[388] Justice O’Connor
delivered the opinion of the Court.
This case presents three issues related to the application of Rule 11 of the Federal Rules of Civil Procedure: whether a district court may impose Rule 11 sanctions on a plaintiff who has voluntarily dismissed his complaint pursuant to Rule 41(a)(1)(i) of the Federal Rules of Civil Procedure; what constitutes the appropriate standard of appellate review of a district court’s imposition of Rule 11 sanctions; and whether Rule 11 authorizes awards of attorney’s fees incurred on appeal of a Rule 11 sanction.*
I
In 1983, Danik, Inc., owned and operated a number of discount men’s clothing stores in the Washington, D. C., area. In June 1983, Intercontinental Apparel, a subsidiary of respondent Hartmarx Corp., brought a breach-of-contract action against Danik in the United States District Court for the District of Columbia. Danik, represented by the law firm of Cooter & Gell (petitioner), responded to the suit by filing a counterclaim against Intercontinental, alleging violations of the Robinson-Patman Act, 49 Stat. 1526, 15 U. S. C. § 13. In March 1984, the District Court granted summary judgment for Intercontinental in its suit against Danik, and, in February 1985, a jury returned a verdict for Intercontinental on Danik’s counterclaim. Both judgments were affirmed on appeal. Danik, Inc. v. Intercontinental Apparel, Inc., 245 U. S. App. D. C. 233, 759 F. 2d 959 (1985) (judgment order); Intercontinental Apparel, Inc. v. Danik, Inc., 251 U. S. App. D. C. 327, 784 F. 2d 1131 (1986) (judgment order).
While this litigation was proceeding, petitioner prepared two additional antitrust complaints against Hartmarx and its [389] two subsidiaries, respondents Hart, Schaffner & Marx and Hickey-Freeman Co. One of the complaints, the one giving rise to the Rule 11 sanction at issue in this case, alleged a nationwide conspiracy to fix prices and to eliminate competition through an exclusive retail agent policy and uniform pricing scheme, as well as other unfair competition practices such as resale price maintenance and territorial restrictions. App. 3-14.
Petitioner filed the two complaints in November 1983. Respondents moved to dismiss the antitrust complaint at issue, alleging, among other things, that Danik’s allegations had no basis in fact. Respondents also moved for sanctions under Rule 11. In opposition to the Rule 11 motion, petitioner filed three affidavits setting forth the prefiling research that supported the allegations in the complaint. Id., at 16-17, 22-23, 24-27. In essence, petitioner’s research consisted of telephone calls to salespersons in a number of men’s clothing stores in New York City, Philadelphia, Baltimore, and Washington, D. C. Petitioner inferred from this research that only one store in each major metropolitan area nationwide sold Hart, Schaffner & Marx suits.
In April 1984, petitioner filed a notice of voluntary dismissal of the complaint, pursuant to Rule 41(a)(1)(i). The dismissal became effective in July 1984, when the District Court granted petitioner’s motion to dispense with notice of dismissal to putative class members. In June 1984, before the dismissal became effective, the District Court heard oral argument on the Rule 11 motion. The District Court took the Rule 11 motion under advisement.
In December 1987, 3½ years after its hearing on the motion and after dismissal of the complaint, the District Court ordered respondents to submit a statement of costs and attorney’s fees. Respondents filed a statement requesting $61,917.99 in attorney’s fees. Two months later, the District Court granted respondents’ motion for Rule 11 sanctions, holding that petitioner’s prefiling inquiry was grossly inade[390] quate. Specifically, the District Court found that the allegations in the complaint regarding exclusive retail agency arrangements for Hickey-Freeman clothing were completely baseless because petitioner researched only the availability of Hart, Schaffner & Marx menswear. In addition, the District Court found that petitioner’s limited survey of only four Eastern cities did not support the allegation that respondents had exclusive retailer agreements in every major city in the United States. Accordingly, the District Court determined that petitioner violated Rule 11 and imposed a sanction of $21,452.52 against petitioner and $10,701.26 against Danik.
The Court of Appeals for the District of Columbia Circuit affirmed the imposition of Rule 11 sanctions. Danik, Inc. v. Hartmarx Corp., 277 U. S. App. D. C. 333, 875 F. 2d 890 (1989). Three aspects of its decision are at issue here.
First, the Court of Appeals rejected petitioner’s argument that Danik’s voluntary dismissal of the antitrust complaint divested the District Court of jurisdiction to rule upon the Rule 11 motion. After reviewing the decisions of other Circuits considering the issue, the Court of Appeals concluded that “the policies behind Rule 11 do not permit a party to escape its sanction by merely dismissing an unfounded case.” Id., at 337, 875 F. 2d, at 894. The court reasoned that because Rule 11 sanctions served to punish and deter, they secured the proper functioning of the legal system “independent of the burdened party’s interest in recovering its expenses.” Id., at 338, 875 F. 2d, at 895. Accordingly, the court held that such sanctions must “be available in appropriate circumstances notwithstanding a private party’s effort to cut its losses and run out of court, using Rule 41 as an emergency exit.” Ibid.
