William and Mary law professors Tara Leigh Grove and Neal Devins have written this article arguing for “a limited congressional power to represent itself in court.” Specifically, they argue that while the House or Senate may enforce subpoenas (including subpoenas directed to the executive branch) in court, neither house may intervene in federal litigation to defend the constitutionality of federal statutes where the executive branch refuses to do so.
Professor Jack Beermann responds to Grove and Devins here. He disagrees with one of their conclusions, noting “there is no constitutional provision that can fairly be interpreted to prohibit Congress or one House of Congress from defending the constitutionality of a duly enacted federal statute.” Moreover, although Beermann agrees with Grove and Devins that the House or Senate may litigate in support of the investigatory power, he largely disagrees with their reasoning on this point also.
I think Beermann has by far the better of this argument, and it is tempting to let the matter rest there. But I think it worthwhile to correct at least one part of their thesis that Beermann accepts largely without challenge. Grove and Devins contend that the history of congressional involvement in litigation supports the distinction they draw between the investigatory power and other types of cases, including the defense of the constitutionality of federal law. They say:
Historical practice supports our argument for a limited congressional power to represent itself in court. From 1789 until modern times, the House and the Senate asserted the power to conduct investigations and to litigate any disputes related to those investigations. By contrast, Congress historically delegated control over all other federal litigation to the executive. That was true even when the executive declined to defend a federal law. Although members of Congress occasionally participated as amici in such cases, neither Congress nor its components asserted the power to intervene on behalf of federal laws. This historical pattern remained unchanged until 1983, when the Supreme Court—with virtually no explanation—permitted intervention by the House and Senate counsel in INS v. Chadha.
This description, however, is misleading. To understand why, it is helpful to focus on Congress’s overall practices with regard to litigation in the pre-Watergate era. During the 1970s, particularly during Watergate itself, Congress became (somewhat) more litigious, reflecting factors such as (1) the increasing litigiousness of society itself, (2) an increasing tendency to see congressional-executive disputes as essentially legal in nature and (3) the development of institutional legal offices in both Houses. The intervention of the House and Senate counsel in the 1983 Chadha case must be seen against that background.
Investigations
In the pre-Watergate era, all congressional litigation was extremely rare. With regard to Congress’s alleged assertion of a power “to litigate any disputes related to [congressional] investigations,” the best place to start is with a case not mentioned by Grove and Devins, Reed v. County Commissioners, 277 U.S. 376 (1928).
Reed involved a Senate select committee investigating allegations of fraud and other irregularities in the 1926 senate election in Pennsylvania. When the county officials in Delaware County, Pa., refused to give the Senate committee ballot boxes and other election records it demanded, the committee petitioned a federal court for an injunction directing the county officials to comply.
To my knowledge, this is first time that any congressional committee (or the House or Senate themselves) asserted a “power to litigate disputes” related to investigations or anything else. Indeed, in asking the Supreme Court to review the lower court’s dismissal of their suit for lack of jurisdiction, the Senate committee noted it was “present[ing] for the first time in the history of this Government, on behalf of the Senate and in pursuance of the powers vested in them by the Senate, a petition to this Court for writ of certiorari in order to maintain and preserve the coordinate authority of the Senate and the Legislative branch of the Government.”
To be sure, on several occasions prior to 1928, beginning with Anderson v. Dunn, 19 U.S. 204 (1821), the House or Senate participated in litigation arising out of their use of the inherent contempt power. In these cases the contemnor, having been imprisoned or otherwise aggrieved by the exercise of that power, challenged the contempt by seeking habeas or other judicial relief. Since Congress and/or its officers were the defendants sued in these proceedings, their (involuntary) participation can hardly be said to reflect the assertion of an “authority to litigate.”
If this point is not self-evident, one has merely to look at the Supreme Court’s opinion in Reed, where it makes clear there is no historical practice of the Senate (or House) litigating in support of its investigatory powers. Declining to interpret the Senate resolutions establishing the select committee as giving it authority to sue, the Court held:
The resolutions are to be construed having regard to the power possessed and customarily exerted by the Senate. It is the judge of the elections, returns, and qualifications of its members. Article I, § 5. It is fully empowered, and may determine such matters without the aid of the House of Representatives or the executive or judicial department. That power carries with it authority to take such steps as may be appropriate and necessary to secure information upon which to decide concerning elections. It has been customary for the Senate, and the House as well, to rely on its own power to compel attendance of witnesses and production of evidence in investigations made by it or through its committees. . . . Authority to exert the powers of the Senate to compel production of evidence differs widely from authority to invoke judicial power for that purpose.
277 U.S. at 388-89 (emphasis added).
As I have argued before (see here, here and here), the Reed case was narrowly written and does not stand for the proposition that there is a constitutional barrier to suits by Congress or its committees. It does, however, show, contra Grove and Devins, there was no historical practice prior to 1928 of Congress seeking to litigate in support of its investigatory authority. Moreover, following the Reed decision, no such litigation would be attempted again until the Watergate era.
