Arnold, Arlene, undated - 1 of 1

    No. 05-3349-cv
3in tlje

■Huitefi sta tes Court of Appeals 
for tlje ^ecouft Circuit

IN RE: INITIAL PUBLIC OFFERING SECURITIES LITIGATION
John G. Miles, Saswata Basu, Michael Huff, Sean Rooney, Krikor Kasbarian, Stathis Pappas, 
James Collins, Diane Collins, Joseph Zhen, Zitto Investments, J. Chris Rowe, Vasanthakumar 
Gangaiah, Frederick Henderson, Barry Lemberg, Anita Budich, Spiros Gianos, Mary Jane Gianos 
AND HARALD ZAGODA,

Plaintiffs-Appellees,
v.

Merrill Lynch & Co., Inc., Goldman, Sachs & Co., Merrill Lynch, Pierce, Fenner & Smith 
Incorporated, Credit Suisse First Boston LLC, Robertson Stephens, Inc., Morgan Stanley & Co., 
Incorporated, Bear Stearns & Co., Inc., The Bear Stearns Companies, Inc., J.P. Morgan Securities 
Inc., Deutsche Bank Securities, Inc., (f/k/a Deutsche Banc Alex. Brown, Inc., DB Alex. Brown 
LLC and BT Alex. Brown Incorporated), Lehman Brothers, Inc., SG Cowen Securities Corp., 
(n/k/a SG Cowen & Co., LLC), RBC Dain Rauscher, Inc. (f/k/a/D ain Rauscher, Inc.) and Prudential 
Securities, Incorporated,

Defendants-Appellants.

Appeal from the United States D istrict Court 
for the Southern D istrict o f  N ew  Y ork

Memorandum of the

Naacp Legal Defense and Educational Fund, Inc.,
Amicus Curiae,

In Support Of Rehearing and Rehearing En  Banc

Theodore M. Shaw 
Director-Counsel 

Jacqueline A. Berrien 
Norman J. Chachkin 
Robert H. Stroup 
NAACP Legal Defense and 

Educational Fund, Inc.
99 Hudson Street, Suite 1600 
New York, NY 10013-2897 
(212) 965-2200

Attorneys for  Amicus Curiae



CORPORATE DISCLOSURE STATEMENT

Pursuant to Rule 26.1 o f  the Federal Rules o f  A ppellate Procedure, amicus curiae 

states that the N A A CP Legal D efense and Educational Fund, Inc. is a non-profit corporation 

exem pt from  taxation pursuant to Section 501(c)(3) o f  the Internal Revenue Code and is not

a publicly held corporation that issues stock.



The N A A CP Legal Defense and Educational Fund, Inc. (LDF), w hich has participated 

in this appeal as an amicus curiae w ith the consent o f  the parties, submits this M em orandum  

in support o f  the grant o f  rehearing requested by the Plaintiffs-A ppellees, and urges the Court 

to reconsider and clarify the panel opinion issued Decem ber 5, 2006.

We filed our initial amicus b rie f

because o f  [our] concern that the legal standard for certification o f a Rule 
23(b)(3) class action urged by  D efendants-A ppellants in this m atter is so 
sweeping, and so broadly formulated, tha t— if  adopted by this C ourt— it would 
have dram atic im pact not only in securities cases, but as w ell in the broad range 
o f civil rights class actions brought pursuant to Rule 23(b)(2) to obtain injunctive 
and other re lie f to end racial and sim ilar invidious discrim ination.

LDF Amicus Br. at 3 [footnote om itted]. The panel opinion issued on D ecem ber 5, 2006 

unfortunately confirm ed our concern— and in fact is potentially m ore dam aging than we had 

anticipated, because o f its lack o f clear standards to guide trial courts in the certification process, 

and because its abandonm ent o f  interpretations o f Rule 23 previously enunciated by this Court 

in Caridad1 and Visa Check2 is unnecessary, unwise, and erroneous. Sweeping in its terms, 

it will, unless m odified, im pair effective realization o f the deterrent and rem edial purposes 

o f  federal civil rights legislation in general —  and o f  the “central statutory purpose o f [Title 

VII o f  the 1964 Civil Rights Act, of] eradicating discrim ination throughout the econom y and 

m aking persons w hole for injuries suffered through past discrim ination,” Albemarle Paper 

Co. v. Moody, 422 U.S. 405,421 (1975), in particular. W e elaborate briefly  on these points 1 2

1 Caridad v. Metro-North Commuter R.R., 191 F.3d 283 (2d Cir. 1999).

2 In re Visa Check Mastermoney Antitrust Lit., 280 F.3d 124 (2d Cir. 2001).

1



in the balance o f  this docum ent and urge the Court, for the reasons set forth herein as well 

as those urged by P laintiffs-A ppellees,3 to grant rehearing.

