Linda S. PHINNEY, etc., et al., Plaintiffs, Appellees, v. WENTWORTH DOUGLAS HOSPITAL, et al., Defendants, Appellees, Craig Paulshock, M.D., Defendant, Appellant, 199 F.3d 1

SELYA, Circuit Judge. After a magistrate judge sanctioned defendant-appellant Craig Paulshock, M.D., for discovery misconduct, he appealed unsuccessfully to the district court. He now appeals to us, asserting that the district court misapplied the standard of review and that the magistrate judge made clearly erroneous findings of fact, exceeded his authority by allowing “complex” medical testimony during the sanctions hearing, improperly imposed sequestration, arbitrarily denied a continuance, and misapprehended the burden of proof. For the reasons set forth below, we affirm.

The facts leading to the institution of suit are largely unrelated to the issues on appeal. Suffice it to say that Kenneth J. Phinney died during surgery and his widow (on her own behalf and on behalf of his heirs and estate) invoked diversity jurisdiction, 28 U.S.C. § 1332(a)(1), and brought a medical malpractice action against a plethora of defendants, including Paulshock (the attending anesthesiologist).

In the course of pretrial discovery, Phin-ney requested certain information related to a particular nurse-anesthetist, Elise Jackson (whom Paulshock had entrusted to administer an anesthetic used in the operation). The defendants produced some documents in response to this request, but they failed to supply Jackson’s most relevant personnel evaluation, at first ignoring its existence and later, when Phinney learned of it, claiming that they could not locate the sheet.

To make a tedious tale tolerably terse, the finger of suspicion eventually pointed to Paulshock, and Phinney amended her complaint to allege that he had spoliated evidence. More than six months after Phinney’s original document request, Paul-shock “found” the evaluation sheet in his house. When he finally turned it over, the document proved damning because it revealed that, prior to the date of the surgery, Paulshock had criticized Jackson’s skills and questioned her integrity. This information not only strengthened Phin-ney’s case but also directly contradicted Paulshock’s deposition testimony.

Citing this and other purported discovery abuses, Phinney filed a motion for sanctions against a defense attorney and various defendants (Paulshock included). Magistrate Judge Muirhead conducted a five-day evidentiary hearing and dismissed several of the charges. The one exception involving Paulshock related to the Jackson evaluation. In that regard, the magistrate found Phinney’s allegations to be substantially true and ruled that Paulshock had intentionally withheld the evaluation sheet and had prevaricated at his deposition concerning its whereabouts. See Phinney v.

Paulshock, 181 F.R.D. 185, 200-02 (D.N.H.1998). He ordered Paulshock to pay a sizable monetary sanction to Phinney. See id. at 205. Paulshock secured new counsel and pressed forward, asking us to rectify the situation. While this appeal was pending, we learned that the underlying case had been settled and that, as part of the settlement, the discovery sanction had been paid and Paulshock had agreed not to seek recoupment. He had, however, reserved the right to appeal the magistrate judge’s factual findings.

We questioned whether this reservation was enforceable or, in the alternative, whether the no-recoupment arrangement rendered the appeal moot. Cf. Cordero v. De Jesus-Mendez, 867 F.2d 1, 21 (1st Cir.1989) (explaining that once a “contempt order has been complied with, no case or controversy remains,” and a pending appeal is no longer maintainable). Paulshock unsuccessfully moved for reconsideration and then asked the district court to overturn the decision.

He filed only a general objection (in narrative form) and, when the district court demanded a concise statement of the grounds under-girding it, he responded at some length that certain findings, i.e., that he withheld a document and then lied about it, were not substantiated and should be set aside. The district court undertook clear-error review and rejected Paulshock’s assevera-tions in an unpublished order. We think that the jurisdictional question is close, but we need not resolve it. Generally speaking, we have the power to treat a notice of appeal as a request for a writ of mandamus. See United States v.

