Merck & Co. v. Reynolds

(Slip Opinion) OCTOBER TERM, 2009 1
Syllabus
NOTE: Where it is feasible, a syllabus (headnote) will be released, as is
being done in connection with this case, at the time the opinion is issued.
The syllabus constitutes no part of the opinion of the Court but has been
prepared by the Reporter of Decisions for the convenience of the reader.
See United States v. Detroit Timber & Lumber Co., 200 U. S. 321, 337.
SUPREME COURT OF THE UNITED STATES
Syllabus
MERCK & CO., INC., ET AL. v. REYNOLDS ET AL.
CERTIORARI TO THE UNITED STATES COURT OF APPEALS FOR
THE THIRD CIRCUIT
No. 08–905. Argued November 30, 2009—Decided April 27, 2010
On November 6, 2003, respondent investors filed a securities fraud ac
tion under §10(b) of the Securities Exchange Act of 1934, alleging
that petitioner Merck & Co. knowingly misrepresented the heart
attack risks associated with its drug Vioxx. A securities fraud com
plaint is timely if filed no more than “2 years after the discovery of
the facts constituting the violation” or 5 years after the violation. 28
U. S. C. §1658(b). The District Court dismissed the complaint as un
timely because the plaintiffs should have been alerted to the possibil
ity of Merck’s misrepresentations prior to November 2001, more than
2 years before the complaint was filed, and they had failed to under
take a reasonably diligent investigation at that time. Among the
relevant circumstances were (1) a March 2000 “VIGOR” study com
paring Vioxx with the painkiller naproxen and showing adverse car
diovascular results for Vioxx, which Merck suggested might be due to
the absence of a benefit conferred by naproxen rather than a harm
caused by Vioxx (the naproxen hypothesis); (2) an FDA warning let
ter, released to the public on September 21, 2001, saying that Merck’s
Vioxx marketing with regard to the cardiovascular results was “false,
lacking in fair balance, or otherwise misleading”; and (3) pleadings
filed in products-liability actions in September and October 2001 al
leging that Merck had concealed information about Vioxx and inten
tionally downplayed its risks. The Third Circuit reversed, holding
that the pre-November 2001 events did not suggest that Merck acted
with scienter, an element of a §10(b) violation, and consequently did
not commence the running of the limitations period.
Held:
1. The limitations period in §1658(b)(1) begins to run once the
plaintiff actually discovered or a reasonably diligent plaintiff would
2 MERCK & CO. v. REYNOLDS
Syllabus
have “discover[ed] the facts constituting the violation”—whichever
comes first. In the statute of limitations context, “discovery” is often
used as a term of art in connection with the “discovery rule,” a doc
trine that delays accrual of a cause of action until the plaintiff has
“discovered” it. The rule arose in fraud cases but has been applied by
state and federal courts in other types of claims, and legislatures
have sometimes codified this rule. When “discovery” is written di
rectly into a statute, courts have typically interpreted the word to re
fer not only to actual discovery, but also to the hypothetical discovery
of facts a reasonably diligent plaintiff would know. Congress in
tended courts to interpret the word “discovery” in §1658(b)(1) simi
larly. That statute was enacted after this Court determined a gov
erning limitations period for private §10(b) actions, Lampf, Pleva,
Lipkind, Prupis & Petigrow v. Gilbertson, 501 U. S. 350, concluding
that such actions “must be commenced within one year after the dis
covery of the facts constituting the violation . . . ,” id., at 364 (empha
sis added). Since then, Courts of Appeals deciding the matter have
held that “discovery” occurs both when a plaintiff actually discovers
the facts and when a hypothetical reasonably diligent plaintiff would
have discovered them. In 2002, Congress repeated Lampf’s critical
language in enacting the present limitations statute. Normally,
when Congress enacts statutes, it is aware of relevant judicial prece
dent. See, e.g., Edelman v. Lynchburg College, 535 U. S. 106, 116–
117, and n. 13. Given the history and precedent surrounding the use
of “discovery” in the limitations context generally as well as in this
provision, the reasons for making this assumption are particularly
strong here. Merck’s claims are evaluated accordingly. Pp. 8–12.
2. In determining the time at which “discovery” occurs, terms such
as “inquiry notice” and “storm warnings” may be useful insofar as
they identify a time when the facts would have prompted a reasona
bly diligent plaintiff to begin investigating. But the limitations pe
riod does not begin to run until the plaintiff thereafter discovers or a
reasonably diligent plaintiff would have discovered “the facts consti
tuting the violation,” including scienter—irrespective of whether the
actual plaintiff undertook a reasonably diligent investigation.
Pp. 12–17.
(a) Contrary to Merck’s argument, facts showing scienter are
among those that “constitut[e] the violation.” Scienter is assuredly a
“fact.” In a §10(b) action, it refers to “a mental state embracing in
tent to deceive, manipulate, or defraud,” Ernst & Ernst v. Hochfelder,
425 U. S. 185, 194, n. 12, and “constitut[es]” an important and neces
sary element of a §10(b) “violation.” See Tellabs, Inc. v. Makor Issues
& Rights, Ltd., 551 U. S. 308, 319. Because the scienter element of
§10(b) fraud cases has special heightened pleading requirements, see
Cite as: 559 U. S. ____ (2010) 3
Syllabus
15 U. S. C. §78u–4(b)(2), unless a §10(b) complaint sets forth facts show
ing that it is “at least as likely as” not that the defendant acted with the
relevant intent, the claim will fail. Tellabs, supra, at 328 (emphasis de
leted). It would frustrate the very purpose of the discovery rule codified in
§1658(b)(1) if the limitations period began to run regardless of whether a
plaintiff had “discover[ed]” any facts suggesting scienter. Pp. 12–14.
(b) The Court also rejects Merck’s argument that, even if “discov
ery” requires facts related to scienter, facts that tend to show a mate
rially false or misleading statement (or material omission) are ordi
narily sufficient to show scienter. Where §10(b) is at issue, the
relation of factual falsity and state of mind is more context specific.
For instance, an incorrect prediction about a firm’s future earnings,
by itself, does not automatically show whether the speaker deliber
ately lied or made an innocent error. Hence, “discovery” of additional
scienter-related facts may be required. The statute’s inclusion of an
unqualified bar on actions instituted “5 years after such violation,”
§1658(b)(2), should diminish Merck’s fear that this requirement will
give life to stale claims or subject defendants to liability for acts
taken long ago. P. 14.
(c) And the Court cannot accept Merck’s argument that the limi
tations period begins at “inquiry notice,” meaning the point where
the facts would lead a reasonably diligent plaintiff to investigate fur
ther, because that point is not necessarily the point at which the
plaintiff would already have “discover[ed]” facts showing scienter or
other “facts constituting the violation.” The statute says that the
plaintiff’s claim accrues only after the “discovery” of those latter
facts. It contains no indication that the limitations period can some
times begin before “discovery” can take place. Merck also argues that
determining when a hypothetical reasonably diligent plaintiff would
have “discover[ed]” the necessary facts is too complicated for judges
to undertake. But courts applying the traditional discovery rule have
long had to ask what a reasonably diligent plaintiff would have
known and done in myriad circumstances and already undertake this
kind of inquiry in securities fraud cases. Pp. 14–17.