Second, the Court of Appeals affirmed the District Court’s determination that petitioner had violated Rule 11. Petitioner’s arguments failed to “cal[l] into doubt” the two fatal deficiencies identified by the District Court. Rather, peti[391] tioner’s “account of [its] efforts d[id] no more than confirm these shortcomings.” Ibid.
Third, the Court of Appeals considered respondents’ claim that petitioner should also pay the expenses respondents incurred in defending its Rule 11 award on appeal. Relying on Westmoreland v. CBS, Inc., 248 U. S. App. D. C. 255, 770 F. 2d 1168 (1985), the Court of Appeals held that an appellant that successfully defends a Rule 11 award is entitled to recover its attorney’s fees on appeal and remanded the case to the District Court to determine the amount of reasonable attorney’s fees and to enter an appropriate award.
II
The Rules Enabling Act, 28 U. S. C. § 2072, authorizes the Court to “prescribe general rules of practice and procedure and rules of evidence for cases in the United States district courts (including proceedings before Magistrates thereof) and courts of appeals.” The Court has no authority to enact rules that “abridge, enlarge or modify any substantive right.” Ibid. Pursuant to this authority, the Court promulgated the Federal Rules of Civil Procedure to “govern the procedure in the United States district courts in all suits of a civil nature.” Fed. Rule Civ. Proc. 1. We therefore interpret Rule 11 according to its plain meaning, see Pavelic & LeFlore v. Marvel Entertainment Group, 493 U. S. 120, 123 (1989), in light of the scope of the congressional authorization.
Rule 11 provides, in full:
“Every pleading, motion, and other paper of a party represented by an attorney shall be signed by at least one attorney of record in the attorney’s individual name, whose address shall be stated. A party who is not represented by an attorney shall sign the party’s pleading, motion, or other paper and state the party’s address. Except when otherwise specifically provided by rule or statute, pleadings need not be verified or accompanied by affidavit. The rule in equity that the averments of [392] an answer under oath must be overcome by the testimony of two witnesses or of one witness sustained by corroborating circumstances is abolished. The signature of an attorney or party constitutes a certificate by the signer that the signer has read the pleading, motion, or other paper; that to the best of the signer’s knowledge, information, and belief formed after reasonable inquiry it is well grounded in fact and is warranted by existing law or a good faith argument for the extension, modification, or reversal of existing law, and that it is not interposed for any improper purpose, such as to harass or to cause unnecessary delay or needless increase in the cost of litigation. If a pleading, motion, or other paper is not signed, it shall be stricken unless it is signed promptly after the omission is called to the attention of the pleader or movant. If a pleading, motion, or other paper is signed in violation of this rule, the court, upon motion or upon its own initiative, shall impose upon the person who signed it, a represented party, or both, an appropriate sanction, which may include an order to pay to the other party or parties the amount of the reasonable expenses incurred because of the filing of the pleading, motion, or other paper, including a reasonable attorney’s fee.”
An interpretation of the current Rule 11 must be guided, in part, by an understanding of the deficiencies in the original version of Rule 11 that led to its revision. The 1938 version of Rule 11 required an attorney to certify by signing the pleading “that to the best of his knowledge, information, and belief there is good ground to support [the pleading]; and that it is not interposed for delay . . . or is signed with intent to defeat the purpose of this rule.” 28 U. S. C., pp. 2616-2617 (1940 ed.). An attorney who willfully violated the rule could be “subjected to appropriate disciplinary action.” Ibid. Moreover, the pleading could “be stricken as sham and false and the action [could] proceed as though the pleading had not [393] been served.” Ibid. In operation, the Rule did not have the deterrent effect expected by its drafters. See Advisory Committee Note on Rule 11, 28 U. S. C. App., pp. 575-576. The Advisory Committee identified two problems with the old Rule. First, the Rule engendered confusion regarding when a pleading should be struck, what standard of conduct would make an attorney liable to sanctions, and what sanctions were available. Second, courts were reluctant to impose disciplinary measures on attorneys, see ibid., and attorneys were slow to invoke the Rule. Vairo, Rule 11: A Critical Analysis, 118 F. R. D. 189, 191 (1988).
To ameliorate these problems, and in response to concerns that abusive litigation practices abounded in the federal courts, the Rule was amended in 1983. See Schwarzer, Sanctions Under the New Federal Rule 11—A Closer Look, 104 F. R. D. 181 (1985). It is now clear that the central purpose of Rule 11 is to deter baseless filings in district court and thus, consistent with the Rules Enabling Act’s grant of authority, streamline the administration and procedure of the federal courts. See Advisory Committee Note on Rule 11, 28 U. S. C. App., p. 576. Rule 11 imposes a duty on attorneys to certify that they have conducted a reasonable inquiry and have determined that any papers filed with the court are well grounded in fact, legally tenable, and “not interposed for any improper purpose.” An attorney who signs the paper without such a substantiated belief “shall” be penalized by “an appropriate sanction.” Such a sanction may, but need not, include payment of the other parties’ expenses. See ibid. Although the Rule must be read in light of concerns that it will spawn satellite litigation and chill vigorous advocacy, ibid., any interpretation must give effect to the Rule’s central goal of deterrence.
