Gonzalez Abreau v. Banco Central

*754 SELYA, Circuit Judge. This appeal raises tantalizing questions concerning the application of the doctrine of res judicata to nonparties. Because we conclude that appellants cannot lawfully be precluded from bringing their action in the circumstances at bar, we reverse the district court’s order of dismissal and remand for further proceedings. I. BACKGROUND In the 1970s, a consortium of real estate developers sold subdivided lots of undeveloped land to approximately 3,000 purchasers, most of whom resided in Puerto Rico. Contrary to the promoters’ glowing representations, the real estate proved to be Florida swampland, unsuitable for development. In 1982, a gaggle of duped purchasers (whom we shall call “the Rodriguez plaintiffs”) commenced a civil action in the United States District Court for the District of Puer-to Rico. They sued the sellers, the banks that financed the project, 1 and several related individuals. The Rodriguez plaintiffs alleged violations of the Interstate Land Sales Full Disclosure Act (“ILSFDA”), 15 U.S.C. § 1703, the Securities Exchange Act of 1934, 15 U.S.C. § 78j, Rule 10b-5 thereunder, 17 C.F.R. § 240.10b-5, and the Racketeering Influenced and Corrupt Organizations Act (“RICO”), 18 U.S.C. §§ 1961-1964. Some of the plaintiffs then assisted in the formation of the Sunrise Litigation Group. The group’s members paid fees that helped defray the costs of the litigation and exchanged information that sometimes proved to be of use in pursuing the litigation. After several years of discovery and numerous amendments to the pleadings, the Rodriguez plaintiffs, 152 strong, sought to convert their suit to a class action. In April of 1987, the district court refused either to certify a class or to permit additional plaintiffs to intervene. Almost immediately thereafter, several prospective plaintiffs who had tried in vain to join the Rodriguez litigation initiated the instant action. The new coalition of claimants (whom we shall call “the Gonzalez plaintiffs”) were represented by the same lawyers who represented the Rodriguez plaintiffs. They sued the same defendants and their complaint mimicked a proposed amended complaint on file (but never allowed) in the Rodriguez litigation. During the next few years, some of the Gonzalez plaintiffs joined the Sunrise Litigation Group. In the same time frame, they prevailed on no fewer than five motions to bring in additional claimants. And on January 16, 1992, the district court allowed the Gonzalez plaintiffs to amend their complaint to include mail fraud as a RICO predicate act, see 18 U.S.C. § 1962(d), and to include claims for breach of contract and fraud under Puerto Rico law, see, e.g., P.R.Laws Ann. tit. 31, § 3018. Despite strong evidence of skullduggery, 2 the Rodríguez plaintiffs frittered away much of their case through a series of pretrial blunders. See, e.g., Rodriguez v. Banco Central Corp., 727 F.Supp. 759, 763-65 (D.P.R.1989) (dismissing claims under ILSFDA as time-barred), aff'd in part and vacated in part, 917 F.2d 664 (1st Cir.1990); id. at 769-70 (dismissing RICO claims premised on federal securities violations); Rodriguez v. Banco Central Corp., 777 F.Supp. 1043, 1047 (D.P.R.1991) (discussing plaintiffs’ failure to plead certain potentially viable claims). The Rodriguez plaintiffs ultimately lost what remained of their case after a seven-week jury trial when Judge Fuste directed verdicts for the defendants on the only surviving claims and this court upheld his ruling on appeal, see Rodriguez v. Banco Central Corp., 990 F.2d 7, 14 (1st Cir.1993). Following the interment of the Rodriguez litigation, renewed attention focused on the Gonzalez litigation (which was pending be *755 fore Judge Laffitte). By then, the Gonzalez plaintiffs were pressing certain claims that replicated those pressed and lost by the Rodriguez plaintiffs, e.g., claims under the ILSFDA, Rule 10b-5, and RICO (premised on securities fraud), and certain additional claims that had been neglected or abandoned by the Rodriguez plaintiffs, e.g., RICO claims premised on mail fraud, state-law claims for fraud, and claims for breach of contract. After silhouetting the Gonzalez plaintiffs’ suit against the backdrop of the completed Rodriguez litigation, Judge Laffitte, by way of an unpublished memorandum opinion, dismissed the action in its entirety on grounds of res judicata. The Gonzalez plaintiffs appeal. We have jurisdiction pursuant to 28 U.S.C. § 1291. II. ANALYSIS Although appellants were not parties to the earlier litigation, the court below applied res judicata in bar of their claims under a theory of privity. The applicability vel non of the doctrine of res judicata presents a question of law over which we exercise plenary appellate review. See E. & J. Gallo Winery v. Gallo Cattle Co., 967 F.2d 1280, 1287 (9th Cir.1992). Federal law governs the res judicata effects of a federal court judgment in a federal question case as applied to a later case that again presents a federal question to a federal court. See Blonder-Tongue Labs., Inc. v. University of Ill. Found., 402 U.S. 313, 324 n. 12, 91 S.Ct. 1434, 1440 n. 12, 28 L.Ed.2d 788 (1971); Kale