Da Silva Moore v. Publicis Groupe, 2012 U.S. Dist. LEXIS ... · MSL GROUP, Defendants. 11 Civ. 1279 (ALC) (AJP) UNITED STATES DISTRICT COURT FOR THE SOUTHERN DISTRICT OF NEW YORK

Predictive Coding Cases

1. Da Silva Moore v. Publicis Groupe, 2012 U.S. Dist. LEXIS 23350 (SDNY, Feb.

24, 2012) 2. Robocast v. Apple, 2012 U.S. Dist. LEXIS 24879 (D. Del. Feb. 24, 2012) 3. In Re: Actos (Pioglitazone) Products Liability Litigation, 2012 US Dist. LEXIS

187519 (WD La. Jul. 27, 2012) 4. Kleen Products v. Packaging Corp. of America, 2012 U.S. Dist. LEXI 139632

(ND Ill. Sep. 28, 2012) 5. EORHB, Inc. v. HOA Holdings, C.A. No. 7409-VCL (Del. Ch. Oct. 15, 2012). 6. U.S. Dept. of Labor v. Subcontracting Concepts, 2013 U.S. Dist. LEXIS 33593

(NDNY, Mar. 11, 2013) 7. Chevron Corporation v. Donzinger, 2013 U.S. Dist. LEXIS 36353 (SDNY, Mar.

15, 2013) 8. In Re: Biomet M2a Magnum Hip Implant Products Liability Litigation, C.A. No.

3:12-MD-2391 (MDL 2391) (ND Ind. Apr. 18, 2013) These cases are being provided for educational purposes only. If you have any questions about the cases, predictive coding, or e-discovery please contact John J. Jablonski, Esq. at 716.566.5469 or [email protected]. For more information about Goldberg Segalla or the firm’s e-discovery capabilities, please visit our E-Discovery Practice Group page.

http://www.goldbergsegalla.com/attorneys/john-j-jablonski

mailto:[email protected]

http://www.goldbergsegalla.com/practice-groups/e-discovery

http://www.goldbergsegalla.com/practice-groups/e-discovery

ktylec

Typewritten Text

ktylec

Typewritten Text

ktylec

Typewritten Text

Click on the case below to jump to that case file

ktylec

Typewritten Text

ktylec

Typewritten Text

ktylec

Typewritten Text

ktylec

Typewritten Text

ktylec

Typewritten Text

MONIQUE DA SILVA MOORE, et al., Plaintiffs, -against- PUBLICIS GROUPE &MSL GROUP, Defendants.

11 Civ. 1279 (ALC) (AJP)

UNITED STATES DISTRICT COURT FOR THE SOUTHERN DISTRICT OFNEW YORK

287 F.R.D. 182; 2012 U.S. Dist. LEXIS 23350; 18 Wage & Hour Cas. 2d (BNA) 1479

February 24, 2012, DecidedFebruary 24, 2012, Filed

SUBSEQUENT HISTORY: Adopted by, Objectionoverruled by Moore v. Publicis Groupe SA, 2012 U.S.Dist. LEXIS 58742 (S.D.N.Y., Apr. 25, 2012)

PRIOR HISTORY: Moore v. Publicis Groupe SA, 2012U.S. Dist. LEXIS 19857 (S.D.N.Y., Feb. 14, 2012)

DISPOSITION: Computer-assisted review wasjudicially-approved for use in appropriate cases.

COUNSEL: [**1] For Monique Da Silva Moore, onbehalf of herself and all others similarly situated,Plaintiff: Jeremy Heisler, LEAD ATTORNEY, Sanford,Wittels & Heisler, LLP, New York, NY; David W.Sanford, PRO HAC VICE, Sanford, Wittels & Heisler,LLP (DC), Washington, DC; Deepika Bains, StevenLance Wittels, Sanford Wittels & Heisler, LLP, NewYork, NY; Janette Lynn Wipper, PRO HAC VICE,SANFORD WITTELS & HEISLER, LLP (SanFrancisco, San Francisco, CA; Siham Nurhussein,Clifford Chance US, LLP (NYC), New York, NY.

For Maryellen O'Donohue, on behalf of herself and allothers similarly situated, Laurie Mayers, on behalf ofherself and all others similarly situated, Heather Pierce onbehalf of herself and all others similarly situated,Katherine Wilkinson, on behalf of herself and all otherssimilarly situated, Plaintiffs: Jeremy Heisler, LEAD

ATTORNEY, Sanford, Wittels & Heisler, LLP, NewYork, NY; David W. Sanford, PRO HAC VICE, Sanford,Wittels & Heisler, LLP (DC), Washington, DC; DeepikaBains, Steven Lance Wittels, Sanford Wittels & Heisler,LLP, New York, NY; Janette Lynn Wipper, PRO HACVICE, SANFORD WITTELS & HEISLER, LLP (SanFrancisco, San Francisco, CA.

For Publicis Groupe, Defendant: Melissa [**2] RuthKelly, LEAD ATTORNEY, Morgan, Lewis & BockiusLLP (New York), New York, NY; Paul Clayton Evans,PRO HAC VICE, Morgan Lewis & Bockius, LLP (PA),Philadelphia, PA; Paul Clayton Evans, Morgan, Lewis &Bockius LLP, Philadelphia, PA.

For MSL Group, Defendant: Noel P. Tripp, Paul J.Siegel, LEAD ATTORNEYS, Jeffrey W. Brecher,Jackson Lewis LLP(Melville N.Y.), Melville, NY; BrettMichael Anders, Jackson Lewis LLP(NJ), Morristown,NJ; Victoria Woodin Chavey, Jackson Lewis LLP,Hartford, CT.

JUDGES: Andrew J. Peck, United States MagistrateJudge.

OPINION BY: Andrew J. Peck

OPINION

Page 1

[*182] OPINION AND ORDER

ANDREW J. PECK, United States MagistrateJudge:

In my article Search, Forward: Will manualdocument review and keyword searches be replaced bycomputer-assisted coding?, I wrote:

To my knowledge, no reported case(federal or state) has ruled on the use ofcomputer-assisted coding. Whileanecdotally it appears that some lawyersare using predictive coding technology, italso appears that many lawyers (and their[*183] clients) are waiting for a judicialdecision approving of computer-assistedreview.

Perhaps they are looking for anopinion concluding that: "It is the opinionof this court that the use of predictivecoding is a proper [**3] and acceptablemeans of conducting searches under theFederal Rules of Civil Procedure, andfurthermore that the software provided forthis purpose by [insert name of yourfavorite vendor] is the software of choicein this court." If so, it will be a long wait.

. . . .

Until there is a judicial opinionapproving (or even critiquing) the use ofpredictive coding, counsel will just have torely on this article as a sign of judicialapproval. In my opinion,computer-assisted coding should be usedin those cases where it will help "securethe just, speedy, and inexpensive" (Fed. R.Civ. P. 1) determination of cases in oure-discovery world.