3. Prior to November 6, 2001, the plaintiffs did not discover, and
Merck has not shown that a reasonably diligent plaintiff would have
discovered, “the facts constituting the violation.” The FDA’s Septem
ber 2001 warning letter shows little or nothing about the here
relevant scienter, i.e., whether Merck advanced the naproxen hy
pothesis with fraudulent intent. The FDA itself described the hy
pothesis as a “possible explanation” for the VIGOR results, faulting
Merck only for failing sufficiently to publicize the less favorable al
ternative, that Vioxx might be harmful. The products-liability com
plaints’ general statements about Merck’s state of mind show little
4 MERCK & CO. v. REYNOLDS
Syllabus
more. Thus, neither these circumstances nor any of the other pre-
November 2001 circumstances reveal “facts” indicating the relevant
scienter. Pp. 17–19.
543 F. 3d 150, affirmed.
BREYER, J., delivered the opinion of the Court, in which ROBERTS,
C. J., and KENNEDY, GINSBURG, ALITO, and SOTOMAYOR, JJ., joined.
STEVENS, J., filed an opinion concurring in part and concurring in the
judgment. SCALIA, J., filed an opinion concurring in part and concur
ring in the judgment, in which THOMAS, J., joined.
Cite as: 559 U. S. ____ (2010) 1
Opinion of the Court
NOTICE: This opinion is subject to formal revision before publication in the
preliminary print of the United States Reports. Readers are requested to
notify the Reporter of Decisions, Supreme Court of the United States, Wash
ington, D. C. 20543, of any typographical or other formal errors, in order
that corrections may be made before the preliminary print goes to press.
SUPREME COURT OF THE UNITED STATES
_________________
No. 08–905
_________________
MERCK & CO., INC., ET AL., PETITIONERS v. RICHARD
REYNOLDS ET AL.
ON WRIT OF CERTIORARI TO THE UNITED STATES COURT OF
APPEALS FOR THE THIRD CIRCUIT
[April 27, 2010]
JUSTICE BREYER delivered the opinion of the Court.
This case concerns the timeliness of a complaint filed in
a private securities fraud action. The complaint was
timely if filed no more than two years after the plaintiffs
“discover[ed] the facts constituting the violation.” 28
U. S. C. §1658(b)(1). Construing this limitations statute
for the first time, we hold that a cause of action accrues (1)
when the plaintiff did in fact discover, or (2) when a rea
sonably diligent plaintiff would have discovered, “the facts
constituting the violation”—whichever comes first. We
also hold that the “facts constituting the violation” include
the fact of scienter, “a mental state embracing intent to
deceive, manipulate, or defraud,” Ernst & Ernst v.
Hochfelder, 425 U. S. 185, 194, n. 12 (1976). Applying this
standard, we affirm the Court of Appeals’ determination
that the complaint filed here was timely.
I
The action before us involves a claim by a group of inves
tors (the plaintiffs, respondents here) that Merck & Co.
and others (the petitioners here, hereinafter Merck) know
ingly misrepresented the risks of heart attacks accompany
2 MERCK & CO. v. REYNOLDS
Opinion of the Court
ing the use of Merck’s pain-killing drug, Vioxx (leading to
economic losses when the risks later became apparent).
The plaintiffs brought an action for securities fraud under
§10(b) of the Securities Exchange Act of 1934. See 48 Stat.
891, as amended, 15 U. S. C. §78j(b); SEC Rule 10b–5, 17
CFR §240.10b–5(b) (2009); Dura Pharmaceuticals, Inc. v.
Broudo, 544 U. S. 336, 341–342 (2005).
The applicable statute of limitations provides that a
“private right of action” that, like the present action,
“involves a claim of fraud, deceit, manipulation, or con
trivance in contravention of a regulatory requirement
concerning the securities laws . . . may be brought not
later than the earlier of—
“(1) 2 years after the discovery of the facts constituting
the violation; or
“(2) 5 years after such violation.” 28 U. S. C. §1658(b).
The complaint in this case was filed on November 6,
2003, and no one doubts that it was filed within five years
of the alleged violation. Therefore, the critical date for
timeliness purposes is November 6, 2001—two years
before this complaint was filed. Merck claims that before
this date the plaintiffs had (or should have) discovered the
“facts constituting the violation.” If so, by the time the
plaintiffs filed their complaint, the 2-year statutory period
in §1658(b)(1) had run. The plaintiffs reply that they had
not, and could not have, discovered by the critical date
those “facts,” particularly not the facts related to scienter,
and that their complaint was therefore timely.
A
We first set out the relevant pre-November 2001 facts,
as we have gleaned them from the briefs, the record, and
the opinions below.
1. 1990’s. In the mid-1990’s Merck developed Vioxx. In
1999 the Food and Drug Administration (FDA) approved it
for prescription use. Vioxx suppresses pain by inhibiting
Cite as: 559 U. S. ____ (2010) 3
Opinion of the Court
the body’s production of an enzyme called COX–2 (cyclooxy-
genase-2). COX–2 is associated with pain and inflamma-
tion. Unlike some other anti-inflammatory drugs in its
class like aspirin, ibuprofen, and naproxen, Vioxx does not
inhibit production of a second enzyme called COX–1
(cyclooxygenase-1). COX–1 plays a part in the functioning
of the gastrointestinal tract and also in platelet aggregation
(associated with blood clots). App. 50–51.
2. March 2000. Merck announced the results of a study,
called the “VIGOR” study. Id., at 291–294. The study
compared Vioxx with another painkiller, naproxen. The
study showed that persons taking Vioxx suffered fewer
gastrointestinal side effects (as Merck had hoped). But
the study also revealed that approximately 4 out of every
1,000 participants who took Vioxx suffered heart attacks,
compared to only 1 per 1,000 participants who took
naproxen. Id., at 296, 306; see Bombardier et al., Com-
parison of Upper Gastrointestinal Toxicity of Rofecoxib
and Naproxen in Patients with Rheumatoid Arthritis, 343
New England J. Medicine 1520, 1523, 1526–1527 (2000).
Merck’s press release acknowledged VIGOR’s adverse
cardiovascular data. But Merck said that these data were
“consistent with naproxen’s ability to block platelet aggre-
gation.” App. 291. Merck noted that, since “Vioxx, like all
COX–2 selective medicines, does not block platelet aggre-
gation[, it] would not be expected to have similar effects.”
Ibid. And Merck added that “safety data from all other
completed and ongoing clinical trials . . . showed no indica-
tion of a difference in the incidence of thromboembolic
events between Vioxx” and either a placebo or comparable
drugs. Id., at 293 (emphasis deleted).
This theory—that VIGOR’s troubling cardiovascular
findings might be due to the absence of a benefit conferred
by naproxen rather than due to a harm caused by Vioxx—
later became known as the “naproxen hypothesis.” In
advancing that hypothesis, Merck acknowledged that the
4 MERCK & CO. v. REYNOLDS
Opinion of the Court
naproxen benefit “had not been observed previously.” Id.,
at 291. Journalists and stock market analysts reported all
of the above—the positive gastrointestinal results, the
troubling cardiovascular finding, the naproxen hypothesis,
and the fact that the naproxen hypothesis was unproved.
See id., at 355–391, 508–557.
