George Fleming and Fleming & Associates, L.L.P. v. Rebecca Wilson

Supreme Court of Texas
══════════
No. 22-0166
══════════

George Fleming and Fleming & Associates, L.L.P.,
Petitioners,
v.
Rebecca Wilson, et al.,
Respondents

═══════════════════════════════════════
On Petition for Review from the
Court of Appeals for the Fourteenth District of Texas
═══════════════════════════════════════

Argued October 24, 2023

JUSTICE YOUNG delivered the opinion of the Court.

Justice Devine and Justice Busby did not participate in the
decision.

We last saw this case after the Fourteenth Court of Appeals had
reversed the trial court’s summary judgment on a technical ground. This
Court reversed and remanded for the court of appeals to reach the merits.
Fleming v. Wilson, 610 S.W.3d 18 (Tex. 2020). On remand, that court
again reversed the summary-judgment order. 669 S.W.3d 450 (Tex.
App.—Houston [14th Dist.] 2021). We granted the ensuing petition for
review and now affirm, but for a reason different from the one stated by
the court of appeals. Specifically, we conclude that petitioners are
judicially estopped from establishing an essential component of their
summary-judgment motion.
I
We begin with an abbreviated background. This case involves a
long-running dispute between a lawyer and his former clients. Many
years ago, George Fleming and his law firm—petitioners in this Court,
to whom we refer jointly as “Fleming”—represented more than 8,000
plaintiffs in a mass-tort action against the manufacturer of a diet pill
known as “fen-phen.” Respondents are about 4,000 of Fleming’s former
clients, whom we call the “Wilson plaintiffs.” Before suing the drug
manufacturer in 2001, Fleming spent roughly $20 million to medically
screen over 40,000 potential claimants. About 20% of them became
Fleming’s clients. In 2006, Fleming settled the case for $339 million.
Fleming reimbursed himself for the costs of the screenings by
deducting that amount from the settlement funds. Based on the clients’
contingency agreements with him, he then distributed their percentage
of what remained. In other words, he charged his clients not just for their
own medical-screening costs but also for those of approximately 32,000
people who never became his clients and who did not participate in the
underlying case. See Fleming, 610 S.W.3d at 19.
This financial choice led to further litigation, now casting Fleming
as the defendant in various actions brought by his former clients. In one,
Sandra Karnes and Carol Tallant sued Fleming in federal court, claiming
that Fleming breached his fiduciary duty to his clients by charging them
for the screening costs. Karnes and Tallant tried to bring their lawsuit
as a class action, but Fleming successfully opposed class certification,

2
convincing the court that certification would be “inappropriate” given how
many distinct issues of law and fact separated the various potential
plaintiffs. Karnes v. Fleming, No. H-07-0620, 2008 WL 4528223, at *8
(S.D. Tex. July 31, 2008) (order denying class certification). Federal
jurisdiction over the state-law claims against Fleming was based on the
Class Action Fairness Act, see 28 U.S.C. § 1332(d), so Fleming’s success
in defeating class certification also led the district court to grant Fleming’s
motion to dismiss the case for lack of jurisdiction.
After the class-certification denial in Karnes, another group of
about 650 former clients—called the “Kinney plaintiffs”—sued Fleming
for breaches of contract and fiduciary duty. In a bellwether trial involving
ten of those plaintiffs, the jury rendered a verdict against Fleming. After
the Kinney verdict, the Wilson plaintiffs—the roughly 4,000 respondents
here—moved for summary judgment on the ground that the Kinney
verdict collaterally estopped Fleming from contesting the merits of the
Wilson plaintiffs’ claims against him. Fleming successfully opposed that
motion on the ground that “the [breach of fiduciary duty] issues presented
by the Kinney Plaintiffs are nowhere near ‘identical’ to those of the Wilson
Plaintiffs or any others.” As support for his argument, Fleming noted that
“the federal court [in Karnes] had denied class certification of breach of
fiduciary duty claims precisely because Plaintiffs had failed to show that
common questions predominated over the individualized issues.” He then
stated that “[n]othing has changed since Karnes.” The trial court denied
the Wilson plaintiffs’ motion for summary judgment without explanation.
After several years had passed, Fleming moved for a trial setting
in this case and proposed a bellwether trial with ten randomly selected

