there there, particularly based on these videos.

I don’t know about the legal analysis, but the videos and the more recent reporting in the Washington Post a few days ago about the advice that Bradley got about the strikes and the delays before the second strike—and apparently there was at least disagreement among the lawyers and they wouldn’t even commit to what the final legal advice was.

It seems to me that there’s unattractive stuff there for the administration. I don’t see them being transparent there unless their backs are completely to the wall, unless they’re coerced into doing so. And I don’t see that that’s going to happen.

I couldn’t agree more. I suspect that if these videos were helpful to the administration, they would have been released days ago. They haven’t been, and there has been equivocation from Secretary Hegseth.

The president—who tends to be rather muscular in his pronouncements—has said that it’s up to Hegseth.

After saying earlier that he thought that they would be.

That there would be no problem releasing the videos. When that earlier statement was repeated back to him by a reporter, he became extremely aggressive toward the reporter and denied having said there would be no problem releasing that video.

I don’t mean to sound conspiratorial about this. I could reach back across other administrations and cite other examples. I think it is vastly worse in this administration, but — we’re not going to see, I think, unless something extraordinary happens here, what they don’t want us to see.

There are simply too many ways for the administration to dodge transparency. And along the way, the president does what he does very well: he creates distractions. When pressure builds on one front, he opens a new front elsewhere and sends everyone chasing in that direction.

Whether he can outrace this problem in that fashion remains to be seen, but it’s a tactic he has used to great effect in the past.

I’ll add one point on the transparency front. While the opinions and the videos haven’t been disclosed, there have been significant leaks—and damaging leaks—and they’ve been ratcheting up. I expect that to continue, because these situations tend to feed on themselves.

That is a form of loss of control that they haven’t really suffered much yet either.

I totally agree with that. I also think—and I’m obviously speculating here—that one other problem is that I think Hegseth has some dedicated foes within the military establishment and the Pentagon. The question of his actions here, even his failure to be forthcoming about it now, are weapons that are going to be used against him in an effort to weaken his position, if not force him from his position altogether.

Okay. Last issue: there’s a story from a couple of days ago about the law firm cases—these executive orders. I’ll let you describe the executive orders and the state of play with the law firms, but the bottom line is that the administration has been slow-walking this despite repeated losses. So explain what’s going on and why it matters.

Yes. I think it goes to the question of a use of presidential power that is illegal, but not necessarily wholly ineffective.

So about nine months ago, I think it’s nine months ago—the first court order against the administration on one of these law firm executive orders was entered. I think in the Perkins Coie case, it’s now been nine months. And the administration has lost every single one of the cases that have been so far decided on these executive orders.

And yet at the same time, it has been clear from reporting in the law firm trade press that law firms are not eager to have these executive orders issued against them, even though they know that in the end they’re going to win. There’s also been some reporting that some law firms are looking to work with the administration to implement the settlements that he reached with a number of law firms, by which they agreed to change some of their policies, say on DEI, but also to provide support for pro bono causes that were important to the administration. So there’s been some reporting on that.

But the administration is appear eager to have the legality of these executive orders fully resolved. So they have not, as they have in other cases, rushed to seek scase or tried to expedite rulings at the appellate level in any way. Rather, they’ve been more than happy to have delay after delay.

Most recently, an additional 45-day delay because the one attorney who apparently is central to the administration’s defense of its executive order advised the court that he had a planned paternity leave coming up. So he needed additional time. And the law firms that have been targeted, the law firms who were defendants, by and large, either have taken no position or have agreed to these extensions.

So what you have is plainly illegal orders that are out there that probably still have to be taken into account by law firms that don’t want to cross the administration. And the administration is happy to leave them in place in all of their illegality as we try to find out what precisely firms are doing to fulfill the settlements of the illegal orders that have been issued against them.

Yeah, and just to underscore the point in a way, this is also, I think, going on in the Harvard litigation where Harvard has been winning in court. The government has not been rushing for appellate review, certainly an emergency review, certainly not the Supreme Court. And the point is that the main goal is not to win on the law.

The main goal is to affect behavior and to change the way law firms act, change the way universities act. And so the law is being used as a tool, even if it’s unlawful, even if courts declare it unlawful, the administration can go slow on the appeal because it doesn’t matter. They’re still imposing costs in virtue of just having the program out there.

Harvard is spending an enormous amount of money defending these cases and investigations and the like. The law firms are, as you suggest, and I don’t know the full scale of what they’re doing, but they’re certainly worried about these EOs, worried about crossing the administration. These are all wins for the administration, whether they win in court or not.

That’s the important point, I think.

Absolutely. And in the meantime, we don’t know much about what’s actually taking place in pursuant to these settlements. There have been lawsuits filed to try to disgorge from the administration details, more information about what precisely has transpired between the administration and the firms that decided to settle these illegal orders.

But those are relatively recent, and it’s not clear where they’re going to go. So this is a depressingly effective tool that the administration has been using to try to cast a pall or introduce some intimidation into the law firm community. And again, if you look at the trade press, it does appear there are law firms that indeed are litigating against interests of the administration, but there’s a lot of cautiousness among the largest law firms that do not want to face these executive orders.

Okay. I think we’ll stop there. I hope you have a great weekend.

You too. Thanks.

Mentioned:

• “The Unitary Artificial Executive” by Alan Z. Rozenshtein (Lawfare, Oct. 30, 2025)

Thumbnail: Abstract virtual artificial Intelligence interface with human head hologram on USA flag. (Pixels Hunter, Shutterstock.)

This is an edited transcript of an episode of “Executive Functions Chat.” You can listen to the full conversation by following or subscribing to the show on Substack, Apple, Spotify, or wherever you get your podcasts.

Jack Goldsmith: Today I’m having a chat with Alan Rozenshtein of the University of Minnesota Law School. Alan gave a lecture at the University of Toledo this fall on executive power and artificial intelligence, and he published the lecture on Lawfare. I had Alan come to a seminar I taught this fall.

The thesis is really interesting, so I thought we could have a good chat about it. So, Alan, why don’t you just tell us about the claim and what follows from it, and we’ll go from there.

Alan Rozenshtein: Sure. So, the claim in a sentence is that near-term—so I’m not talking here about artificial general intelligence or superintelligence, but the next three-to-five-years’ level of artificial intelligence—is going to, or at least has the potential to, massively increase the president’s control over the executive branch, and that has a lot of political and legal and normative implications that are worth unpacking.

But the core descriptive and predictive claim is that—and just to give a little bit more context—so we’re having this conversation on Monday the 8th in the afternoon, and just this morning the Supreme Court heard oral argument in a case about the president’s power to fire heads of putatively independent agencies. And it sounds like the Supreme Court, based on oral argument, is going to overrule Humphrey’s Executor, this really important case—almost 100 years old now—stating that Congress can make heads of certain agencies independent. And this is going to be a really big deal when it happens, and con law professors are going to talk about this forever, and it does matter.

But I think it’s useful to think about what the practical implications of this will be—like what the practical implications of having a, quote, unitary executive, which is this idea that whatever powers the president has under Article II, they’re the president’s powers, and so he or she has to be able to exercise them, and that means firing people basically at will.

And directing their actions and firing them if they don’t.

Exactly, right. So, you know, a few months from now the Court’s probably going to overrule Humphrey’s Executor, and we’re going to have a, quote-unquote, unitary executive. And what is that going to mean? Well, it’s going to mean important things.

It’s going to mean that the president can direct the head of really any agency—probably not the Fed, but that’s a separate conversation—but basically any agency to do what the president wants. And if not, the president can fire that person.

Nevertheless, I think that even under such a world, the president’s actual ability to run the executive branch—not just the cabinet officials or their deputies, but the, I think it’s something like three million civilian employees, and then obviously a couple million more in DOD, in the military—these millions of individuals, the bureaucracy, the civil servants, the quote-unquote deep state—the president’s powers, even under a fully unitary executive system, are still going to be quite limited, right?

And there’s a wonderful quote from Harry Truman. This was, I think, in 1952 on the eve of Eisenhower’s inauguration. Truman was giving a press conference or interview in the Oval Office, and he said something like, “Poor Ike, he’ll sit here, you know, behind the Resolute Desk, and he’ll say, ‘Do this, do that,’ and nothing will happen. It won’t be anything like the Army,” right? And the idea is just: it’s really hard, just as a matter of logistics, to oversee all these people.

And my argument is that this is actually not a legal issue. This is a managerial issue—or to put it another way, a technological, a technical issue. And one thing that AI is going to be able to do is that it’s going to be able to solve that issue. It’s going to take this unitary executive, which has always been kind of more of an academic theory, and it’s going to make it into a reality, really for the first time in American history. And that’s going to have a lot of implications—some good, a bunch bad—but that’s the idea in a nutshell.

So how exactly—I mean, what is the mechanism whereby—first of all, first of all, let me say, I’m not so sure how far the Court is going to go in saying the president has control over every agency all the way down. But set that aside. Let’s just assume that’s so, because that’s not really your point. Just tell us what you mean by artificial intelligence and tell us what the mechanism is whereby the president is going to be able to use this tool to assert more robust unitary control.