Interestingly, though, the Senate’s immediate reaction to the Reed decision was to adopt a standing order, still in effect today, declaring in part that
any committee of the Senate is hereby authorized to bring suit on behalf of and in the name of the United States in any court of competent jurisdiction if the committee is of the opinion that the suit is necessary to the adequate performance of the powers vested in it or the duties imposed on it by the Constitution, resolution of the Senate, or other law.
While one may argue that actions speak louder than words (and no Senate would act on these words during the pre-Watergate era), this resolution suggests the Senate saw no constitutional barrier to suits by its committees in a wide variety of circumstances, not limited to enforcement of subpoenas or other investigatory matters.
Certainly this resolution is a more direct reflection of the Senate’s constitutional views than any of the evidence discussed by Grove and Devins. The fact that Congress almost never sought to participate directly in judicial proceedings during the pre-Watergate era may be some evidence that it doubted its constitutional authority to do so, but it can hardly be taken as a definitive judgment on the subject. As Beermann points out, there is no evidence to connect this history to any particular interpretation of the Constitution, including the interpretation that Grove and Devins advance.
There are in fact a number of other plausible ways to explain Congress’s inaction (apart from the snarky observation that inaction is Congress’s core competency). For one thing, it is doubtful that the early congresses would have even considered appearing in court to be an option. There was no precedent for a legislature (or one house thereof) appearing in court, and the traditional attitude of Parliament toward the courts was along the lines of “you’re not the boss of me.” See J. Chafetz, Democracy’s Privileged Few 27-36 (2007). Had it considered the issue at all, Congress almost certainly would have viewed it as inappropriate to subject itself to the jurisdiction of a coordinate branch. Indeed, it was precisely avoid such unseemly reliance on the other branches that Congress asserted (and the Supreme Court accepted in Anderson v. Dunn) the power to punish contempt directly.
In addition, whatever Congress’s own views on the constitutionality of its initiating or intervening in litigation, it might reasonable have anticipated that the courts would be reluctant to hear such cases and would be tempted to throw them out on constitutional or other grounds (as in Reed). This problem was exacerbated by the fact that Congress lacked institutional counsel to advise or represent it with respect to judicial proceedings during the pre-Watergate era.
It should therefore not be surprising that historically Congress has almost always attempted to vindicate its interest in litigation indirectly, rather than by appearing in court itself. Grove and Devins are right about that, but they are wrong in suggesting that congressional investigations were any exception to this rule. For example, when Congress decided in 1857 that it needed a supplement to its inherent contempt power in order to punish uncooperative congressional witnesses, it enacted a statutory scheme in which either house may refer a contempt to the U.S. attorney, who is then statutorily mandated to present the matter to the grand jury. See 2 U.S.C. § 194.
Defending Federal Statutes
Because Congress almost never sought to appear directly in court for any reason during the pre-Watergate era, its practice with regard to defending the constitutionality of federal statutes is not indicative of a particular aversion to such litigation. If anything, the opposite would seem to be true. There was arguably more direct congressional involvement in the handful of pre-Watergate cases where the executive refused to defend the constitutionality of a statute than in any other type of litigation.
Of course, as Grove and Devins acknowledge, the executive branch never refused to defend the constitutionality of a federal statute until Myers v. United States, 272 U.S. 52 (1926), which involved a statutory restriction on the president’s removal power. Prior to that time, Congress had no occasion to consider whether it should appear in court in defense of a statute’s constitutionality, and obviously nothing can be read into its failure to do something it had no reason to do.
Unquestionably Congress has always viewed defending the constitutionality of federal law as first and foremost the responsibility of the executive branch. Even today, members of Congress believe that the executive branch has a duty to defend the constitutionality of statutes under most, if not all, circumstances. Thus, while Grove and Devins stress the fact that Congress passed a law in the 1930s to enhance the Attorney General’s authority to intervene in any lawsuit for the purpose of defending the constitutionality of a federal statute, there is no contradiction between this action and the view that Congress can step in to defend the constitutionality of a law when the executive branch refuses to do so. Indeed, Congress would do exactly that just a few years after passing the law.
In the post-Watergate era it has become common for members of Congress to submit amicus briefs for or against the constitutionality of statutes even when the executive branch is defending them, but this was likely unheard of prior to Myers, no doubt for some of the same reasons that caused Congress’s general reluctance to get involved in litigation. Myers, however, introduced a new element when the Supreme Court invited Senator George Pepper to participate in the case as an amicus defending the law’s constitutionality. Technically, Pepper was neither representing a party nor Congress, but one presumes that the Court wanted Congress’s institutional perspective on an important constitutional issue that affected the balance of power between the political branches.
Grove and Devins do not cite any evidence that Congress or anyone else thought it inappropriate for Congress’s interests to be represented before the courts in this fashion. Instead, they make three points. First, they state that “Congress has a whole [has] showed little interest in defending its work product against constitutional challenge.” Second, they contend that the members of Congress have not been particularly principled or consistent in deciding what laws to defend; such decisions tend to be driven by partisan or political factors such as the salience of the law in question rather than an institutional interest in upholding Congress’s lawmaking authority in general.