3
plaintiffs. In response, the Wilson plaintiffs proposed a bellwether trial
with six plaintiffs randomly selected by the court: one from the largest
group of settlements, two from the second largest group, and three from
the smallest group. The court adopted the Wilson plaintiffs’ proposal.
Kathy Harpst was the first-named of these six, and for that reason the
bellwether trial is called “the Harpst trial” and the six named plaintiffs
are collectively called “the Harpst plaintiffs.”
At the conclusion of the Harpst trial, the jury found in favor of
Fleming. At that point, Fleming moved for summary judgment, asserting
defensive collateral estoppel against the Wilson plaintiffs. Fleming, who
in the past had successfully emphasized the many alleged differences
among the various plaintiffs, now argued that the Harpst trial resolved
the common issues against each of the approximately 4,000 remaining
plaintiffs. The trial court granted summary judgment for Fleming.
The court of appeals initially reversed on the ground that the trial
court’s own judgment following the Harpst trial had not been
“authenticated” in support of Fleming’s summary-judgment motion and
thus could not be the basis for a judgment against the Wilson plaintiffs.
Wilson v. Fleming, 566 S.W.3d 410, 418 (Tex. App.—Houston [14th Dist.]
2018). We granted Fleming’s petition for review and, in a per curiam
opinion, reversed and remanded for the court of appeals to consider the
merits of Fleming’s collateral-estoppel defense. Fleming, 610 S.W.3d at 22.
On remand, the court of appeals again reversed. The court held that
Fleming could not establish his entitlement to defensive collateral estoppel
because there was no “privity” between the six Harpst plaintiffs and the
larger group of Wilson plaintiffs. 669 S.W.3d at 463. Thus, the court

4
concluded, the claims of the thousands of Wilson plaintiffs other than the
six Harpst trial plaintiffs could not be precluded by the Harpst judgment.
Fleming again sought this Court’s review, which we again granted.
II
We review a trial court’s grant of summary judgment de novo.
Provident Life & Accident Ins. Co. v. Knott, 128 S.W.3d 211, 215 (Tex.
2003). On a traditional motion for summary judgment, the movant—in
this case, Fleming—must conclusively prove every essential element of
his claim or defense as a matter of law. Draughon v. Johnson, 631 S.W.3d
81, 87–88 (Tex. 2021).
Fleming asserted three grounds for summary judgment in his
motion: defensive collateral estoppel, waiver, and release. Fleming raised
only collateral estoppel when he challenged the court of appeals’ first
judgment, so only that ground for summary judgment remains live. To
assert collateral estoppel, the movant “must establish that (1) the facts
sought to be litigated in the second action were fully and fairly litigated in
the first action; (2) those facts were essential to the judgment in the first
action; and (3) the parties were cast as adversaries in the first action.”
Sysco Food Servs., Inc. v. Trapnell, 890 S.W.2d 796, 801 (Tex. 1994).
We confine our analysis to the first of these requirements, which
we find dispositive. To benefit from defensive collateral estoppel, it is
Fleming’s burden to show that further proceedings would entail “the
relitigation of identical issues of facts or law that were actually litigated
and essential to the judgment in a prior suit.” Van Dyke v. Boswell,
O’Toole, Davis & Pickering, 697 S.W.2d 381, 384 (Tex. 1985) (emphasis
added). Fleming sought to carry this burden by asserting in his summary-