III
We first address the question whether petitioner’s dismissal of its antitrust complaint pursuant to Rule 41(a)(1)(i) [394] deprived the District Court of the jurisdiction to award attorney’s fees. Rule 41(a)(1) states:
“(1) By Plaintiff; by Stipulation. Subject to the provisions of Rule 23(e), of Rule 66, and of any statute of the United States, an action may be dismissed by the plaintiff without order of court (i) by filing a notice of dismissal at any time before service by the adverse party of an answer or of a motion for summary judgment, whichever first occurs, or (ii) by filing a stipulation of dismissal signed by all parties who have appeared in the action. Unless otherwise stated in the notice of dismissal or stipulation, the dismissal is without prejudice, except that a notice of dismissal operates as an adjudication upon the merits when filed by a plaintiff who has once dismissed in any court of the United States or of any state an action based on or including the same claim.”
Rule 41(a)(1) permits a plaintiff to dismiss an action without prejudice only when he files a notice of dismissal before the defendant files an answer or motion for summary judgment and only if the plaintiff has never previously dismissed an action “based on or including the same claim.” Once the defendant has filed a summary judgment motion or answer, the plaintiff may dismiss the action only by stipulation, Rule 41(a)(1)(h), or by order of the court, “upon such terms and conditions as the court deems proper,” Rule 41(a)(2). If the plaintiff invokes Rule 41(a)(1) a second time for an “action based on or including the same claim,” the action must be dismissed with prejudice.
Petitioner contends that filing a notice of voluntary dismissal pursuant to this Rule automatically deprives a court of jurisdiction over the action, rendering the court powerless to impose sanctions thereafter. Of the Courts of Appeals to consider this issue, only the Court of Appeals for the Second Circuit has held that a voluntary dismissal acts as a jurisdictional bar to further Rule 11 proceedings. See Johnson [395] Chemical Co. v. Home Care Products, Inc., 823 F. 2d 28, 31 (1987).
The view more consistent with Rule 11’s language and purposes, and the one supported by the weight of Circuit authority, is that district courts may enforce Rule 11 even after the plaintiff has filed a notice of dismissal under Rule 41(a)(1). See Szabo Food Service, Inc. v. Canteen Corp., 823 F. 2d 1073, 1076-1079 (CA7 1987), cert. dism’d, 485 U. S. 901 (1988); Greenberg v. Sala, 822 F. 2d 882, 885 (CA9 1987); Muthig v. Brant Point Nantucket, Inc., 838 F. 2d 600, 603-604 (CA1 1988). The district court’s jurisdiction, invoked by the filing of the underlying complaint, supports consideration of both the merits of the action and the motion for Rule 11 sanctions arising from that filing. As the “violation of Rule 11 is complete when the paper is filed,” Szabo Food Service, Inc., supra, at 1077, a voluntary dismissal does not expunge the Rule 11 violation. In order to comply with Rule ll’s requirement that a court “shall” impose sanctions “[i]f a pleading, motion, or other paper is signed in violation of this rule,” a court must have the authority to consider whether there has been a violation of the signing requirement regardless of the dismissal of the underlying action. In our view, nothing in the language of Rule 41(a)(1)(i), Rule 11, or other statute or Federal Rule terminates a district court’s authority to impose sanctions after such a dismissal.
It is well established that a federal court may consider collateral issues after an action is no longer pending. For example, district courts may award costs after an action is dismissed for want of jurisdiction. See 28 U. S. C. § 1919. This Court has indicated that motions for costs or attorney’s fees are “independent proceeding[s] supplemental to the original proceeding and not a request for a modification of the original decree.” Sprague v. Ticonic National Bank, 307 U. S. 161, 170 (1939). Thus, even “years after the entry of a judgment on the merits” a federal court could consider an award of counsel fees. White v. New Hampshire Dept. of [396] Employment Security, 455 U. S. 445, 451, n. 13 (1982). A criminal contempt charge is likewise “'a separate and independent proceeding at law’” that is not part of the original action. Bray v. United States, 423 U. S. 73, 75 (1975), quoting Gompers v. Bucks Stove & Range Co., 221 U. S. 418, 445 (1911). A court may make an adjudication of contempt and impose a contempt sanction even after the action in which the contempt arose has been terminated. See United States v. Mine Workers, 330 U. S. 258, 294 (1947) (“Violations of an order are punishable as criminal contempt even though . . . the basic action has become moot”); Gompers v. Bucks Stove & Range Co., supra, at 451 (when main case was settled, action became moot, “of course without prejudice to the power and right of the court to punish for contempt by proper proceedings”). Like the imposition of costs, attorney’s fees, and contempt sanctions, the imposition of a Rule 11 sanction is not a judgment on the merits of an action. Rather, it requires the determination of a collateral issue: whether the attorney has abused the judicial process, and, if so, what sanction would be appropriate. Such a determination may be made after the principal suit has been terminated.