A.

The panel opinion rewrites this C ircu it’s interpretation o f  the Rule 23 prerequisites 

for class certification —  both those applicable to all putative class suits (Rule 23(a)), as well 

as those specific to suits brought under Rule 23(b)(3). The opinion reaches these issues in 

spite o f  the facts that (i) this is a Rule 23(b)(3) case, and (ii) the p an el’s disposition rests upon 

its holding, as a m atter o f  law, that the “predom inance” requirem ent o f  Rule 23(b)(3) can never 

be m et in an IPO securities fraud case alleging m arket m anipulation. See infra § B. In the 

process, the panel m isinterprets and abandons prior rulings o f  the Court in both Rule 23(b)(2)4 

and Rule 23(b)(3)5 cases that construed the very different requirem ents o f  Rule 23(a) —  

especially subsections (a)(2) and (a)(3) (usually referred to as “com m onality” and “typicality”).

The panel opinion does so not directly, but through “observations” that: “to w hatever 

extent Caridad m ight have implied [a “some showing”] standard for a Rule 23 requirement, 

that implication is disavow ed,” and “also d isav o w ed  is] the suggestion in Visa Check that 

an expert’s testim ony may establish a com ponent o f  a Rule 23 requirem ent sim ply by not being

3 We agree with, and will not duplicate, the arguments made by Plaintiffs-Appellees 
concerning Rule 23 issues in their rehearing petition. We pretermit comment on the panel’s 
determination of substantive securities law principles that, Plaintiffs-Appellees also argue, demand 
consideration and revision because LDF’s expertise lies outside that area. This Memorandum 
focuses on the implications of the panel’s opinion for civil rights cases brought as class actions under 
Rule 23(b)(2).

4 E.g., Caridad.

5 E.g., Visa Check.

2



fatally flaw ed.” (S lipop. a t7892 .6) (em phasis supplied.) This m odest description understates

the m agnitude o f  the panel opinion’s m odification o f  prior precedent, w hich warrants

consideration by the full Court and/or correction by the panel.

First, the “some showing” language o f Caridad was quite deliberately chosen, and when

read in context, emphasizes the necessity for plaintiffs to make an evidentiary showing to justify

class certification. D efendants-A ppellants and others have consistently read the phrase with

an em phasis on the w ord “som e” (as i f  the opinion actually said “a m inim al”). We submit

that, considered in context, it is m ore sensibly interpreted w ith an em phasis on the word

“show ing,” as the following excerpt from  the Caridad opinion demonstrates.

Though proving that the grant o f  discretionary authority to supervisory employees 
either results in a pattern and practice o f  discrim ination or affects one class o f  
em ployees m ore harshly than others is likely to be extrem ely difficult [citation 
and parenthetical omitted], the fact that the C lass Plaintiffs challenge the 
subjective com ponents o f  com pany-wide em ploym ent practices does not bar 
a finding o f  commonality under either the disparate treatm ent or disparate impact 
m odel. . . . [discussion o f  language in Gen. Tel. Co. v. Falcon, 357 U.S. 147 
(1982) and other cases suggesting or recognizing that challenges to subjective 
decisionm aking could be brought as class actions omitted].

O f course, class certification would not be warranted absent some showing 
that the challenged practice is causally related to a pattern o f  disparate treatm ent 
or has a disparate im pact on A frican-A m erican em ployees at M etro-North.
W here the decision-m aking process is difficult to review  because o f  the role 
o f  subjective assessm ent, significant statistical disparities are relevant to 
determ ining w hether the challenged employment practice has a class-wide 
impact, [citations and parentheticals omitted.] Regardless o f  their ultim ate 
persuasiveness on the issue o f  liability, the statistical report and anecdotal 
evidence submitted by the Class Plaintiffs are sufficient to demonstrate common

6 Citations are to the printed slip opinion.

3



questions o f  fact regarding the discrim inatory im plem entation and effects o f 
M etro-N orth’s com pany-wide policies regarding prom otion and discipline.

