A transcript of today’s podcast appears below. It has been edited for flow and clarity.
Andy Craig: Welcome to The Reconstruction. I’m Andy Craig.
At The UnPopulist, we do a lot of tracking of the daily storm of assaults on freedom and democracy. But sometimes you need to zoom out, take a breath, and look at the bigger picture. So every month I’ll sit down with people who actually know the terrain—scholars, historians, experts—not just pundits with hot takes and opinions.
We’ll explore how we got here and more importantly, where we go next. In other words, a reconstruction of what’s happened; a forensic analysis, but with an eye to rebuilding our institutions for the future.
The Constitution’s text provides that the president “shall appoint by and with the advice and consent of the Senate” various officers, and allows Congress to exempt some inferior officers within the executive branch from that requirement. But the Constitution is silent on who can fire them—i.e., the removal power.
So now we have a president whose catchphrase is, “you’re fired”—and he’s putting that to the test. Emboldened by the Supreme Court, Trump has sought to fire, among others, a member of the Federal Trade Commission and a member of the Federal Reserve Board.
But this hasn’t always been how American government worked. There’s a long history of Congress trying to constrain who the president can fire and under what circumstances, most often by requiring “for cause” protections for certain quasi-independent agencies. Most recently, the Supreme Court has taken up the issue of Trump’s attempted firing of an FTC commissioner. Oral arguments are scheduled for later this year in which the court might do what it has long hinted at: overturning the New Deal-era case Humphrey’s Executor, which also involved a president, in that case FDR, trying to fire an FTC commissioner over policy disagreements.
But underlying Humphrey’s is another little-known case that helped give us the modern supercharged presidency. That’s Myers v. United States, a 1926 opinion written by Chief Justice William Howard Taft, whom you might remember from your bar trivia as the only ex-president to serve on the Supreme Court. As our guest today has argued persuasively, Myers has a lot of problems.
So, to help unpack this issue, I’m happy to be joined today by Noah Rosenblum, Associate Professor of Law at New York University, specializing in constitutional law and legal history, including this exact question. Noah, thanks for joining us.
Noah Rosenblum: Thanks for having me on, Andy.
Craig: As usual with these things, we have to start in Philadelphia in 1787, where the text of the Constitution is unhelpfully silent. This seems like a pretty big oversight. So I guess an initial question is: Why doesn’t the Constitution, which has all these detailed procedures about who can appoint various officers, not say anything about who can fire them?
Rosenblum: It’s a great question, and something that constitutional theorists and legal historians have been debating for a very long time. My co-author on that article about Myers observed that, based on the timing, it seems like they were probably ready to get out of Philadelphia. It was hot. They had been there a long time. They all wanted to get home.
The thing we have to remember, from where we are now, is that when they were drafting the Constitution, they were putting together a document that was part of a culture that is just super different from the highly text-specific, lawyerly culture that we have subsequently developed. And the presidency is a particular site where those differences become apparent.
As people who listen to this podcast may know, up until pretty late in the Constitutional Convention, the election of the president was going to happen by the House of Representatives. And while James Wilson had proposed direct election of the president, he was roundly defeated and his suggestion was rejected. It was really important to people in Philadelphia that the president not be elected by the people. That was one of their key concerns to prevent something like a demagogue.
Meanwhile, when it came to what had been the royal prerogative powers, those were apportioned differently across the different branches and to different actors. Michael McConnell, the wonderful professor at Stanford, has written about this in his great book, The President Who Would Not Be King.
The way that the Constitution was assembled was a lot less, “let’s specify in specific textual detail the contents of three and only three branches, each with carefully defined powers.” If they’d done that, it would obviously be super useful for the kind of legal separation-of-powers formalism that lawyers have to practice before the Supreme Court today. But that just wasn’t what they thought they were trying to do. They were, in the famous phrase, creating a revolution in favor of government. They were trying to design a state that would be more efficacious than the Articles of Confederation. There were a lot of different pieces to that, but in their drafting process we see that the specific words mattered a lot less than the general structure.
As I recall, they leave a ton of stuff to the Committee on Style, after they’ve already made these substantive decisions. Gouverneur Morris changes some things in ways that may or may not be significant. Certainly at the time they were not understood to be fundamentally transformative, these phrases that lawyers today put a lot of weight on—for example, the difference in the Vesting Clauses between the Executive Power Clause and the Legislative and Judicial Vesting Clauses.
This is just a long way of saying that the past is another country. That hot summer in Philadelphia looked very different from what we might expect or want from a constitutional convention. My best guess, even though it’s unsatisfying, is that they genuinely just forgot, that they just didn’t think through the removal power question. They were rushing; they wanted to get out.
Just as you said, they clearly put a lot of thought into the appointments question. They also put a lot of thought into how much they wanted to specify about the structure of government. That’s the other thing that’s totally missing from the Constitution, right? There’s nothing about the removal power, but there’s also nothing about what the staff that’s going to be working to run the government is supposed to look like. But we know that they knew that there would be staff—both because there is this reference to the inferior officers (so, clearly they didn’t expect the president to do it all himself) and there’s the bit where the president gets to request the opinion in writing from the principal officers in each department (even though, of course, there’s not a single department specified in the Constitution). To this day, we don’t really know from the Constitution alone what makes somebody a principal officer.
“Today, the president’s warmaking powers have been so aggrandized that it’s difficult for us to remember just how central Congress was to the Civil War. I have this bust of Abraham Lincoln in my office. He’s an almost larger than life figure, right? But Lincoln is working incredibly closely with Congress in prosecuting the war. There’s a committee of Congress that’s deeply involved there. He’s anxious that they’re constantly looking over his shoulder. That captures the sense in which the 19th century state really was a legislature-dominated state.” — Noah Rosenblum
They presumed all these things about how the government was going to work, but they didn’t put them in the Constitution. We know that when it comes to the executive administration, they thought about including more detail in the Constitution and then they took it out. One of the reasons for taking it out was because they wanted to leave it to the regular political process to decide what that structure would look like. They wanted the statutes to define what the departments would look like, and what the officers would be. So it stands to reason that one of the reasons they left the removal power out was to let it be defined via the ordinary political process—that is to say, through statute.