3. February 2001 to August 2001. Public debate about
the naproxen hypothesis continued. In February 2001, the
FDA’s Arthritis Advisory Committee convened to consider
Merck’s request that the Vioxx label be changed to reflect
VIGOR’s positive gastrointestinal findings. The VIGOR
cardiovascular findings were also discussed. Id., at 392–
395, 558–577. In May 2001, a group of plaintiffs filed a
products-liability lawsuit against Merck, claiming that
“Merck’s own research” had demonstrated that “users of
Vioxx were four times as likely to suffer heart attacks as
compared to other less expensive, medications.” Id., at
869. In August 2001, the Journal of the American Medical
Association wrote that the available data raised a “cau
tionary flag” and strongly urged that “a trial specifically
assessing cardiovascular risk” be done. Id., at 331–332;
Mukherjee, Nissen, & Topol, Risk of Cardiovascular
Events Associated with Selective Cox-2 Inhibitors, 286
JAMA 954 (2001). At about the same time, Bloomberg
News quoted a Merck scientist who claimed that Merck
had “additional data” that were “very, very reassuring,”
and Merck issued a press release stating that it stood
“behind the overall and cardiovascular safety profile . . . of
Vioxx.” App. 434, 120 (emphasis deleted; internal quota
tion marks omitted).
4. September and October 2001. The FDA sent Merck a
warning letter released to the public on September 21,
2001. It said that, in respect to cardiovascular risks,
Merck’s Vioxx marketing was “false, lacking in fair bal
ance, or otherwise misleading.” Id., at 339. At the same
time, the FDA acknowledged that the naproxen hypothesis
Cite as: 559 U. S. ____ (2010) 5
Opinion of the Court
was a “possible explanation” of the VIGOR results. Id., at
340. But it found that Merck’s “promotional campaign
selectively present[ed]” that hypothesis without adequately
acknowledging “another reasonable explanation,” namely,
“that Vioxx may have pro-thrombotic [i.e., adverse cardio
vascular] properties.” Ibid. The FDA ordered Merck to
send healthcare providers a corrective letter. Id., at 353.
After the FDA letter was released, more products
liability lawsuits were filed. See id., at 885–956. Merck’s
share price fell by 6.6% over several days. See id., at 832.
By October 1, the price rebounded. See ibid. On October
9, 2001, the New York Times said that Merck had reexam
ined its own data and “found no evidence that Vioxx in
creased the risk of heart attacks.” App. 504. It quoted the
president of Merck Research Laboratories as positing
“ ‘two possible interpretations’ ”: “ ‘Naproxen lowers the
heart attack rate, or Vioxx raises it.’ ” Ibid. Stock ana
lysts, while reporting the warning letter, also noted that
the FDA had not denied that the naproxen hypothesis
remained an unproven but possible explanation. See id.,
at 614, 626, 628.
B
We next set forth three important events that occurred
after the critical date.
1. October 2003. The Wall Street Journal published the
results of a Merck-funded Vioxx study conducted at Bos
ton’s Brigham and Women’s Hospital. After examining
the medical records of more than 50,000 Medicare pa
tients, researchers found that those given Vioxx for 30-to
90 days were 37% more likely to have suffered a heart
attack than those given either a different painkiller or no
painkiller at all. Id., at 164–165. (That is to say, if pa
tients given a different painkiller or given no painkiller at
all suffered 10 heart attacks, then the same number of
patients given Vioxx would suffer 13 or 14 heart attacks.)
6 MERCK & CO. v. REYNOLDS
Opinion of the Court
Merck defended Vioxx and pointed to the study’s limita
tions. Id., at 165–167.
2. September 30, 2004. Merck withdrew Vioxx from the
market. It said that a new study had found “an increased
risk of confirmed cardiovascular events beginning after 18
months of continuous therapy.” Id., at 182 (internal quo
tation marks omitted). A Merck representative publicly
described the results as “totally unexpected.” Id., at 186.
Merck’s shares fell by 27% the same day. Id., at 185, 856.
3. November 1, 2004. The Wall Street Journal published
an article stating that “internal Merck e-mails and mar
keting materials as well as interviews with outside scien
tists show that the company fought forcefully for years to
keep safety concerns from destroying the drug’s commer
cial prospects.” Id., at 189–190. The article said that an
early e-mail from Merck’s head of research had said that
the VIGOR “results showed that the cardiovascular events
‘are clearly there,’ ” that it was “ ‘a shame but . . . a low
incidence,’ ” and that it “ ‘is mechanism based as we wor
ried it was.’ ” Id., at 192. It also said that Merck had
given its salespeople instructions to “ ‘DODGE’ ” questions
about Vioxx’s cardiovascular effects. Id., at 193.
C
The plaintiffs filed their complaint on November 6,
2003. As subsequently amended, the complaint alleged
that Merck had defrauded investors by promoting the
naproxen hypothesis, knowing the hypothesis was false.
It said, for example, that Merck “knew, at least as early as
1996, of the serious safety issues with Vioxx,” and that a
“1998 internal Merck clinical trial . . . revealed that . . .
serious cardiovascular events . . . occurred six times more
frequently in patients given Vioxx than in patients given a
different arthritis drug or placebo.” Id., at 56, 58–59
(emphasis and capitalization deleted).
Merck, believing that the plaintiffs knew or should have
Cite as: 559 U. S. ____ (2010) 7
Opinion of the Court
known the “facts constituting the violation” at least two
years earlier, moved to dismiss the complaint, saying it
was filed too late. The District Court granted the motion.
The court held that the (March 2001) VIGOR study, the
(September 2001) FDA warning letter, and Merck’s (Octo
ber 2001) response should have alerted the plaintiffs to a
“possibility that Merck had knowingly misrepresented
material facts” no later than October 9, 2001, thus placing
the plaintiffs on “inquiry notice” to look further. In re
Merck & Co. Securities, Derivative & “ERISA” Litigation,
483 F. Supp. 2d 407, 423 (NJ 2007) (emphasis added).
Finding that the plaintiffs had failed to “show that they
exercised reasonable due diligence but nevertheless were
unable to discover their injuries,” the court took October 9,
2001, as the date that the limitations period began to run
and therefore found the complaint untimely. Id., at 424.
The Court of Appeals for the Third Circuit reversed. A
majority held that the pre-November 2001 events, while
constituting “storm warnings,” did not suggest much by
way of scienter, and consequently did not put the plaintiffs
on “inquiry notice,” requiring them to investigate further.
In re Merck & Co. Securities, Derivative & “ERISA” Litiga
tion, 543 F. 3d 150, 172 (2008). A dissenting judge consid
ered the pre-November 2001 events sufficient to start the
2-year clock running. Id., at 173 (opinion of Roth, J.).
Merck sought review in this Court, pointing to dis
agreements among the Courts of Appeals. Compare Theo
harous v. Fong, 256 F. 3d 1219, 1228 (CA11 2001) (limita
tions period begins to run when information puts plaintiffs
on “inquiry notice” of the need for investigation), with
Shah v. Meeker, 435 F. 3d 244, 249 (CA2 2006) (same; but
if plaintiff does investigate, period runs “from the date
such inquiry should have revealed the fraud” (internal
quotation marks omitted)), and New England Health Care
Employees Pension Fund v. Ernst & Young, LLP, 336 F. 3d
495, 501 (CA6 2003) (limitations period always begins to
8 MERCK & CO. v. REYNOLDS
Opinion of the Court
run only when a reasonably diligent plaintiff, after being
put on “inquiry notice,” should have discovered facts con
stituting violation (internal quotation marks omitted)).
We granted Merck’s petition.
