Should Congress Change the Standard for Dismissing a Federal Lawsuit?
|By MICHAEL C. DORF
|Wednesday, July 29, 2009|
Earlier this year, in Ashcroft v. Iqbal, the U.S. Supreme Court ordered that a 9/11 detainee's civil rights suit against the former Attorney General and FBI Director be dismissed because, in the majority's view, bare allegations that these high-ranking government officials ordered discrimination against Arabs and Muslims were not credible. As I explained in a column on this site shortly after the Iqbal ruling, the decision was troubling because the Justices accepted the "few bad apples" narrative of detainee mistreatment that has so far prevailed in the government's half-hearted efforts to assign accountability for abuses in the detention and treatment of terrorism suspects.
Yet, as my earlier column also noted, Iqbal's significance transcends detainee or even civil rights cases. The ruling affects all civil lawsuits in the federal courts. Together with the 2007 case of Bell Atlantic v. Twombly, Iqbal makes it substantially easier for a defendant to have a plaintiff's case dismissed before any factual discovery occurs. And thus, conversely, Twombly and Iqbal make it harder for plaintiffs who might have meritorious cases, but need access to defense witnesses and files, to have their cases heard.
Twombly and Iqbal were interpretations of Rule 8 of the Federal Rules of Civil Procedure. Both decisions have been criticized (including by me) as failing to honor the text of the Rule and past precedent interpreting it. That is now water under the bridge, however. As Judge Sotomayor said in answering questions by Wisconsin Senator Herb Kohl during her hearing before the Judiciary Committee, these cases are themselves now precedents entitled to respect by the lower courts and the Supreme Court itself.
Congress, by contrast, can change the rule, and Pennsylvania Senator Arlen Specter has proposed legislation to do just that. Dubbed the "Notice Pleading Restoration Act of 2009," Senator Specter's bill would change the standard for the dismissal of a federal complaint back to the approach set forth by the Supreme Court in the 1957 case of Conley v. Gibson. Under that case, "a complaint should not be dismissed for failure to state a claim unless it appears beyond doubt that the plaintiff can prove no set of facts in support of his claim which would entitle him to relief."
Should Congress enact Senator Specter's proposal? The balance of this column sets forth considerations relevant to answering that question.
Notice Pleading and Dismissal
As I discussed in a column on this site in the aftermath of Twombly, the Federal Rules of Civil Procedure, which were adopted in 1938 and have been amended many times but not fundamentally altered since then, embody a system of "notice pleading." The goal of the Rules is adjudication on the merits, and so each of the pleadings—the plaintiff's complaint and the defendant's answer—need only put the court and the other side "on notice" of what claims or defenses are at issue.
To be sure, the Federal Rules provide for an opportunity to dismiss a case before any discovery where the allegations do not provide a legal basis for relief. Suppose, for example, that I sue my next-door neighbor for "lack of neighborliness" when he holds a dinner party but does not invite me. My lawsuit can properly be dismissed without discovery because no legal liability attaches to lack of neighborliness. In lawyer-speak, there is no "cause of action" for being a bad neighbor.
Trial courts can also dismiss a complaint under a notice pleading regime where the facts alleged, even if true, do not give rise to legal liability. Suppose now that I sue my employer for age discrimination under the Age Discrimination in Employment Act when a promotion goes to a 55-year-old rather than to me. (I am 45.) In this case I have properly invoked a cause of action: Age discrimination in employment can give rise to a federal lawsuit. However, once again, the complaint would be properly dismissed before discovery because the facts alleged make plain that I am suing for discrimination based on relative youth, rather than on age. Yet the Supreme Court's 2004 decision in General Dynamics Land Systems, Inc. v. Cline holds that the Act only protects relatively old workers, not relatively young ones.
In both of the foregoing examples, pre-discovery dismissal is appropriate because the plaintiff is not legally entitled to win his suit even if the facts as alleged by the plaintiff are true. Discovery and a trial would thus be pointless.
Dismissal Under Twombly and Iqbal
In Twombly and Iqbal, the Supreme Court added an additional circumstance under which a trial court may dismiss a complaint without any discovery: "whether, assuming the factual allegations are true, the plaintiff has stated a ground for relief that is plausible." How could allegations that are assumed to be true not be plausible? It is clear from the context of the two cases that the Court has in mind a two-step process: First, the judge must assume that the specific factual allegations are true; and second, the judge must ask whether those specific factual allegations give rise to a plausible further factual conclusion that is itself a basis for legal liability.
For example, in Iqbal, the Court assumed the truth of the specific allegations that Iqbal was mistreated by his captors, but found it implausible that these allegations supported the further inference that the mistreatment was ordered by the Attorney General or FBI Director on the basis of Iqbal's race or religion.