Horn, 29 F.3d 754, 769 (1st Cir.1994); United States v. Sorren, 605 F.2d 1211, 1215 (1st Cir.1979). More specifically, we have indicated in circumstances analogous to those at bar that if we lack jurisdiction to entertain an appeal from factual findings that potentially harm the putative appellant’s professional reputation, we may in our discretion treat the notice of appeal as a petition for mandamus. See In re Williams, 156 F.3d 86, 93 n. 7 (1st Cir.1998), cert. denied, - U.S. -, 119 S.Ct. 905, 142 L.Ed.2d 904 (1999).

While we would not take this tack in, say, a case involving chastisement resulting in mere embarrassment, or a case involving an unlikely or unsupported claim of serious career damage, Paulshock’s appellate counsel represents that the disputed factual findings already have caused severe collateral consequences and will continue to cause such consequences. These representations are sufficiently specific and detailed that we are prepared, if necessary, to exercise our mandamus jurisdiction. We therefore proceed to consider Paulshock’s arguments. See Jamison v.

Wiley, 14 F.3d 222, 234 (4th Cir.1994) (adopting such a course); United States v. Collamore, 868 F.2d 24, 27 (1st Cir.1989) (same). From Paulshock’s viewpoint, this victory is short-lived. It is a firmly settled rule that a party’s appeal of a magistrate judge’s order to the district court delimits his right to further appellate review. See Thomas v. Arn, 474 U.S. 140, 147, 106 S.Ct. 466, 88 L.Ed.2d 435 (1985); Keating v. Secretary of HHS, 848 F.2d 271, 274-75 (1st Cir.1988) (per curiam); see also Malave v.

Carney Hosp., 170 F.3d 217, 222 (1st Cir.1999) (remarking “the bedrock rule of appellate practice that ... matters not raised in the trial court cannot be hawked for the first time on appeal”). The Civil Rules are quite explicit on this point: Within 10 days after being served with a copy of the magistrate judge’s order [on a nondispositive motion], a party may serve and file objections to the order; a party may not thereafter assign as error a defect in the magistrate judge’s order to which objection was not timely made. Fed.R.Civ.P. 72(a) (emphasis supplied).

We hold, therefore, that an objection to a magistrate judge’s order must apprise the district court of all the objector’s claims of error, and new claims of error on the part of the magistrate judge cannot thereafter be raised in this court. This holding, straightforwardly applied, bars the majority of Paulshock’s arguments, as Paulshock failed to raise them in his objection to the magistrate’s order. Accordingly, we confíne our substantive analysis of the magistrate’s order to the sole issue that has been properly preserved: the supportability of his factual findings.

Like the district court, we review these factual findings under the “clearly erroneous” rubric. See 28 U.S.C. § 686(b)(1)(A); Fed.R.Civ.P. 72(a). This means that we must accept both the trier’s findings of fact and the conclusions drawn therefrom unless, after scrutinizing the entire record, we “form a strong, unyielding belief that a mistake has been made.” Cumpiano v. Banco Santander P.R., 902 F.2d 148, 152 (1st Cir.1990). In this instance, Magistrate Judge Muirhead’s rescript recounts in meticulous detail the evidence on which he based his findings of fact. See Phinney, 181 F.R.D. at 190-95.

Certain uncontested facts lend support to his conclusion: it took Paul-shock eight months to produce the missing personnel evaluation and, when produced, it turned out to have been in his possession all along, despite his repeated statements that it could not be found, that he did not know its whereabouts, and that others should be queried. Moreover, the magistrate found, with record support, that the ultimate production came only after Paul-shock learned that the plaintiff knew he had had the sheet at one time. See id. at 202. In sum, the magistrate’s findings have deep and sturdy roots in the record.

To be sure, the evidence is not totally one-sided. Paulshock’s counsel adroitly marshals the facts in a forgiving manner, emphasizes Paulshock’s explanation of what transpired, and accurately notes that the key to the puzzle is Paul-shock’s subjective intent. This Herculean effort falls short. Although a party’s continuing denials are entitled to some weight in determining points within his personal knowledge, a trial judge is not obliged to accept such protestations at face value. See Aguilar-Solis v.

