Behrens v. Pelletier

Justice Breyer,

with whom Justice Stevens joins,

dissenting.

I do not agree with the Court’s holding that those asserting a defense of qualified immunity are entitled, as a matter of course, to more than one interlocutory appeal. Rather, in my view, the law normally permits a single interlocutory appeal, and not more than one such appeal, from denials of a defendant’s pretrial motions to dismiss a case on grounds of qualified immunity. The “collateral order” doctrine’s basic rationale, this Court’s precedents, and several practical considerations lead to this conclusion.

I

This Court’s basic rationale for permitting an interlocutory appeal of a “collateral order” recognizes that interlocutory appeals are the exception, not the rule. Congress, with statutory exceptions not directly relevant here, has authorized appeals from “final” orders. 28 U. S. C. § 1291 . In that way,

“Congress ... , by forbidding piecemeal disposition on appeal of what for practical purposes is a single controversy, set itself against enfeebling judicial administration [and] . . . the obstruction to just claims that would come from permitting the harassment and cost of a succession of separate appeals from the various rulings to which a litigation may give rise, from its initiation to entry of judgment.” Cobbledick v. United States, 309 U. S. 323, 325 (1940).

Judges have nonetheless created what is, in effect, a non-statutory exception, authorizing a special set of interlocutory *315 appeals, where a trial court’s interlocutory order is a “collateral order” that satisfies the statutory term “final” for purposes of §1291. See Cohen v. Beneficial Industrial Loan Corp., 337 U. S. 541, 545-547 (1949). The trial court’s interlocutory order is “collateral” (and “final”), however, only where it meets certain requirements. It must (1) “conclusively determine [a] disputed question,” (2) “resolve an important issue completely separate from the merits of the action,” and (3) “be effectively unreviewable on appeal from a final judgment.” Coopers & Lybrand v. Livesay, 437 U. S. 463, 468 (1978).

These requirements explain why the courts have created the “collateral order” exception. The “effective unreview-ability” requirement means that failure to review the order on appeal now may cause a litigant permanent harm. The “conclusive determination” requirement means that appellate review now is likely needed to avoid that harm. The “separability” requirement means that review now will not likely force an appellate court to consider the same (or quite similar) questions more than once. Johnson v. Jones, 515 U. S. 304, 311 (1995). Taken together, these requirements, as set forth in the Court’s cases, see, e. g., ibid.; Midland Asphalt Corp. v. United States, 489 U. S. 794, 799 (1989); Gulfstream Aerospace Corp. v. Mayacamas Corp., 485 U. S. 271, 276 (1988), help pick out a class of orders where the error-correcting benefits of immediate appeal likely outweigh the costs, delays, diminished litigation coherence, and waste of appellate court time potentially associated with multiple appeals. See, e. g., Johnson, supra, at 309-311 ; R. Posner, Economic Analysis of Law 585-587 (4th ed. 1992).

In Mitchell v. Forsyth, 472 U. S. 511 (1985), the Court applied this rationale to a District Court order denying a claim of qualified immunity. The Court concluded that the District Court order, by sending the case to trial, could cause the litigant what (in terms of the immunity doctrine’s basic trial-avoiding purpose) would amount to an important harm. *316 See id., at 526-527 . Post-trial appellate review would come too late to avoid that harm. Ibid. And, the legal issue (where purely legal, see Johnson, supra, at 313-318 ) would often prove “separate” enough from the more basic substantive issues in the case to avoid significant duplication of appellate court time and effort. See 472 U. S., at 527-529 ; but see id., at 545-550 (Brennan, J., dissenting). Hence, the “collateral order” doctrine’s basic rationale supported interlocutory appeal.

That same rationale, however, does not support two pretrial interlocutory appeals, the first from a denial of a, motion to dismiss a complaint, the second from a later, postappeal, denial of a motion for summary judgment. Consider the “separability” requirement. Both orders satisfy the literal terms of that requirement because the qualified immunity issues they resolve are both “separate,” in equal measure, from the merits of the plaintiff’s claim. See ante, at 309-310, n. 3. But the reasoned principles and purposes underlying the “separability” requirement are not served by a rule that permits both orders to be appealed because the issues they raise are not normally “separate” one from the other. Rather, they will often involve quite similar issues, likely presented to different appellate court panels, thereby risking the very duplication and waste of appellate resources that the courts intended the “separability” requirement to avoid. See 15A C. Wright, A. Miller, & E. Cooper, Federal Practice and Procedure §3911, pp. 333-334 (2d ed. 1992) (hereinafter Wright & Miller).