Craig: Well, that’s the Necessary and Proper Clause—the congressional power for filling in all the details—right? But they did run into this problem themselves pretty quickly. So, during the First Congress, they’re doing all these things: they’re creating the secretary of state, the secretary of treasury, these first departments. After George Washington is elected president, the first members of Congress have already been elected, but basically nothing else has spun up yet.
Later, we’ll get to how the Supreme Court has interpreted this—but it’s become known as what you refer to as the “so-called” Decision of 1789. So what was the Decision of 1789, and was it actually a decision at all?
Rosenblum: You framed that perfectly. Just as you said, it’s 1789; nothing’s been spun up. All you have is a president and a legislature and this document. The question you have to wrestle with is: What does this mean? How do you actually make a government work?
One of the first things they do is create the main departments of government—and they get into a pickle over exactly what the statutes defining those offices should say. So, the big ones: State, Treasury, and I guess Navy. And one of the questions they start to ask themselves is: How explicit do we have to be about the kinds of power the president has over this body?
Let me tell you how it comes down in the tradition, and then we can talk about what actually happened. Here’s how people talk about it: they say in designing the initial large departments of government, a question arose about whether it should be specified in statute that the president had the power to remove the secretaries of the departments or not. Nobody disagreed that they wanted the president to be able to remove the secretary of state or the secretary of the Navy. But if it were specified by statute that the president could remove them, then it would appear that the power to remove was granted by the statute and not by the Constitution.
Some argued that to do this would be to deny that the Constitution gave the president a power of removal. Their position was that there should be no mention in statute of the ability to remove. Rather, the statute should just reference what would happen if the secretary was removed. And this would in turn make it clear that the Congress believed that the president had the power to remove the secretary, directly from the Constitution, without having to rely on an additional grant of statutory authority.
According to the standard story, these two factions faced off. There was a series of complicated votes, but the meaning of those votes was ultimately that Congress accepted the position that the Constitution gave the president this power to remove. And that decision taken in 1789 came to be respected by all the other branches. It liquidated the Constitution’s meaning, or reflected the Constitution’s original meaning, since there were so many people in the First Congress who had also been part of the drafting and ratification process.
So, the Decision of 1789 establishes that even though the Constitution doesn’t say anything about the removal power, it should be read to give the power of the president to remove at least the heads of departments. That’s the standard story.
What actually happened, when you dig into the votes, is that it just gets weird. The first thing to stress is that nobody thought that this question was going to emerge. That’s not my opinion—I’m actually not a historian of the early republic; that’s the conclusion that my friend, Stanford Law School’s Jonathan Gienapp, who’s got to be the most impressive historian of the early republic of my generation, draws. He’s written an amazing book called Second Creation, which looks at the way new officials, especially the House of Representatives in this new government, worked out the meaning of the Constitution. In that beautiful, complicated [period] of 1789, [when they were trying] to spin up a government, they got into tons of fights over the meaning of the Constitution. Jonathan’s book unpacks those fights. I highly recommend it. He’s also got a great article about the Decision of 1789 in the American Journal of Legal History.
One of the points he makes in that article is that, from reading the debates, we see these representatives stumbling into this question. They didn’t come in with firmly developed ideas—they changed their opinion over the course of the debates. They have to discover that it’s a question they haven’t resolved. Then they realize that they think about it differently and argue over it. That doesn’t mean that they might not have been resolving an important constitutional question. They were certainly debating one. But it gives the lie to the idea that the debate over this particular question was fundamentally different from the other debates in the early Congress.
In particular, he emphasizes how uncertain and contingent it all was. That’s point one. For point two, Jed Shugerman, a law professor and legal historian at Boston University Law, has gone through and reconstructed the debate in a tremendous amount of detail. He wrote an article called “The Indecisions of 1789” where he emphasizes the strategic ambiguity. It’s not as if there were two clean factions that faced off against each other, with one of them saying, “I believe that the Constitution gives the president the power to remove,” and the other side saying, “No, it doesn’t; only the statute does,” and then there was one single vote, and you voted up and down, and the constitutionalists won. That’s sometimes how people talk about it, but that’s not at all what happened.
You had at least three factions, maybe more, and they didn’t come right out and say what they believed. You have to sort of reconstruct what their views were based on the little snippets that they give. It’s clear that there was a range of positions on offer about the relationship between the Constitution and statutory power with respect to removal. And there wasn’t one vote, there were two votes on two different provisions. They were structured in such a way that it makes it very difficult to understand what any individual might have believed based on how they voted. You had different majorities that carried in each of the votes.
One of the votes is for whether the statute should say that the president can remove. That would be the statutory grant. Then there’s another bit that has to do with what happens after a removal. Jed’s article is wonderful—it’s in the Penn Law Review; I can’t recommend it enough, especially if you’re an early republic nerd. But I come away from that article persuaded that it is very difficult to make any clear meaning out of this sequence of votes.
Now that, of course, hasn’t stopped people from trying to do so, although the reasons they rely on this early republic decision have changed over time. Some people have cited it on the grounds that because there were so many people in the First Congress who were part of the ratification process, it gives us privileged insight into the original meaning of the Constitution.
The two problems with that argument are that, first, there were people who were in the early ratification debates and in that early founding moment on both sides of this debate. So there’s no conclusion there for those trying to look for a settled meaning. And, of course, people change their minds on this question. Alexander Hamilton, when he’s writing the Federalist Papers, says, “Of course the president can remove, but given that removal tends to follow the same process as appointment, if someone has been appointed with confirmation of the Senate, then you have to get the concurrence of the Senate to remove them.” That’s what he writes in response to anti-Federalists who are worried that this new constitution is giving the president too much power.
But then when Alexander Hamilton is in Washington’s government, he changes his mind and he writes a letter saying, “Now that I’m actually here, I think maybe the president should be able to fire you without having to go through the Senate.”
If your goal is to use all that as an indicator of original meaning, it’s very hard, because they’re not all on the same side and they change their minds. So what meaning do we make of that? Other people, and I guess this points us towards Taft, rely on it as a decision that others come to rely on and accept. That’s a very different kind of argument. That’s an argument from acquiescence, but I’m getting ahead of myself.
Craig: Lots of fun parliamentary procedure ambiguity. I sometimes wonder if folks who’ve done this have never had the blessing of being in a deliberative body, trying to pass motions and amendments. It’s never quite as definitive as the final result might give you the impression that it was.