Because a Rule 11 sanction does not signify a district court’s assessment of the legal merits of the complaint, the imposition of such a sanction after a voluntary dismissal does not deprive the plaintiff of his right under Rule 41(a)(1) to dismiss an action without prejudice. “[Dismissal . . . without prejudice” is a dismissal that does not “operat[e] as an adjudication upon the merits,” Rule 41(a)(1), and thus does not have a res judicata effect. Even if a district court indicated that a complaint was not legally tenable or factually well founded for Rule 11 purposes, the resulting Rule 11 sanction would nevertheless not preclude the refiling of a complaint. Indeed, even if the Rule 11 sanction imposed by the court were a prohibition against refiling the complaint (assuming that would be an “appropriate sanction” for Rule 11 purposes), the preclusion of refiling would be neither a consequence of the [397] dismissal (which was without prejudice) nor a “term or condition” placed upon the dismissal (which was unconditional), see Rule 41(a)(2).
The foregoing interpretation is consistent with the policy and purpose of Rule 41(a)(1), which was designed to limit a plaintiff’s ability to dismiss an action. Prior to the promulgation of the Federal Rules, liberal state and federal procedural rules often allowed dismissals or nonsuits as a matter of right until the entry of the verdict, see, e. g., N. C. Code § 1-224 (1943), or judgment, see, e. g., La. Code Prac. Ann., Art. 491 (1942). See generally Note, The Right of a Plaintiff to Take a Voluntary Nonsuit or to Dismiss His Action Without Prejudice, 37 Va. L. Rev. 969 (1951). Rule 41(a)(1) was designed to curb abuses of these nonsuit rules. See 2 American Bar Association, Proceedings of the Institute on Federal Rules, Cleveland, Ohio, 350 (1938) (Rule 41(a)(1) was intended to eliminate “the annoying of a defendant by being summoned into court in successive actions and then, if no settlement is arrived at, requiring him to permit the action to be dismissed and another one commenced at leisure”) (remarks of Judge George Donworth, member of the Advisory Committee on Rules of Civil Procedure); id., at 309; see also 9 C. Wright & A. Miller, Federal Practice and Procedure § 2363, p. 152 (1971). Where state statutes and common law gave plaintiffs expansive control over their suits Rule 41(a)(1) preserved a narrow slice: It allowed a plaintiff to dismiss an action without the permission of the adverse party or the court only during the brief period before the defendant had made a significant commitment of time and money. Rule 41(a)(1) was not designed to give a plaintiff any benefit other than the right to take one such dismissal without prejudice.
Both Rule 41(a)(1) and Rule 11 are aimed at curbing abuses of the judicial system, and thus their policies, like their language, are completely compatible. Rule 41(a)(1) limits a litigant’s power to dismiss actions, but allows one dismissal without prejudice. Rule 41(a)(1) does not codify any policy [398] that the plaintiff’s right to one free dismissal also secures the right to file baseless papers. The filing of complaints, papers, or other motions without taking the necessary care in their preparation is a separate abuse of the judicial system, subject to separate sanction. As noted above, a voluntary dismissal does not eliminate the Rule 11 violation. Baseless filing puts the machinery of justice in motion, burdening courts and individuals alike with needless expense and delay. Even if the careless litigant quickly dismisses the action, the harm triggering Rule 11’s concerns has already occurred. Therefore, a litigant who violates Rule 11 merits sanctions even after a dismissal. Moreover, the imposition of such sanctions on abusive litigants is useful to deter such misconduct. If a litigant could purge his violation of Rule 11 merely by taking a dismissal, he would lose all incentive to “stop, think and investigate more carefully before serving and filing papers.” Amendments to Federal Rules of Civil Procedure, 97 F. R. D. 165, 192 (1983) (Letter from Judge Walter Mansfield, Chairman, Advisory Committee on Civil Rules) (Mar. 9, 1982).
We conclude that petitioner’s voluntary dismissal did not divest the District Court of jurisdiction to consider respondents’ Rule 11 motion. Although Rule 11 does not establish a deadline for the imposition of sanctions, the Advisory Committee did not contemplate that there would be a lengthy delay prior to their imposition, such as occurred in this case. Rather, “it is anticipated that in the case of pleadings the sanctions issue under Rule 11 normally will be determined at the end of the litigation, and in the case of motions at the time when the motion is decided or shortly thereafter.” Advisory Committee Note on Rule 11, 28 U. S. C. App., p. 576. District courts may, of course, “adopt local rules establishing timeliness standards,” White v. New Hampshire Dept. of Employment Security, 455 U. S., at 454, for filing and deciding Rule 11 motions.