Similarly, given the law’s promise of one pretrial interlocutory appeal, a litigant’s need for a second is much less pressing. The single interlocutory appeal can avoid much of, though not all of, the harm that Mitchell found. And, the remaining harm, as I shall next discuss, is not of a kind that the law considers important enough to justify an interlocutory appeal.

*317 II

This Court’s precedents justify one interlocutory appeal, but not more, in the ordinary qualified immunity case. When it initially set forth the “collateral order” exception, the Court said that it applied to “that small class” of orders that determine claims of right “too important to be denied [immediate] review.” Cohen, supra, at 546 (emphasis added). In subsequent cases, and again today, the Court has reiterated that, to qualify for interlocutory appeal, the interest being asserted must be an important one. See, e. g., ante, at 308; Digital Equipment Corp. v. Desktop Direct, Inc., 511 U. S. 863, 878-879 (1994) (Cohen inquiry “simply cannot be answered without a judgment about the value of the interests that would be lost through rigorous application of a final judgment requirement”); Coopers & Lybrand, supra, at 468 (disputed question must “resolve an important issue”); Richardson-Merrell Inc. v. Roller, 472 U. S. 424, 436 (1985); see also Lauro Lines s.r.l. v. Chasser, 490 U. S. 495, 502 (1989) (Scalia, J., concurring) (“The importance of the right asserted has always been a significant part of our collateral order doctrine”). Because one pretrial appeal would normally prove sufficient to protect a government defendant’s qualified immunity interest in not standing trial, the right to take multiple interlocutory appeals will normally protect only the defendant’s additional interest in avoiding such pretrial burdens as discovery. Thus, the question, as Justice Scalia has pointed out, is whether this antidiscov-ery interest is “sufficiently important to overcome the policies militating against interlocutory appeals.” Id., at 503 (emphasis added). The relevant precedent indicates that, in the context of qualified immunity, it is not.

For one thing, the Court, when considering the kinds of orders that warrant interlocutory appeal, has identified as “sufficiently important” interests that are considerably more important than the ordinary interest in avoiding discovery. *318 See, e. g., Stack v. Boyle, 342 U. S. 1 (1951) (interest in avoiding imprisonment; Excessive Bail Clause, U. S. Const., Amdt. 8); Abney v. United States, 431 U. S. 651 (1977) (interest in avoiding trial; Double Jeopardy Clause, U. S. Const., Amdt. 5); Helstoski v. Meanor, 442 U. S. 500 (1979) (interest in avoiding trial; Speech or Debate Clause, U. S. Const., Art. I, §6); Puerto Rico Aqueduct and Sewer Authority v. Metcalf & Eddy, Inc., 506 U. S. 139 (1993) (interest in avoiding trial; Eleventh Amendment immunity, U. S. Const., Amdt. 11).

For another thing, the Court has often said that the trouble, expense, and possible embarrassment associated with unnecessary litigation (interests rather like the qualified immunity antidiscovery interest) do not justify interlocutory appeal. See, e. g., Digital Equipment Corp., supra, at 881-882 (no interlocutory review of orders refusing to enforce a settlement agreement); Lauro Lines, supra, at 499 (no interlocutory review of orders refusing to enforce a forum selection clause); Van Cauwenberghe v. Biard, 486 U. S. 517, 524 (1988) (no interlocutory review of orders refusing to dismiss a civil suit on grounds of immunity from civil process or forum non conveniens).

Further, until now litigants have not been able routinely to vindicate, through immediate appeal, a legal right to avoid discovery, 15B Wright & Miller §3914.23, at 123-130, even where the Constitution provides that antidiscovery right, see, e. g., Maness v. Meyers, 419 U. S. 449, 458-461 (1975) (no interlocutory appeal of order refusing to quash subpoena for materials that arguably violated subpoenaed party’s Fifth Amendment privilege against self-incrimination). Although a litigant can sometimes appeal an adverse discovery ruling, to do so, the litigant typically must disobey the discovery order and then appeal a resulting citation for contempt of court. Church of Scientology of Cal. v. United States, 506 U. S. 9, 18, n. 11 (1992); Maness, supra, at 460-461 ; United *319 States v. Ryan, 402 U. S. 530, 532-533 (1971); Cobbledick v. United States, 309 U. S., at 326-330 ; 15B Wright & Miller § 3914.23, at 140-155. But see United States v. Nixon, 418 U. S. 683, 691 (1974) (allowing President Nixon to appeal from a discovery order without first incurring a contempt citation, because “traditional contempt avenue to immediate appeal” would be “peculiarly inappropriate”). This “disobedience and contempt” requirement (somewhat analogous to a one-appeal limitation here) works, in part, because it “encourages reconsideration both by the party resisting discovery and by the party seeking discovery, and in part because it tends to limit appeals to issues that are both important and reasonably likely to lead to reversal.” 15B Wright & Miller § 3914.23, at 154; see also Pennsylvania v. Ritchie, 480 U. S. 39, 50, n. 8 (1987) (disobedience and contempt procedure “rests on an implicit assumption that unless a party resisting discovery is willing to risk being held in contempt, the significance of his claim is insufficient to justify interrupting the ongoing proceedings”).

