Roscoe Pound Warned the Bar About Administrative Absolutism in 1938

Nothing is more dangerous to a favored idea than a friend who changes his mind. The critics of a program can be dismissed as enemies, but when one of the program's own intellectual architects stands up and says that it has gone wrong, the room goes quiet. In 1938 that happened to the American legal establishment, and the man who did it was not a reactionary, and not a corporate lawyer. He was Roscoe Pound, the dean who had spent a career telling the courts to catch up with modern life.

I want to tell his story in order, because it is one of the most useful stories in the history of American government. It explains why the agencies we live with have the procedures they have, and it shows that the warning about unaccountable administrative power did not begin with talk radio. It came from a man at the top of the legal profession, in the language of the bar, to the bar.

A Reformer From Nebraska

Roscoe Pound was born in 1870 in Lincoln, Nebraska, and he came to law by way of botany. He was a serious scientist of plants before he became a serious scholar of law, and the habit of classifying things, of looking at how they actually grow rather than how a treatise says they should, stayed with him all his life. He practiced in Nebraska, taught there, and later joined the faculty at Harvard. In 1916 he became dean of the Harvard Law School, a post he held until 1936, an astonishing twenty years.

His reputation as a reformer was made earlier, in 1906, in an address to the American Bar Association on the causes of popular dissatisfaction with the administration of justice. The speech was a catalog of the ways that courts and procedure had grown clumsy, technical and slow, and it has been remembered ever since as the spark of modern court reform. In his scholarship he was a founder of what he called sociological jurisprudence, the idea that law should be studied by its effects in society, and that a judge ought to look at what a rule does in the world, not only at what the books say about it.

If you wanted a figure to stand for progressive legal thought in the first third of the twentieth century, you could not do better. That is exactly why what he said next mattered.

The New Deal Builds a Government of Agencies

The background is the 1930s. Congress, at the request of President Franklin Roosevelt, created a large number of new agencies, and gave existing ones new powers. The National Labor Relations Board, the Securities and Exchange Commission, the Federal Communications Commission and many others were made the working government of the economy. An agency of this kind typically made rules for an industry, investigated alleged violations, brought charges, and decided the cases, often within the same building and sometimes through the same officials.

There was a theory behind this, and it deserves a fair statement, because the people who held it were intelligent and sincere. The theory was that the problems of a modern industrial economy were too technical and too fast-moving for legislatures and courts. A legislature cannot write a rule for every railroad rate or securities offering. A court cannot supervise an industry one lawsuit at a time. So the agency, staffed by experts and insulated from the clumsiness of the ordinary branches, was the right instrument. The old separation of powers, in this view, was an eighteenth-century device that a twentieth-century state had outgrown.

A conservative sees the trouble immediately. The Founders divided power into three branches precisely because they did not trust any person or body to hold the power to make the law, to enforce it and to judge disputes under it. Madison put it with the finality of a man who had studied the question for years. In Federalist 47 he wrote that "The accumulation of all powers, legislative, executive, and judiciary, in the same hands, whether of one, a few, or many, and whether hereditary, self-appointed, or elective, may justly be pronounced the very definition of tyranny." For a fuller reading of that paper, see Federalist 47 Defines Tyranny as All Powers Gathered Into the Same Hands, and for how the modern agencies hold all three, see Madison Called It the Very Definition of Tyranny, and Agencies Hold All Three. The New Deal agencies had been built, in their daily operation, to do the thing Madison called tyranny, with the best of motives.

The 1938 Report

In the 1930s the American Bar Association appointed a Special Committee on Administrative Law, and Roscoe Pound became its leading voice, the chairman during the committee's critical years. In 1938 the committee presented its report, and the report became the most important statement of the case against unchecked agencies of its time.

Let me describe what the report argued, in terms I am certain of, without putting words in anyone's mouth.

1. It named a danger. The committee, and Pound in his own writing, spoke of "administrative absolutism," a theory that the executive and its agencies should be free from the control of the courts and of general legal rules. The phrase has been attached to Pound's name ever since. He compared the American trend to the English debate that had been opened a decade earlier by Lord Hewart's The New Despotism of 1929, a book by the Lord Chief Justice complaining of the growth of government by departmental rule.
2. It objected to combined functions. An agency that writes the rule, investigates the violation, prosecutes the case and then decides it is not a neutral judge. It is a party with a stake in the outcome. The committee did not say that agencies must not exist. It said that these functions should be separated and that the decision of a contested case should be insulated from the staff who brought it.
3. It wanted real review of facts. If an agency's findings of fact were conclusive whenever some evidence supported them, the courts could not protect a citizen against an agency that found the facts it wanted. The committee asked for stronger judicial review, and a standard that required the findings to be supported by evidence in the whole record.
4. It proposed procedure. The committee wanted uniform procedures across agencies, notice to the parties, an opportunity to be heard and written decisions with reasons.

That is not an argument for abolishing the New Deal. It is an argument that even a good government needs a process, and that the process cannot be left to the people who benefit from skipping it. It is the argument that a lawyer makes: the right to be heard is not a privilege to be handed out by the agency that wants to hear you.

The Case on the Other Side: James Landis

I promised fairness, and the opposing case deserves its best advocate. That advocate was James M. Landis, a gifted lawyer who had been a member and then chairman of the Securities and Exchange Commission, and who in 1937 followed Pound as dean of Harvard Law School. In 1938, the same year as the ABA report, Landis published The Administrative Process. It remains the best defense of the agencies ever written.