v. Combined Ins. Co., 924 F.2d 1161, 1165 (1st Cir.), cert. denied, — U.S. -, 112 S.Ct. 69, 116 L.Ed.2d 44 (1991); see also 18 Charles A. Wright, et al., Federal Practice and Procedure § 4466, at 617-18 (1981) (hereinafter ‘Wright & Miller”). Thus, because both the earlier (ostensibly precluding) suit and the later (ostensibly precluded) suit invoked federal question jurisdiction, see 28 U.S.C. § 1331, the rule of decision here is supplied by federal law. The accepted formulation of res judicata for federal court use teaches that “a final judgment on the merits of an action precludes the parties or their privies from relitigating issues that were or could have been raised in that action.” Allen v. McCurry, 449 U.S. 90, 94, 101 S.Ct. 411, 414, 66 L.Ed.2d 308 (1980). Accordingly, the elements of res judicata are (1) a final judgment on the merits in an earlier suit, (2) sufficient identicality between the causes of action asserted in the earlier and later suits, and (3) sufficient identicality between the parties in the two suits. See Aunyx Corp. v. Canon U.S.A., Inc., 978 F.2d 3, 6 (1st Cir.1992), cert. denied, — U.S. -, 113 S.Ct. 1416, 122 L.Ed.2d 786 (1993); Kale, 924 F.2d at 1165. In the present situation, the first element in this tripartite test provokes no controversy; appellants concede that the earlier (Rodriguez) suit resulted in final judgment on the merits. Thus, we concentrate our energies on the remaining two prongs of the test. A. Identicality of Causes of Action. To determine whether sufficient subject matter identity exists between an earlier and a later suit, federal courts employ a transactional approach. See Kale, 924 F.2d at 1166; Manego v. Orleans Bd. of Trade, 773 F.2d 1, 5 (1st Cir.1985), cert. denied, 475 U.S. 1084, 106 S.Ct. 1466, 89 L.Ed.2d 722 (1986); see also Restatement (Second) of Judgments § 24 (1992). This approach recognizes that a valid and final judgment in an action will extinguish subsequent claims “with respect to all or any part of the transaction, or series of connected transactions, out of which the action arose.” Manego, 773 F.2d at 5 (quoting Restatement § 24). To understand the transactional approach, it is necessary to appreciate that a single transaction or series of transactions can — and often does — give rise to a multiplicity of claims. Phrased another way, “[a] single cause of action can manifest itself in an outpouring of different claims, based variously on federal statutes, state statutes, and the common law.” Kale, 924 F.2d at 1166. The necessary identity will be found to exist if both sets of claims — those asserted in the earlier action and those asserted in the subsequent action — derive from a common nucleus of operative facts. See id. This principle pertains no matter how diverse or prolific the claims themselves may be. See IB J. *756 Moore, Federal Practice ¶ 0.410[1] at 350 (2d ed. 1993) (explaining that “the ‘cause of action’ or ‘claim’ ... is bounded by the injury for which relief is demanded, and not by the legal theory”). It follows that the omission of a particular statement of claim from the original suit is of no great consequence; if the transaction is the same and the other components of the test are satisfied, principles of res judicata will bar all claims that either were or could have been asserted in the initial action. See Kale, 924 F.2d at 1166; Manego, 773 F.2d at 5. The key is to define the underlying injury. This definitional process is not a purely mechanical exercise. “What factual grouping constitutes a ‘transaction’, and what groupings constitute a ‘series’, are [matters that should] be determined pragmatically,” taking into consideration a wide variety of relevant factors, including but not limited to such'things as “whether the facts are related in time, space, origin, or motivation, whether they form a convenient trial unit, and whether their treatment as a unit conforms to the parties’ expectations_” Aunyx, 978 F.2d at 7 (quoting Restatement (Second) of Judgments § 24). Given these criteria, we believe that there is sufficient identicality here between the earlier and later actions to satisfy the requisite standard. Without exception, appellants’ claims stem from the same series of transactions as the claims asserted in the initial litigation. Although the individual sales contracts are different, all of them arise out of a single course of conduct undertaken by a band of allied defendants. By like token, while each purchaser acquired a different lot at a different price, all the lots are part of the same development and all were sold by means of the same ballyhoo. At the very least, the two sets of claims are closely related in time, origin, and geography. Moreover, if merged, the two sets of claims would form a well-integrated unit. The same kinds of land sale contracts that the Rodriguez plaintiffs attacked under ILSFDA and sought to characterize as “securities” for purposes of their RICO claim, see Rodriguez, 990 F.2d at 9, underlie appellants’ current claims. To be sure, appellants have negotiated the procedural minefield more nimbly than their predecessors, and have, therefore, assembled a more varied assortment of legal theories; but their