Rosenblum: As a fellow Robert’s Rules of Order nerd, I appreciate your intervention.
Craig: So let’s leave our Delphic Oracle Founding Fathers behind, and skip forward a few decades, because the next big development in this historical story was during Reconstruction and the aftermath of the Civil War. We could do a whole show on just this incident, but the short version is Andrew Johnson is president after Abraham Lincoln has been assassinated. Johnson is frankly an extremely racist president in a lot of ways. He was not good on wanting to have an aggressive Reconstruction, defending the rights of freedmen in the South, removing the ex-Confederates from power, and all that stuff. Congress was in the hands of the Republicans, particularly the more radical Republicans.
“When they passed the Tenure of Office Act, it’s obviously quite an aggressive act, but they’re doing it in the face of an aggressive president—to protect their own prerogative and to realize, after all, a policy that half a million Americans died to try to bring into being. It’s a really significant moment in American history, and in particular the question of the removal power. What’s interesting is that afterwards, despite questions about its constitutionality from early on, it remains on the books and in some critical ways was expanded. So throughout the second half of the 19th century, we have not only the civilian Tenure of Office Act, but also military tenure of office acts. That ends up being really important for the evolution of the jurisprudence.” — Noah Rosenblum
At this point, half the country or thereabouts was under military occupation. The southern state governments were being run by the Army for the time being. And so Congress passed the Tenure of Office Act, saying that Johnson could not fire a wide range of officers, but most relevant was the Secretary of War, Edwin Stanton. He was in charge of the Army and thus administering Reconstruction, and this law says you can’t fire him without the consent of the Senate. Johnson did so anyway.
There’s some great scenes where Stanton barricades himself in his office. If you’ve seen the recent series, Manhunt, about the aftermath of the assassination of Lincoln, they show this in a scene at the end. I highly recommend it. Johnson was impeached over this—the first president to be impeached. And then he was acquitted by one vote.
This has often been remembered as if the fight wasn’t really about the Tenure of Office Act; that it was some technicality, a pretext. It was really all this other stuff about Reconstruction, and the Tenure of Office Act was unconstitutional. That’s the version I heard in high school when we were doing this in American history. But I take it they were not that disingenuous. There was a serious principle at stake here, that the Republicans in Congress really did believe they had the power to set this restriction.
Rosenblum: Absolutely. As you’ve alluded to, there were so many problems with Andrew Johnson. He was also an alcoholic and he was quite explosive. Today, the president’s warmaking powers have been so aggrandized that it’s difficult for us to remember just how central Congress was to the Civil War. I have this bust of Abraham Lincoln in my office. He’s an almost larger than life figure, right? But Lincoln is working incredibly closely with Congress in prosecuting the war. There’s a committee of Congress that’s deeply involved there. He’s anxious that they’re constantly looking over his shoulder. That captures the sense in which the 19th century state really was a legislature-dominated state.
This is going to remain true, especially afterwards. So, when Johnson is trying to get around the legislature, it’s a real threat. It’s understood to be a real attack on the constitutional order. Not just the “new birth of freedom” that the Radical Republicans are pushing—although obviously that, too. But a tremendous disrespect for what is, after all, the first branch—the one with the power.
So, when he ignores the Tenure of Office Act, he’s not just coming up with a way to implement his will on Reconstruction. It’s a violation of Congress, which is precisely what is at stake in the Tenure of Office Act. It’s the reason why he wants Stanton removed.
Part of what Congress had done at the time, really what the statutes did which Congress passed, is they required the continued military occupation of the South to run through the secretary of war. They didn’t trust Andrew Johnson. They thought that he had broken with them over what the Civil War was for and meant. They wanted to make sure that actors who understood the meaning of the Civil War were in charge of the Reconstruction process. It was in order to subvert Reconstruction that Johnson sought to violate the Tenure of Office Act.
With respect to the high school story ... that high school story isn’t wrong, but the two are more deeply connected than one might imagine. The Tenure of Office Act is a tool that the Radical Republicans are using to implement Reconstruction. When Johnson violates it in order to do his own attempt at Reconstruction, violating the Act is part and parcel of trying to undo the new Reconstruction that the Radical Republicans are pushing. So that’s one part.
Then the other part, which you’re getting at, is it also threatens Congress’s ability to structure the government. If you go all the way back to Ex Parte Hennen, which is an [1830s] case, the Supreme Court there says removal power follows appointment power unless you vary it by constitution or by statute.
There’s by that point a long tradition of Congress being able to control through statute. That is to say, Congress may write the law, but the president signs it. So this is dialogical. It’s not just Congress against the president. It’s the two working together to build out the government. But you can have statutes that specify a term of years. There are statutes going back to the early 19th century in which Congress says you’re going to have these people who serve in office for a given term of years, which is in some ways a limit on the president’s power.
When they passed the Tenure of Office Act, it’s obviously quite an aggressive act, but they’re doing it in the face of an aggressive president—to protect their own prerogative and to realize, after all, a policy that half a million Americans died to try to bring into being. It’s a really significant moment in American history, and in particular the question of the removal power. What’s interesting is that afterwards, despite questions about its constitutionality from early on, it remains on the books and in some critical ways was expanded. So throughout the second half of the 19th century, we have not only the civilian Tenure of Office Act, but also military tenure of office acts. That ends up being really important for the evolution of the jurisprudence.
The key point for us to keep in mind is that even though the Stanton removal is the one that’s most dramatic, in the 1870s and ’80s, you have these cases where you’ve got cadets in the Navy who the president would like to kick out of office. You might expect that this is the high watermark, the president is the commander in chief of the armed forces. But for a variety of reasons we can get into, Congress had written a law, and the statute said that these cadets enjoyed tenure in office, that they were in their positions and could not be removed from them.
The Supreme Court repeatedly backed the cadets against the government, against the executive branch. So that’s much less controversial, a much more bread-and-butter legal conflict. But it’s the same principle as the principle involved in the dramatic fight over the Tenure of Office Act and Johnson’s impeachment but completely run of the mill law.
Craig: So that takes us into the later 19th century. Like you said, the Tenure of Office Act was still on the books, even though Johnson had been acquitted, and we have all these other laws. A lot of this was related to the spoils system, what we might less disparagingly call patronage—the ways the party machines of that era involved a lot of using government offices and positions. This was before we had the modern civil service system. Towards the end of this period, narrow initial versions of the civil service protections come in.