Sure. So by artificial intelligence, I’m referring to the sort of last, I’d say, ten years or so of machine learning, as it is expressed in chatbots like ChatGPT and Claude, but in particular in what are sometimes called agents. So these are artificial intelligence systems that can actually take action.

Now, obviously, right now, the only action they can take is the control of a computer, right? They’re not yet embodied in robots. I think one day they will be. And I think especially when that gets into the military domain, that’s going to raise some really tricky, sticky issues. But let’s put that aside for a second and just talk about control over a computer. That’s what the main labs are trying to do.

And the reason they’re trying to do that and the reason they’re spending hundreds—literally hundreds of billions, if not at this point trillions—of dollars in investment in the ability for an agent to control a computer, I mean, part of it is because they themselves are coders—they want to make their jobs easier. But their view, and I think this is largely correct, is that if you can automate computer work, then you have kind of automated a lot of white-collar work.

And certainly the management of the executive branch happens—or you can imagine the management of the executive branch largely being—the management of a set of computer systems, right? I mean, the executive branch is largely managed through writing memos and sending emails and organizing meetings and that sort of thing. So if you have a system that can do that—and can do that not just over a five- or ten- or fifteen-minute time horizon, as the current agentic systems can do, but can really plan over hours, days, weeks, months—and can do so not over one or two or three people, but over thousands or hundreds of thousands or millions, then you have the ability for the president to really control, at a level of granularity, the executive branch in a way that right now relies on human beings.

So another way of conceptualizing this is: what if instead of having to rely on a bunch of fallible humans, you can rely on much less fallible AI systems, whose preferences you can match much more closely to your own, to not just be your chief of staff at the high level, but get your tendrils down into all parts of government?

So is the idea that—I mean, I still don’t quite understand the claim and maybe a concrete example would help—but is the idea that computers that are going to reflect the president’s will are going to replace human beings in the executive branch, or that the president will be better able to control and direct human beings because he’ll be able to—because the computers will be able to reflect his will better? And what’s an example?

Sure. So I think both of those are true. And so I can give you an example of each one.
Right now, the way decisions—so let’s talk about the first one, which is using computers to actually control what actually happens. So right now, the way that decisions are made in the executive branch is that different bureaucrats at different levels come together and create a memo. And if they can solve that issue at that level, they try to do it at that level. And if not, it kind of bubbles up and up and up. And at some point you get to the deputies committee and principals committee and the cabinet.

And every once in a while, a decision is so important or so difficult that it has to go up to the president’s desk. Now, the higher that decision goes, the more personally the president can take control over that, either himself or through his close subordinates. But the vast majority of decisions are made at a very, very low level. And it’s very hard to oversee those decisions.

So imagine a situation in which you had a set of agentic systems that were trained on the president’s preferences—and we can talk later about sort of what I imagine that looking like. And instead of bureaucrats themselves deciding, “We’re going to resolve this issue by ourselves at our level”—and these are going to be a lot of career bureaucrats, these are people largely not politically appointed and therefore possibly working independently or even averse to the president’s own priorities—what if instead those bureaucrats have to clear their own decisions through this computer system? They have an oracle that they have to consult before they’re allowed to institute any particular policy. So that’s one way in which you could have the kind of machine implementing some of these policies.

So let me make sure I understand. So every bureaucratic decision of any—whatever magnitude—has to get signed off from the oracle, which will be trained on the president’s preferences. Therefore, the president can exercise perfect control over his agents in the executive branch. Is that the basic idea?

Yes. Certainly a lot more control. Now, you could imagine this going even further where you—so my example was kind of about the policymaking functions of government. In fact, the vast majority of what the executive branch does is adjudications of one sort or another, mostly informal. And so probably 99 percent of what the executive branch actually does—not necessarily in importance, but just in volume of action—is that.

Already, there’s a lot of interest in making automated decisions, either because they’re viewed as more efficient or they’re viewed as more accurate. But another reason that a president might want the executive branch to run much more in an automated decision way—and here I’m referring to the person making the social security determination, the person making the immigration determination, the person making the excise tax determination, or whatever the import fee determination is—another reason why the president might prefer a lot of that to be automated is that that can be controlled much more centrally by him.

Now, I think in the extreme case, you still probably need human beings—again, partially because AI systems aren’t well embodied yet. Also, at the end of the day, the courts are probably going to require a person to get up in court and say, “Carbon-based life form here representing the United States.” But certainly, even short of that, you can get a lot of centralized control over the bureaucracy.

So tell me more. I mean, let me give an example. Think about the Venezuela boat strikes. Trump said at some gaggle like a week or ten days ago he wouldn’t have wanted to take the second strike. He said that. Hegseth may or may not have been in the loop. The JSOC commander did the second strike. We still don’t know the facts very well.

I mean, how confident are you that in the next three to five years an AI system trained on the president’s preferences would have been a perfect agent there in instructing the JSOC commander whether to take the second strike or not? Let me put it another way. I’m a little skeptical. I’m a little skeptical that we’re going to have systems in the next three or five years that are going to be able to train on the president’s preferences—I’m not even sure how you do that—and to reach decisions that are reliably reflective of the president’s desires as agents.

Yeah. So I think here it’s worth emphasizing that all of these claims about whether AI is, quote-unquote, good or bad at something—and I probably was sloppy earlier, so I should fix that, so you’re good to push me on that—they’re strongest when they’re comparative claims, right?
So the question is not so much, “Is the AI going to be a perfect agent?” right? Any more than: is Spotify the perfect DJ, or does Netflix perfectly predict the next sci-fi show I want to watch? It’s relative to the existing alternative and, at the cost, are they better? And are they sufficiently better that they’re worth the switching costs? And are they sufficiently better that it’s worth it for people like you and me who care about the executive branch to think this is going to be important going forward?

That’s a great point. You’re right. It’s about which is relatively better, but just play it out for me, just using the example I gave.

In one scenario—we still don’t know quite what happened—but there was pre-planning. This was the first strike. It was September 2nd. There was pre-planning. There’s probably lots of operational planning. There was consultation between the SecDef and the JSOC commander, tons of preparation on the ground. Trump was probably in the loop at some level, but maybe not terribly tightly into the loop.

But the AI would have been trained on the president’s preferences, and I’m just wondering why you think the AI would have done a better job in deciding that second strike in reflecting the president’s preferences than Hegseth and JSOC’s coordinated activities. Just flesh it out.

Yeah. One thing that’s very tricky when you’re thinking of Donald Trump in particular—and he’s the president, so he is naturally the example—is his preferences are unstable.

That’s kind of what I’m getting at.

Right. Yeah. So for issues like this, and especially high—I think AI is going to be least interesting in situations like this, which are very high salience, in which the president’s preferences are probably fairly easily understandable, to the extent that they exist—and again, I’m not sure they even necessarily exist—by the people around him.

I don’t want to be unresponsive to your point. My sense, based on what I know about Donald Trump, is that were it up to him, he would absolutely have taken that second strike, right? The double tap is great. I think he doesn’t like drug smugglers, if these were in fact drug smugglers. He doesn’t have a lot of respect for law, whether of the domestic or international variety. He likes to look tough, all these sorts of things.

But at the same time, he likes to avoid bad PR. And so probably the best outcome for him would have been: take the double tap if you think this won’t get me in trouble, and the blowback won’t come, right? Now, that is a predictive judgment, and there’s no guarantee that any system, human or AI, would have gotten that right.

The question is, when you look across the broad range of activities, and especially those activities that are not so high salience, and right now there aren’t politicals spending a lot of time trying to figure out how to make their principal happy, will AI be better than the currently existing alternative? I think there’s at least a strong possibility that the answer is yes, yes to that.

And so what will the AI train on? What will it look at? Every speech ever given, everything ever written?

Yeah, so what I would imagine is that there would be two main sources for the AI to generate a set of preferences. One is existing public remarks. So again, the whole corpus of speeches, public statements, tweets, that sort of thing. And obviously there’s going to be some filtering because people evolve over time.

The AI can figure that out.

Yeah, it can figure that out. It’s not that hard to weight things that are closer in time versus things that are farther in time. So that’d be one source.

And then the other source I would imagine is a kind of updating feedback loop, where what you would want to do in the White House is you’d want to have someone who samples from the AI decisions that have been made recently, and then presents them to the president and asks for presidential feedback on that. So you would probably need this kind of feedback loop to avoid too much preference drift.

Now, I’ve not been part of the White House policy process, but I assume that something like this has to be happening. I assume that a big part of what the chief of staff does is sit there and make sure that the current system is properly reflecting the president’s preferences, and then makes tweaks to that system as it occurs. I’m just suggesting replacing that with a machine.

But I think the implications of your view are that all deliberation in the executive branch can just go by the wayside. And once the machine is trained on the president’s preferences, then every executive branch order will be knowable.