claims — including both those that replicate the Rodriguez plaintiffs’ claims and those that do not — implicate the same series of interconnected transactions that gave rise to the causes of action litigated in the earlier lawsuit. In short, both sets of claims, though dressed in different legal garb, grow out of a common nucleus of operative facts. No more is exigible. B. Identicality of Parties. Concluding, as we do, that the district court’s analysis passes muster on the first two components of the tripartite test, we turn to the third essential ingredient needed to invoke the doctrine of res judicata: the presence of a sufficient identity between the parties to the earlier and later actions. Short of situations in which precisely the same parties appear in both suits, this element is almost always difficult to gauge. 1. Nonparty Preclusion. We step back to gain a sense of perspective. We are aware that a Supreme Court dictum can be read to suggest that res judicata is inoperative as a matter of law insofar as nonparties are concerned. See Montana v. United States, 440 U.S. 147, 154, 99 S.Ct. 970, 974, 59 L.Ed.2d 210 (1979) (“Preclusion of ... nonparties falls under the rubric of collateral estoppel rather than res judicata because the latter doctrine presupposes identity between causes of action. And the cause of action which a nonparty has vicariously asserted differs by definition from that which he subsequently seeks to litigate in his own right.”) (dictum). We believe it is highly improbable, however, that the Montana Court, whose primary interest lay in molding the contours of the related doctrine of collateral estoppel, 3 meant categorically to banish privity — a *757 time-honored concept that collapses distinctions between form and substance in respect to party status — from use in conjunction with principles of res judicata. This conclusion is firmly supported not only by respectable precedent but also by practical considerations. Notwithstanding the Montana dictum, several courts, including this court, continue to apply res judicata to nonparties when the circumstances warrant. See, e.g., Aunyx, 978 F.2d at 7-8 (applying res judicata to preclude the alter ego of a corporation from relitigating); In re Air Crash at Dallas/Fort Worth Airport, 861 F.2d 814, 816-18 (5th Cir.1988) (applying res judicata to bar decedent’s daughter from re-litigating); see also Restatement (Second) of Judgments §§ 40, 41 (endorsing application of claim preclusion to nonparties in specified circumstances). In the same vein, courts continue routinely to formulate res judicata as a doctrine that bars parties “or their privies” from relitigating claims. See, e.g., Allen, 449 U.S. at 94, 101 S.Ct. at 414; Kale, 924 F.2d at 1165; In re Air Crash, 861 F.2d at 816; United States v. Athlone Indus., Inc., 746 F.2d 977, 983 (3d Cir.1984); Lee v. City of Peoria, 685 F.2d 196, 199 (7th Cir.1982). There are also strong practical considerations that counsel against blind adherence to the Montana dictum. The doctrine of res judicata serves many desirable ends, among them finality and efficiency. See Montana, 440 U.S. at 153, 99 S.Ct. at 973. Logic suggests that the doctrine can achieve its goals only if its preclusive effects occasionally can reach persons who, technically, were not parties to the original action. The pitfalls of a more mechanical rule are obvious; making party status a sine qua non for the operation of res judicata opens the door to countless varieties _ of manipulation, including claim-splitting, suits by proxy, and forum-shopping. Finally, reading Montana’s dictum as categorically eliminating res judicata whenever there are technically distinct parties is at loggerheads with the hoary concept of privity — a concept long since integrated into the legal lexicon and routinely applied in analogous situations. See, e.g., Stacy v. Thrasher, 47 U.S. (6 How.) 44, 51, 12 L.Ed. 337 (1848) (applying privity to determine the binding effect of court judgments); Wallingsford v. Larcon Co., 237 F.2d 904, 906 (8th Cir.1956) (applying privity to determine the extent of the res judicata effect of a prior judgment). We are loath to assume that the Court intended to wrest this concept from the jurisprudence of res judicata by a casual observation, bereft of any meaningful discussion or explanation. As a rule, appellate courts do not operate in so Delphic a fashion. See, e.g., United States v. Zapata, 18 F.3d 971, 977 (1st Cir.1994) (rejecting argument that “an unheralded dictum” in a Supreme Court opinion altered settled Fourth Amendment concepts and thereby “worked a sea change in the law”). We find this combination of precedent, policy, and practicalities to be irresistible. Consequently, we hold that, under federal law, res judicata can sometimes operate to bar the maintenance of an action by persons who, technically, were not parties to the initial action (to which preclusive effect is attributed). Nonetheless, we appreciate that this is a murky corner of the law and caution the district courts to tread gingerly in applying res judicata to nonparties. 