(Don’t) Lock Him Up?

By “him,” of course, I mean this guy:

This is Hudson Snowden Marshall, who served as the United States Attorney for the Southern District of New York from 1915 to 1917. He also has the honor of being one of only two executive branch officials ever arrested by the Sergeant at Arms for contempt of Congress. The significance of Marshall’s case for interpreting the scope of the congressional contempt power is a matter of some contention, to which we will return in future posts. Compare Josh Chafetz, Executive Branch Contempt of Congress, 76 U. Chi. L. Rev. 1083,  1137-39 (2009) (arguing that Marshall’s case supports the proposition that the contempt power extends to actions of executive branch officials) with Todd David Peterson, Contempt of Congress v. Executive Privilege, 14 U. Pa. J. Const. L. 77, 128-30 (2011) (arguing that Marshall’s case has no broader significance because it involved no claim of executive privilege or separation of powers).

For today, however, I want to focus on the procedure used to “lock up” Marshall when the House held him in contempt in 1916. To the extent the House is considering the possibility of including a contempt process in its current impeachment inquiry, Marshall’s story is instructive.

A brief background may be helpful. Marshall’s contempt arose out of an intense conflict with a member of the House of Representatives. Marshall’s office procured an indictment in the Southern District of New York of the member (Frank Buchanan, a representative from Illinois) for violations of the Sherman antitrust act. The indictment alleged that Buchanan was receiving funds from Germany  to foment labor strikes in American factories as part of Germany’s effort during WWI (which America had not yet entered) to disrupt American munitions shipments to the allies. Buchanan, in turn, accused Marshall of all manner of corruption and impeachable offenses, and he succeeded in persuading the House to authorize an investigation of Marshall’s alleged misconduct by the Judiciary Committee. See 6 Cannon’s Precedents § 530.

A special Judiciary subcommittee was then appointed to take evidence on the matter. It proceeded to hold hearings in New York in early March 1916. While these hearings were ongoing, a New York newspaper published an article which stated, among other things, “[i]t is the belief in [Marshall’s] office that the real aim of the Congress investigation is to put a stop to the criminal investigation of the pro-German partisans.” 6 Cannon’s Precedents § 531.

The subcommittee did not take kindly to this allegation and it immediately summoned the author, Leonard R. Holme, to demand that he disclose whether anyone in the U.S. attorney’s office had made this accusation to him. When Holme refused to answer, the Sergeant at Arms was directed to arrest him and keep him in custody until further notice. 6 Cannon’s Precedents § 531.

The Sergeant at Arms did as instructed, but found himself in a bit of a dilemma. How was he going to keep a prisoner in custody, particularly in New York. Apparently he asked the local United States marshal to place Holme in confinement, but the marshal understandably declined the opportunity to get in the middle of a battle between Congress and the U.S. attorney’s office. This forced the subcommittee to reconsider and order Holme released, somewhat to its own embarrassment. 6 Cannon’s Precedents § 531.

Fortunately, the mystery of Holme’s source was solved the next day when a letter arrived from Marshall, who not only identified himself as the source, but amplified his charges against the subcommittee and the House. Among his milder complaints was that it was “irregular and extraordinary” to conduct an impeachment inquiry without a formal House vote authorizing it. (Sound familiar?) He also said that he regarded any member of Congress who would take money from a foreign agent (obviously referring to Buchanan) as a “traitor” and felt “it was a great pity that such a person could only be indicted under the Sherman law, which carries only one year in jail as punishment.” 6 Cannon’s Precedents § 531. He further accused the subcommittee of “a deliberate effort to intimidate any district attorney who had the temerity to present charges against one of your honorable body” and said it “had apparently resolved to prevent prosecution by causing the district attorney in charge to be publicly slandered.” Id.

When the subcommittee returned to DC. the House adopted a resolution to investigate Marshall’s statements as contempt. A committee then recommended that Marshall be found guilty of “a breach of the privileges and a contempt of the House of Representatives.” 6 Cannon’s Precedents § 532. The House resolved that Marshall be brought to the bar of the House to answer these charges, and on June 22, 1916, the Speaker issued a warrant for Marshall’s arrest. Id.

Here is where we get to the point of today’s post. After Marshall was arrested on June 26, he immediately secured a writ of habeas corpus from Judge Learned Hand, who further ordered that pending a hearing Marshall “should be at liberty to go upon his own recognizance.” Journal of the House of Representatives, 64th Cong. 1st sess. 855 (July 5, 1916). Judge Hand then heard the case and issued a decision on July 22, in which he found for the House and dismissed the writ. (We will discuss the merits of Hand’s decision in a future post). Rather than immediately remanding Marshall to the custody of the Sergeant at Arms, however, Hand directed that Marshall remain free on his own recognizance until August 15, unless Marshall had filed an appeal by that date, in which case he would remain free until December 1, at which time he would be remanded to the custody of the Sergeant at Arms. 53 Cong. Rec., Pt. 12, p. 11,691 (July 27, 1916). This procedure allowed Marshall a reasonable but not indefinite period to challenge the lower court’s ruling on appeal.

While I do not know who suggested this procedure, it appears to have been agreeable to both parties. It makes sense that the House would have accepted it because, then as now, the House had little capacity to keep an individual in custody for any significant period of time. This is particularly true if the House could not count on the cooperation of the executive branch.

Marshall did appeal to the Supreme Court, but the case was not argued until December 11, 1916 and it was not decided until April 23, 1917. See Marshall v. Gordon, 243 U.S. 521 (1917). (Spoiler alert: Marshall won). Marshall apparently remained free during this period, presumably based on an understanding between the parties.

There would seem to be no reason why an arrangement like this could not be employed today, with or without the imprimatur of a court. As a condition of release by the Sergeant at Arms, a contumacious witness could agree to return to the custody of the House on a particular date or upon the occurrence of a specified event. In addition to being better for the witness, such a procedure ameliorates the House’s difficulties with keeping prisoners for extended periods. In the case of executive officials, it would also lessen concerns about “the specter of violence between the political branches.” See Andrew McCanse Wright, Constitutional Conflict and Congressional Oversight, 98 Marquette L. Rev. 881, 934 (2014).

There are a couple of ways to ensure that witnesses comply with the conditions of their parole. First, they can be released into the custody of their counsel. That way the lawyer is on the hook, at least reputationally, for her client’s behavior. Second, the witness can be informed that failure to comply with the specified conditions will result in monetary fines. While I am somewhat skeptical about the use of fines as a direct punishment for contempt, such fines would seem to be justifiable when the witness agrees to them as a substitute for confinement. Indeed, the Court noted in Anderson v. Dunn that legislative fines imposed in the past were justified as “mere commutation for confinement.”

This proposal, it should be noted, does not obviate the need for the Sergeant at Arms to arrest contemnors who refuse to surrender voluntarily. Moreover, some witnesses may refuse to agree to the specified conditions for release so the House will still need contingency plans for holding individuals for an indefinite period. As in the case of recent House subpoenas, though, once a few witnesses comply with the specified procedures, pressure will build on others to do so. (This assumes, of course, that these procedures will survive judicial review, which I believe they will).

It is important to note that this is not a plan for mass incarceration. Commitment for contempt still requires the witness to be brought before the bar of the House to show cause why he should not be held in contempt. This imposes a significant cost of time on the House, and is therefore not something it can do lightly or often.

Finally, for reasons we have discussed such a procedure will be more resistant to any type of judicial challenge if it is adopted in the context of an impeachment inquiry. Thus, if the House wishes to exercise this power, the time is now.