Scott v. Twelfth Court of Appeals

843 S.W.2d 439 (1992) Earnest SCOTT v. TWELFTH COURT OF APPEALS. No. D-1418. Supreme Court of Texas. March 13, 1992. Rehearing Overruled April 15, 1992. Dissenting Opinions Filed December 16, 1992. Paul M. Boyd, James T. Foley, Foley & Boyd, P.C., Tyler, for relator. David W. Prasifka, Lorance & Thompson, Houston, Tom Henson, Michael A. Hatchell, Gregory D. Smith, Ramey, Flock, Jeffus, Crawford, Harper & Collins, Tyler, for respondent. OPINION PHILLIPS, Chief Justice. This is an original mandamus proceeding. The parties in the underlying action dispute whether defendant's investigations of the accident are privileged from discovery because made in anticipation of litigation. The trial court ordered discovery of the investigations, but the court of appeals granted mandamus relief reversing that order. Earnest Scott, the party seeking discovery, asks us to reinstate the trial court's ruling. We conditionally grant the writ of mandamus. Scott was injured at work on April 3, 1989, when a load of wood fell on the forklift he was operating. The overhead protective cage on the forklift gave way when struck by the wood. Scott sued Hyster Company, the manufacturer of the forklift, and Stewart & Stevenson, Inc., the seller of the forklift. American International Recovery, Inc., the workers' compensation carrier for Scott's employer, intervened to assert a subrogation claim. The disputed discovery issue is whether Hyster's investigations of the accident conducted *440 between April 12, 1989, the date Hyster learned of the accident, and May 19, 1989, the date the compensation carrier demanded reimbursement from Hyster, are privileged because made in anticipation of litigation. See Tex.R.Civ.P. 166b(3)(d). At the trial court's hearing on this issue, Hyster relied on the affidavit of its risk manager, Marvin Welch, which stated that he was informed of the accident by a written report from Stewart & Stevenson on April 12. The report 1) described the accident, 2) disclosed that after the accident occurred a representative of Scott's employer opined that the guard on the forklift was defective, and 3) advised that the compensation carrier had requested that the forklift parts be impounded. Welch stated that based on this information he was "convinced... that a claim and lawsuit would eventually be pursued by the workers' compensation carrier and by Mr. Scott against Hyster Company." After receiving the report Welch immediately notified Hyster's products-liability attorneys and initiated an investigation. Scott did not refute the information in Welch's affidavit. Instead, Scott pointed out that no one demanded compensation from Hyster until May 19, when the compensation carrier demanded reimbursement for the benefits paid Scott. Scott did not make demand on Hyster until September 12, 1989, and suit was not filed until October 1989. After considering the affidavit and arguments of counsel, the trial court concluded that Hyster did not have good cause to believe a lawsuit would be filed until May 19, when it received the demand letter from the compensation carrier. The trial court thus held that investigations conducted between April 12 and May 19 were discoverable. Upon Hyster's petition, the court of appeals issued a writ of mandamus vacating the trial court's order and rendering Hyster's investigations conducted after April 12 non-discoverable. Scott now asks us to issue a writ of mandamus vacating the court of appeals' action and reinstating the trial court's order. Mandamus should issue only to correct a clear abuse of discretion where there is no adequate remedy by appeal. Johnson v. Fourth Court of Appeals, 700 S.W.2d 916, 917 (Tex.1985). When this Court is reviewing the grant of a writ of mandamus by the court of appeals, the focus of our inquiry should be whether the trial court abused its discretion. If the trial court did not abuse its discretion, then the issuance of mandamus by the court of appeals is improper and constitutes an abuse of discretion. See Flores v. Fourth Court of Appeals, 777 S.W.2d 38 (Tex. 1989); Loftin v. Martin, 776 S.W.2d 145 (Tex.1989). A party contending that an investigation is privileged because made in anticipation of litigation must satisfy a twoprong test: 1) the facts must objectively indicate that litigation is imminent; and 2) the party must subjectively believe that litigation is imminent. Flores v. Fourth Court of Appeals, 777 S.W.2d 38 at 40-41. Determining whether a party has met this test is committed to the discretion of the trial court. Id. The focus in this case is on the first prong. Welch's