4 2. Privity. The most familiar mechanism for extending res judicata to nonparties without savaging important constitutional rights is the concept of privity — a concept that furnishes a serviceable framework for an exception to the rule that res judicata only bars relitigation of claims by persons who were parties to the original litigation. See Meza v. General Battery Corp., 908 F.2d 1262, 1266 (5th Cir.1990); see also NLRB v. Donna-Lee Sportswear Co., 836 F.2d 31, 34 (1st Cir.1987) (applying same exception in connection with issue preclusion). Although privity can be elusive, this case does not require us to build four walls *758 around it. Here, the res judicata defense is based not on some exotic doctrinal refinement but on commonly accepted principles of how privity operates to bring about nonparty preclusion. The theory underlying defendants’ iteration of the defense is that privity exists (and, therefore, nonparty preclusion potentially obtains) if a nonparty either substantially controlled a party’s involvement in the initial litigation or, conversely, permitted a party to the initial litigation to function as his de facto representative. 5 We accept defendants’ theoretical premise, but, after close perscrutation of the record as a whole, we conclude that neither stripe of privity exists here. Substantial Control The doctrine of res judicata rests upon the bedrock principle that, for claim preclusion to apply, a litigant first must have had a full and fair opportunity to litigate his claim. See Fiumara v. Fireman’s Fund Ins. Cos., 746 F.2d 87, 92 (1st Cir.1984); see also 18 Wright & Miller, supra, § 4449, at 417 (noting “[o]ur deep-rooted historic tradition that everyone should have his own day in court”); cf. Blonder-Tongue, 402 U.S. at 328, 91 S.Ct. at 1442 (commenting that it is sufficient to afford a litigant one “full and fair opportunity for judicial resolution” of a particular issue). If a nonparty either participated vicariously in the original litigation by exercising control over a named party or had the opportunity to exert such control, then the nonparty effectively enjoyed his day in court, and it is appropriate to impute to him the legal attributes of party status for purposes of claim preclusion. See United States v. Bonilla Romero, 836 F.2d 39, 44 (1st Cir.1987), ce rt. denied, 488 U.S. 817, 109 S.Ct. 55, 102 L.Ed.2d 33 (1988); see also 18 Wright & Miller, supra, § 4451, at 430 (arguing that “[preclusion is fair so long as the relationship between the nonparty and a party was such that the nonparty had the same practical opportunity to control the course of the proceedings that would be available to a party”); cf. Montana, 440 U.S. at 154, 99 S.Ct. at 974 (finding issue preclusion appropriate “when nonparties assume control over litigation in which they have a direct financial or proprietary interest and then seek to redetermine issues previously resolved”); Restatement (Second) of Judgments § 39 (similar). 6 Substantial control means what the phrase implies; it connotes the availability of a significant degree of effective control in the prosecution or defense of the case — what one might term, in the vernacular, the power— whether exercised or nob — to call the shots. 7 See Rumford Chem. Works v. Hygienic Chem. Co., 215 U.S. 156, 160, 30 S.Ct. 45, 46, 54 L.Ed. 137 (1909) (holding that the concept of substantial control refers to “the right to intermeddle in any way in the conduct of the case”); Hy-Lo Unit & Metal Prods. Co. v. Remote Control Mfg. Co., 83 F.2d 345, 350 (9th Cir.1936) (stating that substantial control means the “right to participate and control such prosecution or defense”); Restatement (Second) of Judgments § 39, comment c, at 384 (stating that control, for purposes of issue preclusion, refers to the right to exercise “effective choice as to the legal theories and proofs to be advanced,” as well as “control over the opportunity to obtain review”); see generally IB Moore, supra, ¶ 0.411[6] at 456-58. As the proverb suggests, a picture is sometimes worth a thousand words. Along these *759 lines, we suspect that the concept of substantial control can be illustrated better by examples than by linguistic constructs. For instance, substantial control has been found in the case of a liability insurer that assumes the insured’s defense, see, e.g., Iacaponi v. New Amsterdam Cas. Co., 379 F.2d 311, 312 (3d Cir.1967), cert. denied, 389 U.S. 1054, 88 S.Ct. 802, 19 L.Ed.2d 849 (1968), an indemnitor who participates in defending an action brought against the indemnitee, see, e.g., Bros, Inc. v. W.E. Grace Mfg. Co., 261 F.2d 428, 430-31 (5th Cir.1958), and the owner of a close corporation who assumes control of litigation brought against the firm, see, e.g., Kreager v. General Elec. Co., 497 F.2d 468, 471-72 (2d Cir.), cert. denied, 419 U.S. 1041, 95 S.Ct. 530, 42 L.Ed.2d 319 (1974). Conversely, courts have refused to find substantial control merely because a nonparty retained the attorney who represented a party to the earlier action, see Freeman v. Lester Coggins Trucking, Inc., 771 F.2d 860, 864 (5th Cir.1985); Ramey v. Rockefeller, 348 F.Supp. 780, 785 (E.D.N.Y.1972), or because the nonparty assisted in financing the earlier action, see Rumford Chem., 215 U.S. at 159-60, 30 S.Ct. at 