And what about—I mean, there’s learning from deliberation outside of what the president says and does. There tends to be learning from interagency processes, from new facts being brought to bear that maybe the president didn’t know about. What happens to all that? I mean, why do we need any deliberation in the executive branch?

So it’s an empirical question. I don’t know if we do. I think if there’s any lesson of machine learning in the last ten years, it’s that one should be very careful about assuming that there’s something ineffably special about how humans do things that AIs cannot replicate.

Now, because every time someone has said, “Oh yes, AIs can do this and this, but they can’t do this other thing,” it’s like three weeks later they can do it. And so you just end up moving the goalpost over and over again. Now, there’s no guarantee that this progresses forever, right? None of my argument depends on us waking up one morning and discovering that Silicon Valley has created the machine god.

So it’s quite possible that there will always be necessary space for human decision makers, right? I think it’s useful to think about the extreme case as a kind of stylized, you know, what is the sort of logical conclusion possible of this? And that is, you know, everything just becomes the sort of oracle and you ask that.

But the audience for my argument is lawyers and executive power types, right? For whom a 10 percent more powerful executive is a massively big deal, right? And so that’s the only point I’m making, right? I don’t have a super strong prior on how far you can push this.

Yeah, that claim seems much more defensible and likely to me—that the president, and I wouldn’t just view it as the president, the oracle via the president. I mean, it might be that—probably would be the case—that a well-informed president might want the AI to train on his—maybe this current administration is not a great example—but his specially chosen expert secretary of defense or especially chosen expert secretary of state, because the president might realize that. And then that secretary of state might want to delegate the AI to be trained on someone lower.

So you can imagine there being lots of humans in the loop and the AI decision-making happening at a lower level, which means that the humans think that there’s—if that’s true, if that’s a possible scenario—it means the humans think that there is, in fact, value in delegation.

Yeah, I think that’s all possible. And let me—let me try to describe the argument in just a different frame. This is one that is, I think, going to be less flashy but maybe more convincing, which I’m fine with.

Which is: if you want to view AI as just the next advance in information technology, where information technology has always increased the power of the presidency, I’m more than happy with that version of the argument, right? If you want to say, “Oh no, this is just a difference in degree, not in kind,” like—sure. I think it’s a big enough difference in degree that it’s worth paying attention to.

But one historical tidbit that I’ve always enjoyed is the fact that Abraham Lincoln was the first president to have the advantage of the telegraph in war. That was the new information technology of the day. And although it seems quaint right now because you’re sitting there typing in Morse code, it was huge. It was maybe the biggest technological advance of the last 200 years, because suddenly Lincoln could sit in his office in Washington, D.C., and actually be a battlefield commander.

He could go and he could send things to George McClellan, who then could happily ignore them—but that’s sort of a different story. I think there’s been interesting—and I’m curious if anyone has written this, and sort of I’m tempted to, though I’m not really a historian—to write the history of the presidency through technological transformation, right?

Oh yeah, there have been. I mean, the president’s control over communication and information gathering is a source of the president’s increased power over time.

And so AI is just the next chapter of this.

So I think that’s a fair point at a general level. I mean, it seems to me that at a minimum, that presidents since at least Roosevelt have been trying to centralize decision-making power in the White House, away from the departments and agencies, and with various degrees of control—but that’s been a clear trend since at least Roosevelt.

And one challenge to that is knowing what’s going on in the executive branch and being able to collect and analyze information. And there’s no doubt that these systems are going to allow greater information extraction from the executive branch, and greater information analysis from the executive branch, and probably more efficient White House control in issuing orders.

Indeed, you can see this as basically what DOGE was trying to do, in the sense of having a White House entity go inside of an agency, extract information, inform the White House, and then the White House can impose the order on the agency—basically a way of circumventing the agency. If that could all be automated, which is basically what you’re saying, it would definitely facilitate White House control.

That’s right. And it’s not just what DOGE was trying to do—it’s what DOGE did in a very abortive fashion, right? They sent out these emails, the famous “what are the five things you did this week?” emails.

And my understanding, based on some reporting—I don’t have any information, just based on public reporting—was that they had set up an open-source model, the Meta Llama 2 model, and they were trying to feed—or were in the process of feeding—these emails through that. I don’t know how far they got. I mean, DOGE was a sort of shambles, largely throughout. But technically, it’s not that hard.

It’s certainly a lot less data than a lot of these. I meant the DOGE as an organizational model, because there was a lot of humans in the loop there.

So here’s a different question: Why is this necessarily a bad thing?

So it’s not necessarily a bad thing if—well, okay. So as long as your president’s great, maybe it’s a great thing.

Well, I’m wondering how much you worry about this if the president is non-virtuous, or you’d madly disagree with the president’s policies, and how much you’re okay with it if you think the president is virtuous and you love the president’s policies. The president’s elected; the bureaucracy isn’t. This is another old theme in presidential history: presidents—Trump is not the first, and he won’t be the last—try to get control of the bureaucracy and impose his will on the bureaucracy.

And a lot of—you know—FDR wanted to do that. Thomas Jefferson wanted to do that. So I’m just wondering: is this necessarily a bad thing, or could it be seen as a good thing depending on the president and the policy aims?

I mean, it depends on whether you care about process or not, right? And, you know, I think one pathology of lawyers sometimes is that they forget that normal people aren’t quite as obsessed with process as lawyer robots are.

You know, I tend to think that it’s a useful thought experiment and a useful disciplining function to ask, okay, this power—if it was wielded by the other guy—how would I feel about that? And I think, at least in our age of partisanship, where people seem to hate the other side much more than they actually like their own side, that any technology that could so massively increase the powers of the president, I think that should be of bipartisan concern. Because, yeah, think of all the good things your guy could do with it—but think of all the really terrible things that the other guy could do with it.

So, yeah—again—I mean, if all you care about is the results within a particular four-year stretch, it’s a great thing, right? But that’s no different. I mean, this is just a special case of: do you like dictatorship or not, right? And sometimes there are good dictators, but the problem is: what about the next guy?

Why dictatorship? The president’s elected, and let’s assume that the AI is constrained by law. If the AI is constrained by law and the president’s elected, I don’t see why the dictatorship argument is right.

It’s really whether you like the unitary executive or not, because this is the unitary executive on steroids.

Yeah, sure. And by the way, every president likes the unitary executive to some degree—to some degree. And so I think the “AI is constrained by law” is doing a lot of work here.

So one problem is that we’re not sure how to do that, right? So there’s this idea—and there are some really interesting, smart academics who have been trying to build this out—of what’s called law-following AI, which is the idea that just as you can train an AI system to refuse to respond to certain prompts or refuse to take certain actions, you should also be able to get it to refuse to take unlawful action, right? Conceptually, that makes a lot of sense. How you implement that is much, much less obvious. These deep questions of alignment and so forth.

So there’s a technical question whether you can have a law-following AI. There’s another question—and this kind of gets back to the unitary executive part here—of whether Congress is going to be able to sufficiently regulate these systems such that they will, in fact, be law-following. I think there’s some interesting constitutional questions over the extent to which Congress can do that. I think there’s certainly going to be political will questions every time Congress will do that.

And then, of course, there is also just the more philosophical question of how comfortable is one with the unitary executive, right? Does one think, as Chief Justice Roberts has said in several of these recent opinions, that because the president is the only individual—other than the vice president, but whatever, no one cares about the VP—since the president is the only person who was elected by the entirety of the nation, he has a special democratic pedigree? Or as Justice Kagan kind of said in her “Presidential Administration” piece before Roberts did?

To what extent do you think that that is the highest essence of democracy, right? Now, you can think that—but then you’re going to have the issue of democracy policy swinging that massively between administrations. Maybe that’s okay.

There is, however, an alternate view that democracy is a much more—the democratic will is a much more complicated thing. It’s much better reflected in a balance between the president and Congress and the quote-unquote deep state—that that is a much better expression of democracy in getting at the people’s preferences—in which case, this is not a great thing.

But at this point, the argument just collapses into what do you think about the unitary executive and democratic theory more generally?

Okay, two more questions. One is, doesn’t your argument depend on assuming away alignment problems in AI?

And don’t we need to know something about that to have confidence? Wouldn’t the president need to know about that to have confidence in using these techniques?

Yeah. So it doesn’t assume away alignment problems, but the assumption it does make—which is a non-trivial one, but I think a much weaker assumption than assuming away alignment—is that AI systems are more aligned than humans, right?
So I just want to keep hammering this point home, which is: the question is never, is the AI system perfect? Or does the AI system meet some very, very high threshold?

The question is, is the AI system better than the current human alternative?
You know, we’re dealing with this debate right now with self-driving cars, for example, right? And people will point out that, you know, we haven’t solved self-driving cars. They get into accidents. They kill people. You know, they killed this cat in San Francisco—Kit Kat—and it was like a big deal. But that’s not the question. The question is relative to humans.

And so the problem for presidents and for rulers generally—which is why there’s a version of this article or this argument that you could write in China or Russia or North Korea or any government across the political spectrum—is it’s a principal-agent problem all the way down, right? The president is just one person, only so many hours in the day, right? So he relies on his close subordinates, and they rely on their subordinates.