affidavit stated that he had a good-faith belief that litigation would ensue, and Scott does not challenge this assertion. Instead, Scott argues that the facts did not objectively support Welch's conclusion. The issue is not whether the court of appeals could have justifiably found that Hyster satisfied the first prong of the Flores test based on an independent analysis of the evidence, but whether the trial court's decision exceeded its discretion. The evidence before the court did not compel a ruling that the test was met on any particular date prior to May 19. Welch knew that a representative of Scott's employer had stated that the forklift guard was defective, but he did not know whether the potential plaintiffs, Scott and the compensation carrier, shared this view. The compensation carrier had requested that the forklift parts be impounded, but this indicated only that it planned to investigate *441 the accident, not that it necessarily would sue Hyster. Neither Scott nor the compensation carrier made demand on Hyster before May 19, and Hyster had no indication that Scott had consulted an attorney before then. Based on the record, the trial court's decision did not constitute an abuse of discretion. Pursuant to Tex.R.App.P. 170, the Court determines that this cause should be submitted without oral argument. The writ of mandamus is conditionally granted. The writ will issue only if the court of appeals refuses to act in accordance with this opinion. GONZALEZ, HECHT and CORNYN, JJ., note their dissent. (Dissenting opinion to follow.) HECHT, Justice, dissenting. Mandamus relief is so frequently sought and granted that to be accurate, we should consider labeling the writ occasional rather than extraordinary. This is a relatively recent development, as illustrated by the fact that the word "mandamus" appears in more reported decisions of Texas courts in the 15 years since 1978 than in well over a century before. Until the last decade or so, mandamus relief was rarely granted, partly because of its disruptive effect on proceedings in the trial court, and partly in recognition of the sheer impossibility of regular appellate review of interlocutory trial court rulings. These considerations are no less compelling today, as we acknowledged earlier this year in Walker v. Packer, 827 S.W.2d 833, 842 (Tex.1992): Mandamus disrupts the trial proceedings, forcing the parties to address in an appellate court issues that otherwise might have been resolved as discovery progressed and the evidence was developed at trial. Moreover, the delays and expense of mandamus proceedings may be substantial.... The impact on the appellate courts must also be considered. We stated in [Braden v. Downey, 811 S.W.2d 922, 928 (Tex.1991)] that "[t]he judicial system cannot afford immediate review of every discovery sanction." It follows that the system cannot afford immediate review of every discovery order in general. Id. (citation omitted). For these reasons, the strict legal standards for granting mandamus have not been relaxed. To the contrary, we reaffirmed in Walker: "`Mandamus issues only to correct a clear abuse of discretion or the violation of a duty imposed by law when there is no other adequate remedy by law.'" Id. at 839 (quoting Johnson v. Fourth Court of Appeals, 700 S.W.2d 916, 917 (Tex.1985)). The recent increase in granting mandamus relief is not due to any change in the reasons for restricting it or the requirements for obtaining it, but rather to a laxity in judicial enforcement of these requirements. As we noted in Walker, while no case ever rejected the prerequisite of an inadequate remedy by appeal, several decisions granting relief failed to explain how, or even whether, that requirement was met. 827 S.W.2d at 840-842. Unless the standards for granting mandamus relief are strictly adhered to, relief is granted more frequently, and thus perceived to be more available, and thus more often requested, and thus more often granted, and so on. Though the standards remain, their inconsistent application leads to an acceleration in granting relief. The merry-go-round has brakes; they just are not always applied. This is a case in point. In the litigation out of which this original mandamus proceeding arises, the trial court held, based upon evidence and argument, that an accident investigation was not made in anticipation of litigation and therefore was not privileged from discovery. See Tex. R.Civ.P. 166b(3)(d). On application for mandamus by the party resisting discovery, a divided court of appeals, after hearing oral argument, came to the opposite conclusion. Maj. op. at 