45-46; General Foods Corp. v. Massachusetts Dep’t of Pub. Health, 648 F.2d 784, 787-88 (1st Cir.1981), or because the nonparty testified as a witness in the earlier action, see Benson & Ford, Inc. v. Wanda Petroleum Co., 833 F.2d 1172, 1174-75 (5th Cir.1987); Ponderosa Devel. Corp. v. Bjordahl, 787 F.2d 533, 536-37 (10th Cir.1986), or because the nonparty procured witnesses or evidence, see Carl Zeiss Stiftung v. V.E.B. Carl Zeiss, Jena, 293 F.Supp. 892, 921 (S.D.N.Y.1968), modified, 433 F.2d 686 (2d Cir.1970), cert. denied, 403 U.S. 905, 91 S.Ct. 2205, 29 L.Ed.2d 680 (1971), or because the nonparty furnished his attorney’s assistance, see Cofax Corp. v. Minn. Mining & Mfg. Co., 79 F.Supp. 842, 844 (S.D.N.Y.1947). In the last analysis, there is no bright-line test for gauging substantial eon-trol. The inquiry must be case-specific, see IB Moore, supra, ¶ 0.411[6] at 458, and fact patterns are almost endlessly variable. The critical judgment cannot be based on isolated facts. Consequently, an inquiring court must consider the totality of the circumstances to determine whether they justify a reasonable inference of a nonparty’s potential or actual involvement as a decisionmaker in the earlier litigation. The nonparty’s participation may be overt or covert, and the evidence of it may be direct or circumstantial — so long as the evidence as a whole shows that the nonparty possessed effective control over a party’s conduct of the earlier litigation as measured from a practical, as opposed to a purely theoretical, standpoint. The burden of persuasion ultimately rests with him who asserts that control (or the right to exercise it) existed to such a degree as would warrant invoking nonparty preclusion. See id. Applying this standard, there is no principled way in which it can be said that the Gonzalez plaintiffs substantially controlled the Rodriguez plaintiffs in regard to the original litigation. The only facts to which the district court alluded in ruling that nonparty preclusion loomed involve the similarity of the complaints at one point in time, the parties’ common legal representation, and the planned use of some discovered materials in both litigations. In our view, these facts do not begin to show that the Gonzalez plaintiffs exercised any meaningful degree of control over the course of the Rodriguez litigation. Nor did they have either the right or the opportunity to demand such control. 8 Moreover, the record contains much additional evidence indicating the absence of substantial control. No useful purpose would be served by marshalling this evidence. We do, however, remark the most telling datum: that the Rodriguez plaintiffs sought to amend their complaint to add those who later be *760 came the Gonzalez plaintiffs a full half-decade after the start of the litigation — a datum strongly suggesting that appellants had no involvement in the initial five years of litigation. This lack of participation at the early stages of the Rodriguez litigation is particularly probative on the issue of substantial control, for it was during this period that many pivotal strategic decisions were made, resulting in the virtual forfeiture of some especially promising causes of action (including the mail fraud and state-law claims). Obviously, appellant had no chance to share in this decisionmaking. Virtual Representation The defendants also attempt to sustain the application of res judicata by employing principles of virtual representation to demonstrate that privity exists. The attempt stalls. Following defendants’ itinerary would require us to imbue the theory of virtual representation with a much greater cruising range than either the law or the facts permit. Although rooted in the eighteenth century law of estates, virtual representation has only recently emerged as a vehicle for general nonparty preclusion. See Robert G. Bone, Rethinking the ‘Day in Court’ Ideal and Nonparty Preclusion, 67 N.Y.U.L.Rev. 193, 206-219 (1992). Its recent jurisprudential history has been characterized by breadth of initial articulation followed by abrupt retrenchment in actual application. These per-errations, and the competing centrifugal and centripetal forces that account for them, are most easily explained by reference to the due process analyses that must guide any effort to place the theory into practice. See, e.g., Meza, 908 F.2d at 1266. The courts that first rode the warhorse of virtual representation into battle on the res judicata front invested their steed with near-magical properties. They suggested that mere identity of interests between party and nonparty warranted application of the theory and, hence, authorized nonparty preclusion. See, e.g., Aerojet-General Corp. v. Askew, 511 F.2d 710, 719 (5th Cir.) (holding that, under federal law, “a person may be bound by a judgment even though not a party if one of the parties to the suit is so closely aligned with his interests as to be his virtual representative”), cer t. denied, 423 U.S. 908, 96 S.Ct. 210, 46 L.Ed.2d 137 (1975). Despite such sweeping generalities, courts soon came to realize that, though virtual representation was not the old gray mare, neither should it be confused with Pegasus; finding