And then you have a massive tree, and suddenly you have three million people at the very end. So it’s not, “Is it perfectly aligned?” It’s, “Is it better aligned than humans?”

And in answering that question—fair point again—I just come back to something I said before to consider, which is: it’s not obvious to me, given that uncertainty, that you don’t want to delegate, and then delegate, and then delegate even with these AI tools, because you might have more confidence in the AI tools’ decision-making authority at a lower level of greater expertise that you trust.

So to the extent that you talked about an oracle that’s deciding everything, it strikes me that even in an optimal world that may not—it may be multiple oracles embedded at different levels of the bureaucracy, based on some calculation made in the White House, probably with the assistance of AI.

Yeah, absolutely. And there’s nothing—I still think that would strengthen presidential control. Because even if you just have the oracle—

My point is, I think that would enhance presidential control over just having the presidential oracle.

Yeah, yeah, absolutely. You don’t need the president—I do think of the presidential oracle. I think of, sort of, the presidential oracle. Yeah, the presidential oracle is least useful in the Executive Office of the Presidency, right? Because by that point, the decision set is a more manageable set of decisions.

You need to train the presidential oracle in the White House because you need the president’s preferences. But yeah, where it has the most bite is when you send it out to, like—you know—the Ramsey County, Texas—or whatever, that’s a state position—but you send it out to sort of the random, kind of deepest, darkest corners of the bureaucracy.

Yep. Last question. I’m very skeptical that the U.S. government has the competence to execute this, especially in the next three to five years.

Am I wrong about that?

I mean, certainly not at the kind of totalizing level that the sort of thought experiment suggests. I think there are going to be some areas in which this is more easily implementable, right?

I think if you take sort of discrete adjudicatory places, you could slot this in, right? You know, right now it would not be rocket science to come up with a model that you feed every asylum application through to do algorithmic scoring. And you can tune that—

You know, you can tune it all the way down, right? Say no asylum if you’re Trump; you can tune it all the way up if you’re President AOC. The point is not about Trump.
So yes, I think it is fair. And I think—look—I think this is a fair critique of breathless claims about AI’s impact generally, which is that AI enthusiasts—of which I am definitely one—get really enamored by the technology, and then they kind of forget about all the sticky frictions.

So it’s totally possible this will take a long time. But again, especially given how overpowered the executive already is, I think, relative certainly to Congress—we can have an interesting conversation about the judiciary—but certainly relative to Congress, even incremental increases in power are quite dramatic.

And then, you know, of course, you never know—maybe the AI gets smart enough that it figures out the one neat trick to self-propagate its way through the entire executive branch.

Okay, Alan, thanks very much. There were a lot of issues in your lecture we didn’t talk about. The lecture can be found on Lawfare. It’s called The Unitary Artificial Executive, October 30th.

Thanks very much.

Thanks for having me, Jack.

Mentioned:

• Walter Stahr, Stanton: Lincoln’s War Secretary (2017)

• William Howard Taft, Our Chief Magistrate and Its Powers (1916)

• Theodore Roosevelt, An Autobiography (1913)

Thumbnail: Republican President candidate William Howard Taft, outside his campaign train in 1908. (Everett Collection, Shutterstock.)

This is an edited transcript of an episode of “Executive Functions Chat.” You can listen to the full conversation by following or subscribing to the show on Substack, Apple, Spotify, or wherever you get your podcasts.

Bob Bauer: Good morning, Walter.

Walter Stahr: Good morning. How are you doing, Bob?

Very well, thank you, and I hope you are too. I want to introduce you to our Executive Functions viewers. Walter Starr is an historian of considerable note.

He has a background as a practicing lawyer, both domestically and internationally, both in private practice and in government. And then, fortunately, he turned to full-time writing of history. He has done so with great success and has written works on William Seward, John Jay, Sam and Chase, and now is working on a biography, which will be published in 2026, possibly on Election Day 2026, about William Howard Taft.

And Walter and I met, actually, through a mutual friend who was very instrumental in my appreciation of history, my high school history instructor, Jack Kearney. And I was very glad that Jack made that introduction. And so that’s what brings us to the current day.

And the Taft story is an important story because what we’ll focus on today is how Taft’s presidency was distinctive, specifically through the way he thought about and later articulated views about the exercise of presidential power. It’s a model of sorts, a different model of thinking about the presidency that I don’t think is generally well understood.

But let me begin, Walter, by asking you the following. The understanding of Taft, of course, is that he had an exceptional career and was both president and chief justice of the United States Supreme Court. But he also seems to have had positions as a legal advisor in the executive branch, virtually any one, any such position that anybody could hold. Can you walk us through that exceptional history? Because I think it bears on what he learned about and then came to understand to be the right way to think about the presidency.

Sure. I should mention that we shared Jack as history teacher, and Jack also was the person who introduced me to Washington in the spring of my senior year at Exeter. He was the head of the Washington intern program, but tasked with the almost impossible task of keeping two dozen young interns out of trouble and gainfully employed.

Taft really was not much older than a college student when he started his career of public service in Cincinnati, Ohio, as a junior prosecutor, and then grows very quickly. He was the youngest person ever to be solicitor general of the United States. And that job in those days included not just what we think of as the solicitor general’s job, representing the United States in the Supreme Court, but also what used to be your job, Bob. There was no White House counsel.

And so, when President Benjamin Harrison wanted a legal opinion, he turned to his friend, the attorney general, but he turned often to his young solicitor general, William Howard Taft, who was in the White House a lot in the Harrison administration. So the Congress, to relieve the crowding in the Supreme Court, creates the Circuit Courts of Appeal. He’s a circuit judge for eight years in Cincinnati. He’s really, throughout his life, he’s sort of a judge in his bones.

Then is our governor of the Philippines for four years. So sees Washington from a very long distance at that point, but then comes back and sees it very up close as Theodore Roosevelt’s war secretary, member of the cabinet, and really kind of almost deputy president for Roosevelt.

You know, like every president down to the late 20th century, Roosevelt didn’t rely on his vice president, almost never saw his vice president, whereas Taft was with him all the time, talking about politics, about the administration, as well as the more narrowly defined job of the War Department. So by the time Taft is elected president in 1908 and takes the oath in early 1909, he’s seen a lot of presidents up close and personal, and he’s had a lot of other experiences as a judge running the Philippines that give him very decided views about, you know, the American constitutional system.

So in summary, and then we’ll get into it in a little bit more detail, how did Taft come to think about the exercise of presidential power? And in particular, since you’ve mentioned Theodore Roosevelt, can you describe, first of all, the difference that Roosevelt saw in their approaches, his approach and Taft’s approach to presidential power, and what you actually thought that difference amounted to?

Yeah. Well, we’re blessed in that not only were, you know, Taft and Roosevelt both presidents, so we can kind of look at how they administered, but they both wrote about this issue. First, Roosevelt puts pen to paper not long after the disastrous, from Taft’s perspective, 1912 presidential campaign on an autobiography, in which Roosevelt derides what he calls the Buchanan-Taft view of the presidency, linking Taft with the person who at that time was viewed as the worst American president, James Buchanan, a view that, in Roosevelt’s view, sort of only was willing to do something if you could find a provision of the constitution or a statute that authorized it.

No, Roosevelt said no. His view was that unless there was a prohibition, a clear prohibition, the president should do what he thought best for the American people.

Taft disagreed with that, first in lectures at Yale, and then in a book that he published not long thereafter. He said, no, no, this notion that there’s some undefined residuum of presidential powers is all wrong. The president can only do what Congress has authorized him to do or what the constitution itself fairly interpreted authorizes.

That term, undefined residuum of presidential power, just to be clear, is the language Taft used to describe Roosevelt’s view.

Correct. He used that to criticize and warn against the Roosevelt approach of sort of really going first for what the people need, and then, unless there’s some prohibition, going ahead and doing it.

The concrete example that they both had in mind was while Roosevelt was president, he set aside, that was the vocabulary, tens of thousands of acres of undeveloped land and said that they could not be developed. There was no statutory authority whatsoever for this. One of the first things that Taft had on his to-do list when he became president in 1913 was to get legislation out of Congress to, in essence, ratify what Roosevelt had done.

He didn’t, as Secretary of War, he didn’t raise his hand and say, no, no, don’t do this. But he thought that this was not going to stand up in court if challenged unless there was a statute. And he did get that statute through Congress.

Let me ask you a question about that move to Congress for authorization. Did it come under any kind of pressure? You had said he had it on his to-do list, but did he feel compelled by any kind of external push? Criticism and the like? Or was it something he thought he should do in, if you will, cleaning up the legal authority behind what Roosevelt had done?

I have not seen any sort of external, and in particular, as best I can tell, there was no lawsuit wending its way through the courts headed up towards the Supreme Court. You know, developer versus Roosevelt. It was just, you know, as a kind of dot the I’s and cross the T’s kind of judge, Taft felt that this was something that needed doing or else there would eventually be some lawsuit and the government would lose.