440. Now, on application for mandamus by the party requesting discovery, this Court directs the appeals court to withdraw its ruling. Unlike both lower courts, this Court does not consider it necessary to hear oral argument. The Court's opinion does not mention its decision in Walker, and it grants relief without explaining how the requirements *442 for issuance of mandamus have been satisfied, the very practice criticized in Walker. As I read the Court's opinion, it rests solely on the conclusion that the trial court reached the right result and the court of appeals did not. Although the Court correctly recites that "[m]andamus should issue only to correct a clear abuse of discretion where there is no adequate remedy by appeal", ante at 440, it never considers whether it has complied with its own rule. Acknowledgment of the two prerequisites for mandamus is but a meaningless onesentence liturgy before the benediction. I have five points to make. First, relators have wholly failed to demonstrate that their appellate remedy is inadequate. The Court does not discuss this requirement at all. Second, the court of appeals has some discretion in determining whether to grant mandamus relief, and it has not clearly abused that discretion in this case. Third, the trial court was incorrect in concluding that the investigation materials in dispute were not prepared in anticipation of litigation. Fourth, even if the appeals court should not have granted mandamus, this is not an extraordinary case warranting further intervention by this Court. And fifth, despite the Court's failure to cite or follow Walker, the Court's decision does not indicate a lack of commitment to that decision, but rather, a flaw in our procedures for issuing opinions without argument. On the first point, relators have failed to show an inadequate appellate remedy. There is, of course, no appeal to this Court from the judgment of an appellate court granting mandamus relief in an original proceeding; such judgments can be reviewed by this Court only on application for mandamus against the court of appeals, as relators have filed in this case. See Goffney v. Lowry, 554 S.W.2d 157 (Tex.1977); Alexander v. Meredith, 152 S.W.2d 732 (Tex. 1941). Nevertheless, when the court of appeals grants mandamus relief against a trial court, it directs that court to take action which may be reviewable on appeal from the final judgment rendered in the case. That is the situation here, where the court of appeals has directed the trial court to rescind its order allowing discovery. When the trial court complies, its ruling may be appealed after final judgment, should relators choose to do so. See State of Texas v. $435,000, 842 S.W.2d 642 (Tex. 1992). Hence, relators unquestionably have a remedy by appeal in these circumstances. The question is whether relators have demonstrated that their appellate remedy is inadequate. If the trial court, like the court of appeals, had ruled contrary to relators that is, if it had denied discovery instead of allowing it, relators could not obtain mandamus relief without demonstrating either that they fall within one of the situations described in Walker when appeal may be inadequate, or that appeal is inadequate for some other reason. Relators have done neither. Without this showing, mandamus should issue only if the inadequate remedy by appeal requirement does not apply when mandamus is sought against a court of appeals, or if the right to appeal is always inadequate in these circumstances. Neither Walker nor any other decision by this Court limits application of this requirement to decisions of trial courts, and there is no rational basis for doing so. There is no reason to think that appeal is somehow less adequate because the last ruling was by the court of appeals rather than the trial court. The issue, adequate redress, is the same in either case. Thus, relators must meet the second requirement for mandamus directed to the court of appeals in the same way as for mandamus directed to the trial court. Here, they have not shown that they can. My second point—that the court of appeals has not clearly abused its discretion in granting mandamus—is more difficult. As a general matter, it is well established that the decision whether to issue mandamus is discretionary. As this Court wrote in Westerman v. Mims, 111 Tex. 29, 227 S.W. 178, 181 (Tex.1921): It is generally held by the courts of last resort in America that the writ of mandamus is a discretionary writ, and it is frequently stated that it is not a writ of right.... *443 However, the court's discretion lies in