II
Before turning to Merck’s arguments, we consider a
more basic matter. The parties and the Solicitor General
agree that §1658(b)(1)’s word “discovery” refers not only to
a plaintiff’s actual discovery of certain facts, but also to
the facts that a reasonably diligent plaintiff would have
discovered. We agree. But because the statute’s language
does not make this interpretation obvious, and because we
cannot answer the question presented without considering
whether the parties are right about this matter, we set
forth the reasons for our agreement in some detail.
We recognize that one might read the statutory words
“after the discovery of the facts constituting the violation”
as referring to the time a plaintiff actually discovered the
relevant facts. But in the statute of limitations context,
the word “discovery” is often used as a term of art in con
nection with the “discovery rule,” a doctrine that delays
accrual of a cause of action until the plaintiff has “discov
ered” it. The rule arose in fraud cases as an exception to
the general limitations rule that a cause of action accrues
once a plaintiff has a “complete and present cause of ac
tion,” Bay Area Laundry and Dry Cleaning Pension Trust
Fund v. Ferbar Corp. of Cal., 522 U. S. 192, 201 (1997)
(citing Clark v. Iowa City, 20 Wall. 583, 589 (1875); inter
nal quotation marks omitted). This Court long ago recog
nized that something different was needed in the case of
fraud, where a defendant’s deceptive conduct may prevent
a plaintiff from even knowing that he or she has been
defrauded. Otherwise, “the law which was designed to
prevent fraud” could become “the means by which it is
Cite as: 559 U. S. ____ (2010) 9
Opinion of the Court
made successful and secure.” Bailey v. Glover, 21 Wall.
342, 349 (1875). Accordingly, “where a plaintiff has been
injured by fraud and remains in ignorance of it without
any fault or want of diligence or care on his part, the bar
of the statute does not begin to run until the fraud is
discovered.” Holmberg v. Armbrecht, 327 U. S. 392, 397
(1946) (internal quotation marks omitted; emphasis
added). And for more than a century, courts have under
stood that “[f]raud is deemed to be discovered . . . when, in
the exercise of reasonable diligence, it could have been
discovered.” 2 H. Wood, Limitation of Actions §276b(11),
p. 1402 (4th ed. 1916); see id., at 1401–1403, and nn. 74–
84 (collecting cases and statutes); see, e.g., Holmberg,
supra, at 397; Kirby v. Lake Shore & Michigan Southern
R. Co., 120 U. S. 130, 138 (1887) (The rule “regard[s] the
cause of action as having accrued at the time the fraud
was or should have been discovered”).
More recently, both state and federal courts have ap
plied forms of the “discovery rule” to claims other than
fraud. See 2 C. Corman, Limitation of Actions §§11.1.2.1,
11.1.2.3, pp. 136–142, and nn. 6–13, 18–23 (1991 and 1993
Supp.) (hereinafter Corman) (collecting cases); see, e.g.,
United States v. Kubrick, 444 U. S. 111 (1979). Legisla
tures have codified the discovery rule in various contexts.
2 Corman §11.2, at 170–171, and nn. 1–9 (collecting stat
utes); see, e.g., 28 U. S. C. §2409a(g) (actions to quiet title
against the United States). In doing so, legislators have
written the word “discovery” directly into the statute. And
when they have done so, state and federal courts have
typically interpreted the word to refer not only to actual
discovery, but also to the hypothetical discovery of facts a
reasonably diligent plaintiff would know. See, e.g., Pea
cock v. Barnes, 142 N. C. 215, 217–220, 55 S. E. 99, 100
(1906); Davis v. Hibernia Sav. & Loan Soc., 21 Cal. App.
444, 448, 132 P. 462, 464 (1913); Roether v. National
Union Fire Ins. Co., 51 N. D. 634, 640–642, 200 N. W. 818,
10 MERCK & CO. v. REYNOLDS
Opinion of the Court
821 (1924); Goldenberg v. Bache & Co., 270 F. 2d 675, 681
(CA5 1959); Mobley v. Hall, 202 Mont. 227, 232, 657 P. 2d
604, 606 (1983); Tregenza v. Great American Communica
tions Co., 12 F. 3d 717, 721–722 (CA7 1993); J. Geils Band
Employee Benefit Plan v. Smith Barney Shearson, Inc., 76
F. 3d 1245, 1254 (CA1 1996).
Thus, treatise writers now describe “the discovery rule”
as allowing a claim “to accrue when the litigant first
knows or with due diligence should know facts that will
form the basis for an action.” 2 Corman §11.1.1, at 134
(emphasis added); see also ibid., n. 1 (collecting cases); 37
Am. Jur. 2d, Fraud and Deceit §347, p. 354 (2001 and
Supp. 2009) (noting that the various formulations of “dis
covery” all provide that “in addition to actual knowledge of
the fraud, once a reasonably diligent party is in a position
that they should have sufficient knowledge or information
to have actually discovered the fraud, they are charged
with discovery”); id., at 354–355, and nn. 2–11 (collecting
cases).
Like the parties, we believe that Congress intended
courts to interpret the word “discovery” in §1658(b)(1)
similarly. Before Congress enacted that statute, this
Court, having found in the federal securities laws the
existence of an implied private §10(b) action, determined
its governing limitations period by looking to other limita
tions periods in the federal securities laws. Lampf, Pleva,
Lipkind, Prupis & Petigrow v. Gilbertson, 501 U. S. 350(1991). Noting the existence of various formulations
“differ[ing] slightly in terminology,” the Court chose the
language in 15 U. S. C. §78i(e), the statutory provision
that governs securities price manipulation claims. 501
U. S., at 364, n. 9. And in doing so, the Court said that
private §10(b) actions “must be commenced within one
year after the discovery of the facts constituting the viola
tion and within three years after such violation.” Id., at
364 (emphasis added). (The Court listed among the vari
Cite as: 559 U. S. ____ (2010) 11
Opinion of the Court
ous formulations the one in 15 U. S. C. §77m, on which the
concurrence relies. See post, at 2–4 (SCALIA, J., concurring
in part and concurring in judgment); Lampf, supra, at 360,
and n. 7 (quoting §77m).)
Subsequently, every Court of Appeals to decide the
matter held that “discovery of the facts constituting the
violation” occurs not only once a plaintiff actually discov
ers the facts, but also when a hypothetical reasonably
diligent plaintiff would have discovered them. See, e.g.,
Law v. Medco Research, Inc., 113 F. 3d 781, 785–786 (CA7
1997); Dodds v. Cigna Securities, Inc., 12 F. 3d 346, 350,
353 (CA2 1993); see In re NAHC, Inc. Securities Litigation,
306 F. 3d 1314, 1325, n. 4 (CA3 2002) (collecting cases).
Some of those courts noted that other limitations provi
sions in the federal securities laws explicitly provide that
the period begins to run “ ‘after the discovery of the untrue
statement . . . or after such discovery should have been
made by [the] exercise of reasonable diligence,’ ” whereas
the formulation adopted by the Court in Lampf from 15
U. S. C. §78i(e) does not. Tregenza, supra, at 721 (quoting
§77m; emphasis added in Tregenza); see Lampf, supra, at
364, n. 9. But, courts reasoned, because the term “discov
ery” in respect to statutes of limitations for fraud has long
been understood to include discoveries a reasonably dili
gent plaintiff would make, the omission of an explicit
provision to that effect did not matter. Tregenza, supra, at
721; accord, New England Health Care, 336 F. 3d, at 499–
500.