Andrew Peck, Search, Forward, L. Tech. News, Oct.2011, at 25, 29. This judicial opinion now recognizes thatcomputer-assisted review is an acceptable way to searchfor relevant ESI in appropriate cases.1

1 To correct the many blogs about this case,initiated by a press release from plaintiffs' vendor-- [**4] the Court did not order the parties to use

predictive coding. The parties had agreed todefendants' use of it, but had disputes over thescope and implementation, which the Court ruledon, thus accepting the use of computer-assistedreview in this lawsuit.

CASE BACKGROUND

In this action, five female named plaintiffs are suingdefendant Publicis Groupe, "one of the world's 'big four'advertising conglomerates," and its United States publicrelations subsidiary, defendant MSL Group. (See Dkt.No. 4: Am. Compl. ¶¶ 1, 5, 26-32.) Plaintiffs allege thatdefendants have a "glass ceiling" that limits women toentry level positions, and that there is "systemic,company-wide gender discrimination against female PRemployees like Plaintiffs." (Am. Compl. ¶¶ 4-6, 8.)Plaintiffs allege that the gender discrimination includes

(a) paying Plaintiffs and other female PRemployees less than similarly-situatedmale employees; (b) failing to promote oradvance Plaintiffs and other female PRemployees at the same rate assimilarly-situated male employees; and (c)carrying out discriminatory terminations,demotions and/or job reassignments offemale PR employees when the companyreorganized its PR practice [**5]beginning in 2008 . . . .

(Am. Compl. ¶ 8.)

Plaintiffs assert claims for gender discriminationunder Title VII (and under similar New York State andNew York City laws) (Am. Compl. ¶¶ 204-25),pregnancy discrimination under Title VII and relatedviolations of the Family and Medical Leave Act (Am.Compl. ¶¶ 239-71), as well as violations of the Equal PayAct and Fair Labor Standards Act (and the similar NewYork Labor Law) (Am. Compl. ¶¶ 226-38).

The complaint seeks to bring the Equal PayAct/FLSA claims as a "collective action" (i.e., opt-in) onbehalf of all "current, former, and future female PRemployees" employed by defendants in the United States"at any time during the applicable liability period" (Am.Compl. ¶¶ 179-80, 190-203), and as a class action on thegender and pregnancy discrimination claims and on theNew York Labor Law pay claim (Am. Compl. ¶¶171-98). Plaintiffs, however, have not yet moved for

Page 2287 F.R.D. 182, *182; 2012 U.S. Dist. LEXIS 23350, **2;

18 Wage & Hour Cas. 2d (BNA) 1479

collective action or class certification at this time.

Defendant MSL denies the allegations in thecomplaint and has asserted various affirmative defenses.(See generally Dkt. No. 19: MSL Answer.) DefendantPublicis is challenging the Court's jurisdiction over it, andthe parties [**6] have until March 12, 2012 to conductjurisdictional discovery. (See Dkt. No. 44: 10/12/11Order.)

COMPUTER-ASSISTED REVIEW EXPLAINED

My Search, Forward article explained myunderstanding of computer-assisted review, as follows:

By computer-assisted coding, I meantools (different vendors use differentnames) that use sophisticated algorithmsto enable the computer to determinerelevance, [*184] based on interactionwith (i.e., training by) a human reviewer.

Unlike manual review, where thereview is done by the most junior staff,computer-assisted coding involves a seniorpartner (or [small] team) who review andcode a "seed set" of documents. Thecomputer identifies properties of thosedocuments that it uses to code otherdocuments. As the senior reviewercontinues to code more sample documents,the computer predicts the reviewer'scoding. (Or, the computer codes somedocuments and asks the senior reviewerfor feedback.)

When the system's predictions and thereviewer's coding sufficiently coincide, thesystem has learned enough to makeconfident predictions for the remainingdocuments. Typically, the senior lawyer(or team) needs to review only a fewthousand documents to train the computer.

Some [**7] systems produce asimple yes/no as to relevance, while othersgive a relevance score (say, on a 0 to 100basis) that counsel can use to prioritizereview. For example, a score above 50may produce 97% of the relevantdocuments, but constitutes only 20% of

the entire document set.

Counsel may decide, after samplingand quality control tests, that documentswith a score of below 15 are so highlylikely to be irrelevant that no furtherhuman review is necessary. Counsel canalso decide the cost-benefit of manualreview of the documents with scores of15-50.

Andrew Peck, Search, Forward, L. Tech. News, Oct.2011, at 25, 29.2

2 From a different perspective, every person whouses email uses predictive coding, even if they donot realize it. The "spam filter" is an example ofpredictive coding.

My article further explained my belief that Daubertwould not apply to the results of using predictive coding,but that in any challenge to its use, this Judge would beinterested in both the process used and the results:

[I]f the use of predictive coding ischallenged in a case before me, I will wantto know what was done and why thatproduced defensible results. I may be lessinterested in the science [**8] behind the"black box" of the vendor's software thanin whether it produced responsivedocuments with reasonably high recall andhigh precision.

That may mean allowing therequesting party to see the documents thatwere used to train the computer-assistedcoding system. (Counsel would not berequired to explain why they codeddocuments as responsive ornon-responsive, just what the coding was.)Proof of a valid "process," includingquality control testing, also will beimportant.

. . . .

Of course, the best approach to theuse of computer-assisted coding is tofollow the Sedona CooperationProclamation model. Advise opposing



counsel that you plan to usecomputer-assisted coding and seekagreement; if you cannot, considerwhether to abandon predictive coding forthat case or go to the court for advanceapproval.

Id.

THE ESI DISPUTES IN THIS CASE AND THEIRRESOLUTION

After several discovery conferences and rulings byJudge Sullivan (the then-assigned District Judge), hereferred the case to me for general pretrial supervision.(Dkt. No. 48: 11/28/11 Referral Order.) At my firstdiscovery conference with the parties, both parties'counsel mentioned that they had been discussing an"electronic [**9] discovery protocol," and MSL'scounsel stated that an open issue was "plaintiff'sreluctance to utilize predictive coding to try to cull downthe" approximately three million electronic documentsfrom the agreed-upon custodians. (Dkt. No. 51: 12/2/11Conf. Tr. at 7-8.)3 Plaintiffs' counsel clarified that MSLhad "over simplified [plaintiffs'] stance on predictivecoding," i.e., that it was not opposed [*185] but had"multiple concerns . . . on the way in which [MSL] planto employ predictive coding" and plaintiffs wanted"clarification." (12/2/11 Conf. Tr. at 21.)

3 When defense counsel mentioned thedisagreement about predictive coding, I statedthat: "You must have thought you died and wentto Heaven when this was referred to me," towhich MSL's counsel responded: "Yes, yourHonor. Well, I'm just thankful that, you know, wehave a person familiar with the predictive codingconcept." (12/2/11 Conf. Tr. at 8-9.)

The Court did not rule but offered the parties thefollowing advice:

Now, if you want any more advice, forbetter or for worse on the ESI plan andwhether predictive coding should be used,. . . I will say right now, what should notbe a surprise, I wrote an article in theOctober Law [**10] Technology Newscalled Search Forward, which sayspredictive coding should be used in theappropriate case.