deciding whether the prerequisites for mandamus relief have been met, not in deciding whether to grant relief irrespective of such prerequisites. Westerman continues: [T]he discretion referred to is a judicial discretion to be exercised in the application of fixed principles. It is because these principles require the determination of the existence of many incidents, some of which are of an unusual nature, as well as the nonexistence of others, as conditions precedent to the writ's issuance, and because such determination is impossible without the exercise of judicial discretion, in the light of the peculiar facts of each proceeding, that the writ is correctly classified as discretionary. Id. Thus, mandamus cannot issue unless an applicant demonstrates a clear entitlement to the relief, id., but it cannot be denied if he makes such a showing, Navarro County v. Tullos, 237 S.W. 982, 986 (Tex.Civ.App.—Dallas 1922, writ ref'd). The court's discretion lies in determining whether the requirements have or have not been met. This determination often involves weighing various considerations, including the certainty of the law as it applies to the issues presented, Walker, 827 S.W.2d at 840, the importance and complexity of the factual context, id. at 839-840, and the conduct of the parties, Westerman, 227 S.W. at 181. Whether a trial court has clearly abused its discretion is often not a fact so empirical as to preclude honest disagreements over its existence. There will be times when one court will conclude that the requirements for mandamus have been met while another court decides, on the same facts, that they have not. The extent of an appellate court's discretion in deciding whether to grant or deny mandamus relief has become both more important and more difficult to define since the Legislature expanded the courts of appeals' jurisdiction concurrent with this Court's to issue writs of mandamus to the district courts. Tex.Gov't Code §§ 22.-002(a), 22.221(b). Ordinarily, a party seeking mandamus relief must first apply to the court of appeals before applying to this Court. Tex.R.App.P. 121(a)(1). Now, in most instances when we are asked to decide whether to grant relief, the court of appeals has already made the same decision. In exercising our concurrent jurisdiction, we must determine what deference to give the court of appeals' decision. We first approached this issue in Johnson v. Fourth Court of Appeals, 700 S.W.2d 916, 917 (Tex.1985), where we explained: Since June 19, 1983, the courts of appeals have exercised concurrent mandamus jurisdiction with the Supreme Court over district judges of this state. As a result of this expansion of jurisdiction, we have increasingly been asked to decide whether a court of appeals has abused its discretion in granting mandamus relief. This issue typically arises when the court of appeals grants mandamus relief to a party based upon an abuse of discretion in the trial court. The party adversely affected by the mandamus judgment seeks review by mandamus in this Court, arguing that the court of appeals abused its discretion in holding that the trial court abused its discretion. The use of the phrase, "abuse of discretion," to describe the alleged misfeasance of both the trial court and the court of appeals is unfortunate because its meaning in each context is not the same. The discretion exercised by a trial court when ruling on an interlocutory matter is ordinarily quite broad, whereas the discretion exercised by an appellate court possessing mandamus power is more confined. Although the writ of mandamus is a discretionary remedy, its use is subject to certain conditions. Namely, the court of appeals may issue writs of mandamus "agreeable to the principles of law governing those writs." Tex.Gov't Code § 22.221(b). Mandamus issues only to correct a clear abuse of discretion or the violation of a duty imposed by law when there is no other adequate remedy by law. State v. Walker, 679 S.W.2d 484, 485 (Tex.1984). The court of appeals, therefore, acts in excess of its writ power (abuses its discretion) when it grants *444 mandamus relief absent these circumstances. Id. (citations omitted). While we noted that an appellate court's discretion in deciding whether to issue mandamus is "more confined", we did not conclude that it is nonexistent. We stated only that a court of appeals cannot issue mandamus except "to correct a clear abuse of discretion or the violation of a duty imposed by law when there is no other adequate remedy by law." Id. In other words, Johnson holds that an appeals court has no discretion to issue mandamus