[399]
IV
Petitioner further contends that the Court of Appeals did not apply a sufficiently rigorous standard in reviewing the District Court’s imposition of Rule 11 sanctions. Determining whether an attorney has violated Rule 11 involves a consideration of three types of issues. The court must consider factual, questions regarding the nature of the attorney’s prefiling inquiry and the factual basis of the pleading or other paper. Legal issues are raised in considering whether a pleading is “warranted by existing law or a good faith argument” for changing the law and whether the attorney’s conduct violated Rule 11. Finally, the district court must exercise its discretion to tailor an “appropriate sanction.”
The Court of Appeals in this case did not specify the applicable standard of review. There is, however, precedent in the District of Columbia Circuit for applying an abuse-of-discretion standard to the determination whether a filing had an insufficient factual basis or was interposed for an improper purpose, but reviewing de novo the question whether a pleading or motion is legally sufficient. See, e. g., International Brotherhood of Teamsters, Chauffeurs, Warehousemen & Helpers of America (Airline Div.) v. Association of Flight Attendants, 274 U. S. App. D. C. 370, 373, 864 F. 2d 173, 176 (1988); Westmoreland v. CBS, Inc., 248 U. S. App. D. C., at 261, 770 F. 2d, at 1174-1175. Petitioner contends that the Court of Appeals for the Ninth Circuit has adopted the appropriate approach. That Circuit reviews findings of historical fact under the clearly erroneous standard, the determination that counsel violated Rule 11 under a de novo standard, and the choice of sanction under an abuse-of-discretion standard. See Zaldivar v. Los Angeles, 780 F. 2d 823, 828 (1986). The majority of Circuits follow neither approach; rather, they apply a deferential standard to all issues raised by a Rule 11 violation. See Kale v. Combined Ins. Co. of America, 861 F. 2d 746, 757-758 (CA1 1988); Teamsters Local Union No. 430 v. Cement Express, Inc., 841 F. [400]*4002d 66, 68 (CA3), cert. denied, 488 U. S. 848 (1988); Stevens v. Lawyers Mutual Liability Ins. Co. of North Carolina, 789 F. 2d 1056, 1060 (CA4 1986); Thomas v. Capital Security Services, Inc., 836 F. 2d 866, 872 (CA5 1988) (en banc); Century Products, Inc. v. Sutter, 837 F. 2d 247, 250 (CA6 1988); Mars Steel Corp. v. Continental Bank N. A., 880 F. 2d 928, 933 (CA7 1989); Adamson v. Bowen, 855 F. 2d 668, 673 (CA10 1988).
Although the Courts of Appeals use different verbal formulas to characterize their standards of review, the scope of actual disagreement is narrow. No dispute exists that the appellate courts should review the district court’s selection of a sanction under a deferential standard. In directing the district court to impose an “appropriate” sanction, Rule 11 itself indicates that the district court is empowered to exercise its discretion. See also Advisory Committee Note on Rule 11, 28 U. S. C. App., p. 576 (suggesting that a district court “has discretion to tailor sanctions to the particular facts of the case, with which it should be well acquainted”).
The Circuits also agree that, in the absence of any language to the contrary in Rule 11, courts should adhere to their usual practice of reviewing the district court’s findings of fact under a deferential standard. See Fed. Rule Civ. Proc. 52(a) (“Findings of fact. . . shall not be set aside unless clearly erroneous, and due regard shall be given to the opportunity of the trial court to judge of the credibility of the witnesses”). In practice, the “clearly erroneous” standard requires the appellate court to uphold any district court determination that falls within a broad range of permissible conclusions. See, e. g., Anderson v. Bessemer City, 470 U. S. 564, 573-574 (1985) (“If the district court’s account of the evidence is plausible in light of the record viewed in its entirety, the court of appeals may not reverse it even though convinced that had it been sitting as the trier of fact, it would have weighed the evidence differently. Where there are two permissible views of the evidence, the factfinder’s choice [401] between them cannot be clearly erroneous”); Inwood Laboratories, Inc. v. Ives Laboratories, Inc., 456 U. S. 844, 857-858 (1982). When an appellate court reviews a district court’s factual findings, the abuse-of-discretion and clearly erroneous standards are indistinguishable: A court of appeals would be justified in concluding that a district court had abused its discretion in making a factual finding only if the finding were clearly erroneous.
The scope of disagreement over the appropriate standard of review can thus be confined to a narrow issue: whether the court of appeals must defer to the district court’s legal conclusions in Rule 11 proceedings. A number of factors have led the majority of Circuits, see supra, at 399-400, as well as a number of commentators, see, e. g., C. Shaffer & P. Sandler, Sanctions: Rule 11 and Other Powers 14-15 (2d ed. 1988) (hereinafter Shaffer & Sandler); American Judicature Society, Rule 11 in Transition, The Report of the Third Circuit Task Force on Federal Rule of Civil Procedure 11, pp. 45-49 (Burbank, reporter 1989), to conclude that appellate courts should review all aspects of a district court’s imposition of Rule 11 sanctions under a deferential standard.
