concurring1 and dissenting.
While a widow plans a funeral, the corporation in whose facility her husband was killed conducts an investigation. While family and friends mourn, the corporation obtains witness statements and prepares reports concerning the circumstances surrounding the death. If this occurrence is ever considered by a judge and jury, they should be able to hear the plain, unvarnished truth — to learn what really happened when memories were fresh and unpolished by counsel.
But now the majority2 puts a stop to all of that; it approves concealment of this *208investigation. As the family buries the victim, the corporation can bury any inconvenient facts it has learned. There is certainly nothing improper about the corporation investigating, but justice may well be defeated if the fruits of that investigation are hidden from the victim as well as other parties who may be forced to defend themselves against charges of wrongdoing. Such unwarranted secrecy defeats the search for truth and violates the previous law of Texas, as the trial judge in Wichita properly recognized. Unfortunately, once again neither an explicit procedural rule nor the prior decisions of this court prevent the continued erection of what is essentially a double standard of justice in Texas. Amply displayed here is the added cost and delay resulting from the majority’s eagerness to intrude rather than willingness to accept our existing law.3
I.
So that the real facts may ultimately be made known, we permit parties discovery before a trial begins. That process is designed to draw no distinction between the weak and the strong. The contrary approach approved today — secrecy—will sometimes benefit one side and sometimes another,4 but it will invariably make truth its first casualty. This court’s commitment to openness has previously been firm.5 Most recently in State v. Lowry, 802 S.W.2d 669 (Tex. 1991, orig. proceeding), an opinion I authored for a unanimous court, discovery worked to the immediate benefit of the most powerful insurance companies in this country. And, under the circumstances of that case, rightly so. There, as in so many of our prior decisions, this court set forth the principle that should govern the present dispute:
Affording parties full discovery promotes the fair resolution of disputes by the judiciary. This court has vigorously sought to ensure that lawsuits are “decided by what the facts reveal, not by what facts are concealed.” Discovery is thus the linchpin of the search for truth, as it makes “a trial less of a game of blind man’s bluff and more a fair contest with the issues and facts disclosed to the fullest practicable extent.” In recent years, we have sought to secure this objective through both revision of the Texas Rules of Civil Procedure and our opinions discouraging gamesmanship and secrecy.
Only in certain narrow circumstances is it appropriate to obstruct the search for truth by denying discovery. Very limited exceptions to the strongly preferred policy of openness are recognized in our state procedural rules and statutes. See Tex.R.Civ.Evid. 501; Tex. R.Civ.P. 166(b)(3).
802 S.W.2d at 671 (case citations omitted). Today, however, this is all just history, as privileges to hide the truth are unreasonably expanded.
Among “the very limited exceptions to [our] strongly preferred policy of openness,” Lowry, 802 S.W.2d at 671, is that for
*209written statements of potential witnesses and parties ... [and] [communications between agents or representatives or the employees of a party to the action or communications between a party and that party’s agents, representatives or employees, when made subsequent to the occurrence or transaction upon which the suit is based and in connection with the prosecution, investigation or defense of the particular suit, or in anticipation of the prosecution or defense of the claims made a part of the pending litigation.
Tex.R.Civ.P. 166b(3)(c) and (d). Given the persistent efforts of skilled attorneys to employ these provisions to suppress information adverse to their clients, Texas courts have frequently written on this subject. In the most recent attempt to prevent these exceptions from swallowing the rule, our court defined an explicit two part test to ascertain whether an investigation has been conducted in anticipation of litigation. Flores v. Fourth Court of Appeals, 111 5.W.2d 38 (Tex.1989, orig. proceeding). There must be first “an objective examination of the facts surrounding the investigation [that] [c]onsider[s] outward manifestations which indicate that litigation is imminent;” and second, a subjective determination that “the party opposing discovery had a good faith belief that litigation would ensue.” Id. at 41. We concluded that “[u]nless there is an abuse of discretion, the trial court’s ruling should not be disturbed.” Id. While admitting no abuse of discretion occurred here, the majority nevertheless “disturbs” this litigation in a highly disturbing manner.