without a clear showing of the requirements for such relief. This is essentially the same rule established in Westerman. Johnson does not address how much discretion an appeals court has to determine whether a clear showing has been made. Since Johnson, we have stated that this Court should make its own determination of whether mandamus is appropriate, irrespective of the court of appeals' decision. In Strake v. First Court of Appeals, 704 S.W.2d 746, 747 (Tex.1986), we said: "In reviewing by writ of mandamus the action of a court of appeals our focus remains on the point of origin" of the case. In Hooks v. Fourth Court of Appeals, 808 S.W.2d 56, 59 (Tex.1991), we wrote: In an original mandamus proceeding, a court of appeals does not possess independent discretion to grant or deny mandamus against a trial court. Any discretion is vested in the trial court, and the question before this court is whether the court of appeals erred in granting the writ of mandamus. Johnson v. Fourth Court of Appeals, 700 S.W.2d 916, 917-18 (Tex.1985). Although our reference to "independent discretion" is unfortunately confusing, our reliance on Johnson as our sole authority indicates that we intended not to deny the court of appeals all discretion in issuing mandamus but only to confine them to the prerequisites for the writ. We repeated the phrase in City of San Antonio v. Fourth Court of Appeals, 820 S.W.2d 762, 764 (Tex.1991), where we said: The appellate court does not exercise any independent discretion in granting or denying a mandamus request against a trial court. In reviewing the court of appeals' decision to grant mandamus, therefore, this court must determine for itself whether the trial court abused its discretion. Johnson v. Fourth Court of Appeals, 700 S.W.2d 916, 917 (Tex.1985). Thus, we have clearly stated that we will determine for ourselves whether the standards for issuance of mandamus have been met. We have not, however, explained how this determination will be made or what deference should be accorded the appeals court's decision on this issue. In San Antonio as in Hooks, our reliance upon Johnson as our sole authority shows that by denying the court of appeals "independent discretion" we meant nothing more than that it must apply the same legal standards we apply. We have not denied the courts of appeals the discretion described in Westerman. We attempted to clarify this in Dancy v. Daggett, 815 S.W.2d 548 (Tex.1991). In an original mandamus proceeding, the court of appeals disapproved the action of the trial court complained of but denied relief on the grounds that Hooks deprived appellate courts of discretion to determine that the trial court abused its discretion. Dancy v. Daggett, 809 S.W.2d 629, 630 (Tex.App — Houston [14th Dist] 1991, orig. proceeding). We granted mandamus relief, explaining: In Hooks, we held that "a court of appeals does not possess independent discretion to grant or deny mandamus against a trial court." Rather, "[a]ny discretion is vested in the trial court, and the question before this court is whether the court of appeals erred in granting the writ of mandamus." Id. (citing Johnson v. Fourth Court of Appeals, 700 S.W.2d 916, 917-18 (Tex.1985)). Our holding in Hooks did not alter the long-standing rule that mandamus may issue to correct a clear abuse of discretion by a trial court. 815 S.W.2d at 549 (citations omitted). Still, Dancy did not define the extent of a court *445 of appeals' discretion in deciding whether to issue mandamus. We have frequently held that a court of appeals abused its discretion in issuing mandamus. Ginsberg v. Fifth Court of Appeals, 686 S.W.2d 105 (Tex.1985); Hill v. Fourteenth Court of Appeals, 695 S.W.2d 554 (Tex.1985) (per curiam); Peeples v. Fourth Court of Appeals, 701 S.W.2d 635 (Tex.1985); Jordan v. Fourth Court of Appeals, 701 S.W.2d 644 (Tex. 1985); Johnson v. Fourth Court of Appeals, 700 S.W.2d 916 (Tex.1985); Strake v. First Court of Appeals, 704 S.W.2d 746 (Tex.1986); Street v. Second Court of Appeals, 715 S.W.2d 638 (Tex.1986) (per curiam); Stringer v. Eleventh Court of Appeals, 720 S.W.2d 801 (Tex.1986) (per curiam); Klein Indep. Sch. Dist. v. Fourteenth Court of Appeals, 720 S.W.2d 87 (Tex.1986); Houston Health Clubs, Inc. v. First Court of Appeals, 722 S.W.2d 692 (Tex.1986) (per curiam); Graham Nat'l Bank v. Fifth Court of Appeals, 747 S.W.2d 370 (Tex.1987) (order); Street v. Second Court of Appeals, 756 S.W.2d 299 (Tex.1988); Hoffman v. Fifth Court of Appeals, 756 S.W.2d 723 (Tex.1988) (per curiam); Champion Int'l Corp. v. Twelfth Court of Appeals, 762 S.W.2d 898 (Tex. 1988) (per