5
judgment motion that “the facts and legal issues in Wilson are identical
to Harpst.” He noted that “[t]he Wilson Plaintiffs assert[ed] the same
claims as the Harpst Plaintiffs” and that the claims all arose “from the
same set of facts.” He also argued that the “breach of contract claims . . .
are governed by the same contractual terms” and that “the breach of
fiduciary duty claim is governed by the same settlement agreements.”
Fleming forcefully defends these arguments in this Court, asserting
that we should adopt § 40 of the Restatement (Second) of Judgments, and
especially comment b and illustration 3. He relies heavily on the U.S.
Supreme Court’s decision in Taylor v. Sturgell, 553 U.S. 880 (2008).
These authorities, he argues, should persuade us that the Wilson
plaintiffs impliedly agreed to be bound by the Harpst judgment, thus
justifying application of nonparty preclusion and allowing us to reinstate
the summary-judgment order favoring Fleming. In response, the Wilson
plaintiffs point us to Fleming’s repeated assertions that his prior clients’
claims were insufficiently common for aggregate treatment. They
explicitly identify Fleming’s successful opposition to Karnes’s motion for
class certification as an example of his contradictory positions.
In a proper case, § 40 of the Restatement and Taylor’s approach to
nonparty preclusion warrant this Court’s serious consideration. But we
conclude that we cannot consider them here, in the context of Fleming’s
motion for summary judgment, because of the doctrine of judicial estoppel.
A
At its most general level, “[j]udicial estoppel is a common law
doctrine that prevents a party from assuming inconsistent positions in
litigation.” Perryman v. Spartan Tex. Six Cap. Partners, Ltd., 546 S.W.3d

6
110, 117 (Tex. 2018) (quoting Kane v. Nat’l Union Fire Ins. Co., 535 F.3d
380, 385 (5th Cir. 2008)). “Its essential function is to prevent the use of
intentional self-contradiction as a means of obtaining unfair advantage.”
Pleasant Glade Assembly of God v. Schubert, 264 S.W.3d 1, 6 (Tex. 2008)
(internal quotation omitted). The primary goal of judicial estoppel is not
to benefit the adverse party, although it will have that consequence.
Judicial estoppel instead aims to protect the integrity of the judicial
system itself. As the Supreme Court has put it, judicial-estoppel
principles target circumstances where a “party has succeeded in
persuading a court to accept that party’s earlier position, so that judicial
acceptance of an inconsistent position in a later proceeding would create
‘the perception that either the first or the second court was misled.’” New
Hampshire v. Maine, 532 U.S. 742, 750 (2001) (quoting Edwards v.
Aetna Life Ins. Co., 690 F.2d 595, 599 (6th Cir. 1982)).
As with other forms of estoppel, therefore, the question is not
whether a party’s current position is correct but whether the party is
entitled to press the position at all. And as with other equitable
principles, judicial estoppel is subject to several important limitations.
For one, “a party cannot be judicially estopped if it did not prevail in the
prior action.” Ferguson v. Bldg. Materials Corp. of Am., 295 S.W.3d 642,
643 (Tex. 2009). A party has “prevailed” if it persuaded the court to adopt
the party’s position and thus grant the relief that the party sought. That
party would “obtain an unfair advantage” if it could prevail in a second
proceeding by directly contradicting its prior successful position. Id. The
prior success is what makes it “unfair” for the party to obtain a new
advantage by contradicting its initial position. So if the party “gained no

7
advantage” from its position in an initial proceeding, “[t]he doctrine of
judicial estoppel simply does not apply” despite “the existence of an
inconsistent position” in later litigation. Id. at 644.
Inconsistency alone, therefore, is not enough. A party who is
unsuccessful in pressing a position does not threaten the integrity of the
judicial process by changing course and later adopting the view of the
prior court. Such “inconsistency” provides no unfair advantage. To the
contrary, it would be unfair to hold a party to a position that was rejected
in prior proceedings or to penalize a party for accepting a position that
the prior court adopted.
Likewise, to satisfy the demands of judicial estoppel, the prior
“statement must be deliberate, clear, and unequivocal.” Am. Sav. & Loan
Ass’n of Hous. v. Musick, 531 S.W.2d 581, 589 (Tex. 1975). After all, “[t]he
doctrine is not intended to punish inadvertent omissions or inconsistencies
but rather to prevent parties from playing fast and loose with the judicial
system for their own benefit.” Ferguson, 295 S.W.3d at 643.
Moreover, judicial estoppel applies only if the successful
representation arose in a different case or, at most, “in another phase” of
the same case. New Hampshire, 532 U.S. at 749 (quoting Pegram v.
Herdrich, 530 U.S. 211, 227 n.8 (2000)). The doctrine, we have said,
precludes “a party who successfully maintains a position in one
proceeding from afterwards adopting a clearly inconsistent position in
another proceeding to obtain an unfair advantage.” Ferguson, 295 S.W.3d
at 643 (emphasis added). “[I]t is not necessary that the party invoking
[the judicial-estoppel] doctrine should have been a party to the former
proceeding.” Long v. Knox, 291 S.W.2d 292, 295 (Tex. 1956). Nothing