In particular, the plum pick that a lot of people had in mind and everybody wanted, if your guy won the Senate election or your guy won the presidential election, was to be in the post office. Because that was one of the largest government agencies that existed in a time when there weren’t a whole lot of very large government agencies.
So, one of these laws that was passed, after Johnson and after the Tenure of Office Act, involved postmasters and giving them a set term and protections. The way it was structured, they could only be removed when the Senate confirmed a successor. If the president wanted to fire them, [he had] to nominate somebody else first, the Senate approves them, then they’re the new postmaster for East Topeka or wherever.
This brings us to Myers. So, how did this case about firing a postmaster—and if I remember right it’s in Oregon somewhere—end up becoming this constitutional landmark case?
Rosenblum: As you’ve just suggested, in the second half of the 19th century, we do keep the Tenure of Office Act on the books, but eventually they repeal it. There’s also this important revision to the statutes in 1878. I talked about this in the article, but, importantly, they rewrote all of the statutes just to codify them.
At the time, the Tenure of Office Act is on the books, but then they repeal it later and they don’t totally harmonize the whole thing. There are all these errors. By the time we get to the turn of the 20th century, it’s confusing. But one thing is really clear, just as you said, the post office remains a huge source of patronage and the law is structured in a way to protect the people who are in it.
You were alluding to it, but I just want to underline this: the protections mean getting the Senate involved. It’s probably about upholding the patronage regime because it means that before you appoint or remove a postman, you have to get the Senate to sign off on it.
Basically it was a pretty standard story that if you ran a political machine, you would eventually get it to put you in the Senate. So a lot of the senators, especially on the East Coast, or actually also in the Mountain West, are the leaders of local patronage machines for which the post office is a critical spot. In part because, just like you said, it’s a huge number of people, but you also get to make deals to carry the mail. You get to subcontract out mail carrying routes, which can be extremely lucrative. Good old-fashioned honest graft, as George Plunkitt of Tammany Hall here in New York might’ve put it.
Okay, so what happens in Myers? Frank Myers is a postman in Portland, Oregon, and he gets the job the way that you get a standard patronage job: he managed some senator’s campaign. Myers is a Democrat, which is convenient because he’s nominated by Woodrow Wilson, also a Democrat. That’s why he gets the job. He clearly gets into fights with people, and maybe he’s part of his own rival political machine. It’s not totally clear. But he must do something to offend whoever his powerful patrons are, because Woodrow Wilson decides that he’s going to be removed.
Although, actually, I should put a small asterisk there. Basically, the law at the time prevented the president from removing someone from a postmaster position without the concurrence of the Senate. The way this usually worked is that the president wouldn’t ask the Senate for permission to remove you. Instead, the president would just nominate someone new to the job. The Senate would confirm that new person, and the confirmation of the new person was understood to be the Senate’s participation in the removal of the other person.
“A real challenge for the Roberts Court has been that they’ve expanded presidential power so much and given the presidency such protections from the rule of law. Think about the immunity case. It’s hard to see how to reconcile the government they’re imagining with the Constitution, or with a president that is supposed to be under the law. So there’s a huge question: If you overturn Humphrey’s, you make it more difficult to understand how to design institutions that involve some degree of insulation from the president. It might still be possible, and it’s certainly the case that there are lots of actors in the government who most of us think should be insulated from presidential removal, even if we disagree about principal officers.” — Noah Rosenblum
Incidentally, this way of removing, by appointing a new person, is what I think the law means when it says that removal is an incident of appointment. If you can appoint someone to a position just by dint of doing so, you remove whoever was in the position before. To go back to your initial question, why didn’t they say more about removal? That could also be part of it. Maybe the Founders were just thinking: We don’t have to specify removal because the easiest way to remove is just to appoint someone new to the same role.
In any case, by the time that Myers’ removal is being advanced, Wilson has suffered a terrible stroke and he’s basically catatonic. This is well described in his biography. His wife, Edith Wilson, is essentially running the government. Woodrow Wilson’s advisors are like, “Half his body doesn’t move.” They’re sort of draping his body in different ways to suggest that he’s still more vital—like a Weekend at Bernie’s kind of thing.
It’s during that time that the order to remove Myers from his job as postmaster comes down. To this day, we don’t have a letter from Woodrow Wilson saying Myers is removed, even though other presidential removals at that time do have the president’s name in the letter signing it. Myers is understandably confused. Not [because] he’s been fired—people get removed and replaced for patronage reasons all the time. But he believes that the law has been violated. This is both because of the long history we were just talking about, and more practically, he writes to various senators that he knows and he says, “I want to plead my case.” The senators write back and say, “Yeah, you’re right. You’ll get to plead your case because Wilson’s going to have to nominate a new postman to get rid of you. We’ll have to hold hearings and when we hold hearings, we’ll bring you in and you’ll tell us why you shouldn’t be removed.”
So Myers is getting geared up for what he understands to be an ordinary political patronage fight. Instead, he just gets this letter saying, “Actually, there’s not going to be a hearing. You’re just kicked out.” So he sues, and he dies while the suit is in process, so ultimately his estate carries on the suit. By the time it gets to the Supreme Court, the political composition of America’s elite has changed. And this gets us to William Howard Taft.
Taft is one of the giants of his era from a law perspective. He is beloved, he’s popular, he’s incredibly judicial. It’s been remarked about Taft that his highest dream in life was to be a judge on the Supreme Court of the United States, not to be president. When he becomes governor general of the Philippines and has to give up his appellate clerkship, he doesn’t want to do it. Teddy Roosevelt, the secretary of state, and his wife all say, “You should do this. Don’t just be on the sidelines. You’re a young man. Go seek out adventure.”
This gives us some insight into how Taft is thinking. He’s very deliberate. When he’s in Ohio before that, he’s very much part of the reform Republicans. The Democratic and Republican parties are transforming at the end of the 19th century because of the patronage dynamics you were describing. You get a bunch of, let’s say, more elite, especially in the Republican Party, politicians who are just scandalized by the patronage corruption. They want to clean things up and they want to bring good government expertise order into government administration.
