Editor: Mr. Katsas, please summarize your career in the Justice Department.
Katsas: I joined the Justice Department in 2001, as the Deputy Assistant Attorney General responsible for overseeing the Appellate Staff of the Civil Division. In 2006, I became Principal Deputy Associate Attorney General, the top advisor to the third-ranking official in the department. For about six months in late 2007 and early 2008, I also served as the Acting Associate Attorney General. Finally, I was nominated by the President and confirmed by the Senate to be the Assistant Attorney General for the Civil Division.
While in these various positions, I was able to handle or supervise many of the federal government's leading civil cases. I personally argued more than 40 appeals, including one Supreme Court case, eight other appeals that eventually went to the Supreme Court, and at least one case in every federal court of appeals.
Editor: What were some of your most challenging cases?
Katsas: I handled many of the leading national security cases in the Bush administration: Boumediene , the case on constitutional habeas rights for detainees at Guantanamo Bay; El Masri , an important appeal on the state secrets privilege in cases challenging alleged covert CIA activities abroad; other appeals involving national security letters and war powers of the President; and cases involving the confidentially of information related to the terrorism investigation following the attacks of September 11, 2001. I handled one appeal involving the Energy Task Force chaired by Vice President Cheney, and another appeal to protect Janet Reno and Eric Holder from personal liability for their role in the raid to seize Elian Gonzalez and remove him to Cuba. I defended the constitutionality of the Pledge of Allegiance in the latest appeal brought by Mike Newdow. Other cases included various hot-button issues ranging from partial-birth abortion to physician-assisted suicide to the Defense of Marriage Act.
Editor: One of the landmark cases decided by the Supreme Court during your government tenure was Ashcroft v. Iqbal . Please provide the background for this case.
Katsas: Iqbal was one of the most important Bivens cases that we handled in the Civil Division, i.e., cases where individuals seek to impose personal liability on government officers for their official acts. Iqbal had been detained in the terrorism investigation that followed September 11, 2001. He pleaded guilty to felony offenses and was removed to Pakistan, but that did not deter him from suing Attorney General Ashcroft and FBI Director Mueller for money damages for alleged constitutional violations. Iqbal claimed that that decisions about his conditions of confinement were motivated by racial and religious discrimination, rather than any legitimate government interest, and that the attorney general and FBI director were personally involved in those decisions. The case turned on how much a defendant needs to plead in order to survive a motion to dismiss and thus commence discovery, which is very burdensome for the government, particularly when directed at high-level officials. The Supreme Court agreed with our position that any civil plaintiff, to survive a motion to dismiss, must allege facts that reasonably support an inference of liability.
Editor: Why was Bell Atlantic Corp v. Twombly so important to Iqbal ?
Katsas: Twombly was another case on how much a civil plaintiff must allege in order to survive a motion to dismiss. In Twombly , the Supreme Court held that the plaintiff must allege facts that support a reasonable inference of liability. Iqbal in my view was a pretty straightforward application of Twombly . And while Iqbal was important because it involved damages claims against high-level government officials, Twombly , which was an antitrust case, was important because it involved a context where discovery is incredibly burdensome and expensive for defendants. The Supreme Court decided that a plaintiff cannot impose those discovery costs on a defendant merely by making conclusory allegations, or by making specific allegations that do not support a reasonable inference of liability. Instead, the Court held, the plaintiff must allege facts that support a reasonable inference that the defendant is liable.
Editor: What had been the Conley dictum that Twombly overturned?
Katsas: Conley v. Gibson , an old 1957 case, contained a one-sentence statement that a plaintiff need only plead any set of facts consistent with liability. That statement cannot literally be true: if it were, a plaintiff could file a complaint alleging only that the defendant violated some law and that the plaintiff was injured as a result. Courts for decades had recognized that this one sentence from Conley is not really an accurate statement of the law. For example, courts repeatedly had said, prior to Twombly , that conclusory pleadings were insufficient, and that courts would accept only reasonable inferences from the specific facts alleged. So, I don't think that Twombly can fairly be characterized as effecting a significant change in the law. Instead, the Twombly Court was simply cleaning up some stray language from a prior opinion, and confirming what the lower courts had been doing for at least a couple of decades.
Editor: Iqbal just extended the reach of Twombly ?
Katsas: Iqbal confirmed that Twombly was not a case limited to antitrust, even though discovery costs are particularly high in that context. Iqbal explained that Twombly states a general principle about pleading under Rule 8 and Rule 12 of the Federal Rules of Civil Procedure, and therefore applies to all civil cases across the board.
Editor: Please explain what the Court meant in stating that a plaintiff's complaint must make factual allegations that are plausible, not merely possible.