The Court has long noted the difficulty of distinguishing between legal and factual issues. See Pullman-Standard v. Swint, 456 U. S. 273, 288 (1982) (“Rule 52(a) does not furnish particular guidance with respect to distinguishing law from fact. Nor do we yet know of any other rule or principle that will unerringly distinguish a factual finding from a legal conclusion”). Making such distinctions is particularly difficult in the Rule 11 context. Rather than mandating an inquiry into purely legal questions, such as whether the attorney’s legal argument was correct, the Rule requires a court to consider issues rooted in factual determinations. For example, to determine whether an attorney’s prefiling inquiry was reasonable, a court must consider all the circumstances of a case. An inquiry that is unreasonable when an attorney has months to prepare a complaint may be reasonable when he has only a [402] few days before the statute of limitations runs. In considering whether a complaint was supported by fact and law “to the best of the signer’s knowledge, information, and belief,” a court must make some assessment of the signer’s credibility. Issues involving credibility are normally considered factual matters. See Fed. Rule Civ. Proc. 52; see also United States v. Oregon State Medical Society, 343 U. S. 326, 332 (1952). The considerations involved in the Rule 11 context are similar to those involved in determining negligence, which is generally reviewed deferentially. See Mars Steel Corp. v. Continental Bank N. A., supra, at 932; see also 9 C. Wright & A. Miller, Federal Practice and Procedure § 2590 (1971); McAllister v. United States, 348 U. S. 19, 20-22 (1954) (holding that the District Court’s findings of negligence were not clearly erroneous). Familiar with the issues and litigants, the district court is better situated than the court of appeals to marshal the pertinent facts and apply the fact-dependent legal standard mandated by Rule 11. Of course, this standard would not preclude the appellate court’s correction of a district court’s legal errors, e. g., determining that Rule 11 sanctions could be imposed upon the signing attorney’s law firm, see Pavelic & LeFlore v. Marvel Entertainment Group, 493 U. S. 120 (1989), or relying on a materially incorrect view of the relevant law in determining that a pleading was not “warranted by existing law or a good faith argument” for changing the law. An appellate court would be justified in concluding that, in making such errors, the district court abused its discretion. “[I]f a district court’s findings rest on an erroneous view of the law, they may be set aside on that basis.” Pullman-Standard v. Swint, supra, at 287. See also Icicle Seafoods, Inc. v. Worthington, 475 U. S. 709, 714 (1986) (“If [the Court of Appeals] believed that the District Court’s factual findings were unassailable, but that the proper rule of law was misapplied to those findings, it could have reversed the District Court’s judgment”).
[403] Pierce v. Underwood, 487 U. S. 552 (1988), strongly supports applying a unitary abuse-of-discretion standard to all aspects of a Rule 11 proceeding. In Pierce, the Court held a District Court’s determination under the Equal Access to Justice Act (EAJA), 28 U. S. C. § 2412(d) (1982 ed.), that “the position of the United States was substantially justified” should be reviewed for an abuse of discretion. As a position is “substantially justified” if it “has a reasonable basis in law and fact,” 487 U. S., at 566, n. 2, the EAJA requires an inquiry similar to the Rule 11 inquiry whether a pleading is “well grounded in fact” and legally tenable. Although the EAJA and Rule 11 are not completely analogous, the reasoning in Pierce is relevant for determining the Rule 11 standard of review.
Two factors the Court found significant in Pierce are equally pertinent here. First, the Court indicated that “‘as a matter of the sound administration of justice,’” deference was owed to the “‘judicial actor . . . better positioned than another to decide the issue in question.’” 487 U. S., at 559-560, quoting Miller v. Fenton, 474 U. S. 104, 114 (1985). Because a determination whether a legal position is “substantially justified” depends greatly on factual determinations, the Court reasoned that the district court was “better positioned” to make such factual determinations. See 487 U. S., at 560. A district court’s ruling that a litigant’s position is factually well grounded and legally tenable for Rule 11 purposes is similarly fact specific. Pierce also concluded that the district court’s rulings on legal issues should be reviewed deferentially. See id., at 560-561. According to the Court, review of legal issues under a de novo standard would require the courts of appeals to invest time and energy in the unproductive task of determining “not what the law now is, but what the Government was substantially justified in believing it to have been.” Ibid. Likewise, an appellate court reviewing legal issues in the Rule 11 context would be required to determine whether, at the time the attorney filed the [404] pleading or other paper, his legal argument would have appeared plausible. Such determinations “will either fail to produce the normal law-clarifying benefits that come from an appellate decision on a question of law, or else will strangely distort the appellate process” by establishing circuit law in “a most peculiar, secondhanded fashion.” Id., at 561.