The D istrict Court relied on the report o f  M etro-N orth’s statistical expert,
Dr. Evans, to conclude that the Class Plaintiffs ’ statistics were inadequate because 
they failed to take into account the fact that various M etro-North positions have 
m aterially different rates o f  discipline and prom otion. Though M etro-N orth’s 
critique o f  the Class P lain tiffs’ evidence m ay prove fatal at the m erits stage, 
the Class Plaintiffs need not dem onstrate at this stage that they will prevail on 
the merits. A ccordingly, this sort o f  “statistical dueling” is not relevant to the 
certification determ ination, [citation omitted.] W e conclude that the Class 
Plaintiffs ’ statistical evidence supports a finding o f  commonality . . . .

Caridad, 191 F .3 d a t2 9 1 -9 3  (em phasis supplied). In Caridad this Court reversed the denial

o f  class certification only after it reviewed the p lain tiffs’ evidentiary showing (statistical

evidence and an expert’s report) and held that the evidence “supports afinding o f  commonality,”

191 F.3d at 293 (em phasis supplied). Accord, Denney v. Deutsche Bank AG, 443 F.3d 253,

270 (2d Cir. 2006), a ff’g  on this point Denney v. Jenkens & Gilchrist, 230 F.R.D. 317, 330

(S.D.N. Y. 2005), w here the district court found the com m onality and typicality requirements

o f Rule 23(a) satisfied based on facts that were either uncontested or supported by plaintiffs’

subm issions.7

Second, the “suggestion” in Visa Check was in fact the ratio decidendum for upholding

the district court’s finding that the requirem ents o f  Rule 23(a) had been met:

A lthough a trial court m ust conduct a ‘rigorous analysis’ to ensure that the 
prerequisites o f  Rule 23 have been satisfied before certifying a class, ‘a m otion 
for class certification is not an occasion for exam ination o f the m erits o f  the

7 Even if the Court were to conclude that the Caridad “some showing” language is too 
imprecise, or insufficiently rigorous, the approach of the panel goes much too far in the other 
direction. See infra § C.

4



case’ [citing Caridad]. A district court m ust ensure that the basis o f  the expert 
opinion is not so flawed that it w ould be inadm issible as a m atter o f  law. . . .
[citing, inter alia, cases granting class certification upon finding experts’ reports 
dem onstrated ‘“ reasonable probability o f  establishing’ p la in tiffs’ claims by 
common proof,” or that “Plaintiffs have dem onstrated at least a ‘colorable 
m ethod’ o f  proving [common injury] at trial.”] However, a district court may 
not w eigh conflicting expert evidence or engage in “statistical dueling” o f 
experts. Caridad, 191 F.3d at 292-93. The question for the district court at the 
class certification stage is w hether p la in tiffs’ expert evidence is sufficient to 
demonstrate common questions o f  fact warranting certification o f  the proposed 
class, not w hether the evidence will ultim ately be persuasive.

To the extent that defendants’ contention is that the court did not 
sufficiently exam ine w hether C arlton’s m ethodology was fatally flawed, and 
thus inadm issible even for class certification purposes, we reject this argum ent 
as m e ritle s s .. . . [discussion o f district court’s review  o f  m ethodology claims 
omitted.] The district court correctly noted that its function at the class 
certification stage was not to determ ine w hether plaintiffs had stated a cause 
o f  action or w hether they w ould prevail on the m erits, but rather w hether they 
had show n, based on m ethodology that was not fatally flawed, that the 
requirem ents o f  Rule 23 were met. . . . [citation omitted.] As fo r  defendant’s 
claim that plaintiffs ’ expert evidence failed to provide a reliable basis fo r  class 
certification, the district court’s finding that Carlton’s methodology was not 
fatally flawed, and therefore, was sufficiently reliable fo r  class certification 
purposes, does not constitute an abuse o f its discretion.

Visa Check, 280 F.3d at 135 (em phasis supplied).

B.