virtual representation based solely on identity of interests, and then deploying the theory to justify nonparty preclusion in a broad spectrum of cases, would threaten the core principles underpinning the due process equation. See Martin v. Wilks, 490 U.S. 755, 761-62, 109 S.Ct. 2180, 2184, 104 L.Ed.2d 835 (1989); Meza, 908 F.2d at 1266. For this reason, contemporary caselaw has placed the theory of virtual representation on a short tether, significantly restricting its range. See Benson & Ford, 833 F.2d at 1175 (observing that the theory of virtual representation must be kept within strict confines); Pollard v. Cockrell, 578 F.2d 1002, 1008-09 (5th Cir.1978) (explicitly limiting Aerojet holding); see generally 18 Wright & Miller, supra, § 4457 at 355 (Supp.1994) (discussing “narrow role” that remains for virtual representation). The upshot is that, today, while identity of interests remains a necessary condition for triggering virtual representation, it is not alone a sufficient condition. More is required to bring the theory to bear. 9 See General Foods, 648 F.2d at 789 (holding that “identity of interests” between a party and a nonparty “does not bind [the nonparty] to the judgment”); Griffin v. Burns, 570 F.2d 1065, 1071 (1st Cir.1978) (explaining that “[m]ere similarity of interests and a quantum of representation” is insufficient to trigger virtual representation); Petit v. City of Chicago, 766 F.Supp. 607, 612 (N.D.Ill.1991) (holding that “identity of interests alone ... is not sufficient to yield a finding of privity”); see also Benson & Ford, 833 F.2d at 1174-76 (declining to find, nonparty preclusion anent *761 an antitrust claim growing out of the same facts where the nonparty plaintiff testified at the earlier trial and had the same attorney); see generally 18 Wright & Miller, supra, § 4457, at 500. To say that a litigant advocating virtual representation, and seeking thereby to preclude a nonparty’s suit, must show more than an identity of interests is to state the nature of the problem, not to solve it. Many of the ensuing questions — questions like “how much more?” and “what comprises ‘more’?” — seem to have no categorical answers. Not surprisingly, then, the cases in which courts have dealt with the doctrine, taken as an array, are resistant to doctrinal rationalization in the form of a single elegant limiting principle of the “one size fits ah” variety. There is no black-letter rule. See Colby v. J.C. Penney Co., 811 F.2d 1119, 1125 (7th Cir.1987) (commenting that “no uniform pattern has emerged from the cases”); Ethnic Employees of Library of Congress v. Boorstin, 751 F.2d 1405, 1411 n. 8 (D.C.Cir.1985) (noting that the virtual representation doctrine has a “highly uncertain scope”); see also Bone, supra, 67 N.Y.U.L.Rev. at 220 (acknowledging absence of clear organizing framework). In the end, virtual representation is best understood as an equitable theory rather than as a crisp rule with sharp corners and clear factual predicates, see 18 Wright & Miller, supra, § 4457 at 502, such that a party’s status as a virtual representative of a nonparty must be determined on a case-by-case basis, see Bonilla Romero, 886 F.2d at 43. Although the need for individualized analysis persists, a common thread binds these variegated cases together: virtual representation has a pronounced equitable dimension. Thus, notwithstanding identity of interests, virtual representation will not serve to bar a nonparty’s claim unless the nonparty has had actual or constructive notice of the earlier litigation, 10 and the balance of the relevant equities tips in favor of preclusion. For example, courts have applied the doctrine in situations in which a nonparty has given actual or implied consent to be bound by the results in a prior action, see, e.g., Boyd v. Jamaica Plain Co-op Bank, 7 Mass.App.Ct. 153, 386 N.E.2d 775, 778-81 (1979); see also Benson & Ford, 833 F.2d at 1176 (finding “tacit agreements]” to be bound characteristic of cases applying virtual representation), or in which there has been “an express or implied legal relationship in which parties to the first suit are accountable to non-parties who file a subsequent suit raising identical issues,” Pollard, 578 F.2d at 1008; see also In re Medomak Canning Co., 922 F.2d 895, 900-01 (1st Cir.1990) (holding that creditors were represented by the trustee in bankruptcy, who had a fiduciary relationship to them), or in which certain types of familial relationships link parties and nonparties, see, e.g., Eubanks v. FDIC, 977 F.2d 166, 170 (5th Cir.1992) (holding wife bound by outcome of bankrupt husband’s prior action); Stone v. Williams, 970 F.2d 1043, 1058-61 (2d Cir.1992) (binding decedent’s son to a prior ruling concerning legacies), cert. denied, — U.S. -, 113 S.Ct. 2331, 124 L.Ed.2d 243, or in which courts have detected tactical maneuvering designed unfairly to exploit technical nonparty status in order to obtain multiple bites of the litigatory apple, see, e.g., Petit, 766 F.Supp. at 611-13; Crane v. Comm’r of Dep’t of Agric., 602 F.Supp. 280, 286-88 (D.Me.1985); see also 18 Wright & Miller, supra, § 4457, at 498-99; Bone, supra, at 222. Implicit in all these scenarios is the existence of actual or constructive notice. 