Is this the appropriate case for it? Youall talk about it some more. And if youcan't figure it out, you are going to getback in front of me. Key words, certainlyunless they are well done and tested, arenot overly useful. Key words along withpredictive coding and other methodology,can be very instructive.

I'm also saying to the defendants whomay, from the comment before, have readmy article. If you do predictive coding,you are going to have to give your seedset, including the seed documents markedas nonresponsive to the plaintiff's counselso they can say, well, of course you arenot getting any [relevant] documents,you're not appropriately training thecomputer.

(12/2/11 Conf. Tr. at 20-21.) The December 2, 2011conference adjourned with the parties agreeing to furtherdiscuss the ESI protocol. (12/2/11 Conf. Tr. at 34-35.)

The ESI issue was next discussed at a conference onJanuary 4, 2012. (Dkt. No. 71: 1/4/12 Conf. Tr.)Plaintiffs' ESI consultant conceded that plaintiffs "havenot taken issue with the use of predictive coding or,frankly, with the confidence levels [**11] that they[MSL] have proposed . . . ." (1/4/12 Conf. Tr. at 51.)Rather, plaintiffs took issue with MSL's proposal thatafter the computer was fully trained and the resultsgenerated, MSL wanted to only review and produce thetop 40,000 documents, which it estimated would cost$200,000 (at $5 per document). (1/4/12 Conf. Tr. at47-48, 51.) The Court rejected MSL's 40,000 documentsproposal as a "pig in a poke." (1/4/12 Conf. Tr. at 51-52.)The Court explained that "where [the] line will be drawn[as to review and production] is going to depend on whatthe statistics show for the results," since"[p]roportionality requires consideration of results as wellas costs. And if stopping at 40,000 is going to leave atremendous number of likely highly responsivedocuments unproduced, [MSL's proposed cutoff] doesn'twork." (1/4/12 Conf. Tr. at 51-52; see also id. at 57-58;Dkt. No. 88: 2/8/12 Conf. Tr. at 84.) The parties agreed tofurther discuss and finalize the ESI protocol by lateJanuary 2012, with a conference held on February 8,2012. (1/4/12 Conf. Tr. at 60-66; see 2/8/12 Conf. Tr.)



Custodians

The first issue regarding the ESI protocol involvedthe selection of which custodians' emails [**12] wouldbe searched. MSL agreed to thirty custodians for a "firstphase." (Dkt. No. 88: 2/8/12 Conf. Tr. at 23-24.) MSL'scustodian list included the president and other membersof MSL's "executive team," most of its HR staff and anumber of managing directors. (2/8/12 Conf. Tr. at 24.)

Plaintiffs sought to include as additional custodiansseven male "comparators," explaining that thecomparators' emails were needed in order to findinformation about their job duties and how their dutiescompared to plaintiffs' job duties. (2/8/12 Conf. Tr. at25-27.) Plaintiffs gave an example of the men beinggiven greater "client contact" or having better jobassignments. (2/8/12 Conf. Tr. at 28-30.) The Court heldthat the search of the comparators' emails would be sodifferent from that of the other custodians that thecomparators should not be included in the emailssubjected to predictive coding review. (2/8/12 Conf. Tr.at 28, 30.) As a fallback position, plaintiffs proposed to"treat the comparators as a separate search," but the Courtfound that plaintiffs could not describe in any meaningfulway how they would search the comparators' emails,even as a separate search. (2/8/12 Conf. Tr. at 30-31.)[**13] Since the plaintiffs likely could develop theinformation needed through depositions of thecomparators, the Court ruled that the comparators' emailswould not be included in phase one. (2/8/12 Conf. Tr. at31.)

[*186] Plaintiffs also sought to include MSL'sCEO, Olivier Fleuriot, located in France and whoseemails were mostly written in French. (2/8/12 Conf. Tr. at32-34.) The Court concluded that because his emails withthe New York based executive staff would be gatheredfrom those custodians, and Fleuriot's emails stored inFrance likely would be covered by the French privacyand blocking laws,4 Fleuriot should not be included as afirst-phase custodian. (2/8/12 Conf. Tr. at 35.)

4 See, e.g., Societe Nationale IndustrielleAerospatiale v. U.S. Dist. Ct. for the S.D. of Iowa,482 U.S. 522, 107 S. Ct. 2542, 96 L. Ed. 2d 461(1987); see also The Sedona Conference,International Principles on Discovery, Disclosure& Data Protection (2011), available athttp://www.thesedonaconference.org/dltForm?did=IntlPrinciples2011.pdf.

Plaintiffs sought to include certain managingdirectors from MSL offices at which no named plaintiffworked. (2/8/12 Conf. Tr. at 36-37.) The Court ruled thatsince plaintiffs had not yet moved [**14] for collectiveaction status or class certification, until the motions weremade and granted, discovery would be limited to offices(and managing directors) where the named plaintiffs hadworked. (2/8/12 Conf. Tr. at 37-39.)

The final issue raised by plaintiffs related to thephasing of custodians and the discovery cutoff dates.MSL proposed finishing phase-one discovery completelybefore considering what to do about a second phase. (See2/8/12 Conf. Tr. at 36.) Plaintiffs expressed concern thatthere would not be time for two separate phases,essentially seeking to move the phase-two custodiansback into phase one. (2/8/12 Conf. Tr. at 35-36.) TheCourt found MSL's separate phase approach to be moresensible and noted that if necessary, the Court wouldextend the discovery cutoff to allow the parties to pursuediscovery in phases. (2/8/12 Conf. Tr. at 36, 50.)

Sources of ESI

The parties agreed on certain ESI sources, includingthe "EMC SourceOne [Email] Archive," the "PeopleSoft"human resources information management system andcertain other sources including certain HR "shared"folders. (See Dkt. No. 88: 2/8/12 Conf. Tr. at 44-45,50-51.) As to other "shared" folders, neither side [**15]was able to explain whether the folders merely containedforms and templates or collaborative working documents;the Court therefore left those shared folders for phase twounless the parties promptly provided information aboutlikely contents. (2/8/12 Conf. Tr. at 47-48.)

The Court noted that because the named plaintiffsworked for MSL, plaintiffs should have some idea whatadditional ESI sources, if any, likely had relevantinformation; since the Court needed to considerproportionality pursuant to Rule 26(b)(2)(C), plaintiffsneeded to provide more information to the Court thanthey were doing if they wanted to add additional datasources into phase one. (2/8/12 Conf. Tr. at 49-50.) TheCourt also noted that where plaintiffs were getting factualinformation from one source (e.g., pay information,promotions, etc.), "there has to be a limit to redundancy"to comply with Rule 26(b)(2)(C). (2/8/12 Conf. Tr. at54.)5

5 The Court also suggested that the best way to



resolve issues about what information might befound in a certain source is for MSL to showplaintiffs a sample printout from that source.(2/8/12 Conf. Tr. at 55-56.)