As used in the first element of this test, the term “outward manifestations” means that the only investigations that can legitimately be considered to have been made “in anticipation of litigation” are those conducted after a claimant has given some objective indication of an intent to sue. See 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); Texaco Ref. & Mktg., Inc. v. Sanderson, 739 S.W.2d 493, 495 (Tex.App.—Beaumont 1987, orig. proceeding); Phelps Dodge Ref. Corp. v. Marsh, 733 S.W.2d 359, 361 (Tex.App.—San Antonio 1987, orig. proceeding). Such actions by the plaintiff could include “commencing an investigation of the accident, retaining an attorney or private investigator and, of course, making a claim or demand for damages.” Phelps, 733 S.W.2d at 361. This well-developed rule had already become known as the “outward manifestations” test when we incorporated that term as a part of the standard announced in Flores. See Enterprise Prod., 759 S.W.2d at 179.
Previously, in Stringer v. Eleventh Court of Appeals, 720 S.W.2d 801 (Tex. 1986, orig. proceeding) (per curiam), we had refused to shield from discovery interviews and an investigation notebook compiled for the Santa Fe Railroad shortly following a train collision which killed a brakeman. Although “any fool” might have known that a lawsuit would result from such an incident, Flores, 777 S.W.2d at 43 n. 1 (Gonzalez, J., dissenting), we nonetheless realized that the compelling need for both parties to have equal access to all the facts requires that any exceptions to open discovery be very narrowly drawn. “The mere fact that an accident has occurred is not sufficient to clothe all post-accident investigations, which frequently uncover fresh evidence not obtainable through other sources, with a privilege.” Stringer, 720 S.W.2d at 802; see also Robinson v. Harkins & Co., 711 S.W.2d 619, 621 (Tex.1986, orig. proceeding) (per curiam) (pre-suit investigation made after collision between train and a truck not made in anticipation of litigation).
Abruptly abandoning these decisions in a continued disregard of Texas precedent,6 *210the majority gropes for euphemisms. Flores is not overruled; it is “modified,” 851 S.W.2d at 195, and “[u]pon further consideration,” id. at 203, “alter[ed]” beyond recognition. Id. at 195. The majority merely “disapprove^] Stringer to the extent [that it applies to] sever[e] accident[s].” Id. at 204.7 Yet, had today’s opinion been in place, the opposite result would have been required in both Flores and Stringer. That jurisprudence, as well as the multiple decisions of our courts of appeals are simply replaced with a new rule — corporate clairvoyance. “[A]n investigation [conducted] before the plaintiff manifests an[y] intent to sue” — indeed an investigation conducted before the victim has been buried — can be completely hidden from view. 851 S.W.2d at 204.
Since our prior decisions concerning it, Rule 166b(3) has been made even more precise in its limitations on privileges to discovery. To come within its scope, the documents involved must have been prepared “in connection with the prosecution, investigation, or defense of the particular suit, or in anticipation of the prosecution or defense of the claims made a part of the pending litigation.” Tex.R.Civ.P. 166b(3)(c) and (d) (emphasis added). The term “particular,” which had already been present in the witness statement portion of Rule 166b(3), was added to the party communications provision to “standardize [Rule 166b(3)’s] language.” Tex.R.Civ.P. 166b, Advisory Committee’s Comment to 1990 Change. Significantly, we standardized by adding the words “particular suit” to the party communications subpart, rather than removing it from the witness statement provision. In so doing, as at least one prominent commentator has noted, we codified our recent decisions regarding the scope of these privileges. “[I]t is now clear that a mere ‘generic anticipation’ of litigation is not sufficient to invoke the witness statement or party communication exemptions....” William Dorsaneo, 3A Texas Litigation Guide § 89A.03[l][c] (1992).