11 We have considered, and rejected, another possible common characteristic. Some courts have suggested that adequacy of *762 representation is also a condition precedent to nonparty preclusion grounded upon virtual representation. See, e.g., Clark v. Amoco Prods. Co., 794 F.2d 967, 973-74 (5th Cir.1986) (suggesting that virtual representation “closely resembles the common law theory of concurrent privity ... which in turn is really only [an] adequate representation of interests analysis”); Delta Air Lines, Inc. v. McCoy Restaurants, Inc., 708 F.2d 582, 587 (11th Cir.1983) (finding no virtual representation because nonparty was not “adequately represented”); cf. 18 Wright & Miller, supra, § 4457, at 355-58 (1994 Supp.) (suggesting somewhat cryptically that “adequate litigation” should “remain[] the central requirement” for nonparty preclusion based on principles of virtual representation). Properly viewed, however, adequacy of representation is not itself a separate and inflexible requirement for engaging principles of virtual representation, 12 although it is one of the factors that an inquiring court should weigh in attempting to balance the equities. 13 Based on these benchmarks, the Gonzalez plaintiffs cannot plausibly be said to have been virtually represented by the Rodriguez plaintiffs notwithstanding the identity of interests between the two groups. Here, the equities counsel very strongly against deploying the theory of virtual representation. In the first place, there has been no showing that the Gonzalez plaintiffs had timely notice of the first suit. 14 In the second place, the parties’ independence — the inescapable fact that the Rodriguez plaintiffs were not legally responsible for, or in any other way accountable to, the Gonzalez plaintiffs — weighs heavily against a finding of virtual representation. 15 See Benson & Ford, 833 F.2d at 1176. In the third place, the lack of a special type of close relationship between the two groups of plaintiffs (who are, for the most part, unrelated lambs purportedly fleeced by the same cadre of unscrupulous sheepherders) also weighs against a finding of virtual representation. See Eubanks, 977 F.2d at 170. Fourth, the fact that the Gonzalez plaintiffs never consented, either explicitly or constructively, to be bound by the verdict in the earlier action is significant, see Benson & Ford, 833 F.2d at 1176, especially since they actually initiated the later action while the earlier action was still pending. And, finally, far from engaging in tactical maneuvering aimed at gaming unfair advantage, appellants sought to join the Rodriguez action — and were thwarted in the effort because the defendants objected and the district court, siding with the defendants, barred appellants’ path. Of course, given the discretionary character of virtual representation, see 18 Wright & Miller, supra, § 4457 at 502, we would not conclude that a case falls outside the theory’s purview solely because it does not fit snugly into some preconceived niche or mirror some established fact pattern. But, here, the sequence of events itself confirms the inappropriateness of bringing virtual representation to the fore in this case. The district court, after refusing to certify a class, prohibited *763 appellants from joining the original suit, yet thereafter precluded them from prosecuting their own action. 16 This whipsawing placed appellants in an untenable position. Short of a class action, with all the concomitant safeguards that class certification portends, see, e.g., Fed.R.Civ.P. 23, we do not think that the Due Process Clause comfortably can accommodate such a paradigm. In any event, on the facts of this case the prospect of depriving these plaintiffs of their day in court offends our collective sense of justice and fair play. Consequently, we hold that the theory of virtual representation cannot be galvanized to preclude appellants from maintaining their suit. III. CONCLUSION We need go no further. Because the appellants were neither parties to the initial action nor in privity with the plaintiffs therein, the district court erred in dismissing their suit under principles of res judicata. Reversed and remanded for further proceedings. Costs to appellants. Footnotes [1] Most of the financing was undertaken by Ban-co Central y Economías and Banco de Economí-as, the predecessors in interest of defendant-appellee Banco Central Corp. [2] Judge Fuste, who presided over the Rodriguez case, believed the plaintiffs "undoubtedly" had been wronged. Even while upholding many of the defendants’ legad arguments, he lamented the seeming injustice "in allowing the ... sellers of swampland to trusting buyers, to walk from this court without so much as a scratch.” Rodriguez v. Banco Central Corp., 777 F.Supp. 1043, 1065 (D.P.R.1991). [3] While the doctrines of res judicata and collateral estoppel have been said to “share a distinct family resemblance,” Fiumara v. Fireman’s Fund Ins. Cos., 746 F.2d 87, 90 n. 1 (1st Cir.1984), they are nonetheless distinct, see Parklane Hosiery Co. v. Shore, 439 U.S. 322, 326 n. 5, 99 S.Ct. 645, 649 n. 5, 58 L.Ed.2d 552 (1979) (delineating