curiam); Packer v. Fifth Court of Appeals, 764 S.W.2d 775 (Tex.1989) (per curiam); Holloway v. Fifth Court of Appeals, 767 S.W.2d 680 (Tex.1989); Flores v. Fourth Court of Appeals, 111 S.W.2d 38 (Tex.1989); Spears v. Fourth Court of Appeals, 797 S.W.2d 654 (Tex. 1990). Far more frequently, however, we have granted mandamus after relief was denied by the court of appeals without characterizing the denial as an abuse of discretion. E.g. Davenport v. Garcia, 834 S.W.2d 4 (Tex. 1992); Eli Lilly & Co. v. Marshall, 829 S.W.2d 157 (Tex.1992) (per curiam); Walker v. Packer, 827 S.W.2d 833 (Tex.1992); Sears, Roebuck & Co. v. Ramirez, 824 S.W.2d 558 (Tex.1992) (per curiam); Owens-Corning Fiberglas Corp. v. Caldwell, 818 S.W.2d 749 (Tex.1991); Dancy v. Daggett, 815 S.W.2d 548 (Tex.1991). Indeed, I can find only one occasion when have we held that a court of appeals abused its discretion in denying mandamus. City of Dallas v. Dixon, 365 S.W.2d 919 (Tex. 1963). From this pattern of our cases one might conclude that granting mandamus relief improperly is always an abuse of discretion, but denying it improperly almost never is; or that the courts of appeals have more discretion to deny mandamus relief than to grant it. Yet that conclusion cannot be supported by principles governing mandamus. Although most applications for mandamus should be denied, given that it is a remedy reserved for extraordinary cases, there is no basis for allowing more discretion to deny relief than to grant it. Denying relief when it should be granted is as much an abuse of discretion as granting it when it should be denied. Our cases do not reflect this, not because the appeals court's discretion is asymmetric, but because we tend to exercise our concurrent jurisdiction over mandamus applications without sufficient regard for the appeals court's decision. If we attempted to explain why the court of appeals clearly abused its discretion in denying mandamus each time we were inclined to disagree with that result, I suspect we would grant relief in fewer cases ourselves. While I recognize that an application for mandamus is an original proceeding and that this Court must make its own determination whether to grant relief, I believe that in doing so we should give greater deference to the court of appeals' decision. If we do not, we should overrule Westerman and Navarro County. There is little to be gained by requiring parties to apply to the court of appeals first for relief unless we are to give respect to that court's decision, and there is little to encourage the court of appeals to give serious consideration to applications filed there if its decision will be of no consequence when the same application is filed in this Court. Thus, I would not disturb the appeals court's decision on application for mandamus unless that court clearly erred in deciding whether the requirements for mandamus had been met, given the considerations which factor into such a decision as recognized in Westerman. *446 The court of appeals committed no clear error in granting mandamus relief in this case because—and this is my third point— the trial court incorrectly held that relator's investigation of an accident involving a forklift it had manufactured was privileged because the investigation was not conducted in anticipation of litigation. See Tex.R.Civ.P. 166b(3)(d). Relator's risk manager, Welch, testified by affidavit that he was convinced when he began the investigation that litigation would ensue. He based his conclusion on his six years' experience as a risk manager, the nature of the accident, and the fact that the worker's compensation carrier covering the injured person had already begun its own investigation of the accident. Since an investigation by the compensation carrier was irrelevant to the injured employee's compensation claim, in which the only issue was the extent of the employee's injury, Welch concluded that the carrier was preparing to assert a subrogation claim against relator. Welch also stated when he communicated his view to relator's attorney, the attorney agreed. Whether an investigation is conducted in anticipation of litigation depends upon "an objective examination of ths facts surrounding the investigation", giving consideration "to outward manifestations which indicate litigation is imminent", and a subjective inquiry into whether the party conducting the investigation had "a good faith belief that litigation would ensue." Flores v. Fourth Court of Appeals, 777 S.W.2d 38, 40-41 (Tex.1989). The italicized words, "giving consideration", have been ignored in the application of the Flores