II.
Limiting discovery by expanding privileges for party communications and witness statements, we are told, will “promote the truthful resolution of disputes through the adversarial process by encouraging complete and thorough investigation of the facts by both sides.” 851 S.W.2d at 203. This is basically a “finders keepers” theory of information gathering: its objective is not that of revealing the whole truth but only selective bits and pieces in the course of a good fight. It wrongly assumes that the widow, who has no permanent accident investigation team, who has no insurance adjuster to summon, and whose access to the explosion site is barred, has an equal opportunity to conduct a “complete and thorough investigation.” Id.
In its frequent reliance on federal precedent, the majority is oblivious to an original objective of formal discovery rules. The drafter of the discovery components of the 1938 Federal Rules of Civil Procedure concluded that they
mark the highest point so far reached in the English speaking world in the elimination of secrecy in the preparation for trial. Each party may in effect be called upon by his adversary or by the judge to lay all his cards upon the table, the important consideration being who has the stronger hand, not who can play the cleverer game.
Edson Sunderland, Discovery Before Trial Under the New Federal Rules, 15 Tenn. L.Rev. 737, 739 (1939). Similarly, a Special Assistant to the Attorney General in charge of monitoring federal decisions in*211terpreting the new rules, said that they were written
with a view to departing as far as possible from “the sporting theory” of justice and to fulfilling that concept of litigation which conceives a lawsuit as a means for ascertaining the truth, irrespective of who may be temporarily in possession of the pertinent facts.
Alexander Holtzhoff, New Federal Procedure and the Courts 7 (1940). Unfortunately, due to the excessive gamesmanship by advocates entrenched in the very antagonistic roles today celebrated by the majority, discovery rules have not achieved their intended goals. As this court has previously observed:
[The goals] of the discovery process [are] often frustrated by the adversarial approach to discovery. The “rules of the game” encourage parties to hinder opponents by forcing them to utilize repetitive and expensive methods to find out the facts.
Garcia v. Peeples, 734 S.W.2d at 347 (citing Wayne Brazil, The Adversary Character of Civil Discovery: A Critique and Proposals for Change, 31 Yand.L.Rev. 1295, 1303-15 (1978) [hereinafter Adversary Discovery ]). While the adversarial system is “an excellent system” for evaluating information, because it “forces fact-finders to think hard before making up their minds,” it is often “a lousy method of information development.’’ Kathleen Waits, Work Product Protection for Witness Statements: Time for Abolition, 1985 Wis.L.Rev. 305, 338-39 (1985) (emphasis added). Brazil summarizes the reasons:
In short, adversarial investigation ... enables counsel to play the games of deception, concealment, and manipulation that defeat the purposes discovery was intended to serve.
Adversary Discovery, supra, at 1319.
III.
Rather than striving to promote the goals of discovery through efficient distribution of information, the majority actively undercuts them by adopting a rule which will increase litigants’ costs, will give parties additional opportunity to conceal the truth, and will unfairly advantage institutional litigants, which have access to greater information than the individual citizens against whom they are often aligned. Failing to confine properly the party communication and witness statement privileges, generally adds unnecessary costs to the discovery process. Limiting each litigant to preparing its own separate investigation wastes resources by causing duplication of effort, while creating complex rules of privilege generates disputes apart from the merits of the case. See Elizabeth Thorn-burg, Rethinking Work Product, 77 Va. L.Rev. 1515, 1561 (1991); Waits, supra, at 319, 322-24. Since litigation occurs in a taxpayer-funded forum and is invested with a public interest, these costs are imposed on not only litigants, but also on society as a whole. See Thornburg, supra, at 1571.