differences). [4] The perils of nonparty preclusion are real. Prominent among them is the prospect that an overly expansive arrangement of the concept, or too free use of it, may endanger constitutional rights. See Meza v. General Battery Corp., 908 F.2d 1262, 1266 (5th Cir.1990) (approving concept but noting the due process concerns implicit in the ideal that, in general, every party is entitled to her own "day in court”). [5] The sobriquet "virtual representation" frequently is used to describe this type of de facto representation. It fits equally well under the label "representation by proxy." [6] We do not think that comment b to section 39, Restatement (Second) of Judgments § 39, comment b, at 383-84 (limiting scope of section to issue preclusion, not claim preclusion), indicates that substantial control can never serve as the basis for a finding of privity when res judicata is in play. Rather, we interpret the comment as suggesting that substantial control has somewhat different dimensions for purposes of issue preclusion than for purposes of claim preclusion — a proposition with which we agree. [7] Some courts and commentators have suggested that, at a minimum, substantial control is the quantum of involvement expected of a co-party. See, e.g., American Postal Workers Union, Etc. v. U.S. Postal Serv., 736 F.2d 317, 319 (6th Cir.1984); IB Moore, supra, ¶0.411[6], at 456. With respect, we do not find this mode of measurement particularly enlightening and, hence, we decline to install it. [8] Admittedly, some plaintiffs in each camp also belonged to an informal litigation group that helped to finance the Rodriguez litigation and disseminated information relevant to members' claims. Yet this link, whether taken by itself or in combination with the circumstances noted by the district court, is far too fragile to support a finding of substantial control. See, e.g., Jenkins v. Hartford Acc. & Indem. Co., 733 F.2d 1090, 1091 (4th Cir.1984) (holding that limited participation will not bind a nonparty); General Foods, 648 F.2d at 788 (noting that merely helping to finance litigation will not bind a nonparty); McKeown v. Wheat, 231 F.2d 540, 543 (5th Cir.1956) (similar). [9] This remains the modem rule despite an occasional dictum that a determined advocate might read to the contrary. See, e.g., In re Medomak Canning Co., 922 F.2d 895, 901 (1st Cir.1990) (suggesting that "privity may be established by identification of interests, even where representation of those interests is not authorized”). [10] Notice is a very important factor. With the possible exception of Aerojet, 511 F.2d 710 (a case that has since been narrowed by the Fifth Circuit), counsel have cited us to no case in which a court has precluded a nonparty, based on a theory of virtual representation, where the nonparty had not received timely notice (actual or constructive) of the initial litigation. [11] To be sure, the Restatement does not require actual notice when nonparty preclusion stems from a preexistent relationship between party and nonparty. See Restatement (Second) of Judgments § 41, at 393. We suggest that the requirement is omitted in such a situation because the formation of the underlying relationship, in and of itself, embodies what amounts to constructive notice of all ensuing litigation. [12] A contrary view would fly in the teeth of the general rule that, in civil litigation, the sins of the lawyer routinely are visited upon the client. See, e.g., Link v. Wabash R.R., 370 U.S. 626, 633-36, 82 S.Ct. 1386, 1390-92, 8 L.Ed.2d 734 (1962); Thibeault v. Square D Co., 960 F.2d 239, 242 (1st Cir.1992). We do not understand why a nonparty who comes within the doctrinal framework for virtual representation- — a framework in which party and nonparty share identical interests, and that provides for notice and a weighing of equitable considerations — should be treated differently from a party in this regard. [13] We are confident that the cases discussing the importance of adequate representation can be reconciled with this analysis. For instance, in McCoy, the prior action was voluntarily dismissed, not determined on the merits as res judi-cata requires. See McCoy Restaurants, 708 F.2d at 587. And in Clark, the court pointed out that the nonparties whose suit defendant sought to preclude did not have fair notice of the prior litigation. See Clark, 794 F.2d at 973-74. [14] The first explicit reference to any of the Gonzalez plaintiffs in the papers of the Rodriguez case occurred on April 10, 1987, when the Rodriguez plaintiffs sought leave to add them as parties. The district court denied this motion on April 27, 1987. See supra p. 754. There is nothing to indicate that, prior thereto, any of the Gonzalez plaintiffs either knew about the pen-dency of the Rodriguez action or had retained the Rodriguez plaintiffs’ lawyers as their counsel. [15] In this connection, it must be emphasized that the district court, in the person of Judge Fuste, refused to certify the Rodriguez case as a class action. [16] Though two different judges made these rulings, that fact is not of legal consequence. We might add parenthetically that it is also cold consolation to appellants.