The Predictive Coding Protocol

The parties agreed to use a [**16] 95% confidencelevel (plus or minus two percent) to create a randomsample of the entire email collection; that sample of2,399 documents will be reviewed to determine relevant(and not relevant) documents for a "seed set" to use totrain the predictive coding software. (Dkt. No. 88: 2/8/12Conf. Tr. at 59-61.) An area of disagreement was thatMSL reviewed the 2,399 documents before the partiesagreed to add two additional concept groups (i.e., issuetags). (2/8/12 Conf. Tr. at 62.) MSL suggested that sinceit had agreed to provide all 2,399 documents (and MSL'scoding of them) to plaintiffs for their review, plaintiffscan code them for the new issue tags, and MSL willincorporate that coding into the system. (2/8/12 Conf. Tr.at 64.) Plaintiffs' vendor agreed to that approach. (2/8/12Conf. Tr. at 64.)

To further create the seed set to train the predictivecoding software, MSL coded certain [*187] documentsthrough "judgmental sampling." (2/8/12 Conf. Tr. at 64.)The remainder of the seed set was created by MSLreviewing "keyword" searches with Boolean connectors(such as "training and Da Silva Moore," or "promotionand Da Silva Moore") and coding the top fifty hits fromthose searches. [**17] (2/8/12 Conf. Tr. at 64-66, 72.)MSL agreed to provide all those documents (exceptprivileged ones) to plaintiffs for plaintiffs to reviewMSL's relevance coding. (2/8/12 Conf. Tr. at 66.) Inaddition, plaintiffs provided MSL with certain otherkeywords, and MSL used the same process withplaintiffs' keywords as with the MSL keywords,reviewing and coding an additional 4,000 documents.(2/8/12 Conf. Tr. at 68-69, 71.) All of this review tocreate the seed set was done by senior attorneys (notparalegals, staff attorneys or junior associates). (2/8/12Conf. Tr. at 92-93.) MSL reconfirmed that "[a]ll of thedocuments that are reviewed as a function of the seed set,whether [they] are ultimately coded relevant or irrelevant,aside from privilege, will be turned over to" plaintiffs.(2/8/12 Conf. Tr. at 73.)

The next area of discussion was the iterative roundsto stabilize the training of the software. MSL's vendor'spredictive coding software ranks documents on a score of

100 to zero, i.e., from most likely relevant to least likelyrelevant. (2/8/12 Conf. Tr. at 70.) MSL proposed usingseven iterative rounds; in each round they would reviewat least 500 documents from different concept [**18]clusters to see if the computer is returning new relevantdocuments. (2/8/12 Conf. Tr. at 73-74.) After the seventhround, to determine if the computer is well trained andstable, MSL would review a random sample (of 2,399documents) from the discards (i.e., documents coded asnon-relevant) to make sure the documents determined bythe software to not be relevant do not, in fact, containhighly-relevant documents. (2/8/12 Conf. Tr. at 74-75.)For each of the seven rounds and the final quality-checkrandom sample, MSL agreed that it would show plaintiffsall the documents it looked at including those deemed notrelevant (except for privileged documents). (2/8/12 Conf.Tr. at 76.)

Plaintiffs' vendor noted that "we don't at this pointagree that this is going to work. This is new technologyand it has to be proven out." (2/8/12 Conf. Tr. at 75.)Plaintiffs' vendor agreed, in general, thatcomputer-assisted review works, and works better thanmost alternatives. (2/8/12 Conf. Tr. at 76.) Indeed,plaintiffs' vendor noted that "it is fair to say [that] we arebig proponents of it." (2/8/12 Conf. Tr. at 76.) The Courtreminded the parties that computer-assisted review"works better than most [**19] of the alternatives, if notall of the [present] alternatives. So the idea is not to makethis perfect, it's not going to be perfect. The idea is tomake it significantly better than the alternatives withoutnearly as much cost." (2/8/12 Conf. Tr. at 76.)

The Court accepted MSL's proposal for the seveniterative reviews, but with the following caveat:

But if you get to the seventh round and[plaintiffs] are saying that the computer isstill doing weird things, it's not stabilized,etc., we need to do another round or two,either you will agree to that or you willboth come in with the appropriate QCinformation and everything else and [maybe ordered to] do another round or two orfive or 500 or whatever it takes to stabilizethe system.

(2/8/12 Conf. Tr. at 76-77; see also id. at 83-84, 88.)

On February 17, 2012, the parties submitted their



"final" ESI Protocol which the Court "so ordered." (Dkt.No. 92: 2/17/12 ESI Protocol & Order.)6 Because this isthe first Opinion dealing with predictive coding, theCourt annexes hereto as an Exhibit the provisions of theESI Protocol dealing with the predictive coding searchmethodology.

6 Plaintiffs included a paragraph noting itsobjection to the [**20] ESI Protocol, as follows:

Plaintiffs object to this ESIProtocol in its entirety. Plaintiffssubmitted their own proposed ESIProtocol to the Court, but it waslargely rejected. The Court thenordered the parties to submit ajoint ESI Protocol reflecting theCourt's rulings. Accordingly,Plaintiffs jointly submit this ESIProtocol with MSL, but reserve theright to object to its use in thiscase.

(ESI Protocol ¶ J.1 at p. 22.)

[*188] OBSERVATIONS ON PLAINTIFF'SOBJECTIONS TO THE COURT'S RULINGS

On February 22, 2012, plaintiffs filed objections tothe Court's February 8, 2012 rulings. (Dkt. No. 93: Pls.Rule 72(a) Objections; see also Dkt. No. 94: NurhusseinAff.; Dkt. No. 95: Neale Aff.) While those objections arebefore District Judge Carter, a few comments are inorder.

Plaintiffs' Reliance on Rule 26(g)(1)(A) is Erroneous

Plaintiffs' objections to my February 8, 2012 rulingsassert that my acceptance of MSL's predictive codingapproach "provides unlawful 'cover' for MSL's counsel,who has a duty under FRCP 26(g) to 'certify' that theirclient's document production is 'complete' and 'correct' asof the time it was made. FRCP 26(g)(1)(A)." (Dkt. No.93: Pls. Rule 72(a) Objections at 8 n.7; accord, [**21] id.at 2.) In large-data cases like this, involving over threemillion emails, no lawyer using any search method couldhonestly certify that its production is "complete" -- butmore importantly, Rule 26(g)(1) does not require that.Plaintiffs simply misread Rule 26(g)(1). The certificationrequired by Rule 26(g)(1) applies "with respect to adisclosure." Fed. R. Civ. P. 26(g)(1)(A) (emphasis

added). That is a term of art, referring to the mandatoryinitial disclosures required by Rule 26(a)(1). Since theRule 26(a)(1) disclosure is information (witnesses,exhibits) that "the disclosing party may use to support itsclaims or defenses," and failure to provide suchinformation leads to virtually automatic preclusion, seeFed. R Civ. P. 37(c)(1), it is appropriate for the Rule26(g)(1)(A) certification to require disclosures be"complete and correct."