By requiring the party claiming privilege to show that a potential opponent had taken objectively identifiable steps toward filing suit, Flores gave trial courts a relatively straightforward and easy-to-apply standard by which to determine whether materials were gathered in anticipation of litigation. In contrast, the majority’s amorphous standard will only produce more time-consuming evidentiary hearings and additional disagreement among the courts regarding its scope. More importantly, by significantly reducing the secreting of relevant information, Flores served the underlying purposes of our discovery rules. As we recognized in Stringer, 720 S.W.2d at 802, immediate post-accident investigations can uncover fresh evidence from witnesses and the scene that will often not be available at any other time. Subsequent discovery is usually a most imperfect substitute:
Such interviews and depositions ... often are taken weeks, months, or even years after the incident. The witnesses’ recollections may have weakened over time or may have been enhanced or shaped by intervening conversations with opposing counsel.
Thornburg, supra, at 1556; see also Waits, supra, at 318 (decisions shielding contemporaneous witness statements from discovery “disregard the rapid deterioration of *212human memory and sacrifice highly probative and reliable evidence on the altar of work product”); Edward Cooper, Work Product of the Rulesmakers, 53 Minn. L.Rev. 1269, 1322 (1969) (access to witness statements “may provide the only meaningful opportunity to recall legitimately vanished memory_[and] invaluable in ... paving the way for effective cross-examination”). The majority’s so-called “common sense” definition of “in anticipation of litigation” makes no sense at all in terms of our traditional Texas commitment to open discovery; henceforth every institutional litigant will claim and often obtain privilege for post-accident reports. Neither the thousands of Texans harmed nor the jurors constitutionally assigned to decide their cases will ever know the whole truth about the matters at issue.
Nor is the majority’s approach mitigated by the possibility that a litigant might occasionally surmount the hurdle erected today by showing a “substantial need” for the materials, and an inability to obtain equivalent information except through “undue hardship.” Tex.R.Civ.P. 166b(3)(e). Unfairly reversing the burden of proof by creating a presumption in favor of secrecy, this exception requires parties to articulate a need for materials when there has not even been an opportunity to unearth their existence. “Simply stated, it is difficult to prove that you have substantial need of evidence and cannot obtain it without undue hardship unless you know what the evidence is.” David Keltner, Texas Discovery § 3:200.1 (1992); see also Lowry, 802 S.W.2d at 673 (noting the inherent difficulty in showing a particularized need for documents when their contents are unknown). Although the substantial need and undue hardship exception is entirely undeveloped in Texas, federal courts require the party seeking discovery to show that the privileged materials are “essential to the preparation of one’s case.” See Hickman v. Taylor, 329 U.S. 495, 511, 67 S.Ct. 385, 394, 91 L.Ed. 451, 462 (1947). Such a narrow standard provides no significant relief from today's expansion of privilege.
IV.
Today’s opinion further bars access to admittedly relevant information that even without litigation would have been generated in the ordinary course of business. This twist on the majority’s position provides an additional affront to prior Texas caselaw and our commitment to open discovery. Several courts have noted that documents prepared in the ordinary course of business are not within the scope of the party communication or witness statement privileges. See Wiley v. Williams, 769 S.W.2d 715, 717 (Tex.App.—Austin 1989, orig. proceeding); Evans v. State Farm Mut. Auto Ins. Co., 685 S.W.2d 765, 767 (Tex.App.—Houston [1st Dist.] 1985, writ ref’d n.r.e.); Axelson, Inc. v. McIlhany, 755 S.W.2d 170, 173 (Tex. App.—Amarillo 1988, orig. proceeding), overruled in part on other grounds, 798 S.W.2d 550 (Tex.1990, orig. proceeding).
Among the biggest winners from today’s writing are casualty insurance companies. Since claim investigations could also reasonably be considered as preparations for likely future litigation, they are probably henceforth shielded from discovery.8 One court recently condemned the “minority rule” adopted here9 as “clashpng]” with the objective of “ascertaining] the truth” *213by permitting access to “hardly any document authorized by or for an insurer.” Langdon v. Champion, 752 P.2d 999, 1007 (Alaska 1988).10 But today’s ruling covers far more than insurance companies; indeed, every business record containing facts that could bear on a mishap will be unavailable without a special showing.