8
turns on whether the adverse parties were different, as they typically are.
By contrast, “[c]ontradictory positions taken in the same proceeding may
raise issues of judicial admission but do not invoke the doctrine of
judicial estoppel.” Schubert, 264 S.W.3d at 6 (emphases added).*
We also emphasize that while the prior inconsistent statement
must have been intentionally made, the doctrine does not require a court
to conclude that a party aimed to purposefully deceive either the first or
the second court. Purposeful deception, of course, would implicate even
more serious concerns. We make no accusation of any such impropriety
here. Judicial estoppel turns not on a party’s subjective intent but on the
objective inconsistency of that party’s statements in different
proceedings. In the face of such an inconsistency, the “perception” that
the courts have been misled is at its height. New Hampshire, 532 U.S. at
750. The doctrine does not seek to police ill intent; it rather aims to
protect the integrity of the judicial system as a whole. Judicial estoppel
does not “punish” anyone for anything—it estops a party from

* Same-case inconsistencies are often wholly unproblematic. Parties may

maintain alternative positions as a case unfolds. But some contradictions made
earlier in the same proceeding may warrant relief when the contradiction
prejudices the adverse party, who must demonstrate detrimental reliance on an
earlier admission and “protect it by objecting to the introduction of evidence
contrary to the admission and by objecting to the submission of any issue bearing
on the fact or facts admitted.” Hurlbut v. Gulf Atl. Life Ins. Co., 749 S.W.2d 762,
765 (Tex. 1987). “[J]udicial oversight and procedural safeguards of a trial make
it unlikely that inconsistent positions taken and resolved in the same proceeding
will taint the judicial process.” Mark J. Plumer, Judicial Estoppel: The
Refurbishing of a Judicial Shield, 55 Geo. Wash. L. Rev. 409, 418 (1987). The
doctrine of judicial estoppel is thus both inapplicable within and “unnecessary
to protect the integrity of a single proceeding.” Id. Those same-case safeguards,
however, cannot prevent inconsistency across cases and in different tribunals.
Judicial estoppel fills that void.

9
contradicting a prior position that falls within the doctrine’s scope. It is
neither a punishment nor unfair to hold a party to its prior position
when the first court adopted that position and, because of that adoption,
the party obtained the result it sought.
On the other hand, while judicial estoppel does not turn on a
party’s good faith, it is still “an equitable doctrine.” Perryman, 546
S.W.3d at 117. Courts have discretion in applying judicial estoppel. Id.;
accord New Hampshire, 532 U.S. at 750. A court may decline to apply
the doctrine if, for example, it determines that a prior inconsistency was
wholly inadvertent, played a comparatively small role in the earlier
litigation, or provided a minimal or nominal benefit. See Perryman, 546
S.W.3d at 117; Schubert, 264 S.W.3d at 6. In such cases, the sense that
the party is winning both coming and going—and, correspondingly, any
hint of manipulation of the judicial system—is lessened or even
eliminated. On the other hand, an indication of bad faith or an especially
blatant inconsistency might warrant a court’s exercise of its discretion to
apply the doctrine even if the prior benefit was small, so long as the
doctrine’s requirements are otherwise satisfied.
But “[d]iscretion is not whim, and limiting discretion according to
legal standards helps promote the basic principle of justice that like cases
should be decided alike.” In re Rudolph Auto., LLC, 674 S.W.3d 289, 308
(Tex. 2023) (quoting Martin v. Franklin Cap. Corp., 546 U.S. 132, 139
(2005)). Consistent application of judicial estoppel follows from respect
for the requirements and limitations we have described. Specifically, the
party to be estopped must have obtained a benefit by making a clear and
unambiguous statement that convinced a prior court to adopt a position