Rule 26(g)(1)(B) is the provision that applies todiscovery responses. It does not call for certification thatthe discovery response is "complete," but ratherincorporates the Rule 26(b)(2)(C) proportionalityprinciple. Thus, Rule 26(g)(1)(A) has absolutely nothingto do with MSL's obligations to respond to plaintiffs'discovery requests. [**22] Plaintiffs' argument is basedon a misunderstanding of Rule 26(g)(1).7

7 Rule 26(g)(1) provides:

(g) Signing Disclosures and DiscoveryRequests, Responses, and Objections.

(1) Signature Required; Effect ofSignature. Every disclosure underRule 26(a)(1) or (a)(3) and everydiscovery request, response, orobjection must be signed by atleast one attorney of record in theattorney's own name . . . . Bysigning, an attorney or partycertifies that to the best of theperson's knowledge, information,and belief formed after areasonable inquiry:

(A) with respectto a disclosure, it iscomplete andcorrect as of thetime it is made; and

(B) withrespect to adiscovery request,response, orobjection, it is:

(i)consistent



withtheserulesandwarrantedbyexistinglawor byanonfrivolousargumentforextending,modifying,orreversingexistinglaw,orforestablishingnewlaw;

(ii)notinterposedforanyimproperpurpose,suchas toharass,causeunnecessarydelay,orneedlesslyincreasethecostoflitigation;and

(iii)neitherunreasonable

norundulyburdensomeorexpensive,consideringtheneedsofthecase,priordiscoveryinthecase,theamountincontroversy,andtheimportanceoftheissues[**23]atstakeintheaction.

Fed. R. Civ. P. 26(g)(1) (emphasis added).

Rule 702 and Daubert Are Not Applicable toDiscovery Search Methods

Plaintiffs' objections also argue that my acceptanceof MSL's predictive coding protocol "is contrary toFederal Rule of Evidence 702" and "violates thegatekeeping function underlying Rule 702." (Dkt. No. 93:Pls. Rule 72(a) Objections at 2-3; accord, id. at 10-12.)8

8 As part of this argument, plaintiffs complainthat although both parties' experts (i.e., vendors)



spoke at the discovery conferences, they were notsworn in. (Pls. Rule 72(a) Objections at 12: "Tohis credit, the Magistrate [Judge] did ask theparties to bring [to the conference] the ESI expertsthey had hired to advise them regarding thecreation of an ESI protocol. These experts,however, were never sworn in, and thus thestatements they made in court at the hearings werenot sworn testimony made under penalty ofperjury.") Plaintiffs never asked the Court to havethe experts testify to their qualifications or besworn in.

Federal Rule of Evidence 702 and the SupremeCourt's Daubert decision9 deal with the trial court's roleas gatekeeper to exclude unreliable expert testimony frombeing [**24] submitted [*189] to the jury at trial. Seealso Advisory Comm. Notes to Fed. R. Evid. 702. It is arule for admissibility of evidence at trial.

9 Daubert v. Merrell Dow Pharms., Inc., 509U.S. 579, 113 S. Ct. 2786, 125 L. Ed. 2d 469(1993).

If MSL sought to have its expert testify at trial andintroduce the results of its ESI protocol into evidence,Daubert and Rule 702 would apply. Here, in contrast, thetens of thousands of emails that will be produced indiscovery are not being offered into evidence at trial asthe result of a scientific process or otherwise. Theadmissibility of specific emails at trial will depend uponeach email itself (for example, whether it is hearsay, or abusiness record or party admission), not how it was foundduring discovery.

Rule 702 and Daubert simply are not applicable tohow documents are searched for and found in discovery.

Plaintiffs' Reliability Concerns Are, At Best,Premature

Finally, plaintiffs' objections assert that "MSL'smethod lacks the necessary standards for assessingwhether its results are accurate; in other words, there isno way to be certain if MSL's method is reliable." (Dkt.No. 93: Pls. Rule 72(a) Objections at 13-18.) Plaintiffs'concerns may be appropriate [**25] for resolution duringor after the process (which the Court will be closelysupervising), but are premature now. For example,plaintiffs complain that "MSL's method fails to includean agreed-upon standard of relevance that is transparent

and accessible to all parties. . . . Without this standard,there is a high-likelihood of delay as the parties resolvedisputes with regard to individual documents on acase-by-case basis." (Id. at 14.) Relevance is determinedby plaintiffs' document demands. As statistics show,perhaps only 5% of the disagreement among reviewerscomes from close questions of relevance, as opposed toreviewer error. (See page 18 n.11 below.) The issueregarding relevance standards might be significant ifMSL's proposal was not totally transparent. Here,however, plaintiffs will see how MSL has coded everyemail used in the seed set (both relevant and notrelevant), and the Court is available to quickly resolveany issues.

Plaintiffs complain they cannot determine if "MSL'smethod actually works" because MSL does not describehow many relevant documents are permitted to be locatedin the final random sample of documents the softwaredeemed irrelevant. (Pls. Rule 72(a) [**26] Objections at15-16.) Plaintiffs argue that "without any decision aboutthis made in advance, the Court is simply kicking the candown the road." (Id. at 16.) In order to determineproportionality, it is necessary to have more informationthan the parties (or the Court) now has, including howmany relevant documents will be produced and at whatcost to MSL. Will the case remain limited to the namedplaintiffs, or will plaintiffs seek and obtain collectiveaction and/or class action certification? In the finalsample of documents deemed irrelevant, are any relevantdocuments found that are "hot," "smoking gun"documents (i.e., highly relevant)? Or are the only relevantdocuments more of the same thing? One hot documentmay require the software to be re-trained (or some othersearch method employed), while several documents thatreally do not add anything to the case might not matter.These types of questions are better decided "down theroad," when real information is available to the partiesand the Court.

FURTHER ANALYSIS AND LESSONS FOR THEFUTURE

The decision to allow computer-assisted review inthis case was relatively easy -- the parties agreed to itsuse (although disagreed about [**27] how best toimplement such review). The Court recognizes thatcomputer-assisted review is not a magic,Staples-Easy-Button, solution appropriate for all cases.The technology exists and should be used where



appropriate, but it is not a case of machine replacinghumans: it is the process used and the interaction of manand machine that the courts needs to examine.

The objective of review in ediscovery is to identifyas many relevant documents as possible, while reviewingas few non-relevant documents as possible. Recall is thefraction of relevant documents identified during a review;precision is the fraction of identified documents that arerelevant. Thus, recall is a measure of completeness, whileprecision is [*190] a measure of accuracy orcorrectness. The goal is for the review method to result inhigher recall and higher precision than another reviewmethod, at a cost proportionate to the "value" of the case.See, e.g., Maura R. Grossman & Gordon V. Cormack,Technology-Assisted Review in E-Discovery Can BeMore Effective and More Efficient Than ExhaustiveManual Review, Rich. J.L.& Tech., Spring 2011, at 8-9,available at http://jolt.richmond.edu/v17i3/article11 .pdf.