These observations seem fair enough, though their relevance to Congress’s constitutional authority to defend a statute is debatable. As already noted, Congress has always been reluctant to get involved in litigation. In cases where a component of Congress overcomes its natural inclination to stay out of court, there is almost always a strong political motivation. This is no less true in cases involving the investigatory power than defending the constitutionality of a statute. As Grove and Devins themselves point out in footnote 153, “a multimember body like Congress will tend to have more difficulty than the executive in adopting a unified, coherent litigation strategy.” This explains why congressional defense of statutes tends to be episodic and why it is nearly impossible to get “Congress as a whole” (as opposed to a single house or committee) to participate.
Their third point is that “even when individual members (or components) of Congress sought to participate in litigation, they did so as amicus curiae.” In other words, prior to Chadha no congressional litigants tried to intervene as a party in cases where the executive refused to defend the constitutionality of a federal law.
Here it should be noted that Grove and Devins are dealing with a rather small data set. Apart from Myers, Grove and Devins identify only two other pre-Watergate cases, Humphrey’s Executor v. United States, 295 U.S. 602 (1935), and United States v. Lovett, 328 U.S. 303 (1946), in which the executive refused to defend a law’s constitutionality. In Humphrey’s Executor, which like Myers involved the extent to which Congress could constitutionally limit the president’s removal power, Congress did not appear in defense of the challenged law, although at least one Senate Republican argued it should. Grove and Devins ascribe Congress’s inaction to partisanship (the Democratic majorities in Congress did not want to oppose the Roosevelt administration) and do not suggest it had anything to do with whether Congress would intervene or only appear as an amicus.
Grove and Devins also point to The Pocket Veto Case, 279 U.S. 655 (1929), in which the House Judiciary Committee, through its chairman, appeared as an amicus in support of a narrow construction of the president’s pocket veto power. Here there is also no evidence suggesting that Congress (or its components) considered the possibility of intervening as a party or that there would have been any advantage in doing so. In light of the Reed decision, the House Judiciary Committee might have been leery of seeking to intervene without explicit authorization from the House, but it is sheer speculation whether this had any impact on its manner of proceeding.
This leaves Lovett, a case in which the executive branch refused to defend the constitutionality of an appropriations rider which forbade paying the salaries of three named federal officials. The House adopted a resolution appointing a “Special Counsel” to defend the rider’s constitutionality, and Congress later passed a statutory provision to authorize and fund the Special Counsel’s activities. Grove and Devins make much of the fact that the Special Counsel only appeared as an amicus in the case, but it is not clear whether the decision to be an amicus only was made by Congress or by the Special Counsel himself.
Whoever made the decision, it seems almost certain it was tactical in nature. Indeed, if Congress does not care much about the constitutionality of the laws it passes (as Grove and Devins imply), it is rather far-fetched to imagine that it cared about (if it was even aware of) the constitutional distinction between appearing as a amicus or as a party.
In any event, the decision to appear as an amicus had no discernable impact on congressional interests in the Lovett case. When the lower court ruled in favor of the plaintiff officials seeking to recover their salaries, the Solicitor General sought certiorari at the request of the Special Counsel even though the executive branch agreed with the ruling. The Supreme Court questioned neither this procedure nor the existence of a justiciable controversy despite the fact that the nominal plaintiffs and the nominal defendant (the executive branch) were on the same side. Instead, the Court treated “counsel for Congress,” as it repeatedly referred to the Special Counsel in its opinion, as the de facto defendant in support of the statute.
In the wake of Myers and especially Lovett, one would reasonably conclude that the Court welcomed congressional participation in cases where the executive branch refused to defend the constitutionality of a federal law. The technical distinction between appearing as an amicus or as a party/intervenor was apparently of little or no significance to anyone. As one reasonably well-informed observer would later remark:
United States v. Lovett might more accurately have been reported as Lovett et al. and the Executive Branch v. Congress. For on December 6, [1943,] Attorney General Biddle informed the Speaker of the House and the Vice President that because he concurred with the President’s judgment that section 304 was unconstitutional, he found it “impossible to advocate with conviction the views of the Congress.” He therefore suggested that Congress employ its own attorney to argue in favor of the statute. Congress adopted the necessary legislation and hired John C. Gall as “Special Counsel.” Though officially an amicus curiae, functionally he was counsel for the defendant (which is to say the statute). All the ingredients of a real adversary proceeding were present.
John Hart Ely, “United States v. Lovett: Litigating the Separation of Powers.=,” 10 Harv. C.R.-C.L. L. Rev. 1 (1975).
Conclusion
In short, during the pre-Watergate period Congress was reluctant to get directly involved in any kind of litigation. This may in part have reflected doubts about the constitutional propriety of doing so, as well as concerns about how the courts would react to its involvement. There is, however, no direct evidence tying Congress’s practice to any particular constitutional judgment, and there are equally plausible non-constitutional explanations for this practice. There is certainly no reason to believe that Congress was more confident in its authority to litigate in support of the investigatory power than in defense of statutes; if anything, its experience in Reed and Lovett suggested the exact opposite.