So let’s talk for a second about Taft as lawyer—very engaged with judging, very engaged with lawyering—and Roosevelt, who also went to law school but apparently hated it. The question I’m sort of moving to here is, did Roosevelt have a legal advisory apparatus that he proposed any trust in or cared to consult with any degree of regularity?

Well, he had, you know, the same people, you know, they were not the same people, but you know, there was an attorney general, there was a solicitor general. But unlike Taft, who would, you know, as a former Justice Department official, would, you know, refer questions to the Justice Department and take their advice, Roosevelt, as best I can tell, didn’t rely heavily upon lawyers, didn’t value legal advice terribly highly.

One example that springs to mind, there was a rebellion or revolution really going on in Cuba and Roosevelt was about to send Taft down to sort it out. Taft wrote him a three-page letter kind of explaining why he, he, Taft, didn’t think it necessary to seek authorization from Congress. There was already something in place called the Platt Amendment that Taft thought sufficient. But he ends by saying, you probably want to consult with your attorney general and perhaps even with Congress about this matter or key congressional people.

To which Roosevelt responds, I shall not confer with either the attorney general or with Congress. No, if an action becomes necessary, I shall take action and establish a precedent for presidents in the future. And that, I think, is probably, you know, a good example of the general Roosevelt attitude towards lawyers.

Again, we should be clear, we’re talking here about Theodore, not his distant cousin, Franklin.

Yeah, I should mention his distant cousin, Franklin, wasn’t all that crazy about lawyers either. And his attorney general, Robert Jackson, later on the Supreme Court of the United States, of course, wrote in a book of memoirs that Jackson had no particular use for lawyers, not that he didn’t involve them or consult them, but he did not view them as particularly useful contributors to the pursuit of the policies that he was interested in.

Let me then ask a question that will come up in anybody’s consideration of Taft. And that is that he’s very prominently associated, of course, because he wrote the opinion in a case on presidential removal power, the Myers case. And there, and that case is frequently cited, as indeed it would be, for a very, very robust view of the president’s authority to fire senior officials in the government.

And so somebody might ask, well, he wrote that opinion, doesn’t that imply, maybe he would not be in contemporary terms an advocate of a unitary executive view as we currently understand it, but it certainly seems like he had in that respect a opportunity to curtail presidential power. And in fact, he did not. He authored an opinion that’s cited by presidents who were looking to expand on that power.

Can you give us some background on the Myers case and how that doesn’t in fact affect your analysis of the Taftian view of the presidency?

So Taft becomes chief justice in 1921. And one of the cases that, I mean, indeed the most important single case of his time on the court is this Myers case, which actually, as is often the case with cases from that era, it has a long history because cases took a long time to get to the court and then win their way through the court.

And as you say, the opinion is full of things that reflect his own extensive experience as both a subordinate and then president. He talks about how it’s impossible for the president to do everything himself and he has to rely on subordinates. He has to rely on reliable subordinates. It just can’t be, Taft said, that those who drafted the constitution meant to hobble the president with advisors upon whom he could not rely.

So as you say, it does in that respect reflect a broad view of presidential power. I don’t think though that it’s inconsistent with his view that you have to trace to a statute or the constitution. He does spend a lot of time both with the language of the constitution and with the how it was understood by those who had drafted and were tasked with first implementing it.

There’s an interesting bit that’s not in the Myers opinion that I think shows that he didn’t envisage the president as having the power to remove every single member of the executive branch. In the late stages of the process, as those in the majority were reading the dissents, one of those in the majority, Harlan Stone justice and later chief justice, sends a note to Taft saying, look, let’s deal with this argument in the dissents about the civil service laws by simply saying that the president has the ability to remove anyone regardless even of the civil service laws.

And Taft, unfortunately, we don’t have a written answer from Taft to that memo, but we sort of know what Taft thinks from the opinion. Taft disagrees with that. He’s been a long champion of civil service reform going back to the days when as a junior revenue officer, he was besieged with requests to hire what he thought were unsuitable people for his revenue office.

And so he does not think that the Myers principle, if you will, invalidates federal civil service protections and exactly where the border is is left unclear in the Myers opinion. I mean, they don’t have to draw the border precisely. They simply draw it such that Myers, who was what was then known as a first class postmaster, which were sort of the eyes and ears of the president around the country, that Congress could not prevent the president from removing him without Senate consent.

Tell us a little bit about another episode that could also shed some light on how he approached the issue in Myers and what it means generally for his view of presidential removal power. He addressed a case you’re very deeply familiar with, I’m not, that involved removal of the members of an entity known as the Board of General Appraisers, I believe. Can you talk a little bit about how he approached that and how do we build that into how we understand his conception of presidential authority?

Sure. So the Board of General Appraisers was a quasi-judicial body. These were not Article III judges. They did not have life tenure. They could be removed by the president and it’s a variation of that fairly standard wording about inefficiency, neglect of duty, et cetera.

And when he heard rumors, this is in the summer of 1912, so the last year of his presidency, when he heard rumors of this, he didn’t just immediately fire those accused of mismanagement. No, he set up a little committee of three government lawyers and when one member of this committee got sick, he appointed someone whose name is very familiar to us today, although it was utterly unknown at the time, Felix Frankfurter.

Frankfurter was at the time a junior lawyer in the War Department and this committee took evidence, it held hearings, it actually heard from those accused of misconduct and recommended that two members of the board should be removed by the president and on the last full day of his presidency, so like the eve of the inauguration of Woodrow Wilson, Taft goes ahead and does that.

In one case, the person was apparently an alcoholic and thus frequently absent for meetings and in the other case, and the bit that really got the headlines, there was an official who was both writing opinions in a way that benefited his son who was a New York City lawyer specializing in tariff cases and also requiring the Baltimore and Ohio Railroad to stop its trains at his summer house so that he could get on and off at convenient moments. And that may not be the worst thing that federal officials have ever done in the long history of federal corruption, but for some reason, that really grabbed the attention of the reading public.

Anyway, Taft accepts the recommendations of the Frankfurter committee and removes these two people on his last full day in office. Again, I think it doesn’t directly translate, but it certainly indicates an adherence to process in removal in Taft’s view of how these things should be done.

And on his part, it seems he took seriously that Congress has specified certain grounds for removal and so he should go about in a judicious way determining whether those grounds in that particular case existed. So it was Taft as lawyer again.

And Taft as judge.

And Taft as judge. Yes.

Right at the same time at the very end of his presidency, there’s a statute passed to increase the requirements for immigrants—in essence to impose an English literacy requirement for all immigrants to the United States, and the newspapers analogize this to a court hearing. Taft organizes a half day of a judicial hearing at the White House in which he’s not in robes, but he’s almost in robes, listening to arguments on both sides.

He ultimately vetoes that bill and the veto is very narrowly sustained, actually mainly with Democratic votes. He’s a Republican president, but he gets that veto sustained with the votes of Democrats who are, I think, more rightly alive to the sort of racist and xenophobic motivations of that literacy requirement.

So let’s turn now to another, of course, both of contemporary significance, lasting significance in all debates about presidential authority, and that’s the war-making power. And I’d be interested in Taft’s history and in particular, we’ve talked about his differences with Roosevelt. Let’s also bring into that, if we could, his differences with Woodrow Wilson.

Sure. Well, and we also ought to bring in his qualms and differences with William McKinley over the Spanish-American War. In the spring of 1898, when the Maine explodes and war fever really rages in America. I think it’s fair to say that Taft is way on the uninfected and cautious side of the spectrum.

There’s a letter from him to McKinley. So this is from a sitting federal judge to the president, sort of praising McKinley’s caution and hesitancy and saying that, you know, he should remain cautious and hesitant. This is on the very same day that Roosevelt writes to a family member saying that he (Roosevelt being at the time the assistant secretary of the Navy), saying that he had told the president and the whole cabinet that the only course was war.

So he’s cautious and in that, during his own presidency, there are a number of incidents where we could have, you know, viewed it as legitimate cause for war with Mexico. He says, no, no, we are not going to war with Mexico.

And then during the Woodrow Wilson presidency, when Wilson really does go to war with Mexico, although in a limited way, but with both the invasion of Veracruz and then the Pershing expedition in Northern Mexico, Taft is quite critical in private letters, although he does largely observe the conventions about, you know, former presidents not criticizing the current presidents, at least on matters of foreign policy.

He’s also critical in the run up to the First World War. And here he is a little bit publicly critical. Wilson spends a long time before he finally sends out a note to Congress saying, okay, okay, come to Washington and we’ll talk.

Taft, who’s been spending the time sort of traveling around the country, thinks that that showed a lack of respect for Congress and a failure to appreciate that he wasn’t just going to need a declaration of war, he was going to need various bits of legislation, draft legislation, armaments legislation, et cetera. And this is going to take time. And so he felt that Wilson had failed in that and at some other junctures to accord sufficient respect to Congress.