In 2002, when Congress enacted the present limitations
statute, it repeated Lampf’s critical language. The statute
says that an action based on fraud “may be brought not
later than the earlier of . . . 2 years after the discovery of
the facts constituting the violation” (or “5 years after such
violation”). §804 of the Sarbanes-Oxley Act, 116 Stat. 801,
codified at 28 U. S. C. §1658(b) (emphasis added). (This
statutory provision does not make the linguistic distinc
12 MERCK & CO. v. REYNOLDS
Opinion of the Court
tion that the concurrence finds in a different statute,
§77m, and upon which its argument rests. Cf. 29 U. S. C.
§1113(2) (statute in which Congress provided that an
action be brought “three years after the earliest date on
which the plaintiff had actual knowledge of the breach or
violation” (emphasis added)).) Not surprisingly, the
Courts of Appeals unanimously have continued to inter
pret the word “discovery” in this statute as including not
only facts a particular plaintiff knows, but also the facts
any reasonably diligent plaintiff would know. See, e.g.,
Staehr v. Hartford Financial Servs. Group, Inc., 547 F. 3d
406, 411 (CA2 2008); Sudo Properties, Inc. v. Terrebonne
Parish Consolidated Govt., 503 F. 3d 371, 376 (CA5 2007).
We normally assume that, when Congress enacts stat
utes, it is aware of relevant judicial precedent. See, e.g.,
Edelman v. Lynchburg College, 535 U. S. 106, 116–117,
and n. 13 (2002); Commissioner v. Keystone Consol. Indus
tries, Inc., 508 U. S. 152, 159 (1993). Given the history
and precedent surrounding the use of the word “discovery”
in the limitations context generally as well as in this
provision in particular, the reasons for making this as
sumption are particularly strong here. We consequently
hold that “discovery” as used in this statute encompasses
not only those facts the plaintiff actually knew, but also
those facts a reasonably diligent plaintiff would have
known. And we evaluate Merck’s claims accordingly.
III
We turn now to Merck’s arguments in favor of holding
that petitioners’ claims accrued before November 6, 2001.
First, Merck argues that the statute does not require
“discovery” of scienter-related “facts.” See Brief for Peti
tioners 19–28. We cannot agree, however, that facts about
scienter are unnecessary.
The statute says that the limitations period does not
begin to run until “discovery of the facts constituting the
Cite as: 559 U. S. ____ (2010) 13
Opinion of the Court
violation.” 28 U. S. C. §1658(b)(1) (emphasis added).
Scienter is assuredly a “fact.” In a §10(b) action, scienter
refers to “a mental state embracing intent to deceive,
manipulate, or defraud.” Ernst & Ernst, 425 U. S., at 194,
n. 12. And the “ ‘state of a man’s mind is as much a fact as
the state of his digestion.’ ” Postal Service Bd. of Gover
nors v. Aikens, 460 U. S. 711, 716 (1983) (quoting Edging
ton v. Fitzmaurice, [1885] 29 Ch. Div. 459, 483).
And this “fact” of scienter “constitut[es]” an important
and necessary element of a §10(b) “violation.” A plaintiff
cannot recover without proving that a defendant made a
material misstatement with an intent to deceive—not
merely innocently or negligently. See Tellabs, Inc. v.
Makor Issues & Rights, Ltd., 551 U. S. 308, 319 (2007);
Ernst & Ernst, supra. Indeed, Congress has enacted
special heightened pleading requirements for the scienter
element of §10(b) fraud cases. See 15 U. S. C. §78u–4(b)(2)
(requiring plaintiffs to “state with particularity facts
giving rise to a strong inference that the defendant acted
with the required state of mind” (emphasis added)). As a
result, unless a §10(b) plaintiff can set forth facts in the
complaint showing that it is “at least as likely as” not that the
defendant acted with the relevant knowledge or intent, the
claim will fail. Tellabs, supra, at 328 (emphasis deleted). It
would therefore frustrate the very purpose of the discovery
rule in this provision—which, after all, specifically applies
only in cases “involv[ing] a claim of fraud, deceit, manipu
lation, or contrivance,” §1658(b)—if the limitations period
began to run regardless of whether a plaintiff had discov
ered any facts suggesting scienter. So long as a defendant
concealed for two years that he made a misstatement with
an intent to deceive, the limitations period would expire
before the plaintiff had actually “discover[ed]” the fraud.
We consequently hold that facts showing scienter are
among those that “constitut[e] the violation.” In so hold
ing, we say nothing about other facts necessary to support
14 MERCK & CO. v. REYNOLDS
Opinion of the Court
a private §10(b) action. Cf. Brief for United States as
Amicus Curiae 12, n. 1 (suggesting that facts concerning a
plaintiff’s reliance, loss, and loss causation are not among
those that constitute “the violation” and therefore need not
be “discover[ed]” for a claim to accrue).
Second, Merck argues that, even if “discovery” requires
facts related to scienter, facts that tend to show a materi
ally false or misleading statement (or material omission)
are ordinarily sufficient to show scienter as well. See
Brief for Petitioners 22, 28–29. But we do not see how
that is so. We recognize that certain statements are such
that, to show them false is normally to show scienter as
well. It is unlikely, for example, that someone would
falsely say “I am not married” without being aware of the
fact that his statement is false. Where §10(b) is at issue,
however, the relation of factual falsity and state of mind is
more context specific. An incorrect prediction about a
firm’s future earnings, by itself, does not automatically tell
us whether the speaker deliberately lied or just made an
innocent (and therefore nonactionable) error. Hence, the
statute may require “discovery” of scienter-related facts
beyond the facts that show a statement (or omission) to be
materially false or misleading. Merck fears that this
requirement will give life to stale claims or subject defen
dants to liability for acts taken long ago. But Congress’
inclusion in the statute of an unqualified bar on actions
instituted “5 years after such violation,” §1658(b)(2), giv
ing defendants total repose after five years, should dimin
ish that fear. Cf. Lampf, 501 U. S., at 363 (holding compa
rable bar not subject to equitable tolling).
Third, Merck says that the limitations period began to
run prior to November 2001 because by that point the
plaintiffs were on “inquiry notice.” Merck uses the term
“inquiry notice” to refer to the point “at which a plaintiff
possesses a quantum of information sufficiently suggestive
of wrongdoing that he should conduct a further inquiry.”
Cite as: 559 U. S. ____ (2010) 15
Opinion of the Court
Brief for Petitioners 20. And some, but not all, Courts of
Appeals have used the term in roughly similar ways. See,
e.g., Franze v. Equitable Assurance, 296 F. 3d 1250, 1254
(CA11 2002) (“[I]nquiry notice [is] “ ‘the term used for
knowledge of facts that would lead a reasonable person to
begin investigating the possibility that his legal rights had
been infringed’ ”). Cf. Dodds, 12 F. 3d, at 350 (“duty of
inquiry” arises once “circumstances would suggest to an
investor of ordinary intelligence the probability that she
had been defrauded”); Fujisawa Pharmaceutical Co. v.
Kapoor, 115 F. 3d 1332, 1335–1336 (CA7 1997) (“The facts
constituting [inquiry] notice must be sufficien[t] . . . to
incite the victim to investigate” and “to enable him to tie
up any loose ends and complete the investigation in time
to file a timely suit”); Great Rivers Cooperative of South
eastern Iowa v. Farmland Industries, Inc., 120 F. 3d 893,
896 (CA8 1997) (“Inquiry notice exists when the victim is
aware of facts that would lead a reasonable person to
investigate and consequently acquire actual knowledge of
the defendant’s misrepresentations” (emphasis added)).