The panel finds it unnecessary to rem and the case “to the D istrict Court for 

reconsideration o f  the class certification m otion under the proper [Rule 23(a)] standards as 

we have explained them ” because “Plaintiffs ’ own allegations and evidence dem onstrate that 

the Rule 23 [(b)(3)] requirement o f  predominance o f common questions over individual questions 

cannot be m et under the standards as we have explicated them ” (Slip op. at 7892). The panel 

holds —  as a m atter o f  law —  that “the m arket for IPO shares is not efficient” (id. at 7893), 

precluding plaintiffs from showing reliance except through individual-by-individual p roof

5



(id. at 7892-93).8 Thus, no m atter how  the Rule 23(a) evidentiary standards w ere phrased, 

the disposition o f  this case w ould be unaffected. N either the panel opinion’s treatm ent o f  the 

predom inance requirem ent nor its judgm ent in the m atter, therefore, illum inates in any 

m eaningful w ay the application o f  the standards that it announces (in dicta) apply to the 

requirem ents o f  Rule 23(a).

It is true that the panel opinion does briefly m ention the “num erosity” requirem ent o f  

Rule 23(a)( 1 ),9 but, with respect, that requirem ent is rarely controversial or dispositive in class 

suits. The panel opinion does not deal w ith the Rule 23(a) requirem ents o f  com m onality and 

typicality, w hich are far m ore often central to the certification question in discrim ination suits 

under Rule 23(b)(2). Lacking any discussion o f  those issues, the panel opinion ( if  not altered 

by the panel or the Court en banc) is likely to stimulate m ore frequent appeals seeking further 

guidance. It will also m arkedly reduce Rule 2 3 ’s beneficial effects, increase the cost o f  class 

certification proceedings and dissipate the constructive im pact o f  civil rights class actions, 

see LDF Amicus Br. at 6-8 & nn. 5, 6; infra § C.

This problem  is exacerbated by the broad authority that the panel opinion gives trial 

courts to curtail a class p la in tiffs  discovery and evidentiary presentation opportunities (see 

Slip op. at 7891) and thereafter to deny class certification on the grounds that one o f  the Rule

8 The panel announces similar conclusions with regard to other elements o f a securities fraud 
claim in this case alleging market manipulation. See Slip op. at 7894-96 (proposed class members’ 
lack of knowledge of scheme); id. at 7896-98 (determination of what portion o f secondary market 
share prices were attributable to various alleged “overcompensation” arrangements).

9The panel opinion states that

in considering whether the numerosity requirement is met, a judge might need to 
resolve a factual dispute as to how many members are in a proposed class. Any 
dispute about the size of the proposed class must be resolved, and a finding of the 
size o f the class, e.g., 50, 100, or more than 200, must be made. At that point, the 
judge would apply the legal standard governing numerosity and make a ruling as to 
whether that standard, applied to the facts as found, establishes numerosity.

(Slip op. at 7889 [footnote omitted].)

6



23 prerequisites has not been sufficiently “established” (id. at 7888-89,7891).10 These rulings 

are then review ed under a deferential “abuse o f discretion” standard, see Slip op. at 7890. 

There are no lim iting principles adum brated in the panel opinion. Especially in (b)(2) class 

actions— the m ost com m on basis for seeking class certification in civil rights cases, in which 

“com m onality” and “typicality” questions often overlap substantially w ith the merits —  

therefore, the effect o f  the panel opinion ( if  not reconsidered and altered) could be disastrous 

even in meritorious cases. For this reason, independent o f  the correctness o f  the panel’s holdings 

as to the law o f  securities fraud, the decision deserves panel and/or en banc review.

C.

W e turn to the problem s that the panel opinion is likely to create in civil rights 

discrim ination law suits sought to be m aintained as class actions. The opinion summarizes 

the governing Rule 23 law that it announces as follows:

(1) a district judge m ay certify a class only after m aking determinations that each 
o f the Rule 23 requirem ents [“e.g., numerosity, commonality, or predom inance”
(slip op. at 7888)] has been met;

(2) such determ inations can be m ade only if  the judge resolves factual disputes 
relevant to each Rule 23 requirem ent and finds that w hatever underlying facts 
are relevant to a particular Rule 23 requirem ent have been established and is 
persuaded to rule, based on the relevant facts and the applicable legal standard, 
that the requirem ent is met;