As several scholars have pointed out, in practice the increased protection from discovery of post-accident investigations, while superficially neutral, decidedly favors repeat institutional litigants over those who find themselves in court once as a result of being injured; the rule adopted here is clearly designed to favor defendants over plaintiffs. Waits, supra, at 313; Thorn-burg, supra, at 1561. This is because
institutional defendants have a superior ability to structure their dealings so as to create more work product, and they benefit more from delay. In addition, because plaintiffs have the burden of proof, they are more likely to be harmed by a denial of access to information.
Id. at 1562. One empirical study demonstrates the skewed effects of the party communication and witness statement privileges; it found that corporate defense counsel are more likely both to resist discovery and to use it for purposes of delay than attorneys for individuals. See Wayne Brazil, Civil Discovery: Lawyers’ Views of Its Effectiveness, Its Principal Problems and Abuses, 1980 Am.B.Found.Res.J. 787, 853 (1980); Wayne Brazil, Views from the Front Lines: Observations by Chicago Lawyers About the System of Civil Discovery, 1980 Am.B.Found.Res.J. 219, 243 n. 45 (1980). Today the majority provides further incentive for resistance and delay. Creation of such inherently unfair procedures takes an enormous toll on our system of justice:
Academics and even lawyers tend to forget that every opinion is more than words on a page or a citation in a treatise. It represents expense and probably misery for everyone involved, and for society at large.... In the case of work product, the need for the doctrine will have to be awfully strong to warrant all the fuss.
Waits, supra, at 324. Here the “fuss” raised was in no way warranted. The decisions of both the trial court and the court of appeals were entirely proper under the prior law of Texas, as conceded by the majority in. its hesitation now to issue a writ of mandamus.
Much is revealed in the majority’s declaration that National Tank Company has no adequate remedy by appeal because the information in the contested statements “could have a significant impact on the assignment of liability.” 851 S.W.2d at 207. Thus, while expressly acknowledging the potentially critical relevance of the facts in these documents, the majority contends that such relevance provides greater reason to keep them hidden from other litigants and the factfinder. Because this approach only provides more privileges for the privileged, I dissent.11
. While I concur in the judgment that mandamus not issue here, I vigorously dissent from the reasoning of the plurality opinion.
. Although Chief Justice Phillips’ writing represents only a plurality opinion, I refer here to the "majority" because my disagreement also applies to the writings of Justices Gonzalez and Hecht.
. The lethal explosion occurred on August 23, 1990; Judge Brotherton properly applied Texas law to permit discovery by an order of July 25, 1991; the court of appeals promptly and appropriately rejected mandamus on September 27, 1991. After according National Tank emergency relief in November 1991, this court heard oral argument on March 10, 1992 and now obstructs access to information that could “significantly place the blame for the explosion on ... National Tank Company_” 851 S.W.2d at 213.
. Indeed, while the effect of the majority’s effort is to slant the litigation process, individual plaintiffs are hardly the only ones who suffer. Here two small businesses, Bonded Inspections, Inc. and Helm Inspection Services, Inc., oppose mandamus as real parties in interest seeking access to the truth to use in their own defense.
.See Axelson v. McIlhany, 798 S.W.2d 550, 553 (Tex. 1990, orig. proceeding) (abuse of discretion to deny discovery of potentially relevant documents without reviewing them in camera); Garcia v. Peeples, 734 S.W.2d 343, 347-48 (Tex.1987, orig. proceeding) (abuse of discretion to grant a blanket protective order against sharing discovery with other litigants); Peeples v. Hon. Fourth Supreme Judicial Dist., 701 S.W.2d 635, 637 (Tex.1985, orig. proceeding) (burden is on party asserting a privilege from discovery to produce evidence concerning its applicability); Jampole v. Touchy, 673 S.W.2d 569, 573 (Tex.1984, orig. proceeding) (abuse of discretion to deny discovery of alternative designs of product).