The slightly more [**28] difficult case would bewhere the producing party wants to use computer-assistedreview and the requesting party objects.10 The question toask in that situation is what methodology would therequesting party suggest instead? Linear manual review issimply too expensive where, as here, there are over threemillion emails to review. Moreover, while some lawyersstill consider manual review to be the "gold standard,"that is a myth, as statistics clearly show that computerizedsearches are at least as accurate, if not more so, thanmanual review. Herb Roitblatt, Anne Kershaw, andPatrick Oot of the Electronic Discovery Instituteconducted an empirical assessment to "answer thequestion of whether there was a benefit to engaging in atraditional human review or whether computer systemscould be relied on to produce comparable results," andconcluded that "[o]n every measure, the performance ofthe two computer systems was at least as accurate(measured against the original review) as that of humanre-review." Herbert L. Roitblatt, Anne Kershaw &Patrick Oot, Document Categorization in LegalElectronic Discovery: Computer Classification vs.Manual Review, 61 J. Am. Soc'y for Info. Sci. [**29] &Tech. 70, 79 (2010).11

10 The tougher question, raised in Klein Prods.LLC v. Packaging Corp. of Am. before MagistrateJudge Nan Nolan in Chicago, is whether theCourt, at plaintiffs' request, should order thedefendant to use computer-assisted review to

respond to plaintiffs' document requests.11 The Roitblatt, Kershaw, Oot article noted that"[t]he level of agreement among human reviewersis not strikingly high," around 70-75%. Theyidentify two sources for this variability: fatigue("A document that they [the reviewers] mighthave categorized as responsive when they weremore attentive might then be categorized [whenthe reviewer is distracted or fatigued] asnon-responsive or vice versa."), and differences in"strategic judgment." Id. at 77-78. Another studyfound that responsiveness "is fairly well defined,and that disagreements among assessors arelargely attributable to human error," with only 5%of reviewer disagreement attributable toborderline or questionable issues as to relevance.Maura R. Grossman & Gordon V. Cormack,Inconsistent Assessment of Responsiveness inE-Discovery: Difference of Opinion or HumanError? 9 (DESI IV: 2011 ICAIL Workshop onSetting Standards for [**30] Searching Elec.Stored Info. in Discovery, Research Paper),available athttp://www.umiacs.umd.edu/~oard/desi4/papers/grossman3.pdf.

Likewise, Wachtell, Lipton, Rosen & Katz litigationcounsel Maura Grossman and University of Waterlooprofessor Gordon Cormack, studied data from the TextRetrieval Conference Legal Track (TREC) and concludedthat: "[T]he myth that exhaustive manual review is themost effective -- and therefore the most defensible --approach to document review is strongly refuted.Technology-assisted review can (and does) yield moreaccurate results than exhaustive manual review, withmuch lower effort." Maura R. Grossman & Gordon V.Cormack, Technology-Assisted Review in E-DiscoveryCan Be More Effective and More Efficient ThanExhaustive Manual Review, Rich. J.L.& Tech., Spring2011, at 48.12 The technology-assisted reviews in theGrossman-Cormack article also demonstrated significantcost savings over manual review: "Thetechnology-assisted reviews require, on average, humanreview of only 1.9% of the documents, a fifty-foldsavings over exhaustive manual review." Id. at 43.

12 Grossman and Cormack also note that "not[**31] all technology-assisted reviews . . . arecreated equal" and that future studies will beneeded to "address which technology-assisted



review process(es) will improve most on manualreview." Id.

Because of the volume of ESI, lawyers frequentlyhave turned to keyword searches to cull email (or otherESI) down to a more manageable volume for furthermanual review. Keywords have a place in production ofESI -- indeed, the parties here used keyword searches(with Boolean connectors) to find documents for theexpanded seed set to train the predictive coding software.In too [*191] many cases, however, the way lawyerschoose keywords is the equivalent of the child's game of"Go Fish."13 The requesting party guesses whichkeywords might produce evidence to support its casewithout having much, if any, knowledge of theresponding party's "cards" (i.e., the terminology used bythe responding party's custodians). Indeed, theresponding party's counsel often does not know what is inits own client's "cards."

13 See Ralph C. Losey, "Child's Game of 'GoFish' is a Poor Model for e-Discovery Search," inAdventures in Electronic Discovery 209-10(2011).

Another problem with keywords is that they oftenare over-inclusive, [**32] that is, they find responsivedocuments but also large numbers of irrelevantdocuments. In this case, for example, a keyword searchfor "training" resulted in 165,208 hits; Da Silva Moore'sname resulted in 201,179 hits; "bonus" resulted in 40,756hits; "compensation" resulted in 55,602 hits; and"diversity" resulted in 38,315 hits. (Dkt. No. 92: 2/17/12ESI Protocol Ex. A.) If MSL had to manually review allof the keyword hits, many of which would not be relevant(i.e., would be false positives), it would be quite costly.

Moreover, keyword searches usually are not veryeffective. In 1985, scholars David Blair and M. Maroncollected 40,000 documents from a Bay Area RapidTransit accident, and instructed experienced attorney andparalegal searchers to use keywords and other reviewtechniques to retrieve at least 75% of the documentsrelevant to 51 document requests. David L. Blair & M. E.Maron, An Evaluation of Retrieval Effectiveness for aFull-Text Document-Retrieval System, 28 Comm. ACM289 (1985). Searchers believed they met the goals, buttheir average recall was just 20%. Id. This result has beenreplicated in the TREC Legal Track studies over the pastfew years.

Judicial decisions [**33] have criticized specifickeyword searches. Important early decisions in this areacame from two of the leading judicial scholars inediscovery, Magistrate Judges John Facciola (District ofColumbia) and Paul Grimm (Maryland). See UnitedStates v. O'Keefe, 537 F. Supp. 2d 14, 24 (D.D.C. 2008)(Facciola, M.J.); Equity Analytics, LLC v. Lundin, 248F.R.D. 331, 333 (D.D.C. 2008) (Facciola, M.J.); VictorStanley, Inc. v. Creative Pipe, Inc., 250 F.R.D. 251, 260,262 (D. Md. 2008) (Grimm, M.J.). I followed their leadwith Willaim A. Gross Construction Associates, Inc.,when I wrote:

This Opinion should serve as a wake-upcall to the Bar in this District about theneed for careful thought, quality control,testing, and cooperation with opposingcounsel in designing search terms or"keywords" to be used to produce emailsor other electronically stored information("ESI").

. . . .

Electronic discovery requirescooperation between opposing counsel andtransparency in all aspects of preservationand production of ESI. Moreover, wherecounsel are using keyword searches forretrieval of ESI, they at a minimum mustcarefully craft the appropriate keywords,with input from the ESI's custodians as to[**34] the words and abbreviations theyuse, and the proposed methodology mustbe quality control tested to assure accuracyin retrieval and elimination of "falsepositives." It is time that the Bar -- eventhose lawyers who did not come of age inthe computer era -- understand this.