10
that contradicts the party’s current position. The doctrine applies unless,
under the circumstances, the court finds that its application would be
unjust or contrary to the doctrine’s underlying purposes.
B
The Wilson plaintiffs amply invoked the principles of judicial
estoppel here. They highlighted at length Fleming’s repeated assertions
that his prior clients’ claims were insufficiently common for aggregate
treatment and, as an example of his contradictory claims, included
explicit references to Fleming’s opposition to Karnes’s motion for class
certification. The Wilson plaintiffs did not use the phrase “judicial
estoppel,” but they said more than enough to put Fleming on notice of
their contention that his prior inconsistent statements should preclude
relief for him here. We generally accord relief to litigants not on form,
after all, but on substance, so long as the pleadings “provide[] the
opposing party sufficient information to enable that party to prepare a
defense or a response.” First United Pentecostal Church of Beaumont v.
Parker, 514 S.W.3d 214, 225 (Tex. 2017).
We conclude that the doctrine applies. Fleming’s assertions in
prior litigation clearly and unequivocally contradict his summary-
judgment motion’s assertions regarding whether the Wilson plaintiffs’
legal and factual positions are materially identical to those of the Harpst
plaintiffs. In opposing class certification in Karnes, Fleming argued “that
common questions would not predominate” because of “the substantial
choice of law questions that need to be resolved” for each plaintiff.
Karnes, 2008 WL 4528223, at *2. Supporting this argument, Fleming
submitted an affidavit claiming that there were “significant differences in

11
the retainer contracts with respect to the charging of litigation expenses.”
Id. at *7. Fleming also argued that “the individualized contractual
provisions for . . . the treatment of expenses” made class certification
improper. Id. at *2. All of this reinforced his contention that class
certification would be “inappropriate.” Id. at *4.
The federal court was persuaded. Id. at *8. By successfully
arguing that the legal and factual issues brought by his former clients
were far from identical, Fleming obtained the result he desired: denial of
class certification. Needless to say, avoiding a class action is a valuable
result. See, e.g., Am. Campus Cmtys. v. Berry, 667 S.W.3d 277, 286 (Tex.
2023) (acknowledging how significantly class certification “raises the
stakes of a lawsuit”). Moreover, Fleming avoided not just the class action
but the federal forum altogether. He obtained dismissal because, without
a class action, there was no federal jurisdiction.
We do not doubt the good faith of Fleming’s statements opposing
class certification or his current contrary position. But the class-
certification representations were clear and purposeful, unambiguous
and specific. They were neither inadvertent nor misunderstood. The
federal court relied on those representations to grant Fleming the relief
that he sought—denial of class certification—which was far from
nominal. Most importantly, the arguments on which the court relied to
reject class certification directly conflict with Fleming’s current position,
which is that the Wilson and Harpst plaintiffs’ factual and legal issues
are materially identical.
This current position is essential for Fleming to prevail on his
affirmative defense today. If the plaintiffs are not all situated identically,

12
both within the Wilson plaintiffs’ group and relative to the Harpst
plaintiffs, Fleming cannot show that the facts sought to be litigated by
the Wilson plaintiffs were fully and fairly litigated in the Harpst trial.
His summary-judgment motion depends on that foundation, and he has
powerfully advocated the correctness of that position before this Court.
Fleming argues, with significant force, that the Wilson plaintiffs
themselves have repeatedly urged that their claims are sufficiently alike
that a win for one could support offensive collateral estoppel for all. He
contends that their hope of invoking offensive nonmutual collateral
estoppel shows that they have impliedly agreed to be bound by the results
of the trial no matter what they were. But even setting aside any
distinction between offensive and defensive collateral estoppel, Fleming
must carry his own summary-judgment burden to establish his
affirmative defense. The doctrine of judicial estoppel forecloses his ability
to do so because he is estopped from asserting that the thousands of
remaining plaintiffs’ claims are materially indistinguishable.
We therefore have no occasion to resolve the correctness of
Fleming’s current position—correct or not, he is estopped from asserting
it. We thus reserve for future cases the extent to which, under Texas law,
one may impliedly agree to be bound by a judgment to which one is not a
party such that nonparty preclusion applies.

* * *
The judgment of the court of appeals is affirmed.

Evan A. Young
Justice
OPINION DELIVERED: May 17, 2024

13