Second, Pierce noted that only deferential review gave the district court the necessary flexibility to resolve questions involving “‘multifarious, fleeting, special, narrow facts that utterly resist generalization.'” Id., at 561-562. The question whether the Government has taken a “substantially justified” position under all the circumstances involves the consideration of unique factors that are “little susceptible . . . of useful generalization.” Ibid. The issues involved in determining whether an attorney has violated Rule 11 likewise involve “fact-intensive, close calls.” Shaffer & Sandler 15. Contrary to petitioner’s contentions, Pierce v. Underwood is not distinguishable on the ground that sanctions under Rule 11 are mandatory: That sanctions “shall” be imposed when a violation is found does not have any bearing on how to review the question whether the attorney’s conduct violated Rule 11.
Rule 11’s policy goals also support adopting an abuse-of-discretion standard. The district court is best acquainted with the local bar’s litigation practices and thus best situated to determine when a sanction is warranted to serve Rule 11’s goal of specific and general deterrence. Deference to the determination of courts on the front lines of litigation will enhance these courts’ ability to control the litigants before them. Such deference will streamline the litigation process by freeing appellate courts from the duty of reweighing evidence and reconsidering facts already weighed and considered by the district court; it will also discourage litigants from pursuing marginal appeals, thus reducing the amount of satellite litigation.
Although district courts’ identification of what conduct violates Rule 11 may vary, see Schwarzer, Rule 11 Revisited, [405] 101 Harv. L. Rev. 1013, 1015-1017 (1988); Note, A Uniform Approach to Rule 11 Sanctions, 97 Yale L. J. 901 (1988), some variation in the application of a standard based on reasonableness is inevitable. “Fact-bound resolutions cannot be made uniform through appellate review, de novo or otherwise.” Mars Steel Corp. v. Continental Bank N. A., 880 F. 2d, at 936; see also Shaffer & Sandler 14-15. An appellate court’s review of whether a legal position was reasonable or plausible enough under the circumstances is unlikely to establish clear guidelines for lower courts; nor will it clarify the underlying principles of law. See Pierce, supra, at 560-561.
In light of our consideration of the purposes and policies of Rule 11 and in accordance with our analysis of analogous EAJA provisions, we reject petitioner’s contention that the Court of Appeals should have applied a three-tiered standard of review. Rather, an appellate court should apply an abuse-of-discretion standard in reviewing all aspects of a district court’s Rule 11 determination. A district court would necessarily abuse its discretion if it based its ruling on an erroneous view of the law or on a clearly erroneous assessment of the evidence. Here, the Court of Appeals determined that the District Court “applied the correct legal standard and offered substantial justification for its finding of a Rule 11 violation.” 277 U. S. App. D. C., at 339, 875 F. 2d, at 896. Its affirmance of the District Court’s liability determination is consistent with the deferential standard we adopt today.
V
Finally, the Court of Appeals held that respondents were entitled to be reimbursed for attorney’s fees they had incurred in defending their award on appeal. Accordingly, it remanded to the District Court “to determine such expenses and, ultimately, to enter an appropriate award.” Id., at 341, 875 F. 2d, at 898. This ruling accorded with the decisions of the Courts of Appeals for the First and Seventh Circuits, see [406] Muthig v. Brant Point Nantucket, Inc., 838 F. 2d, at 607, and Hays v. Sony Corp. of America, 847 F. 2d 412, 419-420 (CA7 1988), and conflicted with the decisions of the Fourth and Ninth Circuits, see Basch v. Westinghouse Electric Corp., 777 F. 2d 165, 175 (CA4 1985), cert. denied, 476 U. S. 1108 (1986), and Orange Production Credit Assn. v. Frontline Ventures Ltd., 801 F. 2d 1581, 1582-1583 (CA9 1986).
On its face, Rule 11 does not apply to appellate proceedings. Its provision allowing the court to include “an order to pay to the other party or parties the amount of the reasonable expenses incurred because of the filing of the pleading, motion, or other paper, including a reasonable attorney’s fee” must be interpreted in light of Federal Rule of Civil Procedure 1, which indicates that the Rules only “govern the procedure in the United States district courts.” Neither the language of Rule 11 nor the Advisory Committee Note suggests that the Rule could require payment for any activities outside the context of district court proceedings.
Respondents interpret the last sentence of Rule 11 as extending the scope of the sanction to cover any expenses, including fees on appeal, incurred “because of the filing.” In this case, respondents argue, they would have incurred none of their appellate expenses had petitioner’s lawsuit not been filed. This line of reasoning would lead to the conclusion that expenses incurred “because of” a baseless filing extend indefinitely. Cf. W. Keeton, D. Dobbs, R. Keeton, & D. Owens, Prosser and Keeton on Law of Torts § 41, p. 264 (5th ed. 1984) (“In a philosophical sense, the consequences of an act go forward to eternity. . . . As a practical matter, legal responsibility must be limited to those causes which are so closely connected with the result and of such significance that the law is justified in imposing liability” (footnote omitted)). Such an interpretation of the Rule is overbroad. We believe Rule 11 is more sensibly understood as permitting an award only of those expenses directly caused by the filing, logically, those at the trial level. A plaintiff’s filing requires the de[407] fendant to take the necessary steps to defend against the suit in district court; if the filing was baseless, attorneys’ fees incurred in that defense were triggered by the Rule 11 violation. If the district court imposes Rule 11 sanctions on the plaintiff, and the plaintiff appeals, the expenses incurred in defending the award on appeal are directly caused by the district court’s sanction and the appeal of that sanction, not by the plaintiff’s initial filing in district court.