Teddy Roosevelt is one of the exemplars of this. He’s one of the civil service commissioners down in Washington before he becomes a governor of New York, before he becomes police commissioner, and eventually rises in his meteoric political career. Taft is also involved in anti-patronage efforts back in Ohio before he enters into federal judicial service and eventually government work. They’re part of a whole movement of folks who are convinced that the old patronage world is bad, anti-democratic, corrupt, venal, and needs to be changed and reformed. The locus of that venality is these political machines controlled by these terrible senators.
So when Taft is governor general of the Philippines, he gets to set up a government institution, a civil service, a well-functioning administration. He can do it without having to worry about these dumb, corrupt political machines or legislatures intervening and trying to prevent him from doing what he wants. This helps inform how he thinks government should work.
There are a bunch of other tendencies at the same time that lead a group of these elite reformers—especially Republicans, but not just Republicans—to become really suspicious of this legislature-led approach to government. To imagine that a better way of doing government, a more responsible way of doing government, is to give the president more power over the way administration.
Woodrow Wilson is a great example because, when he writes Congressional Government in 1888, he’s deeply critical of legislative dominance. But there, he seems to be pushing something like a return to federalism. He’s not yet imagining presidential primacy. By the time he writes Constitutional Government, 30 years later, he says, “I’ve now realized that in fact the way to do democracy is just to give the power to the president.” He says the president can be “as big a man as he can” with the party behind him.
Now, in Wilson’s case, he’s probably megalomaniacal. I’ve mentioned some books on the podcast already, but I’ll mention one more. Patrick Weil’s The Madman in the White House, which is about Woodrow Wilson, does a wonderful job of detailing Woodrow Wilson’s megalomania: his conviction that he was appointed by God to be the leader of the American people, his sense that he’s a world historical figure.
When the senators are threatening not to ratify the Treaty of Versailles, when they don’t ratify the United States joining the League of Nations, Wilson wants to dismiss them and call for new elections. He’s got this very parliamentary vision in which I am the leader and you guys are accountable to me—which is obviously foreign to the Constitution.
Wilson’s thinking that way, and there’s Teddy Roosevelt with the “stewardship theory” of government. Taft is coming in, in a way that’s much more careful and thoughtful and judicious, but part of a similar current—reforming government to make it more efficacious and accountable by concentrating powers in the hands of the president, informed in part by his experience as governor general of the Philippines. As he later writes in Myers, he was able to see just how much you can get done if you don’t have to worry about this kind of meddling interference from the legislature. It’s under those circumstances that he ends up writing the opinion that decides Frank Myers’ case.
Craig: He writes a rather lengthy decision. He takes a long time to do it. There had been this line of cases doing basically statutory interpretation. The pre-Myers law was: “What did Congress mean to do?'“ There wasn’t a whole lot of challenging the idea that this was a question for Congress to decide. But with Myers, we get the first appearance of this idea that, no, this is constitutional—that, under the Constitution, because the president has the Executive Vesting Clause, he has to be in charge of everything in the executive branch. He has to have pretty much unfettered power to fire at will because they’re his subordinates, and he can’t really be in charge of them if he’s not able to fire them effectively at will.
This is also happening against the backdrop of the Progressive Era, where the federal government is expanding. We’re getting a lot more new regulatory agencies, new laws, antitrust enforcement, the Interstate Commerce Commission, all these sorts of things.
So that’s what Myers says. But then there’s a little bit of walking it back not even a full decade out. Political times have changed a lot as we get to the New Deal. Franklin Roosevelt has a very robust view of presidential power in a lot of ways—even beyond the Teddy Roosevelt, Taft, Wilson vision of the presidency. Though I don’t think he ever thought he could dissolve Congress, that one I hadn’t heard about Wilson.
So we get this case, Humphrey’s Executor, in 1935, where FDR tried to fire, pretty much exactly like is happening right now, a member of the Federal Trade Commission. The law says they can only be removed for cause, which is, you’re taking bribes or incompetent or not showing up to work. But Roosevelt tries to say, “No, we just have policy disagreements.” This is another case where the guy died before it reaches the Supreme Court. That’s why it’s Humphrey’s executor.
What exactly is the rule that it lays out? Because it seems like it’s a little bit of a fudge. But it’s been the backdrop against which all these agencies we’re familiar with now—the FTC, the SEC, the FCC, basically anything that has “commission” in the name—is building off of what was found in Humphrey’s.
Rosenblum: You got it. Let me say one last thing about Myers before we jump to Humphrey’s: I hope it comes across that I have a lot of warmth for Taft in particular. Even though Taft pushes this more presidentialist vision, he doesn’t give up on the independence of the civil service or the importance of expertise. So, the Myers opinion is this really fascinating, complicated document that’s got both strands in it. But you’re absolutely right that at the time the presidentialist strand is the one that people really focus on.
“This is part of why the Fed exceptionalism thing makes a lot of us really uncomfortable. Because the Federal Reserve is special, sure, but it’s also just like ... one agency, man. There are a lot of agencies that regulate industries and are absolutely critical to the American economy. Would it be okay to create, I don’t know, a Securities and Exchange Commission to regulate stocks that was structured the same way as the Federal Reserve? And if it was, would that agency be able to benefit from the same kind of independence protections? In some ways, I think the Court doesn’t want to have to wrestle with that question. They want to have their cake and eat it, too. They want to keep the independent Fed and protect the monetary supply, but also allow for strong presidentialism. But that’s just not really how the law works, because what’s good for the goose is good for the gander.” — Noah Rosenblum
That’s part of why, when Roosevelt fires Commissioner Humphrey, he believes that he can get away with it. He’s thinking, “I’ve got this Myers case on my side. It’s going to support me even though there’s this statute that says I can only fire for cause.” At the time, Roosevelt was working closely with Congress, but many people believed that he was asserting too much power. He famously gives the speech where he says, “Don’t worry, I’m not trying to be a dictator. Besides, I wouldn’t be a very good dictator if I tried,” which is not what you want to say when you’re trying to reassure the public that you don’t have dictatorial ambitions. So, when Humphrey’s is decided, subsequent commentators read it through the lens of this anxiety about overweening presidential power.
Okay, the ground of Humphrey’s is that there are certain positions that are quasi-legislative or quasi-judicial. If you’re in one of those positions, it’s totally fine for a statute to prevent you from being removed by the president, even if it might appear on its face as if that’s the kind of office a president should be able to control. Now, if you and I had been talking last year, I would have said, “Boy, ‘quasi-legislative’ and ‘quasi-judicial,’ what a fudge. What does that even mean?” But over the course of the last year, a group of us have been doing some research into these terms. It turns out that, in fact, they were terms of art, and we’ve just forgotten what they meant.