INS, 168 F.3d 565, 570-71 (1st Cir.1999) (stating that when the judge “who saw and heard a witness makes an adverse credibility determination ... an appellate court ordinarily should accord it significant respect”). While we, if writing on a pristine page, might have resolved the critical issue of intent differently, we cannot say that the magistrate committed clear error in finding as he did. See United States v. Lara, 181 F.3d 183, 195 (1st Cir.) (explaining that “when the evidence gives rise to competing

interpretations, each plausible, the factfinder’s choice between them cannot be clearly erroneous”), cert. denied, -- U.S.--, 120 S.Ct. 432, 145 L.Ed.2d 338 (1999); see also Carr v. PMS Fishing Corp., 191 F.3d 1, 7 (1st Cir.1999) (holding that, for the most part, credibility determinations are grist for the trier’s mill). We must touch one last base. The narrowness of Paulshock’s objection to the magistrate’s adverse ruling does not restrict his right to complain about what the district court subsequently did (or failed to do).

In this vein, he avers that the district court compounded the magistrate’s alleged error in two respects. First, he contends that the district court should have afforded the magistrate’s order plenary review. His reasoning runs along the following lines. The motion for sanctions sought (or, alternatively, could have yielded) a variety of remedies, including in an extreme case a default judgment. See United States v. Palmer, 956 F.2d 3, 6-7 (1st Cir.1992). In this sense, the motion was potentially dis-positive and thus (Paulshock says) the district court should have reviewed the ensuing order de novo.

This reasoning is flawed. Congress enacted the Federal Magistrates Act in 1968. The 1976 amendments to the Act conferred explicit authority upon district courts to designate magistrate judges to hear pretrial motions. See 28 U.S.C. § 636(b)(1). Congress placed an express limitation on this authority, however, providing that certain enumerated motions could not be determined by a magistrate. See 28 U.S.C. § 636(b)(1)(A).

Such motions may be referred to a magistrate for “proposed findings of fact and recommendations for ... disposition,” but must then be returned to the district judge for final action. 28 U.S.C. § 636(b)(1)(B). Each of the two classes of motions comes equipped with its own standard of review. If a referred motion is within the ambit of

section 636(b)(1)(

A) and not excepted (i.e., to be decided by the magistrate), then the district court can revise it only when a timely objection has been filed and “it has been shown that the magistrate’s order is clearly erroneous or contrary to law.” 28 U.S.C. § 636(b)(1)(A). If, however, the motion is excepted from the customary operation of

section 636(b)(1)(A) (i.e., to engender only a magistrate’s recommendation), the district court is directed, upon the filing of a timely objection, to “make a de novo determination of those portions of the report or specified proposed findings or recommendations to which objection is made.” 28 U.S.C. § 636(b)(1)(C). The court thereafter “may accept, reject, or modify, in whole or in part, the findings or recommendations made by the magistrate.” Id. For many years, federal courts operated under the Federal Magistrates Act without any further guidance.

In 1983, however, these statutory provisions were augmented by the adoption of Fed.R.Civ.P. 72. In framing the rule, the drafters used the term “[njondispositive matters” to refer to all matters (including motions) that a magistrate judge can hear and determine, Fed. R.Civ.P. 72(a), and the term “[dispositive motions” to refer to those motions as to which a magistrate judge can only make a recommendation rather than a ruling, Fed. R.Civ.P. 72(b). Rule 72 mirrors the standard-of-review taxonomy described in

section 636(b)(1). It provides that a magistrate’s order on a nondispositive motion shall be modified or set aside by the district court only if “found to be clearly erroneous or contrary to law.” Fed. R.Civ.P. 72(a). Conversely, if a party contests a magistrate’s proposed findings and recommendations on a dispositive motion, the district judge must “make a de novo determination.” Fed.R.Civ.P. 72(b). We think that the terms dispositive and nondispositive as used in Rule 72 must be construed in harmony with the classifications limned in

section 636(b)(1). See generally In re San Juan Dupont Plaza Hotel Fire Litig., 994 F.2d 956, 962 (1st Cir.1993) (construing Fed.R.Civ.P. 54(

d) in light of 28 U.S.C. § 1920); cf. Marlowe v. Bottarelli, 938 F.2d 807, 813 (7th Cir.1991) (acknowledging the rule that, whenever possible, courts should construe statutes and implementing regulations in pari materia). This does not mean, of course, that dispositive motions are those excepted motions specifically enumerated in

section 636(b)(1)(A), and no others. Rather, that enumeration informs the classification of other motions as dispositive or nondisposi-tive. Motions for sanctions premised on alleged discovery violations are not specifically excepted under 28 U.S.C. § 636(b)(1)(