test. Outward manifestations indicative of imminent litigation are a factor to be given consideration, but they are not the only factor in making an objective examination of the relevant facts. Flores does not require that litigation be imminent, only that consideration be given to whether it was. Flores requires "an objective examination of the facts". Those facts must indicate that an ordinary person would reasonably have anticipated that litigation would ensue. Additionally, the subjective element requires that the person conducting the investigation do so with possible litigation in mind. In this case, the facts in Welch's uncontroverted affidavit meet both parts of Flores. The compensation carrier's investigation was clearly in contemplation of litigation against relator, and Welch initiated his own investigation in anticipation of that litigation. The Court holds that relator failed to show objective evidence that litigation was imminent when Welch began his investigation. Not only is this the wrong test, as shown above, the Court misapplies it. The Court bases its holding on its conclusion that Welch did not know whether the potential plaintiffs suspected that the weld on the forklift guard was defective. Ante, at 440. It is not clear how the Court knows what Welch knew, or how it expects Welch to have known what the potential plaintiffs knew. If the objective prong of Flores cannot be satisfied unless the person claiming privilege can show that he knew that a potential plaintiff knew of a basis for litigation, it will be virtually impossible to meet this requirement without direct evidence, such as a statement by the plaintiff to the desired effect. Moreover, the Court's reasoning begins to blur the distinction between an objective and a subjective test. Under an objective test, the issue is not only what Welch or the potential plaintiffs knew, but what they should have known, given the circumstances. Even if it were assumed, however, that the trial court did not abuse its discretion, it does not follow that the court of appeals granted mandamus improperly. Whether the trial court clearly abused its discretion in this case does not admit of a single or certain determination. Reasonable minds can differ, and have differed. The court of appeals considered this case to be a close one, and divided on the result. At this point, five appellate judges—two on the appeals court and three on this Court— have concluded that the trial court clearly abused its discretion, and seven—one on the appeals court and six on this Court— have concluded that it did not. If the court of appeals has any discretion at all to determine whether the requirements for mandamus *447 have been met, I fail to see how it could have abused it in this case. My fourth point is that the Court's decision in this case is yet another example of our disinclination to apply mandamus standards carefully. In Walker, we stated: "Mandamus is intended to be an extraordinary remedy, available only in limited circumstances. The writ will issue `only in situations involving manifest and urgent necessity and not for grievances that may be addressed by other remedies.'" 827 S.W.2d at 840 (citation omitted). By no stretch of language or imagination can the case before us conform to this requirement. This is a routine discovery dispute which can be addressed on appeal. There is no conceivable "manifest and urgent necessity" for us to intervene in the proceedings before this trial court. We also stated in Walker: "Additionally, this Court will not grant mandamus relief unless we determine that the error is of such importance to the jurisprudence of the state as to require correction." Id. at 839 n. 7. I cannot fathom how this case can meet this standard. Even if the court of appeals should not have granted mandamus, I see no justification whatever for our doing so. Finally, my fifth point is that although the Court does not follow Walker, I do not regard this case as a significant departure from Walker. The Court's decision is attributable, I think, to the summary attention the case received. This case was decided on the briefs alone. I doubt the Court would have reached the same conclusion had this case been argued. Rule 170 of the Texas Rules of Appellate Procedure permits the Court to decide cases without benefit of oral argument when six members of the Court agree. The procedure is useful when the determinative issue in a case is so clear that argument beyond the briefs is not likely to affect the decision. This is not such a case. The correctness of the trial court's ruling is a close issue, as the