Landis argued, in substance, that the administrative process was a practical response to the problems of the modern economy. Government had taken on tasks that required continuous attention and technical knowledge, and the three-branch model, as applied to them, was a formalism. Agencies were the means by which expertise, flexibility and sustained attention could be brought to bear. He was candid that agencies combined functions, and his claim was that the combination was the point. An agency that could make the rule, watch the industry and apply the rule to a case could respond to a changing problem faster than any court or legislature. He did not think the old categories of the Constitution described the real work of modern government, and he did not think they should be allowed to stop it.

Notice what makes Landis a worthy opponent. He does not deny the facts that Pound describes. He accepts them and calls them virtues. Pound saw a combination of powers that endangered liberty, and Landis saw a combination of powers that made government effective. The question between them was not what the agencies do. It was whether the thing they do is compatible with a free constitution.

Why Pound Was Right Where It Counts

Here is the point-by-point part of the argument, the answer to the strongest objection.

The strongest objection, in Landis's voice, is that Pound's insistence on process will paralyze the agencies and that the economy cannot wait for courtroom procedures. Here is the answer, in order.

1. Speed is not the same as legitimacy. A government that acts quickly and wrongly is not an improvement over one that acts carefully and rightly. The Founders deliberately made lawmaking slow, because slowness is a protection for the minority, and the minority in an agency proceeding is a particular company or individual with a house, a payroll and a name.
2. Expertise is a reason to consult, not a reason to command. Experts can tell a regulator what will happen if a rule is adopted. They cannot tell the public whether it should be. The first is a question of fact, the second a question of value, and a free people keeps the second for itself.
3. A procedural right is cheap and a lost right is dear. Giving a citizen notice, a hearing, an impartial decision-maker and a reasoned opinion costs the agency time. Withholding them costs the citizen the one thing he cannot get back, which is his standing as a person with rights rather than a subject with a file.
4. Power that has no check expands. Pound was not being sentimental about courts. He was observing what any student of institutions would predict: an agency that answers only to itself will find that its purpose requires more power, and more power requires more purpose.

From the Report to the Statute

What happened to Pound's warning is the part of the story that people forget. It was not ignored. It was absorbed, and in a compromised form it became law.

In the years after the report, the American Bar Association backed legislation to reform agency procedure. In 1940 Congress passed the Walter-Logan bill, which would have imposed extensive judicial review and procedural requirements on the agencies. President Roosevelt vetoed it in December of that year, and the veto was sustained. But the pressure did not go away. Roosevelt had already, in 1939, appointed the Attorney General's Committee on Administrative Procedure, which studied the agencies and published its report in 1941. Its recommendations, together with the ABA's, shaped the statute that finally passed, the Administrative Procedure Act of 1946, signed by President Truman on June 11 of that year.

The Act is a truce, which is how I describe it in The Administrative Procedure Act of 1946 Was a Truce, and It Still Shows. It did not abolish the agencies and it did not restore the strict separation of powers. What it did was provide, for the first time, a general code. It set requirements for notice and comment in rulemaking, standards for formal adjudication, a role for independent hearing officers, who today are administrative law judges, and a right to judicial review in which a court may set aside agency action that is arbitrary, capricious or unsupported by substantial evidence. Five years later, in Universal Camera Corp. v. NLRB (1951), the Supreme Court read the substantial-evidence standard to require a review of the whole record, which was what Pound's committee had wanted.

So Pound lost the argument in 1938 and won a good deal of it by 1946. That is a lesson for anyone who thinks a lost argument is a lost cause. He did not persuade the administration, and he did not persuade the Congress that wrote the Walter-Logan bill to stay in the fight. But he put the argument on the record, in the language that lawyers use, and the legislators who finished the work had to answer him.

What Pound's Work Still Tells Us

The same fight is still with us, under new names. A citizen whose case is heard by the agency that is also his prosecutor is in Pound's courtroom. The Supreme Court's 2024 insistence that certain claims belong before a jury and an Article III judge, which I discuss in SEC v. Jarkesy Put the Jury Back Between Citizens and Agency Courts, is a late echo of the same point. The debate over how far the President may direct the commissions is the other half of the question, since the structure that Pound worried about is the structure described in Humphrey's Executor Created a Fourth Branch the Constitution Never Mentions.

And there is a human lesson in the man. Pound was a reformer who was not afraid to be a critic of his own side's reforms. Many people believe that to be progressive is to be in favor of every new instrument of government, and that to be conservative is to be against them. Pound's career shows the better position: that a reform is to be judged by whether it serves liberty and law, not by the date of its invention. A conservative reading him today should take the lesson that the best critics of concentrated power are often those who have seen it up close, and should be gracious about welcoming them.

The Case in Brief

1. Roscoe Pound, dean of Harvard Law School from 1916 to 1936 and a founder of progressive legal thought, became the sharpest critic of unchecked New Deal agencies.
2. The 1938 report of the American Bar Association's Special Committee on Administrative Law warned against "administrative absolutism" and asked for separated functions, real judicial review of facts and uniform procedures.
3. James Landis's The Administrative Process (1938) made the best opposing case, that combined functions and expertise were the point of agencies, and the Constitution's separation of powers did not forbid them.
4. Madison's definition of tyranny in Federalist 47 sides with Pound on the principle, and Landis's side prevailed on the existence of the agencies.
5. Pound's concerns flowed into the Administrative Procedure Act of 1946, which gave agencies a code of procedure and gave citizens the right to ask a court to set aside action that is arbitrary or unsupported.

What you can do: read the text of the Administrative Procedure Act, found in Title 5 of the U.S. Code, sections 551 to 559 and 701 to 706, and check which of its protections applies the next time you or someone you know deals with an agency.