10It is o f no comfort to class plaintiffs that “[a] trial judge’s finding on a merits issue for 
purposes o f a Rule 23 requirement no more binds the court to rule for the plaintiff on the ultimate 
merits of that issue than does a finding that the plaintiff has shown a probability o f success for 
purposes o f a preliminary injunction” (Slip op. at 7886 [emphasis in original]; see also id. at 7890). 
In the reverse situation, that is, where a court limits discovery and then finds that class plaintiffs 
failed to meet their burden of demonstrating commonality or typicality, in most cases there will be 
no “second bite at the apple.” For example, in employment discrimination suits few individual 
plaintiffs can afford to shoulder the costs of discovery against large employers, even though such 
discovery may reveal patterns o f action that support the individual claim, see, e.g., Duke v. Univ. o f 
Texas at El Paso, 729 F.2d 994 (5th Cir. 1984) (cited in LDF Amicus Br. at 10-11). And it is as tme 
today as it was 30 years ago that attorneys also cannot afford to bear such costs and remain in 
practice; thus, “[e]conomic reality dictates that [an individual plaintiff s discrimination] suit proceed 
as a class action or not at all,” Eisen v. Carlisle & Jacquelin, 417 U.S. 156, 161 (1974).

7



(3) the obligation to make such determinations is not lessened by overlap between 
a Rule 23 requirem ent and a merits issue, even a m erits issue that is identical 
w ith a Rule 23 requirement;

(4) in m aking such determ inations, a district judge should not assess any aspect 
o f  the m erits unrelated to a Rule 23 requirem ent; and

(5) a district judge has ample discretion to circum scribe both  the extent o f 
discovery concerning Rule 23 requirem ents and the extent o f  a hearing to 
determ ine w hether such requirem ents are m et in order to assure that a class 
certification m otion does not becom e a pretext for a partial trial o f  the merits.

(Slip op. at 7891 [spacing added].) How w ould such a regim e w ork in a civil rights

discrim ination suit? It is instructive to use the Caridad litigation as an example.

In Caridad, “the members o f  the proposed class challenge[d] the subjective components

o f  com pany-w ide policies and procedures governing employee discipline and prom otion” as

racially discrim inatory, 191 F.3d at 286, on both disparate treatm ent (pattern and practice)

and disparate im pact grounds, id. at 291. This Court recognized that “ [i]n m any [such] cases,

the only available avenue o f p ro o f [on the merits] is the use o f  racial statistics,” id., citing Int 7

Bhd. o f Teamsters v. U.S., 431 U.S. 324 ,340  n.20 (1977), and in the subsequent appeal from

the district cou rt’s decision on rem and in Caridad, expanded on this discussion, Robinson

v. Metro-North Commuter R. Co., 267 F.3d 147, 158-62 (2d Cir. 2001). A t the same time,

the Caridad court also recognized that at the certification stage, “significant statistical disparities

are relevant to determ ining w hether the challenged em ploym ent practice has a class-wide

im pact,” 191 F.3d at 292.

8



The expert reports subm itted in Caridad created a dispute about that very question. 

The p lain tiffs’ expert conducted “num erous m ultiple regression analyses” and “concluded 

that the ‘ effect o f  being black on num ber o f disciplinary charges over the 1990-to -1994 period 

was positive and highly significant statistically.’ . . . The results o f  her regression analyses 

w ith respect to prom otion were similar; being Black reduced an em ployee’s likelihood o f 

prom otion by approxim ately 33 percent.” 191 F .3 da t288 . The defendant’s expert criticized 

the m ethodology o f  the p la in tiffs’ expert and opined that “ [a]s a result o f  [M etro-N orth’s] 

decentralization [o f “disciplinary and prom otion processes”] an organization-w ide pattern 

and practice o f  discrim ination is . . . im plausible.” Id. P la in tif fs  expert prepared a rebuttal 

report responding to the methodological criticisms, reporting on the results o f  further regression 

analyses conducted in light o f  the criticisms, and indicating that they confirm ed her initial 

view s. Id. at 289.

Because the Caridad court adhered to the understanding that the class certification 

decision did not require resolution o f this ‘“ statistical dueling,” ’ id. at 292, it considered whether 

the p lain tiffs’ evidence was sufficiently probative o f  “class-w ide im pact” o f  the subjective 

decision-m aking system  to dem onstrate that “there are questions o f  law or fact common to 

the class,’’Fed. R. Civ. P. 23(a)(2)11 and concluded that it was, with respect to both the discipline

"The Court in Visa Check engaged in a similar analysis in sustaining “the district court’s 
finding that [the p laintiffs expert’s] methodology was not fatally flawed, and therefore, was 
sufficiently reliable for class certification purposes,” 280 F.3d at 135. See supra pp. 4-5. Cf. 
Bazemore v. Friday, 478 U.S. 385, 400 n.10 (1986) (opinion of Brennan, J., joined by all members 
of the Court, concurring in part) (“There may of course, be some regressions so incomplete as to be 
inadmissible as irrelevant; . . .”).