The crucial move in Iqbal—and what marks the case as applying a more stringent pleading standard than would be applied under the pure notice pleading approach—is the Court's unwillingness to assume the truth of the plaintiff's conclusory allegation of complicity by the high-ranking Justice Department officials. Likewise, in Twombly, the Court refused to assume the truth of the plaintiff's conclusory allegation of a conspiracy among telephone service providers. In both instances, the Court demanded additional factual detail—even as it denied that it was applying a "heightened" pleading requirement.
Should Conley Be Restored?
Would we be better off under Conley? In part, that depends on who "we" are. There is a tradeoff between the costs and benefits of any pleading regime.
Liberal notice pleading ensures that plaintiffs with meritorious, but difficult to prove, cases have an opportunity to avail themselves of discovery in order to obtain the evidence they need. However, notice pleading also permits plaintiffs with non-meritorious or even frivolous claims to impose potentially large discovery costs on defendants, thus inducing some of those defendants to settle the litigation for its nuisance value.
Conversely, the stricter regime of Twombly and Iqbal reduces the damage that can be done by frivolous suits, but it also prevents some plaintiffs with meritorious claims from ever having their day in court.
How should these costs and benefits be balanced? We might try to answer that question by tallying up the dollar value of wasted discovery costs imposed by Conley and comparing that figure with the dollar value of uncompensated legal wrongs imposed by Twombly and Iqbal.
In fact, however, that exercise would be very subjective, and so the question will instead likely be treated as more ideological: The plaintiffs' bar, civil rights organizations, and liberals more broadly favor restoring notice pleading; the defense bar, business groups, and conservatives more broadly are happy with the stricter limits of Twombly and Iqbal. Given the political stakes, it is entirely appropriate for Congress to make the decision whether to change the law.
Senator Specter Should Re-Draft the Bill
Suppose that one were to conclude that the more liberal notice pleading regime of Conley v. Gibson should be restored. The Specter bill would still be problematic. Rather than specifying a standard for dismissing lawsuits, the Specter bill simply incorporates the Conley standard by reference.
That would be fine if the Conley standard were well-understood, but it is not. Indeed, in Twombly, the majority correctly reported that for many years the lower courts had frequently ignored or "not taken literally" Conley's "no set of facts" language. There is also reason to think that the Conley Court itself never meant this language to cover conclusory factual allegations but was instead simply making clear that pleadings should be construed generously to invoke any applicable legal theory. The Twombly Court "retired" the liberal language of Conley but otherwise purported to keep faith with that decision.
Accordingly, it is possible that even after the enactment of Senator Specter's proposed Notice Pleading Restoration Act, courts would still require "plausible" pleadings per Twombly and Iqbal. They would be able to say that Twombly and Iqbal merely clarified, rather than overruled, Conley, and that by following Twombly and Iqbal, they are thereby also following Conley. True, such a move would not be consistent with Senator Specter's intent, but courts often follow what they take to be the meaning of the words of a statute, rather than the intent of its sponsor.
Senator Specter's proposed legislation may raise two other, more technical difficulties as well. First, the bill would make Conley the standard for dismissal of a plaintiff's complaint, but would do nothing with respect to the standard for judging a defendant's answer. Although both Twombly and Iqbal involved plaintiffs' complaints under Rule 8(a), some lower federal courts have already held that the new approach also applies to defendants' complaints. Thus, the Specter proposal risks saddling defendants' answers, but not plaintiffs' complaints, with the more rigorous approach of Twombly and Iqbal.
This asymmetry might be explained ideologically. Perhaps Senator Specter and the backers of his bill simply want to help plaintiffs rather than defendants, although even then, toying with the pleading standards is an odd way to do so. Whether a party is technically a plaintiff or a defendant may turn on who files first: A potential defendant who files a declaratory judgment action will be denominated a plaintiff and, under the Specter bill, receive the benefit of the looser pleading standard, while the antitrust or civil rights claimant sued in this manner will be considered a defendant, and thus relegated to the stricter standard of Twombly and Iqbal. There can be no justification for that set of results or, frankly, for the different treatment of plaintiffs and defendants more generally. I suspect that Senator Specter and his staff simply were not thinking about the potential application of their bill to defendants' pleadings in drafting the proposal.
The second technical difficulty is even more, well, technical. Even in the days of Conley, there were select circumstances in which pleaders were required to set forth specific facts, rather than general allegations. Rule 9(b) states that a party alleging fraud or mistake "must state with particularity the circumstances constituting fraud or mistake." By forbidding the dismissal of a complaint except under the Conley standard, the Specter bill would arguably eliminate the heightened specificity requirement for fraud or mistake—but again, only for plaintiffs.
There is something to be said for repealing Rule 9(b). It has always been a bit of a mystery why just fraud and mistake, but not other easy-to-allege-but-difficult-to-prove facts, are covered by the Rule. Yet, if Rule 9(b) is to be repealed, it should be repealed expressly, and for defendants as well as plaintiffs.
Whether Congress should overrule Twombly and Iqbal is an important question with multiple policy dimensions. But if Congress chooses to do so, it can do a better job than the current proposal does. For one alternative, I commend readers (and members of Congress!) to my blog.