Now, when Roosevelt criticizes Taft’s conception of the presidency, he, as you said earlier, you know, he puts himself in Lincoln’s camp, he puts Taft in Buchanan’s camp. Would you say as we, and of course it’s unimaginable how much has been written about Lincoln, and it’s a huge topic, but is Roosevelt right to claim the Lincoln legacy?

One view, of course, has been, yes, Lincoln certainly aggressively exercised the powers of the office, but he did so in what was an existential set of circumstances for the country and the question of whether the union would survive and whether slavery would be brought to an end. Would you say that Roosevelt, if you will, is skipping over that point or that he does have a conception of the presidency that is comparable to Lincoln’s, even again, you know, setting aside the emergency wartime conditions?

As you say, Lincoln is a big topic. And I’ve contributed a certain amount to that pile of books, although I haven’t written about Lincoln directly, obviously there’s a fair amount, I’ve written about three members of his cabinet, so there’s a fair amount about Lincoln and the war.

In a book that you didn’t mention about Edwin Stanton, the Secretary of War, I spend a fair amount of time dealing with what I find very troubling arrests of editors and other opponents of the war effort. And, you know, these are not, you know, we don’t have like a little piece of paper from Abraham Lincoln to the war secretary saying, please arrest Clement Vallandigham, Vallandigham being an Ohio former member of Congress who’s thinking of running for governor. He’s actually thinking of challenging Lincoln for the 1864 presidential nomination.

Vallandigham’s arrested by a union general, but Lincoln is well aware of the arrest and he doesn’t send a note saying, you know, release him, he didn’t do anything other than give a speech criticizing me. No, he allows the trial to proceed and Vallandigham is sentenced to spend the rest of the war in prison and then Lincoln, you know, commutes the sentence to exile.

But we still, would we really consider it, how to put it, a proper way of treating one’s critics for the president to allow them to be tried by military commissions and exiled to Canada? We, you know, I think a lot of us left and right would find that very troubling and yet that’s part of Lincoln’s legacy.

So long-winded way of saying, I think Theodore Roosevelt has a point that Lincoln is a key figure in the line of what we might call the more aggressive presidents, you know, which you might trace from Lincoln through Theodore Roosevelt to Woodrow Wilson to FDR and down to the imperial presidency.

Well, this is a topic that obviously is going to be debated forever and today. I do want to close here by asking you to talk a little bit because I think it is exceptional about Taft’s writings on the topic and just so people who want to pursue it can look for them. These are not, as I understand it, you know, just casual observations interspersed with memoirs.

I mean, he really attempted, I take it, to articulate a view of presidential power, the executive, and I just wanted to put an exclamation mark on that so that people were aware of it and those who were interested in this, again, as one model of thinking about the presidency that they could go look for it and read it for themselves.

Sure. Well, the particular book I was referencing is called Our Chief Magistrate and Its Powers. It was published in 1916 and so it’s out of copyright so you can find it relatively easily on Hathitrust or other places.

And actually, that’s not the only, he tackled that topic again. I think there’s another book from that period between the White House and the Chief Justiceship with a similar title. And these, the general flavor of these books, I mean, it’s interesting. They are largely about the limits on the powers of the president and this former president is talking about how those limits are a good thing.

So, for example, on the subject of congressional investigations, which presidents by and large are not a fan of congressional investigations. No, former president Taft says that it’s a good thing that Congress has the power to investigate the administration. And indeed, one of his Supreme Court cases, a decade later, the Doherty case, is one of the sort of key foundations of the law on that subject.

So it’s not all that hard to find Taft’s writings on these subjects.

So I’m not aware, check me if I’m wrong. I mean, I don’t know offhand that maybe I’m missing something. There is any other president who has written after leaving office, a book essentially of constitutional theory, much less one that is focused on limits on the power of the office that he previously held.

Am I missing something?

No, I can’t think of anyone. Let me, I suppose in fairness to our former presidents, many of them either died in office or died not long after leaving office. So they didn’t have a lot of time for writing.

But yes, it is interesting that what he writes in his post-presidential years is really not a conventional presidential memoir, you know, great big thick book in which the first person predominates. As you say, he writes books that are, you know, sort of learned constitutional books. And he writes on other topics too, but he spends a lot of time thinking about the presidency.

And he has actually some, you know, really excellent suggestions. For example, one of his suggestions, which he makes both in speeches and then in print, rather than having two terms. So the president from day one is focused on his re-election. Why not amend the constitution? So we have, we elect a president, we elect a president for six years and he or she has no right to run or serve again.

You know, if you think about it, a lot of problems in American history, Watergate, you know, might not have happened if we had, if that rule had been in place because we would have had president Nixon for six years and he wouldn’t have had to run a re-election campaign.

It is a remarkable legacy and I really appreciate you taking the time to discuss it with me today. I think Taft gets overlooked and a big part of that has to do with the particular roles and circumstances in which, you know, we think about Woodrow Wilson, Theodore Roosevelt, but in any event, a remarkable figure and a president that was prepared to think hard about the office and not only in, if you will, completely self-interested terms. That is, as an opportunity to exercise the most possible power that he could amass.

So thank you very much, Walter. And again, this Taft biography is going to be published next year, possibly on Election Day, 2026. And so I’m looking forward to reading it. Thank you for coming.

Thanks for having me, Bob.

Mentioned:

• “What Are We Living Through?” by David Pozen & Jedediah Britton-Purdy (Boston Review, Oct. 15, 2025)

• “Hardball and/as Anti-Hardball” by David Pozen (Lawfare, Oct. 11, 2018)

Thumbnail: Donald Trump speaks at an event in Maryland in February 2024. (Jonah Elkowitz, Shutterstock.)

This is an edited transcript of an episode of “Executive Functions Chat.” You can listen to the full conversation by following or subscribing to the show on Substack, Apple, Spotify, or wherever you get your podcasts.

Jack Goldsmith: Today I’m going to have a chat with David Pozen of Columbia Law School who wrote an article with Jed Britton-Purdy called “What Are We Living Through” in the Boston Review about 10 days ago, I think. I thought it was a terrific essay and I thought we should discuss it.

David, in the first paragraph you ask, I’m just going to read a few sentences, what kind of moment are we living through? Is our system of government undergoing a cyclical swing, an existential transformation, or something in between? Nine months into the second Trump administration, Americans confront three very different answers to these questions. And then you offer three answers.

So if you could just tell us what those are, we can start off with the conversation. And by the way, this is a question that lots of people are asking, so it’s great that you have three answers.

David Pozen: Thanks, Jack, for having me on. The essay grew out of Jed and I having conversations, trying to get our bearings and experiencing disorientation where some relatives and colleagues telling you that you have to flee the country, others telling you that that’s Trump derangement syndrome. And it’s always hard to tell what kind of moment you’re living in, but the degree of disorientation seems qualitatively greater now.

So we basically identify three main camps or schools of thought about what’s been happening in the past nine months or so. And should I run them through?

Yeah, just run through each of them and then we’ll talk about them.

Okay, great. So the first, we call authoritarian crisis. And this has quickly become the conventional wisdom among the liberal and centrist establishment. Think The New York Times or The Atlantic. And the idea here is that there’s an authoritarian playbook, as it’s often called. It’s been used by leaders like Erdogan in Turkey or Orban in Hungary. And Trump has followed it line by line. He has persecuted political opponents. Think of Letitia James or Jim Comey or Adam Schiff. He has rewarded political friends. Think of Eric Adams or the pardons for the January 6th riders. He has targeted sites of countervailing power in civil society, like the media and universities and law firms. He’s demonized enemies and vulnerable groups like immigrants and trans people. He’s declared endless emergencies. He’s sidelined the legislature. He’s manipulated official data. He’s preparing to rig elections. He’s preparing to use the military domestically, on and on.

And there are a lot of variants on this diagnosis. But on all of these accounts, there’s been a profound shift in American law and democracy. And both the rule of law and democracy are on the ropes. Some versions emphasize the kind of self-enrichment or corruption aspects of what’s happening and prefer to describe it as kleptocracy or patrimonialism or sultanism. Others emphasize the violence of ICE or the iconography of DHS or the personality cult around Trump and think it’s amounting to fascism. We use authoritarian crisis to stand in for this array of views that democracy is in grave peril here. Should I go on?

Yeah, good. I think we should get all three out on the table.

I’ll be quicker. The second view, we call it more of the same. And if the first view, authoritarian crisis, is mainly associated with liberals and centrists, this is more associated with the left. And it’s the idea that Trump is a kind of lurid symbol and apotheosis of long-running pathologies rather than an agent of historical rupture.

Some of what he’s done is just standard Republican fare taken further—tax cuts for the wealthy and corporations, hostility to abortion and affirmative action, economic environmental deregulation. This is familiar, even if more extreme. Other parts of his agenda intensify bipartisan patterns. Harsh immigration enforcement was familiar from, say, the Obama administration and his record-setting deportation program. The consolidation of presidential power is, of course, a long-running trend you yourself have written extensively about. The growth of the national security state is another long-running trend.