Katsas: Under Twombly and Iqbal , what the plaintiff has to do is allege some facts that support a reasonable inference of liability. That is not an incredibly high burden. It is not a heightened pleading requirement under Rule 9. The facts alleged don't have to be elaborate, and the inference of liability doesn't have to be probable. The Court used the word "plausible" to describe the likelihood of the necessary inference. But in both cases, the Court made clear that a "plausible" inference of liability is simply a reasonable inference of liability based on the specific facts alleged, which are assumed to be true. That is a standard quite favorable to plaintiffs. Critics of these decisions contend that no plaintiff will be able to survive a motion to dismiss unless the plaintiff has an open-and-shut case before discovery has even begun. I don't think that that is a fair reading of Twombly and Iqbal , nor do I think that more recent cases applying those decisions support the view of the critics.
Editor: What did Justice Kennedy, speaking for the majority in Iqbal , say about imposing personal liability on top government officials?
Katsas: Justice Kennedy confirmed the importance of qualified immunity for all government officials, including the importance of immunity from civil discovery. He repeated what the Supreme Court had said time and again: discovery against government officials sued in their personal capacity is distracting and potentially damaging in any context, especially in cases against high-ranking officials and especially in cases involving national security. He understood that if qualified immunity from discovery is to have any meaning at all, the courts must police the sufficiency of civil complaints at the outset of cases.
Editor: What did Justice Kennedy say about the possibility for limited discovery?
Katsas: One argument made for lax review of motions to dismiss is that the district courts can limit discovery so as to mitigate its costs. Justice Kennedy rightly expressed skepticism about that possibility. It sounds good in the abstract, but in practice discovery inevitably tends to expand.
Let me give you just one example from the national security context. I was actively involved in the Guantanamo Bay habeas litigation. The government asserted that habeas review of the wartime detention of enemy combatants outside the United States would impair national security and impose substantial burdens on the Department of Defense and the intelligence community. The courts responded that national security could be protected, and discovery burdens minimized, through limited discovery carefully managed by the district courts. Unfortunately, what has happened in the Guantanamo litigation is that the courts have required the government to open up its classified files to detainees' counsel in order to give them literally every document in the files bearing on the enemy combatant determination. That has imposed tremendous practical burdens and, more importantly, national security risks on the Defense Department and the intelligence community during the middle of two ongoing wars.
Editor: Can a case be made for replacing notice pleading with fact pleading, as advocated by the American College of Trial Lawyers?
Katsas: Yes, although I think those two models are not conceptually distinct. The hard question is one of degree: how many facts should a complaint have to allege? Even the plaintiffs' bar, which advocates so-called notice pleading under Conley , acknowledges that the plaintiff at a minimum has to plead enough facts to put a defendant on fair notice about the contours of the claim. In that sense, everyone agrees that there is some degree of fact pleading. On the other hand, I think defense counsel would agree that it is inappropriate to make a plaintiff allege in a complaint every minute detail of his ultimate case. So really it is a question of degree, and the courts in cases like Twombly and Iqbal have been drawing some reasonable lines to give plaintiffs a fair amount of leeway, but not an unlimited degree of leeway.
Editor: How do Twombly and Iqbal affect the burdens of e-discovery?
Katsas: The effect of those cases is to screen out some modest number of implausible claims prior to discovery. In the context of complex litigation, doing so is really important because discovery costs are so incredibly high. In particular, the runaway cost of e-discovery is perhaps the single most important problem in civil litigation today. The electronic storage of information has exponentially increased the raw amount of information subject to discovery requests. Searching that information can be time-consuming and technologically challenging, particularly if a firm's information technology has changed between the time of storage and the time of attempted retrieval. So, a business subject to e-discovery demands often has to hire IT specialists to run extensive computer searches for possibly responsive records. It has to tie up its own computer systems and its own personnel in the process. When the searches are done, it has to hire lawyers to review thousands if not millions of individual documents for relevance and privilege determinations. The discovery process can drag on for months, if not years. And in even moderately complex litigation, it can easily cost the defendant millions, if not tens of millions, of dollars of out-of-pocket and non-reimbursable expense, not counting the defendant's additional and non-reimbursable opportunity costs. As Twombly recognized, defendants in such a position will be sorely tempted to settle even meritless claims for large sums of money, simply to avoid the various costs of discovery. Twombly and Iqbal mitigate the possibility of such extorted settlements by foreclosing discovery unless the plaintiff can at least give some reasonable reason for thinking that he might have a valid claim.
Editor: Do you have a panacea for bringing e-discovery under control?
Katsas: Unfortunately, no. Twombly and Iqbal are only part of the solution. Furthermore, I would hope that the Civil Rules Committee would consider changes to the discovery rules themselves, with a view towards further controlling the costs that those rules impose on plaintiffs and defendants alike.
Editor: Finally, tell us about your recent testimony before Congress about the federal pleading standards.
Katsas: There has been a move in Congress to overrule Twombly and Iqbal and to codify the problematic "any set of facts" statement from Conley . Senator Specter introduced a bill in the Senate, and the House Judiciary Committee recently began to hold hearings. I was invited to testify as the only witness in favor of Twombly and Iqbal ; the other side had three witnesses. At the hearing, Representative Jerrold Nadler, who chairs the Constitution Subcommittee of the House Judiciary Committee, announced that the Committee would move forward with its own bill to overrule those cases, but I and others are doing our best to oppose the bill as much as we can.