It seems highly anomalous for the law to deny a routine interlocutory appeal where the Constitution of the United States protects an antidiscovery interest, but to permit a routine appeal where the legal doctrine of qualified immunity protects a similar interest. Yet, today’s holding will either create just such an anomaly, or, as is more likely, it will generate many new interlocutory appeals as lower courts apply its principle wherever the Constitution, or other important legal doctrine, offers a litigant special anti-discovery protection.

The majority suggests that the importance of the anti-discovery interest protected by qualified immunity has already been “settled” by such precedents as Mitchell v. Forsyth, 472 U. S. 511 (1985), and Harlow v. Fitzgerald, 457 U. S. 800 (1982). See ante, at 308. These cases do say that the qualified immunity defense, in its modern formulation, *320 was meant, in part, “to protect public officials from the ‘broad-ranging discovery’ that can be ‘peculiarly disruptive of effective government.’” Anderson v. Creighton, 483 U. S. 635, 646, n. 6 (1987) (quoting Harlow, supra, at 817 ). But the Court’s decision in Mitchell (that district court orders denying qualified immunity are immediately appealable) was concerned primarily with preserving defendants’ immunity from trial, not discovery. See 472 U. S., at 525 (“At the heart of the issue before .us is the question whether qualified immunity ... is in fact an entitlement not to stand trial”); see also Van Cauwenberghe, supra, at 524 (“The critical question, following Mitchell , is whether ‘the essence’ of the claimed right is a right not to stand trial”). The Court has never before suggested, much less “settled,” that the government defendant’s antidiscovery interest — independent of his interest in avoiding trial — is so important that it must be safeguarded by interlocutory appellate review.

Finally, this Court and its individual Members have, in recent years, cautioned against expanding the class of orders eligible for interlocutory appeal. See, e. g., Digital Equipment Corp. v. Desktop Direct, Inc., 511 U. S., at 868 (opinion of Souter, J.) (“[T]he ‘narrow’ exception should stay that way and never be allowed to swallow the general rule”); Gulfstream Aerospace Corp. v. Mayacamas Corp., 485 U. S., at 292 (Scalia, J., concurring) (“[The Court’s] finality jurisprudence is sorely in need of further limiting principles, so that Cohen appeals will be, as we originally announced they would be, a ‘small class [of decisions].. . too important to be denied review’”); Richardson-Merrell Inc. v. Roller, 472 U. S., at 440 (opinion of O’Connor, J.) (“[W]e decline to ‘transform the limited exception carved out in Cohen into a license for broad disregard of the finality rule imposed by Congress in § 1291’ ”) (quoting Firestone Tire & Rubber Co. v. Risjord, 449 U. S. 368, 378 (1981)). Caution would seem especially appropriate where the Court is considering not one interlocutory appeal in a single case, but two.

*321 III

Several important practical considerations also favor limiting the number of interlocutory qualified immunity appeals to one. The majority finds the necessary special harm in the fact that the qualified immunity doctrine protects public officials against discovery as well as trial; and it finds “separability” in the fact that a postdiscovery summary judgment motion likely asks a legal question that is conceptually distinct from the legal question posed by a prediscovery motion to dismiss a complaint. But, given this rationale, can one limit the number of appeals to just one or two? Would it not, in principle, justify several appeals where discovery, proceeding in stages, continuously turns up new facts, or where, after the close of the plaintiff’s case, an immediate appeal would avoid the litigation burden of presenting an entire defense case.

Still, even two pretrial appeals risk what Justice Story called “very great delays, and oppressive expenses,” Canter. v. American Ins. Co., 3 Pet. 307,318 (1830), which can “ossify civil rights litigation,” Abel v. Miller, 904 F. 2d 394, 396 (CA7 1990) (Easterbrook, J.). The defendant in the present case, for example, so far has spent more than four years (of seven since the complaint’s filing) fighting, through interlocutory appeal, a case that he might well have won more quickly and easily either in the trial court or on appeal from an initially adverse judgment on the merits. Cf. Pelletier v. Federal Home Loan Bank of San Francisco, 968 F. 2d 865, 872-873 (CA9 1992) (expressing doubt that plaintiff’s complaint could survive a summary judgment motion). I concede that every added interlocutory appeal will serve the interests that underlie qualified immunity to some extent, for each will help a government defendant terminate meritless litigation. But each added appeal likely would serve those interests to an ever-diminishing degree while posing an ever-increasing threat to the appearance of evenhanded justice in civil rights cases. See Coopers & Lybrand v. Livesay, 437 U. S., at 476 *322 (no immediate appeal of prejudgment order denying class certification, in part because such appeals would “operat[e] only in favor of plaintiffs”).