. See, e.g., Texas Ass’n of Bus. v. Texas Air Control Bd., 852 S.W.2d 440, 446 n. 10 (Tex.1993) (attempting to explain that the opinion only overrules one prior decision of this court, rather than four); Boyles v. Kerr, 1992 WL 353277 (Tex.1992) (overruling 1987 decision creating cause of action for negligent infliction of emotional distress), Walker v. Packer, 827 S.W.2d 833, 841-42 (Tex.1992, orig. proceeding) (expressly "disapproving" a large body of Texas caselaw); Carrollton-Farmers Branch Indep. Sch. Dist. v. Edgewood Indep. Sch. Dist., 826 *210S.W.2d 489, 520 n. 37 (Tex.1992) (disavowing a prior opinion signaling the constitutionality of consolidating school district tax bases); Stewart Title Guar. Co. v. Sterling, 822 S.W.2d 1, 5-6 (Tex.1991) (ignoring recent precedent, looking instead to overruled case).
. This is becoming a familiar pattern. See Texas Ass'n of Bus. v. Texas Air Control Bd., 852 S.W.2d at 446 n. 10 (attempting to show that only one, rather than four, prior decisions of this court overruled in deciding an issue not even raised by the parties); Walker v. Packer, 827 S.W.2d at 842 (noting the “disapproval” of three decisions of this court).
. However, today’s opinion should not be construed to shield post-accident reports from discovery in litigation involving the insurer’s breach of its duty of good faith and fair dealing. Because such a suit is based on the denial of a claim, investigations conducted between the accident and the time of denial are not made "subsequent to the occurrence or transaction upon which the suit is based.” Tex.R.Civ.P. 166b(3)(c) and (d); Jackson v. Downey, 817 S.W.2d 858 (Tex.App.—Houston [1st Dist.] 1991, orig. proceeding); Gilbert v. Black, 722 S.W.2d 548 (Tex.App.—Austin 1987, orig. proceeding).
. Today’s writing concerning ordinary course of business is practically indistinguishable from the severely criticized minority rule. Despite reliance on the tenuous requirement that investigations be made “primarily" for litigation purposes in order to come within the privilege, see 851 S.W.2d at 206 n. 13, the majority fails to offer any example in which materials gathered both in the ordinary course of business and "in anticipation of litigation” under its new standard will not be privileged.
. Such reports are generally held discoverable as having been prepared in the ordinary course of business. See National Farmers Union Prop. & Cas. Co. v. District Court for the City and County of Denver, 718 P.2d 1044, 1047-48 (Colo. 1986) (en banc) (allowing discovery of investigations by attorney for insurer, as made in the ordinary course of business); Henry Enter., Inc. v. Smith, 225 Kan. 615, 592 P.2d 915, 921 (1979) (insurer’s initial investigations of claim discoverable); Soeder v. General Dynamics Corp., 90 F.R.D. 253, 255 (D.Nev.1980) (aircraft manufacturer’s investigations of crash discoverable); Atlanta Coca-Cola Bottling Co. v. Transamerica Ins. Co., 61 F.R.D. 115, 118 (N.D.Ga.1972) ("the obviously incongruous result of [this rule] would be that the major part of the files of an insurance company would be insulated from discovery"); Thomas Organ Co. v. Jadranska Slobodna Plovidba, 54 F.R.D. 367, 373 (N.D.Ill. 1972) ("under [this] theory, the amendments to the discovery rules which were believed to be a liberalization of the scope of discovery would be a foreclosure of almost all internal documents of insurance companies relating to the claims of insureds”).
. I do agree with the result reached in today’s opinion regarding the clear limitation placed on the scope of the attorney client privilege by Texas Rule of Civil Evidence 503.