dissenting justice in the court of appeals stated, and the substantive issue— whether an investigation was made in anticipation of litigation—involves considerable controversy. Indeed, our jurisprudence concerning what constitutes anticipation of litigation is becoming almost surreal. Our rules and their underlying policies are the subject of substantial criticism. See, e.g., Alex W. Albright, The Texas Discovery Privileges: A Fool's Game?, 70 Tex.L.Rev. 781 (1992). They are at issue in at least one pending case, National Tank Co. v. Brotherton (writ granted). We have appointed a special Task Force to study discovery issues, including application of the anticipation of litigation standard. The short shrift the Court gives the arguments in this case is in contrast to the extraordinary attention we have directed to the same issues in other contexts. This case is, I believe, simply an aberration. It neither undermines Walker nor further misconstrues Flores; it speaks only to the manner in which this Court decides some of its cases, and the difficulty we continue to have in limiting mandamus to truly extraordinary situations. Accordingly, I dissent. GONZALEZ and CORNYN, JJ., join in this Dissenting Opinion. GONZALEZ, Justice, dissenting. I join Justice Hecht's dissenting opinion except his discussion of Flores v. Fourth Court of Appeals, 111 S.W.2d 38 (Tex. 1989). The two-prong test of Flores is not used in any other jurisdiction that has interpreted the phrase "anticipation of litigation" and the test has proved unworkable.[1]*448 The test adopted by virtually every other court considering this question inquires only whether the investigation was in fact done for the purpose of preparing for potential litigation.[2] The investigatory privileges are intended to promote the adversarial process by encouraging full investigation of the underlying facts. The privileges should therefore apply whenever an investigation is in fact done to prepare for potential litigation. Flores has been criticized because it "ignores the realities of preparation for the contingency of litigation" by denying protection to early and routine preparation. Alex A. Albright, The Texas Discovery Privileges: A Fool's Game?, 70 Texas L.Rev. 781, 813 (1992). Rather than attempting to salvage Flores' two-prong test through continued modification and explanation, I would overrule it and bring Texas in line with the overwhelming majority of other jurisdictions that have addressed this issue. NOTES [1] For example, some courts have held that retention of an attorney or private investigator is an outward manifestation which indicates that litigation is imminent. Boring & Tunneling Co. of America, Inc. v. Salazar, 782 S.W.2d 284, 287 (Tex.App.—Houston [1st Dist.] 1989, orig. proceeding); Mole v. Millard, 762 S.W.2d 251, 255 (Tex.App.—Houston [1st Dist.] 1988, orig. proceeding); Enterprise Prod. Co. v. Sanderson, 759 S.W.2d 174, 179 (Tex.App.—Beaumont 1988, orig. proceeding); Foster v. Heard, 757 S.W.2d 464, 465 (Tex.App.—Houston [1st Dist.] 1988, orig. proceeding); H.E. Butt Grocery Co. v. Williams, 751 S.W.2d 554, 556 (Tex.App.—San Antonio 1988, orig. proceeding); Phelps Dodge Ref. Corp. v. Marsh, 733 S.W.2d 359, 361 (Tex. App.—El Paso, 1987) (orig. proceeding)). Others have not. Toyota Motor Sales, U.S.A., Inc. v. Heard, 774 S.W.2d 316, 318 (Tex.App.—Houston [14th Dist.] 1989, orig. proceeding); Morris v. TEIA, 759 S.W.2d 14, 15 (Tex.App.—Corpus Christi 1988, writ denied). See also Flores, 777 S.W.2d 38, 44 (Gonzalez J., dissenting). [2] See, e.g., Binks Mfg. v. National Presto Indust., Inc., 709 F.2d 1109, 1120 (7th Cir.1983); United States v. El Paso Co., 682 F.2d 530, 542 (5th Cir.1982), cert. denied, 466 U.S. 944, 104 S.Ct. 1927, 80 L.Ed.2d 473 (1984); In re Grand Jury Proceedings, 604 F.2d 798, 803 (3rd Cir.1979); Diversified Indust, Inc. v. Meredith, 572 F.2d 596, 604 (8th Cir.1977); State ex rel. Corbin v. Weaver, 140 Ariz. 123, 680 P.2d 833, 839 (App. 1984); Hawkins v. District Court, 638 P.2d 1372, 1379 (Colo. 1982) (en banc); Mullins v. Vakili, 506 A.2d 192, 197-198 (Del.Super.1986); American Bldgs. v. Kokomo Grain Co., 506 N.E.2d 56, 62-63 (Ind.App.1987); Ashmead v. Harris, 336 N.W.2d 197, 200 (Iowa 1983); Kaarup v. St. Paul Fire & Marine Ins. Co., 436 N.W.2d 17, 21 (S.D.1989). See also 8 Wright and Miller, Federal Practice and Procedure, § 2024 at 198 (1970) ("[T]he test should be whether, in light of the nature of the document and the factual situation in the particular case, the document can fairly be said to have been prepared or obtained because of the prospect of litigation.").