So on this view, the more of the same view, seeing Trump as some kind of shock to the system flatters the pre-existing system, which has long been marked by democratic dysfunction, executive abuses, and racial repression.

And then the third view is the most sympathetic view to Trump and is in a sense the official story of the Trump administration that his advisors and supporters often channel, which is that we’re undergoing a constitutional regime change, we call it. Not an authoritarian crisis, but a shift in the constitutional order.

And this view emphasizes that the U.S. Constitution is the hardest to amend in the world, and so major constitutional transformations typically happen here through political and social movements. And MAGA is such a movement. On this view, just as FDR revolutionized the constitutional order in the New Deal, breaking a lot of norms and grabbing a lot of power in the process, so is Trump.

And the campaign previewed a lot of what Trump is now doing, from cracking down on illegal immigration, to shrinking the federal workforce, to eliminating DEI. And if liberals are shocked and appalled by some of these changes, well, conservatives were shocked and appalled by some of the changes wrought by the New Deal and the Civil Rights Revolution, the Warren Court, and its aftermath.

So in short, liberals waged successful constitutional revolutions in the early to mid and mid to late 20th centuries, seizing hegemonic control over law and culture, and now the right is doing the same. And this isn’t breaking the Constitution; it’s just ending an era. It’s a transition to a new constitutional regime after the prior one lost its legitimacy.

And I map, Jack, I map these three positions not to say that any is right, and they’re certainly not morally equivalent, but it’s an attempt just to parse where people seem to be in making sense of the moment.

Great. So I want to ask you lots of questions about these, including in some of the points you made about them. The first is, so I’ve been struggling with this question: where are we? What is the significance of what’s going on?

And the first thing that struck me when I read your essay is that there are elements of all three of these things going on. And so one question is, can all three be true? And so I guess that’s the question.

Yeah, we—

In some respects?

In some respects, yes, I think all three can be true. And more than that, capture a core political truth. And so it makes it hard to dismiss any of them out of hand, even if they all have vulnerabilities.

The constitutional regime change view, which I think is the hardest for liberals and progressives to get their heads around, I think has an important point—that reconstructive presidencies can and have changed the constitutional order, that some kind of liberal hegemony is not inevitable. It’s a category error to mistake a current constitutional regime or settlement for the Constitution. And the fact that a new administration comes in and outrages political opponents and disrupts settled norms doesn’t in itself mean it’s illegitimate. And indeed, progressives may well want a preferred president to do a version of the same in the future.

So I think the constitutional regime change story has an important point about how constitutional change happens in a country where formal amendments are extraordinarily rare.

At the same time, the authoritarian crisis story, I think, captures an important point that even if a new political regime starts out pursuing constitutional change in more or less familiar ways, it doesn’t mean it won’t end up shattering both constitutionalism and democracy. Authoritarianism is a real possibility even in a mature democracy. There are unmistakable signs of its spread that I began by cataloging. And so it’s not exculpatory for the Trump administration that it’s pursuing regime change if the form it’s now taking is authoritarian.

And then on the more of the same critique, I think the crucial point is that defeating authoritarianism, at least in the U.S. context, demands that we reckon with the underlying forces, social conflict, and economic struggles that brought Trump into power, and just retreating into the prior dispensation isn’t enough. Trump is, to some degree, a symptom, not a cause, of democratic breakdown—although we can debate exactly how much. And so we really need to address the deeper roots of the discontent that led us here.

So ultimately, we think that there’s something important about holding all three positions in mind, even if we personally are more associated with some combination of the first and second.

Yeah. You also say in the essay—I’m reading from it now—like other aspects of American polarization, the competing scripts reflect differences in worldview that go beyond what any fact-checking could resolve. So, while I see elements of all three going on and I think it’s useful to bring all three lenses or scripts to bear, there’s also a sense in which for many people they’re mutually incompatible. I mean, the centrist liberals who think this is authoritarianism and the Trumpists who totally buy the regime change thing and think that they’re basically FDR in reverse—these are, in some sense, despite what we just said, incompatible worldviews, I think. Yes or no?

Okay. I think they may reflect and stem from deep disagreements. I don’t know, “incompatible” maybe—maybe too strong.

A bit strong.

But I guess I would put it this way: I think all of these positions on where we are now implicitly tell a story about what came before.

This may go to the difference in worldviews you’re noting. The stories they tell about what came before are very different, and therefore about where we may be heading. So on the authoritarian crisis account, at least most versions, what we had prior to Trump was a reasonably well-functioning democracy that was very different from authoritarianism, so that we’ve entered a qualitatively new period in our political history.

A lot of the more of the same commentators don’t accept that view of recent American history. They point out that there’s been rampant inequality for many years. There’s been brutal treatment of immigrants and poor minority communities for many years. This whole literature on racial authoritarianism, which has in some sense now been generalized. We have a notoriously undemocratic constitutional system in which the preferences of most Americans basically don’t get factored into the outputs of the government. I could go on.

But on this account, basically, Democrats and Republicans weren’t that different. We basically had a rotation of neoliberal elites prior to Trump. Trump may be making things worse or heightening the contradictions, but we’ve always been a flawed democracy with authoritarian tendencies.

And so these great-man theories of change aren’t helpful or interesting. We really need to address the deeper forces here.

And then on the constitutional regime change view, we also had a flawed democracy prior to Trump—but not necessarily because of the factors I just cited, but because of all the ways in which liberals exerted ideological and institutional control over conservatives and used the courts and universities and the media to push their own hegemonic agenda.

So I would say the incompatible worldviews mainly are a feature of what came before. But as far as where we are now, I do think it’s possible to hold all three and think that Trump is trying to effect the regime change, that he has exhibited unmistakable authoritarian tendencies and designs, and that this transcends Trump and is a story about a larger constitutional rot and democratic breakdown.

I want to pick up on the “trying to affect the regime change.” You noted, quite correctly—I’ve been making this point also—that it’s still very early. I actually think last week may come to be viewed as a pivot point with the tariff case not going well for the president, with the New York mayoral election, with the Nick Fuentes stuff, and other things that have been happening.

I also think that while Trump has been massively disruptive and done some terrible things, his positive reconstructive program—I guess it’s a negative program—but it’s not clear to me how lasting it’s going to be. So I just wanted to get your comments on the significance of “it’s early.”

And then you also had a point that most constitutional revolutions or constitutional moments happen with much more sweeping elections, with congressional support. Trump doesn’t have either one of those things. So I guess I’m wondering—you alluded to whether he’s going to succeed in this kind of revolution. Just give me your thoughts on this.

Yeah, I think there’s radical uncertainty here looking forward too, for some obvious reasons and maybe a few less obvious ones. It’s still less than a year into the presidency. The president’s party tends to lose ground in the midterm elections. We maybe got a preview of that with last week’s elections.

Trump has never enjoyed majority support, and his popularity with independents is at an all-time low, I believe. So all those factors suggest that this could break any number of ways.

But beyond that, there are some political scientists who will tell you that there has never been democratic backsliding of the kind the U.S. is now experiencing in a democracy with high per capita income and a history of peaceful rotations of power—at least four election cycles of peaceful transitions. And in light of that, no statistical model would have predicted what’s already happened: that we would get the kind of democratic backsliding we’ve had. And that makes it harder to predict where we’re going to go. We’re kind of operating in new terrain here—what’s going on.

So where might this lead? And then you added the point that generally when we talk about constitutional regime changes, we talk about what Bruce Ackerman calls constitutional moments, where you have large and sustained majorities that can legitimate constitutional change even in the absence of a formal amendment. And that generally involves transformative legislation, which I don’t think the “big, beautiful bill” is at all.

And so historically and democratically, Trump is on really shaky ground in saying he’s pursuing a revolution or counterrevolution, given the degree of unilateral action he’s taking and the lack of either—not even supermajority support, even simple majority support—in the population and in a big legislative program.

So where is this going to go? I, of course, don’t know. I could see one future where Democrats take back the House in 2026, the presidency in 2028, and basically by 2029 we’re restoring norms gradually and government capacity and important social programs.

But of course, we could have Trump’s handpicked successor winning in 2028 under less-than-free and fair conditions, and the kind of oligarchic repression deepens. Or we could have Trump engineering some version of the Reichstag fire and declaring an emergency and deploying the military domestically, and we’re galloping toward fascism in a few months.

So I think the wildly discrepant diagnoses on where we’ve been are paired with extreme uncertainty about where we’re heading.

Let’s do one more question about radical constitutional change, constitutional moments, reconstructive presidents, and whether he does need Congress. His program is negative. It’s destructive. It’s deregulatory. It’s incapacitating. He’s trying to delegitimate institutions.

I think he’s succeeding in that score, and I’m wondering whether that might add up to a type of—I mean, I don’t know where it leaves us—but whether he can make those kinds of fundamental constitutional changes without having to have Congress on board with it, without even having to have the court on board for all of it. It just seems to me to be a different kind of constitutional regime change.

That’s interesting. I haven’t thought enough about that. I think you’re right that there’s an asymmetry there in what presidents can do on their own in building or breaking. And some of his moves will have long-term consequences, especially in stripping agency capacity.