A) and, in general, they are not of the same genre as the enumerated motions. We hold, therefore, that such motions ordinarily should be classified as non-dispositive. See Maisonville v. F2 Am., Inc., 902 F.2d 746, 747-48 (9th Cir.1990); Thomas E. Hoar, Inc. v. Sara Lee Corp., 900 F.2d 522, 525 (2d Cir.1990); see also 14 James Wm. Moore, et al., Moore’s Federal Practice ¶ 72.02(7)(b) (3d ed.1999). Withal, we caution that a departure from this general rule may be necessary in those instances in which a magistrate judge aspires to impose a sanction that fully disposes of a claim or defense.

See, e.g., Ocelot Oil Corp. v. Sparrow Indus., 847 F.2d 1458, 1462 (10th Cir.1988) (striking of plaintiffs’ pleadings as discovery sanction reviewed de novo); North Am. Watch Corp. v. Princess Ermine Jewels, 786 F.2d 1447, 1450 (9th Cir.1986) (dis-. missal of counterclaim as a discovery sanction reviewed de novo). This case falls within the general rule, not within the rare exception to it. Magistrate Judge Muirhead — whatever he might theoretically have done — -in fact imposed only a monetary sanction. His order, therefore, was nondispositive and the “clearly erroneous” standard obtains.

See 28 U.S.C. § 636(b)(1)(A); Fed.R.Civ.P. 72(a). Indeed, Paulshock himself apparently recognized as much at an earlier stage of the litigation: when he filed his objection to the magistrate’s order in the district court, he denominated it as one filed “pursuant to Fed.R.Civ.P. 72(a).” The second facet of Paulshock’s argument also lacks force. He calumnizes the district court for concentrating its efforts on the testimony of three witnesses and short-shifting the balance of the compendious record. We reject this initiative.

For one thing, the bulk of the five-day hearing dealt with matters other than the Jackson evaluation; none of those matters resulted in the imposition of sanctions against Paul-shock — and none of them formed any part of Paulshock’s assignment of error to the district court. Thus, the district court had every right Jo eschew detailed consideration of them. For another thing, regardless of which evidence the district court deemed most worthy of attention, the acid test is whether, upon our review of the entire record, the findings manifest clear error. See Smith v. F.W.

Morse & Co., 76 F.3d 413, 420 (1st Cir.1996); Cumpiano, 902 F.2d at 152. Here, they do not. We need go no further. Concluding, as we do, that the magistrate’s findings were not clearly erroneous and that the district court properly performed its review function, we reject Paulshock’s appeal. Affirmed. . As matters turned out, the settlement actually had been consummated prior to the completion of proceedings in the district court. . Congress further amended the Act in 1979 to provide that, with the consent of the parties, a magistrate judge may preside in lieu of a district judge. See 28 U.S.C. § 636(c).

Because no such consent occurred here, we exempt such cases from the ensuing discussion. . In civil cases, the excepted motions are motions "for injunctive relief, for judgment on the pleadings, for

summary judgment, ... to dismiss or to permit maintenance of a class action, to dismiss for failure to state a claim upon which relief can be granted, and to involuntarily dismiss an action.” 28 U.S.C. § 636(b)(1)(A). . At least one court has elected to treat an extremely large monetary sanction imposed by a magistrate judge as a recommendation. See Litton Sys., Inc. v. American Tel. & Tel. Co., 90 F.R.D. 410, 420-21 (S.D.N.Y.1981). Because Paulshock has not argued this point, we need not ponder it (and, at any rate, the sanction levied here was by no means extreme).