If the term “inquiry notice” refers to the point where the
facts would lead a reasonably diligent plaintiff to investi
gate further, that point is not necessarily the point at which
the plaintiff would already have discovered facts showing
scienter or other “facts constituting the violation.” But the
statute says that the plaintiff’s claim accrues only after the
“discovery” of those latter facts. Nothing in the text sug
gests that the limitations period can sometimes begin before
“discovery” can take place. Merck points out that, as we
have discussed, see supra, at 8–9, the court-created “discov
ery rule” exception to ordinary statutes of limitations is not
generally available to plaintiffs who fail to pursue their
claims with reasonable diligence. But we are dealing here
with a statute, not a court-created exception to a statute.
Because the statute contains no indication that the limita
tions period should occur at some earlier moment before
16 MERCK & CO. v. REYNOLDS
Opinion of the Court
“discovery,” when a plaintiff would have begun investigat
ing, we cannot accept Merck’s argument.
As a fallback, Merck argues that even if the limitations
period does generally begin at “discovery,” it should none
theless run from the point of “inquiry notice” in one par
ticular situation, namely, where the actual plaintiff fails to
undertake an investigation once placed on “inquiry no
tice.” In such circumstances, Merck contends, the actual
plaintiff is not diligent, and the law should not “effectively
excuse a plaintiff’s failure to conduct a further investiga
tion” by placing that nondiligent plaintiff and a reasonably
diligent plaintiff “in the same position.” Brief for Petition
ers 48.
We cannot accept this argument for essentially the same
reason we reject “inquiry notice” as the standard gener
ally: We cannot reconcile it with the statute, which simply
provides that “discovery” is the event that triggers the 2
year limitations period—for all plaintiffs. Cf. United
States v. Mack, 295 U. S. 480, 489 (1935) (“Laches within
the term of the statute of limitations is no defense at
law”). Furthermore, the statute does not place all plain
tiffs “in the same position” no matter whether they inves
tigate when investigation is warranted. The limitations
period puts plaintiffs who fail to investigate once on “in
quiry notice” at a disadvantage because it lapses two years
after a reasonably diligent plaintiff would have discovered
the necessary facts. A plaintiff who fails entirely to inves
tigate or delays investigating may well not have discov
ered those facts by that time or, at least, may not have
found sufficient facts by that time to be able to file a §10(b)
complaint that satisfies the applicable heightened plead
ing standards. Cf. Young v. Lepone, 305 F. 3d 1, 9 (CA1
2002) (“[A] reasonably diligent investigation . . . may
consume as little as a few days or as much as a few years
to get to the bottom of the matter”).
Merck further contends that its proposed “inquiry no
Cite as: 559 U. S. ____ (2010) 17
Opinion of the Court
tice” standard is superior, because determining when a
hypothetical reasonably diligent plaintiff would have
“discover[ed]” the necessary facts is too complicated for
judges to undertake. But courts applying the traditional
discovery rule have long had to ask what a reasonably
diligent plaintiff would have known and done in myriad
circumstances. And courts in at least five Circuits already
ask this kind of question in securities fraud cases. See,
e.g., Rothman v. Gregor, 220 F. 3d 81, 97 (CA2 2000); New
England Health Care, 336 F. 3d, at 501; Young, supra, at
1, 9–10; Sterlin v. Biomune Systems, 154 F. 3d 1191, 1201
(CA10 1998); Marks v. CDW Computer Centers, Inc., 122
F. 3d 363, 367–368 (CA7 1997). Merck has not shown this
precedent to be unworkable. We consequently find that
the “discovery” of facts that put a plaintiff on “inquiry
notice” does not automatically begin the running of the
limitations period.
We conclude that the limitations period in §1658(b)(1)
begins to run once the plaintiff did discover or a reasona
bly diligent plaintiff would have “discover[ed] the facts
constituting the violation”—whichever comes first. In
determining the time at which “discovery” of those “facts”
occurred, terms such as “inquiry notice” and “storm warn
ings” may be useful to the extent that they identify a time
when the facts would have prompted a reasonably diligent
plaintiff to begin investigating. But the limitations period
does not begin to run until the plaintiff thereafter discov
ers or a reasonably diligent plaintiff would have discov
ered “the facts constituting the violation,” including sci
enter—irrespective of whether the actual plaintiff
undertook a reasonably diligent investigation.
IV
Finally, Merck argues that, even if all its other legal
arguments fail, the record still shows that, before Novem
18 MERCK & CO. v. REYNOLDS
Opinion of the Court
ber 6, 2001, the plaintiffs had discovered or should have
discovered “the facts constituting the violation.” In re
spect to scienter Merck primarily relies upon (1) the FDA’s
September 2001 warning letter, which said that Merck
had “ ‘minimized’ ” the VIGOR study’s “ ‘potentially serious
cardiovascular findings’ ” and (2) pleadings filed in prod
ucts-liability actions in September and October 2001
alleging that Merck had “ ‘omitted, suppressed, or con
cealed material facts concerning the dangers and risks
associated with Vioxx’ ” and “purposefully downplayed
and/or understated the serious nature of the risks associ
ated with Vioxx.” Brief for Petitioners 36–37 (quoting
App. 340, 893).
The FDA’s warning letter, however, shows little or
nothing about the here-relevant scienter, i.e., whether
Merck advanced the naproxen hypothesis with fraudulent
intent. See Part I–A(4), supra. The FDA itself described
the pro-Vioxx naproxen hypothesis as a “possible explana
tion” for the VIGOR results, faulting Merck only for failing
sufficiently to publicize the alternative less favorable to
Merck, that Vioxx might be harmful. App. 340.
The products-liability complaints’ statements about
Merck’s knowledge show little more. See Part I–A(3),
supra. Merck does not claim that these complaints con
tained any specific information suggesting the fraud al
leged here, i.e., that Merck knew the naproxen hypothesis
was false even as it promoted it. And, without providing
any reason to believe that the plaintiffs had special access
to information about Merck’s state of mind, the complaints
alleged only in general terms that Merck had concealed
information about Vioxx and “purposefully downplayed
and/or understated” the risks associated with Vioxx—the
same charge made in the FDA warning letter. App. 893.
In our view, neither these two circumstances nor any of
the other pre-November 2001 circumstances that we have
set forth in Part I–A, supra, whether viewed separately or
Cite as: 559 U. S. ____ (2010) 19
Opinion of the Court
together, reveal “facts” indicating scienter. Regardless of
which, if any, of the events following November 6, 2001,
constituted “discovery,” we need only conclude that prior
to November 6, 2001, the plaintiffs did not discover, and
Merck has not shown that a reasonably diligent plaintiff
would have discovered, “the facts constituting the viola
tion.” In light of our interpretation of the statute, our
holdings in respect to scienter, and our application of
those holdings to the circumstances of this case, we must,
and we do, reach that conclusion. Thus, the plaintiffs’ suit
is timely. We need not—and do not—pass upon the Court
of Appeals’ suggestion that the November 2003 Brigham
and Women’s study might have triggered the statute of
limitations. The judgment of the Court of Appeals is
Affirmed.
Cite as: 559 U. S. ____ (2010) 1
Opinion of STEVENS, J.
SUPREME COURT OF THE UNITED STATES
_________________
No. 08–905
_________________
MERCK & CO., INC., ET AL., PETITIONERS v. RICHARD
REYNOLDS ET AL.