So, at the risk of mentioning more people for your listeners to go track down, there are two young scholars: one is Beau Baumann at Yale, and the other is a fellow here at NYU with us, Nathaniel Donahue. Nathaniel has a great blog post about this. Beau has an article. Nathaniel and I are writing an article together on this question.
It turns out that the terms “quasi-legislative” and “quasi-judicial” go back, just as you were saying, to these commissions, things like the Interstate Commerce Commission. And they trace back to municipal government. It’s clear that these terms are developed in the late 19th century to describe offices that exercise power under a statute, but that are not operating as alter egos of the head of the executive branch. They are Article I judges, or they are commissions in charge of rate setting, or they are cities that are enacting ordinances pursuant to state law, but pursuant to their own processes.
It’s a way the law developed for talking about offices that unlike the postman who is, in the late 19th century, an extension of the president insofar as they’re part of this patronage apparatus. What are postmasters doing? They’re just exercising executive power. That’s not what the court says what the ICC is up to, or somebody setting rates for railroads. They’re making a kind of expert-driven calculation. Ultimately, what the court decided in Humphrey’s was this included the Federal Trade Commission, even though Roosevelt wanted to use it as a tool for realizing his policy. Roosevelt was like, “I disagree with you, Humphrey, about how to do antitrust stuff, and I want to do it my way.”
The responsibility of the commissioners was not primarily to the president, the Court said. It was to the statute. The statute told them what to do and they were supposed to exercise their own judgment in implementing the statute. So the Humphrey’s case ends up saying quasi-legislative and quasi-judicial roles can be insulated from presidential removal. And that was understood to be, just as you said, a repudiation of Myers.
Although there is this puzzle that I was trying to research this summer. I still haven’t answered it. Myers is a case that’s decided 6-3. Of the six judges in the majority in Myers, four are still on the Court when Humphrey’s is decided nine years later. And all four of those judges vote with the majority in Humphrey’s. So it does seem as if the judges in Myers did believe that Humphrey’s and Myers were compatible—even though you’re totally right, we read it now as a repudiation, and that repudiation, well ... maybe I’m getting ahead of our story again, but that’s what Antonin Scalia really holds onto in his dissent in Morrison, which is sort of the last chapter of how we get to today.
Craig: Yes, and that does bring us up to the next bit of this story.
We went through the larger part of the 20th century and Humphrey’s seemed to be relatively settled. There wasn’t a whole lot of active dispute about this. All these commissions, presidents got their appointments, but they were staggered and there was gradual turnover and it wasn’t a point of too much contention. Until we get to Morrison, which was this case involving the independent counsel statute. That was where Congress said for certain things, particularly when you’re investigating the president himself or misconduct in the executive branch, we’re going to have this independent special prosecutor and the president can’t just go fire them willy-nilly. This got challenged and all of the Supreme Court justices said “No, that’s fine”—except one. But that one happened to be Antonin Scalia who would go on to be more influential in a lot of ways in the Roberts Court era and where we are now. So in going back to Myers as Scalia did, how did he recast that into, what we would call today, unitary executive theory?
Rosenblum: In his solo dissent in Morrison, Scalia says the Constitution gives the president “the” executive power, which must mean all of the executive power. This statute that Congress drafted, the independent counsel statute, which creates a prosecutor who exercises executive power but is not responsible to the president, clearly violates this provision.
We talked about this a little at the beginning. That provision, the “the” in it, was put there by Gouverneur Morris and the Committee on Style. It wasn’t exactly the kind of thing that the convention debated, but now here it is 200 years later coming back and Scalia is making it a real cornerstone of his argument. He says, “What’s really happening in this case is that the majority, my colleagues, are ignoring the Constitution, relying on Humphrey’s Executor to repudiate the Myers case.”
“But the Humphrey’s Executor case,” he says, “didn’t grapple enough with Taft’s excellent analysis in Myers.” Then Scalia writes an essay called, “Originalism: the Lesser Evil.” That essay is really important. It’s probably Scalia’s first statement about this new methodology that he’s pushing called originalism. He says when we read the Constitution, we’ve got to look at its text, but it’s not just like looking at the text of a statute, because we need to understand what it meant—that is, what the people who wrote it intended when they did it.
That’s a really challenging thing to do. But if you want to see how to do it, the person who did it best was William Howard Taft in Myers, according to Scalia. Taft looked at the meaning of these constitutional provisions, looked at the Decision of 1789, and said this tells us the original meaning of the Constitution, and the original meaning was that the president had this removal power.
This essay that Scalia writes—“The Lesser Evil”—is right after Morrison. So in this one moment, Scalia writes the solo dissent in Morrison saying, “I’ve read the Constitution and based on my reading of the Constitution, Taft was right in Myers, not the majority in Humphrey’s.” In other words, it’s not appropriate to insulate executive officers from presidential control. Then, he writes a scholarly essay saying, “It’s not just that I look to Myers in this one case. I actually think Taft is showing us a method for how to decide cases.”
Now, I just have to point out, because I am a law professor, I think this misreads the Myers opinion. When Taft relies on the Decision of 1789, it’s not because he thinks it tells us what the original Constitution means. He doesn’t say, “Oh, look, all these guys were Founders.” He says, “This was a new construction of the Constitution, but it came to be accepted by all the other branches. And in the separation of powers, we should take seriously that notion of acquiescence.”
So I think Scalia misreads Taft’s method. I also think Scalia misreads Myers. Scalia says the court in Humphrey’s just rejected the careful analysis in Myers. Taft wrote 60 pages. The Myers opinions, when you include the dissent, is like 300 pages. Taft worked on it for two years. The Humphrey’s case is like 12 pages—it’s not a hard case for them.
Scalia looks at this and says, “They just dismissed Myers. They didn’t take it seriously.” But remember, of the four Myers judges left on the Court in Humphrey’s, they all vote for the Humphrey’s majority. Clearly they don’t think that they’re breaking with their prior decision—or, if they are, it doesn’t bother them. So I think Scalia is mistaken.