William A. Gross Constr. Assocs., Inc. v. Am. Mfrs.Mut. Ins. Co., 256 F.R.D. 134, 134, 136 (S.D.N.Y. 2009)(Peck, M.J.).

Computer-assisted review appears to be better thanthe available alternatives, and thus should be used inappropriate cases. While this Court recognizes thatcomputer-assisted review is not perfect, the Federal Rulesof Civil Procedure do not require perfection. See, e.g.,Pension Comm. of Univ. of Montreal Pension Plan v.



Banc of Am. Sec., 685 F. Supp. 2d 456, 461 (S.D.N.Y.2010). Courts and litigants must be cognizant of the aimof Rule 1, to "secure the just, speedy, and inexpensivedetermination" of lawsuits. Fed. R. Civ. P. 1. That goal isfurther reinforced by the proportionality doctrine set forthin Rule 26(b)(2)(C), which provides that:

On [**35] motion or on its own, thecourt must limit the frequency or extent ofdiscovery [*192] otherwise allowed bythese rules or by local rule if it determinesthat:

(i) the discovery sought isunreasonably cumulative orduplicative, or can beobtained from some othersource that is moreconvenient, lessburdensome, or lessexpensive;

(ii) the party seekingdiscovery has had ampleopportunity to obtain theinformation by discovery inthe action; or

(iii) the burden orexpense of the proposeddiscovery outweighs itslikely benefit, consideringthe needs of the case, theamount in controversy, theparties' resources, theimportance of the issues atstake in the action, and theimportance of the discoveryin resolving the issues.

Fed. R. Civ. P. 26(b)(2)(C).

In this case, the Court determined that the use ofpredictive coding was appropriate considering: (1) theparties' agreement, (2) the vast amount of ESI to bereviewed (over three million documents), (3) thesuperiority of computer-assisted review to the availablealternatives (i.e., linear manual review or keyword

searches), (4) the need for cost effectiveness andproportionality under Rule 26(b)(2)(C), and (5) thetransparent process proposed by [**36] MSL.

This Court was one of the early signatories to TheSedona Conference Cooperation Proclamation, and hasstated that "the best solution in the entire area ofelectronic discovery is cooperation among counsel. ThisCourt strongly endorses The Sedona ConferenceProclamation (available atwww.TheSedonaConference.org)." William A. GrossConstr. Assocs., Inc. v. Am. Mfrs. Mut. Ins. Co., 256F.R.D. at 136. An important aspect of cooperation istransparency in the discovery process. MSL'stransparency in its proposed ESI search protocol made iteasier for the Court to approve the use of predictivecoding. As discussed above on page 10, MSL confirmedthat "[a]ll of the documents that are reviewed as afunction of the seed set, whether [they] are ultimatelycoded relevant or irrelevant, aside from privilege, will beturned over to" plaintiffs. (Dkt. No. 88: 2/8/12 Conf. Tr.at 73; see also 2/17/12 ESI Protocol at 14: "MSL willprovide Plaintiffs' counsel with all of the non-privilegeddocuments and will provide, to the extent applicable, theissue tag(s) coded for each document . . . . If necessary,counsel will meet and confer to attempt to resolve anydisagreements regarding the coding [**37] applied to thedocuments in the seed set.") While not all experiencedESI counsel believe it necessary to be as transparent asMSL was willing to be, such transparency allows theopposing counsel (and the Court) to be more comfortablewith computer-assisted review, reducing fears about theso-called "black box" of the technology.14 This Courthighly recommends that counsel in future cases bewilling to at least discuss, if not agree to, suchtransparency in the computer-assisted review process.

14 It also avoids the GIGO problem, i.e.,garbage in, garbage out.

Several other lessons for the future can be derivedfrom the Court's resolution of the ESI discovery disputesin this case.

First, it is unlikely that courts will be able todetermine or approve a party's proposal as to whenreview and production can stop until thecomputer-assisted review software has been trained andthe results are quality control verified. Only at that pointcan the parties and the Court see where there is a cleardrop off from highly relevant to marginally relevant to



not likely to be relevant documents. While cost is a factorunder Rule 26(b)(2)(C), it cannot be considered inisolation from the results of the [**38] predictive codingprocess and the amount at issue in the litigation.

Second, staging of discovery by starting with themost likely to be relevant sources (including custodians),without prejudice to the requesting party seeking moreafter conclusion of that first stage review, is a way tocontrol discovery costs. If staging requires a longerdiscovery period, most judges should be willing to grantsuch an extension. (This Judge runs a self-proclaimed"rocket docket," but informed the parties here of theCourt's willingness to extend the discovery cutoff ifnecessary to allow the staging of custodians and otherESI sources.)

[*193] Third, in many cases requesting counsel'sclient has knowledge of the producing party's records,either because of an employment relationship as here orbecause of other dealings between the parties (e.g.,contractual or other business relationships). It issurprising that in many cases counsel do not appear tohave sought and utilized their client's knowledge aboutthe opposing party's custodians and document sources.Similarly, counsel for the producing party often is notsufficiently knowledgeable about their own client'scustodians and business terminology. Another [**39]way to phrase cooperation is "strategic proactivedisclosure of information," i.e., if you are knowledgeableabout and tell the other side who your key custodians areand how you propose to search for the requesteddocuments, opposing counsel and the Court are more aptto agree to your approach (at least as phase one withoutprejudice).

Fourth, the Court found it very helpful that theparties' ediscovery vendors were present and spoke at thecourt hearings where the ESI Protocol was discussed. (Atediscovery programs, this is sometimes jokingly referredto as "bring your geek to court day.") Even where as herecounsel is very familiar with ESI issues, it is very helpfulto have the parties' ediscovery vendors (or in-house ITpersonnel or in-house ediscovery counsel) present atcourt conferences where ESI issues are being discussed.It also is important for the vendors and/or knowledgeablecounsel to be able to explain complicated ediscoveryconcepts in ways that make it easily understandable tojudges who may not be tech-savvy.

CONCLUSION

This Opinion appears to be the first in which a Courthas approved of the use of computer-assisted review.That does not mean computer-assisted review [**40]must be used in all cases, or that the exact ESI protocolapproved here will be appropriate in all future cases thatutilize computer-assisted review. Nor does this Opinionendorse any vendor (the Court was very careful not tomention the names of the parties' vendors in the body ofthis Opinion, although it is revealed in the attached ESIProtocol), nor any particular computer-assisted reviewtool. What the Bar should take away from this Opinion isthat computer-assisted review is an available tool andshould be seriously considered for use inlarge-data-volume cases where it may save the producingparty (or both parties) significant amounts of legal fees indocument review. Counsel no longer have to worry aboutbeing the "first" or "guinea pig" for judicial acceptance ofcomputer-assisted review. As with keywords or any othertechnological solution to ediscovery, counsel must designan appropriate process, including use of availabletechnology, with appropriate quality control testing, toreview and produce relevant ESI while adhering to Rule1 and Rule 26(b)(2)(C) proportionality.Computer-assisted review now can be consideredjudicially-approved for use in appropriate cases.