9



and prom otion claim s. Id. at 292-93. But the determ ination that a case involves factual and

legal questions that necessarily arise in considering the claims o f  the m em bers o f  the putative

class (questions w hose proper resolution is a m atter o f  dispute betw een the p lain tiff and

defendant parties) is very different from  resolving those or subsidiary disputes at the class

certification stage.12 SeeDukesv. Wal-Mart Stores, Inc., 222 F.R.D. 1 3 7 ,164(N .D . Cal. 2004),

appeal pending , N os. 04-16688 & -16720 (9th Cir.):

Defendant’s arguments, which go to the weight o f  the evidence, m erely highlight 
the presence o f  a significant issue affecting all class m em bers w hich supports, 
rather than defeats, granting class certification. M oreover, w hile a ju ry  may 
or m ay no t ultim ately find Dr. D rogin’s approach m ore probative than Dr. 
H aw orth’s, it is at least reasonable. As such, Dr. D rogin’s m ethod is sufficient 
to create an inference o f  discrim ination for the purposes o f  this m otion [for class 
certification].

The language o f  Rule 23(a)(2) does not require a court to choose betw een competing 

expert views o f  the m erits, and in fact the panel opinion’s dicta to the contrary are inconsistent 

w ith the text o f  that subsection o f the Rule (“there are questions . . .  [that are] co m m o n . . . ”).

12 As the description in Caridad of the expert reports indicates, a myriad o f subsidiary factual 
issues are nearly inseparable from cases involving statistical presentations —  such as matters of 
appropriate sample size, variables to be included in regressions, aggregation o f groups o f data for 
analysis and display, etc. The Supreme Court said with respect to one such question, “Normally, 
failure to include variables will affect the analysis’ probativeness, not its admissibility.. . .  Whether, 
in fact, such a regression analysis does carry the plaintiffs’ ultimate burden will depend in a given 
case on the factual context o f each case in light of all the evidence presented by both the plaintiff and 
the defendant.” Bazemore, supra n .l 1, at 400. If such matters go to weight, not consideration vel 
non, and are to be evaluated in the context of the entire body o f evidence on the merits, it is difficult 
to imagine that they are to be resolved in isolation, or on a limited record as the panel opinion 
contemplates, at the certification stage. While the Court in Bazemore remanded for findings as to 
each of the requirements o f Rule 23(a), it made no such suggestion. “Of course, that this case may 
be one in which it is proper to certify a class, is distinct from the question whether county variations 
serve as a basis for the demonstrated disparities between the salaries of black and white employees. 
Bazemore, 478 U.S. at 406 n . l7.

10



That is w hy in Caridad, the ultim ate determination w hether the p lain tiffs’ or defendant’s 

expert’s view s better captured the reality o f  M etro-N orth’s em ploym ent practices was left to 

the fact-finder at the merits stage. The Court allow ed the suit to proceed as a class action 

because “in such cases [where the issues ‘“ turn on questions o f  law applicable in the same 

m anner to each m em ber o f  the c lass’”] ‘“ the class-action device saves the resources o f  both 

the courts and the parties by perm itting an issue potentially affecting every [class member] 

to be litigated in an econom ical fashion under Rule 2 3 .’” Gen. Tel. Co. v. Falcon, 457 U.S. 

at 155, quoting Califano v. Yamasaki, 442 U.S. 682, 700-01 (1979).

Under the panel decision in this case, however, the experts’ differing views about whether 

data from M etro-N orth’s records13 demonstrated a pattern o f  discipline and promotion decisions 

produced by discrim ination —  the ultim ate question on the m erits —  m ust be resolved at the 

class certification stage. And, in light o f  the im portance o f class certification (see the Supreme 

Court’s language in Eisen quoted supra p. 7 n. 10), the approach o f the panel opinion will compel 

plaintiffs to request the broadest possible discovery so that their experts m ay prepare the most 

com prehensive statistical analyses —  with the result o f  m aking illusory the panel opinion’s 

hope o f  “av o id in g ] the risk that a Rule 23 hearing will extend into a protracted mini-trial o f 

substantial portions o f  the underlying litigation” (Slip op. at 3 5 ).14 The panel’s proposed

13 As we said in our amicus brief, “evidence relevant to an employer’s policies and practices 
with respect to such matters as hiring, assignment, promotion, workplace conditions and termination 
lies almost completely within records that are in the custody and control o f the employer.” LDF 
Amicus Br. at 10.