Further, as mentioned above, the majority’s rationale threatens added appeals, not simply in qualified immunity cases, but wherever an immunity-type doctrine (or any other important legal rule) seeks to protect litigants from trial. See, e.g., Puerto Rico Aqueduct and, Sewer Authority v. Metcalf & Eddy, Inc., 506 U. S. 139 (1993) (Eleventh Amendment immunity); Nixon v. Fitzgerald, 457 U. S. 731 (1982) (absolute immunity); Abney v. United States, 431 U. S. 651 (1977) (double jeopardy guarantee against successive prosecutions). It thereby threatens busy appellate courts with added numbers of essentially similar, if not repetitive, appeals, at a time when overloaded dockets threaten the federal appellate system. See Remarks of Chief Justice William H. Rehnquist, Tenth Annual Judicial Conference of the United States Court of Appeals for the Federal Circuit, 146 F. R. D. 256, 257 (Apr. 30, 1992) (“One of the chief needs of our generation is to deal with the current appellate capacity crisis in the Federal Courts of Appeals. New could argue about the existence of such a crisis, born of spiraling federal filings and an increased tendency to appeal District Court decisions”); Judicial Conference of the United States, Long Range Plan for the Federal Courts 132 (Dec. 1995) (“[I]f conditions seriously deteriorate in the courts of appeals, it may be necessary to consider some limitations on the right to appeal”). See generally T. Baker, Rationing Justice on Appeal: The Problems of the U. S. Courts of Appeals 31-51 (1994).

Finally, as a practical matter, where the benefits of immediate appellate review predominate in an individual case, a party still can seek court leave to appeal immediately under 28 U. S. C. § 1292 (b) (permitting immediate review of nonfinal orders that involve a controlling and controversial question of law, the appellate resolution of which “may materially *323 advance the ultimate termination of the litigation”). This Court has- frequently observed that the availability of § 1292(b) review counsels against expanding other judicial exceptions to the rule against piecemeal appeals. See, e. g., Swint v. Chambers County Comm’n, 514 U. S. 35, 45-47 (1995); Digital Equipment Corp., 511 U. S., at 883 ; Van Cauwenberghe v. Biard, 486 U. S., at 529-530 ; Richardson-Merrell Inc., 472 U. S., at 435 ; Firestone Tire & Rubber Co., 449 U. S., at 378, n. 13 ; Coopers & Lybrand, supra, at 474-475, and n. 27; see also Parkinson v. April Industries, Inc., 520 F. 2d 650, 658-660 (CA2 1975) (Friendly, J., concurring). We should be especially reluctant to identify new categories of “collateral orders” now that Congress has, by adding 28 U. S. C. § 2072 (c) to the Rules Enabling Act, “designated]... the rulemaking process as the way to define or refine when a district court ruling is ‘final’ and when an interlocutory order is appealable.” Swint, supra, at 48 .

IV

In sum, purpose, precedent, and practicality all argue for one interlocutory qualified immunity appeal per case and no more. I believe that the Court, following Mitchell , should simply hold that qualified immunity interests, while important enough to justify one interlocutory appeal, are not important enough to justify two. It is not necessary to argue about whether the defendant “waived” a second appeal, see Kaiter v. Boxford, 836 F. 2d 704, 708 (CA1 1988); nor, since the matter turns on “importance,” not conclusiveness, need the Court decide just how the timing of an interlocutory appeal affects the “finality” of the trial court’s denial of a motion to dismiss the complaint. See ante, at 307-308. Rather, a defendant asserting qualified immunity would remain free, as at present, to appeal from a denial of a motion to dismiss the complaint, or the defendant could wait, move for summary judgment, and appeal the motion’s denial, but he could not do both — either because the interest asserted *324 in a first pretrial appeal is insufficiently important if the possibility remains of a second pretrial appeal, or because the interest asserted in a second pretrial appeal is insufficiently important if there has already been a first pretrial appeal.

As I said, precedent permits this result because, under that precedent, the importance of the interest (an interlocutory appeal is needed to protect) is one necessary requirement for application of the technical legal labels “final” or “collateral order.” More importantly, meaning in law depends upon an understanding of purpose. Law’s words, however technical they may sound, are not magic formulas; they must be read in light of their purposes, if we are to avoid essentially arbitrary applications and harmful results. For the reasons I have set forth, precedent, read in this way, does more than permit — it requires — a single interlocutory appeal. I therefore dissent.