Joke’s on me, because even though at the time Scalia wrote a solo dissent, that solo dissent proves to be unbelievably influential. Scalia’s law clerks, many of whom become law professors, draw from it. As Federalist Society judges get confirmed, they really cling to the dissent in Morrison as having been correct. Even though the author of the majority opinion in Morrison, [William] Rehnquist, a very conservative judge, writes a wonderful little takedown of Scalia’s dissent. He’s saying, “I don’t know what’s going on with my brother Scalia. He’s trying to put more on the text than the words will bear.”
But this all comes to a head under the Roberts Court where, in a series of cases, starting with the Free Enterprise Fund case, the Roberts Court turns what had been a dissent in Morrison into the theory of the majority.
That’s where we are today. The key of Scalia’s Morrison dissent was, as my con law teacher taught me, you need a clever textual argument and a strong structural theory. Scalia’s Morrison dissent had both. The clever textual argument was the Vesting Clause, and the Take Care Clause. This means the president has to be responsible for running the government.
But the powerful structural theory wasn’t different from Taft’s structural theory. It was that the president is elected by the whole American people. The president represents the will of the nation and to be a democratic polity, we need the president to be able to control the government. Taft says that in Myers, Scalia says that in his dissent in Morrison, and John Roberts writes it into a majority opinion in Seila Law, decided just five years ago.
Craig: So there are a couple of ironies here. One is that, as much as this is couched in originalism—it’s kind of the wellspring of what became the originalist revolution—it’s very bad history.
You see this in the dissents in Myers and you see it in Rehnquist’s criticism of Scalia. There is this long history after 1789, in the antebellum era, through the Civil War, throughout the 19th century, everywhere except maybe this nine year period between Myers and Humphrey’s [in which] Congress passed laws and presidents mostly acquiesced in this position that Congress actually does have the power to structure the government.
But aside from the bad history angle, I want to zero in on that structural argument because that goes to, well, why does this matter? I think it really is fundamental to how we got this supercharged imperial presidency, as it’s been called—the idea that the president is this unique tribune of the people, even though he’s not actually directly elected nationwide.
My understanding of how the Framers would have talked about this is, “no, actually Congress and particularly the House of Representatives are the representatives of the people. They’re the ones who are much more local. They’re the only ones originally required to be directly elected.“
So that brings us to the Trump era. I don’t think it’s too much of an exaggeration to say he and a lot of people in the administration have been on a purge mentality of people who are not willing to get on board with their policy agenda. This includes a lot of very legally questionable things.
This brings us to the case that is currently pending for the Supreme Court, which is widely expected to overturn Humphrey’s. The way the question is presented and reading the tea leaves, it’s not been subtle. The Roberts Court has been building up to this, that they’re probably going to overturn Humphrey’s altogether. And where will that put us? It’s kind of Myers on steroids.
Obviously, we don’t have the opinion yet. We don’t know exactly what they’re going to say. And there’s this weirdness about, no, the Federal Reserve is different. We’re going to carve it out, maybe. But if Humphrey’s is overturned as expected, does this mean essentially Congress has no power to provide any kind of tenure or removal protections? There’s no ability to have any independent agencies? Unless you’re an Article III judge or a member of Congress, the president’s your boss and he can fire you whenever he wants?
Rosenblum: I would say the biggest challenge is uncertainty as opposed to absolute prohibition. I think your analysis is exactly right. You’re completely on point. A real challenge for the Roberts Court has been that they’ve expanded presidential power so much and given the presidency such protections from the rule of law. Think about the immunity case. It’s hard to see how to reconcile the government they’re imagining with the Constitution, or with a president that is supposed to be under the law.
So there’s a huge question: If you overturn Humphrey’s, you make it more difficult to understand how to design institutions that involve some degree of insulation from the president. It might still be possible, and it’s certainly the case that there are lots of actors in the government who most of us think should be insulated from presidential removal, even if we disagree about principal officers.
Chris Walker at Michigan, a wonderful colleague, has got a great article. He’s got great co-authors too, Aaron Nielsen, Melissa Wasserman. But he’s written a lot about adjudicators—and Emily Bremer at Notre Dame has done this, too—that is, the people inside the government who are basically like judges, even though they’re not Article III judges. I think all of us would be weirded out if these administrative law judges are supposed to decide cases on the basis of a phone call from the president. And if they don’t do what they’re told, they’ll be fired. Soviet justice was also known as telephone justice, right? I don’t think any of us believe that the Constitution creates that. With Humphrey’s Executor on the books, it’s easy to explain why officers like that can be independent. If you overturn Humphrey’s, you’re going to need some sort of a new theory, or a new account. We might be able to develop one.
There’s another case after Humphrey’s called Weiner that relies on Humphrey’s, but that you might be able to rely on independent of Humphrey’s if you needed to. The key point, though, is that imagine you’re a representative in Congress and you want to design an institution. Right now, you know what the rules are. And in fact, the rules have been basically the same for what, 120 years? If the Supreme Court steps in now, and declares this unconstitutional, first, you’ve got a whole bunch of institutions that might now be struck down—ones that Congress undoubtedly would have designed differently if you had just told them that the rules were different. So there’s a little bit of sandbagging. That doesn’t seem fair.
“I do think one real problem with our current arrangements is that so much of the initiative and policymaking capacity have been concentrated in the executive branch. If we’re going to do democracy, but the only organ of government that has the energy and the political will to solve our most pressing problems is the executive branch, then regardless of what the court says, we’ll trend in an increasingly presidentialist direction. That is dangerous. We’re seeing some of the dangers now.” — Noah Rosenblum
Suppose you want to fix things going forward. Right now you just don’t know where the lines are. That’s always a real challenge if you’re a judge, to communicate clearly where the lines are—so that when other actors are dependent on you, which is to say legislators and lower court judges, they know what to do.
For what it’s worth, this is part of why the Fed exceptionalism thing makes a lot of us really uncomfortable. Because the Federal Reserve is special, sure, but it’s also just like ... one agency, man. There are a lot of agencies that regulate industries and are absolutely critical to the American economy. Would it be okay to create, I don’t know, a Securities and Exchange Commission to regulate stocks that was structured the same way as the Federal Reserve? And if it was, would that agency be able to benefit from the same kind of independence protections? In some ways, I think the Court doesn’t want to have to wrestle with that question. They want to have their cake and eat it, too. They want to keep the independent Fed and protect the monetary supply, but also allow for strong presidentialism.