14 Because comprehensive discovery is necessary in order to meet the panel opinion’s 
certification standards, and given the imbalance in access to relevant information, see supra n.13,

11



solution: restriction o f  the scope o f  discovery and evidentiary presentation at the class

certification stage, is one-sided, fundamentally unfair, and could well raise dueprocess concerns;

for as the Suprem e Court said in Eisen, 417 U.S. at 178,

A dditionally, we m ight note that a prelim inary determ ination o f the m erits m ay 
result in substantial prejudice to a [party], since o f  necessity it is not accom panied 
by the traditional rules and procedures applicable to civil trials. The court’s 
tentative findings, m ade in the absence o f  established safeguards, m ay color 
the subsequent proceedings and place an unfair burden on the [party].15

These problem s are real, not imaginary. The panel opinion did not deal w ith them

because it resolved the case at hand on other grounds. But its opinion, i f  left unreview ed and

undisturbed  by the panel or by the Court en banc, is likely ultim ately to frustrate the well-

established national policy o f  rem edying and eradicating racial and other invidious

discrim ination. Further review  is imperative.

Conclusion

For the foregoing reasons, as well as those advanced by P laintiffs-A ppellees in the 

Petition for Rehearing and Rehearing En Banc, the panel and/or the full Court should review  

and reconsider the approach o f  the panel opinion in this matter.

it is highly unlikely that class plaintiffs’ discovery requests would be used as a “pretext for a partial 
trial o f the merits,” as the panel opinion hypothesizes (Slip op. at 7891).

15As we have noted previously, the fact that the resolution o f a factual dispute at the class 
certification stage is not binding at the merits stage is not a practical answer to the concerns raised 
in Eisen. See supra p. 7 n. 10.

12



Respectfully subm itted,

Theodore M. Shaw 
D irector-Counsel 

Jacqueline A. Berrien 
Norman J. Chachkin 
Robert H. Stroup 
NA A CP Legal D efense and 

Educational Fund, Inc.
99 Hudson Street, Suite 1600 
New York, N Y  10013-2897 
(212) 965-2200

Attorneys fo r  Am icus Curiae

13



Certificate o f  Service

I hereby certify  that on this 5th day o f  January, 2 0 0 7 ,1 served a copy o f  the foregoing 

M em orandum  o f  the N A A CP Legal D efense and Educational Fund, Inc., Amicus Curiae, In 

Support o f  R ehearing and Rehearing En Banc upon the follow ing counsel for each party to 

this appeal separately represented, by depositing the same in the U nited States m ail, first-class 

postage prepaid, addressed as follows:

M elvin I. W eiss, Esq.
M ilberg W eiss B ershad & Schulm an LLP 
One Pennsylvania Plaza 
N ew  York, N Y  10119 
{Liaison Counsel fo r  Plaintiffs- 
Appellees)

Jack C. A uspitz, Esq.
M orrison & Foerster LLP 
1290 Avenue o f  the Americas 
N ew  York, N Y  10104 
{Liaison Counsel fo r  Issuer 
Defendants)

M ark Holland, Esq.
C lifford Chance US LLP 
31 W est 52nd Street 
N ew  York, N Y  10019 
{Attorneys fo r  Merrill Lynch & Co., 
Inc. and Merrill Lynch, Pierce, Fenner 
& Smith Incorporated in iXL 
Enterprises, Inc., Sycamore Networks, 
Inc., and Corvis Corp.)