ON WRIT OF CERTIORARI TO THE UNITED STATES COURT OF
APPEALS FOR THE THIRD CIRCUIT
[April 27, 2010]
JUSTICE STEVENS, concurring in part and concurring in
the judgment.
In my opinion the Court’s explanation of why the com
plaint was timely filed is convincing and correct. Ante, at
12–19. In this case there is no difference between the time
when the plaintiffs actually discovered the factual basis
for their claim and the time when reasonably diligent
plaintiffs should have discovered those facts. For that
reason, much of the discussion in Part II of the Court’s
opinion, see ante, at 8–12, is not necessary to support the
Court’s judgment. Until a case arises in which the differ
ence between an actual discovery rule and a constructive
discovery rule would affect the outcome, I would reserve
decision on the merits of JUSTICE SCALIA’s argument, post,
at 1–7 (opinion concurring in part and concurring in
judgment). With this reservation, I join the Court’s excel
lent opinion.
Cite as: 559 U. S. ____ (2010) 1
Opinion of SCALIA, J.
SUPREME COURT OF THE UNITED STATES
_________________
No. 08–905
_________________
MERCK & CO., INC., ET AL., PETITIONERS v. RICHARD
REYNOLDS ET AL.
ON WRIT OF CERTIORARI TO THE UNITED STATES COURT OF
APPEALS FOR THE THIRD CIRCUIT
[April 27, 2010]
JUSTICE SCALIA, with whom JUSTICE THOMAS joins,
concurring in part and concurring in the judgment.
Private suits under §10(b) of the Securities Exchange
Act of 1934, 15 U. S. C. §78j(b), must be brought within
“(1) 2 years after the discovery of the facts constituting the
violation” or “(2) 5 years after such violation,” whichever
comes first. 28 U. S. C. §1658(b)(1). I agree with the
Court that scienter is among the “facts constituting the
violation” that a plaintiff must “discove[r]” for the limita
tions period to begin. Ante, at 12–14 (internal quotation
marks omitted). I also agree that respondents’ suit is
timely, but for a reason different from the Court’s: Merck
has not shown that respondents actually “discover[ed]”
scienter more than two years before bringing suit.
In ordinary usage, “discovery” occurs when one actually
learns something new. See Webster’s New International
Dictionary of the English Language 745 (2d ed. 1957)
(defining “discovery” as “[f]inding out or ascertaining
something previously unknown or unrecognized”). As the
Court notes, however, ante, at 8–10, in the context of
statutes of limitations “discovery” has long carried an
additional meaning: It also occurs when a plaintiff, exer
cising reasonable diligence, should have discovered the
facts giving rise to his claim. See, e.g., Wood v. Carpenter,
101 U. S. 135, 140–142 (1879); 2 H. Wood, Limitations of
2 MERCK & CO. v. REYNOLDS
Opinion of SCALIA, J.
Actions §276b(11)–(13), pp. 1401–1408 (4th ed. 1916);
Dawson, Undiscovered Fraud and Statutes of Limitation,
31 Mich. L. Rev. 591, 619, and n. 77 (1933). Read in isola
tion, “discovery” in §1658(b)(1) might mean constructive
discovery.
In context, however, I do not believe it can. Section 13
of the Securities Act of 1933, 48 Stat. 84, explicitly estab
lished a constructive-discovery rule for claims under §§11
and 12 of that Act:
“No action shall be maintained to enforce any liability
created under section 77k or 77l(a)(2) of this title
unless brought within one year after the discovery of
the untrue statement or the omission, or after such
discovery should have been made by the exercise of
reasonable diligence . . . .” 15 U. S. C. §77m.
“[D]iscovery” in §77m obviously cannot mean constructive
discovery, since that would render superfluous the phrase
“or after such discovery should have been made by the
exercise of reasonable diligence.” Ibid. With §77m al
ready on the books, Congress added limitations periods in
the 1934 Act, 15 U. S. C. §§78i(e), 78r(c), that did not
contain similar qualifying language; instead, each estab
lished a time bar that runs from “discovery” simpliciter.
When Congress enacted §1658(b)(1) in 2002, establishing
a limitations period for private actions for “fraud, deceit,
manipulation, or contrivance in contravention of a regula
tory requirement concerning the securities laws,” specifi
cally including the 1933 and 1934 Acts, see 15 U. S. C.
§78c(a)(47), it likewise included no constructive-discovery
caveat. To interpret §1658(b)(1) as imposing a construc
tive-discovery standard, one must therefore assume, con
trary to common sense, that the same word means two
very different things in the same statutory context of
limitations periods for securities-fraud actions under the
1933 and 1934 Acts.
Cite as: 559 U. S. ____ (2010) 3
Opinion of SCALIA, J.
True, the sensible presumption that a word means the
same thing when it appears more than once in the same
statutory context—or even in the very same statute—is
rebuttable. See General Dynamics Land Systems, Inc. v.
Cline, 540 U. S. 581, 595–596 (2004). Context may make
clear that in one instance the word carries one meaning,
and in a second instance another. See, e.g., id., at 596–
597. But nothing in the context of §77m or §1658(b)(1)
suggests that is the case. Both provisions impose limita
tions periods for federal-law claims based on various false
statements or omissions involving securities. The former
applies to false statements or omissions in registration
statements, §77k, and offers to sell securities, §77l(a)(2);
the broad language of the latter (“claim[s] of fraud, deceit,
manipulation, or contrivance in contravention of a regula
tory requirement concerning the securities laws”) covers
other “manipulative or deceptive device[s] or contriv
ance[s]” made “in connection with the purchase or sale” of
a security in violation of Securities and Exchange Com
mission regulations, §78j(b), including SEC Rule 10b–5, 17
CFR §240.10b–5(b) (2009). There is good reason, more
over, for providing an actual-discovery rule for private
§10(b) claims but providing (explicitly) a constructive
discovery rule for claims governed by §77m: The elements
of §10(b) claims, which include scienter, are likely more
difficult to discover than the elements of claims under
§77k or §77l(a)(2), which do not, see Herman & MacLean
v. Huddleston, 459 U. S. 375, 382 (1983); Ernst & Ernst v.
Hochfelder, 425 U. S. 185, 208–209 (1976); In re Morgan
Stanley Information Fund Securities Litigation, 592 F. 3d
347, 359 (CA2 2010). And a constructive-discovery stan
dard may be easier to apply to the claims covered by §77m.
Determining when the plaintiff should have uncovered an
untrue assertion in a registration statement or prospectus
is much simpler than assessing when a plaintiff should
have learned that the defendant deliberately misled him
4 MERCK & CO. v. REYNOLDS
Opinion of SCALIA, J.
using a deceptive device covered by §10(b).1
Unable to identify anything in the statutory context that
warrants giving “discovery” two meanings, the Court
relies on the historical treatment of “discovery” in limita
tions periods (particularly for fraud claims) as incorporat
ing a constructive-discovery rule. Ante, at 8–10, 12. But
that history proves only that “discovery” can carry that
technical meaning, and that without §77m it would be
reasonable (other things equal) to read it that way here.
It does not show what “discovery” means in §1658(b)(1) in
light of §77m’s codification of a constructive-discovery
rule. In my view, the meaning of “discovery” in the
broader context of limitations provisions is overcome by its
meaning in the more specific context of the federal securi
ties laws.
The Court’s other reason for rejecting the more natural
reading of §1658(b)(1) rests on a consensus among the
Courts of Appeals before the provision’s enactment. Ante,
at 11–12. In Lampf, Pleva, Lipkind, Prupis & Petigrow v.