But that’s just not really how the law works, because what’s good for the goose is good for the gander. Lawyers reason by analogy. We want to know how the analogies are going to work. So having the justices say, “Don’t worry, we’re not going to threaten the Fed”—that doesn’t cut it. Insofar as I have a 401k, that makes me sleep a little bit easier thinking, “good, they’re not going to inflate all my money away.” But insofar as I’m a lawyer, I start thinking, “wait, that doesn’t do me any good at all.” A “good for one trip only” exception doesn’t actually provide any stability in administrative law. Does that make sense?
Craig: Yes it does. That’s where things stand as we wait to see where the justices come down—exactly what contortions, if you’re being ungenerous, they’ll engage in to reach their result here. But with an eye to the future: one way or another, Donald Trump will not be president forever. And eventually there will be a new majority on the Supreme Court.
So, taking in the grand sweep of this history, what should the law be? Both from the constitutional text, what’s the correct interpretation? And can Congress do essentially whatever it wants, is that the desirable outcome?
As a normative matter, do we want to have a more divided executive branch, where we have more of these independent agencies? And if so, how should we go about getting there? Because obviously we don’t want to go back to the spoils system. That’s not what anybody necessarily wants, though in some ways we kind of are moving back to it.
Rosenblum: I think there are some people who would like us to go back to the spoils system.
Craig: Yeah. So ... Chief Justice Rosenblum, you’ve just been appointed and you’ve got eight lackeys willing to do what you want. Next time this comes around in 2040 or something, how do you write that case? Either overturning Slaughter, which is the current case pending, or restoring Humphrey’s, or overturning Myers, or how should we approach that?
Rosenblum: I really appreciate the question, and let me give you three different kinds of answers: a more historical answer, a more legal answer, and then a more institutional answer.
The historical answer would be, I think it’s pretty clear the Constitution intended to leave to us, the living, most of the decisions about how to govern ourselves. This takes us back to the very beginning of our conversation. The deliberate decision not to include the structure of the government inside the Constitution; the repeated references to the powers of the legislature during the ratification debates; putting Congress as Article I; as you said, the fact that the House of Representatives is the only one that was specified to be directly elected, and was supposed to be the most democratic, in the way that it was put together ... I see so many places—in the Constitution and in our history—where there is a real commitment to letting the living figure out how to govern themselves. It strikes me that a real challenge for the Court would be to vindicate that principle, which they’ve moved further and further away from.
When it comes to the constitutional law of administration, they have increasingly adopted the view that there are only a very small number of ways in which the government can be structured, and those are specified by the Constitution. I actually think they’ve got it exactly backwards. The Constitution was designed to be an eminently practical document. The Founders were eminently practical people. And it’s supposed to be a document that enables us to have liberty and equality while governing ourselves efficaciously. That would be the historian’s point.
The lawyerly point would be that the Court—and this is building on that—has just moved way too far in using the Constitution to limit what can be done under statute. They often frame this as a conflict between Congress and the president. But as I said earlier, that’s just not true. All of this stuff gets done by statute, which is to say by law, and those laws are negotiations between the president and the legislature.
And in some ways those laws involve all of us in the negotiation, too, because we vote for the president, we vote for the legislature, and we know the judges will review them. Doing things through law, as opposed to doing them through the Constitution, is I think a much more practical, responsive way of doing government.
The independent agency question, to me, is secondary from allowing us the flexibility we need to design the governance arrangements that work. This was clear even at the founding. So Christine Chabot has a wonderful article about the “sinking fund,” which was a brand new arrangement. This is at the Founding, it’s the beginning of the Republic. It’s a commission that involves representatives from all three branches to manage the debt. What she’s doing there is recovering a form of governance that was designed to solve one really hard problem.
These legislators in the late 18th century thought, here’s a new institution that would help us solve that. You and I could spitball, if we had the time, about the many new problems we’ll be facing today. Whether it’s AI companies, or industry concentration on a scale that really is unrivaled since the Gilded Era, or changing geopolitical situations, or social media—pick your poison. It just seems obvious, to me at least, that the governance regimes for these different problems could very well look different.
I don’t have any special insight there. It’s just that as a legal historian, I can tell you that in the past, we solved problems like this by developing different governance institutions for different issues. If I were a justice drafting an opinion, I would want to make sure I preserved the flexibility that the Constitution gave to our existing political leaders, to design the institutions adequate to those problems. And you can do that with Humphrey’s.
You could also do that in a new form, drawing the lines differently, not relying on quasi-legislative and quasi-judicial. But it’s very hard to do if you believe that the entire government is an extension of the personality of the presidency.
I’m going to add a last point, which is the institutional point. I do think one real problem with our current arrangements is that so much of the initiative and policymaking capacity have been concentrated in the executive branch. If we’re going to do democracy, but the only organ of government that has the energy and the political will to solve our most pressing problems is the executive branch, then regardless of what the court says, we’ll trend in an increasingly presidentialist direction. That is dangerous. We’re seeing some of the dangers now.
How could we rebalance that? To do that, we have to think: What are the conditions that make it possible for other institutions to play a dynamic role and a responsive role in democratic governance? I think that a lot of the criticisms of Congress are overwrought. Congress is operating basically exactly the way it should be operating, given the institutional incentives for congressmen and Congres. If we want the legislature to push back more against the presidency, if we want Congress to be more engaged, if we’re not going to have more presidential power, we need other institutions like Congress to step back up. For that, we have to be thinking, how do we change the institutional incentives and arrangements so that congressmen and Congress are better able to serve the governance goals we have for them?
Craig: This has been a fascinating discussion and I wish we had more short-term optimism, but as always, maybe there’s some case for long-term optimism.
Rosenblum: Andy, you’re talking to a historian. Long-term optimism is my jam. So I think you put it very well.
Craig: Things that can’t go on forever will eventually end, right? Something to that effect.
Rosenblum: The problem of the republic and time is the oldest problem of secular government. And just as no one seems to have solved it yet, I don’t think we should count ourselves out of the experiment too soon.
© The UnPopulist, 2025
Follow us on Bluesky, Threads, YouTube, TikTok, Facebook, Instagram, and X.
We welcome your reactions and replies. Please adhere to our comments policy.