Stanley D. Bernstein, Esq.
B ernstein L iebhard & Lifshitz, LLP 
10 East 40th Street 
N ew  York, N Y  10016 
{Liaison Counsel fo r Plaintiffs- 
Appellees)

Gandolfo V. D iBlasi, Esq.
Sullivan & Crom well LLP
125 Broad Street
N ew  York, N Y  10004
{Attorneys fo r  Goldman, Sachs & Co.
in Engage Technologies, Inc. and
Sycamore N etw orks, Inc. and Liaison
Counsel fo r  Underwriter Defendants)

Randy M astro, Esq.
Gibson, Dunn & Crutcher LLP 
200 Park Avenue 
N ew  York, N Y  10166 
{Attorneys fo r  Bear, Stearns & Co., 
Inc. And The Bear Stearns Companies, 
Inc. in Engage Technologies, Inc. and 
iXL Enterprises, Inc.)

i



Robert B. M cCaw , Esq.
W ilm er C utler Pickering Hale and 

D orr LLP 
399 Park A venue 
N ew  York, N Y  10022 
{Attorneys fo r  Credit Suisse First Boston 
LLC in Corvis Corp., V A  Linux Systems, 
Inc., Engage Technologies, Inc., and 
Sycamore N etw orks, Inc.)

Andrew B. C lubok, Esq.
K irkland & Ellis LLP 
655 Fifteenth Street, N .W .,
Suite 1200
W ashington, DC 20005 
{Attorneys fo r  Morgan Stanley & Co., 
Incorporated in Sycam ore N etw orks,
Inc. and iXL Enterprises, Inc.)

Joseph M. M cLaughlin, Esq.
Sim pson Thacher & Bartlett LLP 
425 Lexington A venue 
N ew  York, N Y  10017 
{Attorneys fo r  J.P. Morgan Securities 
Inc., individually in Sycamore N etworks, 
Inc. and as successor in interest to 
Hambrecht & Quist LLC in Engage 
Technologies, Inc.)

A Robert P ietrzak, Esq.
Sidley A ustin B row n & W ood LLP 
787 Seventh A venue 
N ew  York, N Y  10019 
{Attorneys fo r  Deutsche Bank Securities 
Inc. (f/k/a Deutsche Banc Alex. Brown, 
Inc., DB Alex. Brown LLC and B T  Alex. 
Brown Incorporated) in Engage 
Technologies, Inc.)

Stephen M. Shapiro, Esq.
M ayer, Brown, Rowe & M aw LLP
71 South W acker Drive
Chicago, IL 60606
{Attorneys fo r  Merrill Lynch & Co.,
Inc. and Merrill Lynch, Pierce, Fenner
6  Smith Incorporated in iXL 
Enterprises, Inc., Sycamore Networks, 
Inc.)

Andrew J. Frackm an, Esq.
O ’M elveny & M yers LLP 
Times Square Tower
7 Times Square 
New York, N Y  10036 
{Attorneys fo r  Robertson Stephens,
Inc., in F irepond, Inc., Corvis Corp., 
iXL Enterprises, Inc., and Sycamore 
N etw orks, Inc.)

M oses Silverm an, Esq.
Paul, W eiss, R ifkind, W harton & 

Garrison LLP
1285 A venue o f  the A m ericas 
N ew  York, N Y  10019 
{Attorneys fo r  Lehman Brothers Inc. 
in Corvis Corp. and Sycamore 
N etw orks, Inc.)

Jay B. Kasner, Esq.
Skadden, Arps, Slate, M eagher &

Flom LLP 
Four Times Square 
N ew  York, N Y  10036 
{Attorneys fo r  SG Cowen Securities 
Corp. (n/k/a SG Cowen & Co., LLC) in 
Firepond, Inc.)

n



Stewart D. A aron, Esq.
A rnold & Porter LLP
399 Park Avenue
N ew  York, N Y  10022
(Attorneys fo r  RBC Dain Rauscher, Inc.
(f/k/a Dain Rauscher, Inc.) in Engage
Technologies, Inc., Firepond, Inc. and
Sycam ore N etw orks, Inc.)

Robin S. Conrad, Esq.
N ational Cham ber L itigation Center 
1615 H Street, N.W .
W ashington, D.C. 20062 
{Attorneys fo r  Amicus Curiae Chamber 
o f  Commerce o f  the United States)

Stephen L. Ratner, Esq.
Proskauer Rose LLP
1585 Broadw ay
N ew  York, N Y  10036
{Attorneys fo r  Prudential Securities
Incorporated in Engage Technologies,
Inc.)

Gary A. Orseck, Esq.
Robbins, Russell, Englert, Orseck,

& U ntereiner LLP 
1801 K  Street, N .W ., Suite 411 
W ashington, D.C. 20006 
{Attorneys fo r  Amicus Curiae Chamber 
o f  Commerce o f  the United States)

N orm an J./Chachkin