Gilbertson, 501 U. S. 350 (1991), the Court notes, we
explicitly adopted the terms of §78i(e)—which like
§1658(b)(1) refers only to discovery with no mention of
reasonable diligence—as the limitations period for the
private §10(b) cause of action we created. Id., at 364, and
——————
1 The Court appears to believe that §77m’s distinction between actual
and constructive discovery has no bearing on §1658(b)(1)’s meaning
because the latter does not itself draw the same distinction. Ante, at
11–12. The point, however, is that both provisions use the same word
(“discovery”) with no contextual clue that it carries different meanings;
and its use in §77m makes clear that the meaning is actual discovery.
The Court suggests that usages of the same word in other statutes
are irrelevant, ante, at 11–12, but of course it does not believe that. Its
entire argument rests on the meaning courts have ascribed to “discov
ery” in other limitations provisions (some enacted decades ago by state
legislatures), ante, at 8–10. Yet while the Court considers that broader
context, it provides no explanation for ignoring the more specific
context of securities-fraud claims under the 1933 and 1934 Acts.
Cite as: 559 U. S. ____ (2010) 5
Opinion of SCALIA, J.
n. 9.2 Since every Circuit to address the issue between
Lampf and §1658(b)(1)’s enactment 11 years later had held
constructive discovery applicable to §10(b) claims—and
since Congress copied §78i(e)’s key text into §1658(b)(1)
with no indication it intended to adopt a contrary rule—the
Court assumes Congress meant to codify (or at least not to
disturb) that consensus. Ante, at 11–12.
Even assuming that Congress intended to incorporate
the Circuits’ views—which requires the further unrealistic
assumption that a majority of each House knew of and
agreed with the Courts of Appeals’ opinions—that would
be entirely irrelevant. Congress’s collective intent (if such
a thing even exists) cannot trump the text it enacts, and in
any event we have no reliable way to ascertain that intent
apart from reading the text. See Graham County Soil and
Water Conservation Dist. v. United States ex rel. Wilson,
559 U. S. ___, ___ (2010) (SCALIA, J., concurring in part
and concurring in judgment) (slip op., at 1).
The only way in which the Circuits’ pre-2002 decisions
might bear on §1658(b)(1)’s meaning is if all (or nearly all)
of the Circuits had interpreted “discovery” in §78i(e) to
mean constructive discovery. If that were true, one could
say that those decisions had established the public mean
ing of the term in this context—whether Congress knew of
(or agreed with) that meaning or not. Jerman v. Carlisle,
McNellie, Rini, Kramer & Ulrich LPA, 559 U. S. ___, ___,
n. 1 (2010) (SCALIA, J., concurring in part and concurring
——————
2 The Court notes that Lampf chose §78i(e)’s limitations period as the
time bar for §10(b) claims, even though it was aware of §77m, 501 U. S.,
at 360, and n. 7, 364, and n. 9; see ante, at 10–11. But I fail to see how
that provides any support for the Court’s interpretation. To the con
trary, the fact that in enacting §1658(b)(1) Congress did not copy
§77m’s constructive-discovery proviso—but decreed instead that “dis
covery” alone starts the clock (as it had done in §78i(e), which we
borrowed in Lampf)—is what makes equating §77m and §1658(b)(1) so
implausible.
6 MERCK & CO. v. REYNOLDS
Opinion of SCALIA, J.
in judgment) (slip op., at 2, n. 1).
But as amici note, that is not so. See Brief for Faculty
at Law and Business Schools as Amici Curiae 23–29 (here
inafter Faculty Brief). Some circuit cases cited by the
Court and amici can conceivably be read as interpreting
the language Lampf adopted from §78i(e) as imposing
some form of constructive discovery. See Theoharous v.
Fong, 256 F. 3d 1219, 1228 (CA11 2001); Menowitz v.
Brown, 991 F. 2d 36, 41 (CA2 1993) (per curiam); Howard
v. Haddad, 962 F. 2d 328, 329–330 (CA4 1992); Anixter v.
Home-Stake Production Co., 947 F. 2d 897, 898–899 (CA10
1991), vacated on other grounds, 503 U. S. 978 (1992).
Others, however, cannot be so construed. Two were not
interpreting §78i(e) at all, but looked directly to §77m,
despite Lampf’s explicit selection of §78i(e)’s terms. Great
Rivers Cooperative of Southeastern Iowa v. Farmland
Industries, Inc., 120 F. 3d 893, 896 (CA8 1997); Topalian
v. Ehrman, 954 F. 2d 1125, 1135 (CA5 1992). Another
court candidly acknowledged that §78i(e)’s text—unlike
§77m’s—forecloses constructive discovery, but it nonethe
less held that courts remain “free to apply to [§78i(e)] the
judge-made doctrine of inquiry notice” as a “modest and
traditional . . . exercise of judicial creativity,” since “Con
gress could not have known when it enacted [§78i(e)] that
this section would someday provide the statute of limita
tions for a wide range of securities frauds.” Tregenza v.
Great American Communications Co., 12 F. 3d 717, 721–
722 (CA7 1993) (Posner, J.).
The rest of the Circuits apparently had not decided the
issue before §1658(b)(1)’s enactment. See Betz v. Trainer
Wortham & Co., 519 F. 3d 863, 874 (CA9 2008); New
England Health Care Employees Pension Fund v. Ernst &
Young, LLP, 336 F. 3d 495, 500–501, and n. 3 (CA6 2003);
In re NAHC, Inc. Securities Litigation, 306 F. 3d 1314,
1325 (CA3 2002); see also Cooperativa de Ahorro y Credito
Aguada v. Kidder, Peabody & Co., 129 F. 3d 222, 224 (CA1
Cite as: 559 U. S. ____ (2010) 7
Opinion of SCALIA, J.
1997) (applying pre-Lampf rule under 15 U. S. C. §78aa–
1). And of those that were undecided, two had cast doubt
on a constructive-discovery view in dicta—of which the
omniscient Congress of the Court’s imagining should also
have been aware. See Berry v. Valence Technology, Inc.,
175 F. 3d 699, 703–705 (CA9 1999); Gruber v. Price Water
house, 911 F. 2d 960, 964, n. 4 (CA3 1990).
This motley assortment of approaches comes nowhere
near establishing that the word “discovery” in §78i(e)
meant constructive rather than actual discovery despite
§77m. Absent any textual or contextual reason to read
“discovery” differently in §1658(b)(1) and §77m, I would
hold that only actual discovery suffices to start the limita
tions period for §10(b) claims. Since Merck points to no
evidence showing respondents actually discovered scienter
more than two years before bringing this suit, I agree with
the Court that the suit was not time barred.
Respondents suggested at oral argument, Tr. of Oral
Arg. 29, and their amici imply, see Faculty Brief 33–34,
that in fraud-on-the-market cases there is little if any
difference between actual and constructive discovery
because of the presumption of reliance applicable in such
cases, see Basic Inc. v. Levinson, 485 U. S. 224, 247 (1988).
It seems to me Basic has no bearing on the question dis
cussed here. A presumption of reliance upon market-price
signals is not a presumption of knowledge of all public
information, much less knowledge of nonpublic informa
tion that a reasonably diligent investor would have inde
pendently uncovered. In any event, whether or not a
constructive-discovery standard will in many cases yield
the same result, actual discovery is what §1658(b)(1)
requires to start the limitations period.