That said, I guess it still seems to me that in the absence of bipartisan support and major legislation that the court then upholds, what he’s doing is more easily unwound by a future administration than if he were able to muster an actual New Deal–style reform program.

Let me give you an example, and we don’t need to continue on this, but the Justice Department. I think it’s going to be very hard to reconstitute the Justice Department in the kind of idealized fashion that we imagined it after Watergate for 50 years—where it really was separate from White House control, where it really did have a type of independence, where really its ambitions and largely its success was to remain an independent, quasi-independent, apolitical institution.

I think that’s going to be extremely difficult to reconstitute. And I think there are going to be other institutions like that that are going to be difficult to reconstitute. That’s kind of what I meant. I agree—maybe the Democrats, if the politics turn out well, can reconstitute USAID, etc. But I just think some things are going to be hard to put back together and will amount to something close to constitutional differences, at least in terms of how Article II operates.

But I still think—right, that could break a number of ways—even if it’s right that the Justice Department, as it’s been known, will be hard to reproduce. You could imagine a future effort at a kind of third reconstruction, if you will, where we basically rethink these institutions and don’t just try to restore them. And I’m not sure this works as well for the Justice Department, but it’s just to say it may be true that, given the damage Trump has done, we’re going to need to not just revert back to the status quo ante but reimagine structures of government.

But we’ve done that before. That happened, in a way, in the area of national security and war powers in the 1960s and 70s—particularly the 70s—when we got the War Powers Resolution and the rise of the intelligence committees and FOIA and FACA and transparency laws and all those other reforms. But just to say, after a moment of perceived breakdown there with the Vietnam War and Nixon, we didn’t just reproduce or try to bring back the old institutions, but we remade them. And that may be a necessary project to get beyond this in the future.

I just want to make one point—you don’t need to comment unless you want to—and then I have a couple more questions. It really depends on what—let’s say the Democrats win in 2028—what happens next. There seems to me to be growing evidence that, again, way too early to tell—it seems to me to be plausible to think that Democrats in charge of the presidency, especially a populist Democratic president, are not going to be engaged in imaginative reconstruction. They’re going to use the tools that Trump has shown them how to use, I predict, to weaponize or at least use the presidency aggressively to achieve their aims against their enemies.

This is a prediction. My point is, you may be right. A lot of it depends on what the next round looks like.

Yeah, I think that’s undeniably true. I’ll just say right now, I think a tension that progressives or Democrats or liberals face—opponents of Trump face—is that some of the standard prescriptions for responding to authoritarianism, what some people call the “anti-authoritarian playbook,” are precisely about limiting possibilities for this kind of escalating cycle of tit for tat.

So they emphasize: defend the courts at all costs—those are the bulwarks of the rule of law. Try to limit government capacity to do damage. Try to form broad bipartisan coalitions with any ‘Never Trump’ conservatives you can find. And basically have a minimalist politics built around stopping the bad man, Trump.

That set of prescriptions is just very different from what I think a standard progressive response to the more of the same diagnosis is, which is: The system is badly broken. The problems are fundamental. So we need fundamental reform. We need to get rid of the filibuster. We need to do court reform. We need economic populism in some ambitious sense.

And that kind of much bolder set of responses to Trumpism potentially plays into some of what you’re describing.

Yeah. And I’ve been hearing—again, not doing any scientific analysis—but hearing both, and mostly the latter, from the people in my... Eric Holder was on a couple of days ago saying, we have to pack the courts when we get the power back, for example. Anyway, a couple more—but I’m not suggesting that’s going to happen. I’m just saying it’s a possible future. And I think you’re suggesting that that would not be the optimal future. Is that what I’m hearing you say?

Well, I mean, I would say in the piece, Jed and I talk about how what seems to be needed for those opposed to Trump is a merger of the anti-authoritarian—and it is happening already, I think—the anti-authoritarian and the more of the same camps, where the more of the same camp recognizes that something genuinely authoritarian is happening here. We do need to shore up the rule of law and basic democratic institutions. But the authoritarian crisis people, for their part, recognize that there are deeper roots here.

And so we have to address plutocracy and gerrymandering and some of the obvious sources of our current dysfunction. And so that doesn’t yield a clear set menu of reforms like what you do with the courts, but it suggests we’ll probably need a combination of preservative, classic anti-authoritarian moves and bolder reforms that actually speak to the legitimate grievances Trump has been able to exploit.

You say in the piece, I think it’s maybe related to what you just said, any hope of containing its authoritarian threat in the United States depends in part on driving a wedge between Trump’s general policy agenda, which has plurality support in key respects, and his most extreme abuses of power. I’m not sure I understand what that looks like.

I know that’s vague. I think the idea we’re getting at there is some aspects of Trump’s program are neither clearly authoritarian nor unpopular. Shoring up the border, even if what he’s done with ICE domestically has been highly unpopular. Some of his critiques of wokeism, some of his economic program, are seen within the sphere of legitimate policy changes that presidents make all the time.

And what is really distinctively both authoritarian and extremely unpopular, if polls are to be believed, are some of his more—his cruder—power grabs, some of the corruption and getting his private jet from Qatar, or his trying to deport people without any due process or hearing, flirting with defying judicial orders. These are all wildly unpopular, if polls are to be believed.

And so the idea is basically to isolate what’s truly a distinctive threat to the rule of law and democracy here. And that is something you can build a supermajoritarian coalition to oppose while conceding what is legitimate in the Trump program. I know that’s vague, Jack.

No, that’s good. I’m going to read one more sentence, which is kind of a wrapping-up point, and give you the last word. You say, “it is even harder at present to envision a constitutional modus vivendi spanning all three camps, with recognition of legitimate disagreement bounded by accepted limits on the stratagems of power. In the absence of one or the other form of common ground, there will continue to be sharp and at times surreal disputes over the most basic civic questions, including whether the republic is flourishing or disintegrating.”

For me, that kind of resonates with the beginning of the essay. And I’m just—where are you right now? Where does this leave us?

Right. I’ll just say two things maybe about that final passage. So one, the point about the surreal situation at the very end is a reference to a deeper form of polarization, if you will, that we’re now living in—where beyond just our disagreements about particular policies or candidates or parties, we no longer can even agree on the basic shape of society and the kind of moment we’re living in. So that epistemic polarization bordering on epistemic derangement is the proposition there.

On the how to get out of this, we basically just gesture toward two routes to overcome Trumpism. One would be a kind of broad coalition making common cause around some probably modest agenda that would sideline Trump and his worst abuses. That we say is just hard to imagine right now, given the degree of polarization we’re living under, Trump’s degree of control over the Republican Party.

The other route would be that the first two camps, the anti-authoritarian and anti-oligarchy positions, find a way to work together and construct a lasting majority themselves with a program that marries the core critiques of both. And that seems more plausible to me, although also extraordinarily hard to get there.

Maybe, Jack, since you had emailed me about gerrymandering, I’ll try to make it more concrete by just a note on that. So what do we do about gerrymandering? Trump has, of course, encouraged gerrymandering in red states, and blue states are now responding in kind. So the first solution alluded to at the end of that passage you read, where Democrats and Republicans get together, would look like a kind of détente where everyone agrees not to gerrymander. That’s almost impossible to imagine right now.

The other version, though, of overcoming rampant partisan gerrymandering and all the distrust in elections and public institutions that that tends to foster, would probably involve—this is Joey Fishkin and Rick Hasen have been writing about this recently—Democrats playing hardball. And if they were to control, as Hasen and Fishkin write, Congress and the presidency in 2029, abolishing the filibuster so as to force through an anti-gerrymandering reform—some kind of independent commission model for all of the states. And so that would be another way to get beyond one specific problem associated with gerrymandering. But it would involve aggressive, coordinated action by the left and the center rather than a bipartisan compromise.

And I’ll just say, I think it was for Lawfare some years ago, I wrote a piece on hardball and anti-hardball, where I suggested generally that anti-hardball are policies that reduce the stakes of partisan conflict and forestall or foreclose tit-for-tat cycles of escalation. And I say generally the best case for hardball is if it’s in service of anti-hardball.

And so that might be one way to think about it. I wouldn’t favor Democrats ever pursuing retribution for its own sake or imitating some of the vices of Trumpism. But if there’s a case to be made for aggressive action on the Democratic side that’s pushing boundaries, I think it would be best justified in the register of anti-hardball—that we need to get beyond partisan warfare, and some wildly popular moves like independent commissions are unattainable in the absence of aggressive action to get there.

So, Jack, this doesn’t add up to any kind of set of prescriptions or how to get beyond where we are now. It’s much—Jed and I felt like we’re much more able to map current divisions than to offer a way out of them. So these are really speculative thoughts.

Everything in the essay is super interesting. I highly recommend it. It’s What Are We Living Through? It’s in the October—it’s dated October 15th, 2025—in the Boston Review.

David, thank you so much. That was great.

Thanks, Jack. Appreciate it.