The Federal Rules of Appellate Procedure place a natural limit on Rule 11’s scope. On appeal, the litigants’ conduct is governed by Federal Rule of Appellate Procedure 38, which provides: “If a court of appeals shall determine that an appeal is frivolous, it may award just damages and single or double costs to the appellee.” If the appeal of a Rule 11 sanction is itself frivolous, Rule 38 gives appellate courts ample authority to award expenses. Indeed, because the district court has broad discretion to impose Rule 11 sanctions, appeals of such sanctions may frequently be frivolous. See 9 J. Moore, B. Ward, & J. Lucas, Moore’s Federal Practice ¶ 238.03[2], pp. 38-13, 38-14 (2d ed. 1989) (“[W]here an appeal challenges actions or findings of the district court to which an appellate court gives deference by judging under an abuse of discretion or clearly erroneous standard, the court is more likely to find that the appellant’s arguments are frivolous”). If the appeal is not frivolous under this standard, Rule 38 does not require the appellee to pay the appellant’s attorney’s fees. Respondents’ interpretation of Rule 11 would give a district court the authority to award attorney’s fees to the appellee even when the appeal would not be sanctioned under the appellate rules. To avoid this somewhat anomalous result, Rules 11 and 38 are better read together as allowing expenses incurred on appeal to be shifted onto appellants only when those expenses are caused by a frivolous appeal, and not merely because a Rule 11 sanction upheld on appeal can ultimately be traced to a baseless filing in district court.
[408] Limiting Rule 11’s scope in this manner accords with the policy of not discouraging meritorious appeals. If appellants were routinely compelled to shoulder the appellees’ attorney’s fees, valid challenges to district court decisions would be discouraged. The knowledge that, after an unsuccessful appeal of a Rule 11 sanction, the district court that originally imposed the sanction would also decide whether the appellant should pay his opponent’s attorney’s fee would be likely to chill all but the bravest litigants from taking an appeal. See Webster v. Sowders, 846 F. 2d 1032, 1040 (CA6 1988) (“Appeals of district court orders should not be deterred by threats [of Rule 11 sanctions] from district judges”). Moreover, including appellate attorney’s fees in a Rule 11 sanction might have the undesirable effect of encouraging additional satellite litigation. For example, if a district court included appellate attorney’s fees in the Rule 11 sanction on remand, the losing party might again appeal the amount of the award.
It is possible that disallowing an award of appellate attorney’s fees under Rule 11 would discourage litigants from defending the award on appeal when appellate expenses are likely to exceed the amount of the sanction. There is some doubt whether this proposition is empirically correct. See American Judicature Society, Rule 11 in Transition, The Report of the Third Circuit Task Force on Federal Rule of Civil Procedure 11, p. 51 (Burbank, reporter 1989). The courts of appeals have ample authority to protect the beneficiaries of Rule 11 sanctions by awarding damages and single or double costs under Rule 38—which they may do, as we have noted, when the appellant had no reasonable prospect of meeting the difficult standard of abuse of discretion. Beyond that protection, however, the risk of expending the value of one’s award in the course of defending it is a natural concomitant of the American Rule, i. e., that “the prevailing litigant is ordinarily not entitled to collect a reasonable attorneys’ fee from the loser.” Alyeska Pipeline Service Co. v.
Wilderness Society, 421 U. S. 240, 247 (1975). Whenever [409] damages awards at the trial level are small, a successful plaintiff will have less incentive to defend the award on appeal. As Rule 11 is not a fee-shifting statute, the policies for allowing district courts to require the losing party to pay appellate, as well as district court attorney’s fees, are not applicable. “A movant under Rule 11 has no entitlement to fees or any other sanction, and the contrary view can only breed appellate litigation.” American Judicature Society, supra, at 49.
We affirm the Court of Appeals’ conclusion that a voluntary dismissal does not deprive a district court of jurisdiction over a Rule 11 motion and hold that an appellate court should review the district court’s decision in a Rule 11 proceeding for an abuse of discretion. As Rule 11 does not authorize a district court to award attorney’s fees incurred on appeal, we reverse that portion of the Court of Appeals’ judgment remanding the case to the district court for a determination of reasonable appellate expenses. For the foregoing reasons, the judgment of the court below is affirmed in part and reversed in part.
It is so ordered.