SO [**41] ORDERED.

Dated: New York, New York

February 24, 2012

/s/ Andrew J. Peck

Andrew J. Peck

United States Magistrate Judge

EXHIBIT

PARTIES' PROPOSED PROTOCOLRELATING TO THE PRODUCTION OFELECTRONICALLY STORED INFORMATION("ESI")

A. Scope

1. General. The procedures and protocols outlinedherein govern the production of electronically storedinformation ("ESI") by MSLGROUP Americas, Inc.("MSL") during the pendency of this litigation. Theparties to this protocol will take reasonable steps to



comply with this agreed-upon protocol for the productionof documents and information existing in electronicformat. Nothing in this protocol will be interpreted torequire disclosure of documents or information protected[*194] from disclosure by the attorney-client privilege,work-product product doctrine or any other applicableprivilege or immunity. It is Plaintiffs' position thatnothing in this protocol will be interpreted to waivePlaintiffs' right to object to this protocol as portions of itwere mandated by the Court over Plaintiffs' objections,including Plaintiffs' objections to the predictive codingmethodology proposed by MSL.

2. Limitations and No-Waiver. This protocolprovides a general framework for [**42] the productionof ESI on a going forward basis. The Parties and theirattorneys do not intend by this protocol to waive theirrights to the attorney work-product privilege, except asspecifically required herein, and any such waiver shall bestrictly and narrowly construed and shall not extend toother matters or information not specifically describedherein. All Parties preserve their attorney client privilegesand other privileges and there is no intent by the protocol,or the production of documents pursuant to the protocol,to in any way waive or weaken these privileges. All

documents produced hereunder are fully protected andcovered by the Parties' confidentiality and clawbackagreements and orders of the Court effectuating same.

3. Relevant Time Period. January 1, 2008 throughFebruary 24, 2011 for all non-email ESI relating to topicsbesides pay discrimination and for all e-mails. January 1,2005 through February 24, 2011 for all non-e-mail ESIrelating to pay discrimination for New York Plaintiffs.

B. ESI Preservation

1. MSL has issued litigation notices to designatedemployees on February 10, 2010, March 14, 2011 andJune 9, 2011.

C. Sources

1. The Parties have identified the [**43] followingsources of potentially discoverable ESI at MSL. Phase Isources will be addressed first, and Phase II sources willbe addressed after Phase I source searches are complete.Sources marked as "N/A" will not be searched by theParties.

Data Source Description Phase

a EMC SourceOne I

Archive Archiving System used to capture and

store all incoming and outbound e-mails

and selected instant message

conversations saved through IBM Sametime

(see below).

b Lotus Notes Active corporate system that provides N/A

E-mail e-mail communication and calendaring

functions.

c GroupWise Legacy corporate system that provided N/A

E-mail e-mail communication and calendarin,

functions.

d IBM Sametime Lotus Notes Instant Messaging and N/A

collaboration application.

e Home Directories Personal network storage locations on II

the file server(s) dedicated to



individual users. (With the exception of

2 home directories for which MSL will

collect and analyze the data to

determine the level of duplication as

compared to the EMC SourceOne Archive.

The parties will meet and confer

regarding the selection of the two

custodians.)

f Shared Folders Shared network storage locations on the II

file server(s) that are accessible by

individual users, groups of users or

entire departments. (With the exception

of the following Human Resources shared

folders which will be in Phase I:

Corporate HR, North America HR and New

York HR.)

g Database Servers Backend databases (e.g. Oracle, SQL, N/A

MySQL) used to store information for

front end applications or other

purposes.

h Halogen Software Performance management program provided I

by Halogen to conduct performance

evaluations.

i Noovoo Corporate Intranet site. II

j Corporate E-mail addresses that employees may I

Feedback utilize to provide the company with

comments, suggestions and overall

feedback.

k Hyperion Oracle application that offers global N/A

Financial financial consolidation, reporting and

Management analysis.

("HFM")

l Vurv/Taleo Talent recruitment software. II

m ServiceNow Help Desk application used to track N/A

employee computer related requests.

n PeopleSoft Human resources information management I

system.



o PRISM PeopleSoft component used for time and I

billing management.

p Portal A project based portal provided through II

Oracle/BEA Systems.

q Desktops/Laptops Fixed and portable computers provided to II

employees to perform work related

activities. (With the exception of 2

desktop/laptop hard drives for which MSL

will collect and analyze the data to

determine the level of duplication as

compared to the EMC SourceOne Archive.

The parties will meet and confer

regarding the selection of the two

custodians.)

r Publicis Benefits Web based site that maintains II

Connection information about employee benefits and

related information.

s GEARS Employee expense reporting system. II

t MS&L City Former corporate Intranet. N/A

u Adium Application which aggregates instant N/A

messages.

y Pidgin Application which aggregates instant N/A

message.

w IBM Lotus Mobile device synchronization and N/A

Traveler and security system.

MobileIron

y Mobile Portable PDAs, smart phones, tablets N/A

Communication used for communication.

Devices

z Yammer Social media and collaboration portal. N/A

aa SalesForce.com Web-based customer relationship N/A

management application.

bb Removable Portable storage media, external hard N/A

Storage Devices drives, thumb drives, etc. used to store

copies of work related ESI.

[*196] a. [**44] EMC SourceOne - MSL uses



SourceOne, an EMC e-mail archiving system thatcaptures and stores all e-mail messages that pass throughthe corporate e-mail system. In addition, if a user choosesto save an instant messaging chat conversation from IBMSametime (referenced below), that too would be archivedin SourceOne. Defendant MSL also acknowledges thatcalendar items are regularly ingested into the SourceOnesystem. The parties have agreed that this data source willbe handled as outlined in section E below.

b. Lotus Notes E-mail - MSL currently maintainsmultiple Lotus Notes Domino servers in various datacenters around the world. All e-mail communication andcalendar items are journaled in real time to the EMCSourceOne archive. The parties have agreed to not collectany information from this data source at this time.

c. GroupWise E-mail -- Prior to the implementationof the Lotus Notes environment, GroupWise was used forall e-mail and calendar functionality. Before thedecommissioning of the GroupWise servers, MSLcreated backup tapes of all servers that housed theGroupWise e-mail databases. The parties have agreed tonot collect any information from this data source at thistime.

d. [**45] IBM Sametime -- MSL providescustodians with the ability to have real time chatconversations via the IBM Sametime application that ispart of the Lotus Notes suite of products.

e. Home Directories -- Custodians with corporatenetwork access at MSL also have a dedicated and securednetwork storage location where they are able to save files.MSL will collect the home directory data for 2 custodiansand analyze the data to determine the level of duplicationof documents in this data source against the datacontained in the EMC SourceOne archive for the samecustodians. (The parties will meet and confer regardingthe selection of the two custodians.) The results of theanalysis will be provided to Plaintiffs so that adetermination can be made by the parties as to whetherMSL will include this data source in its production of ESIto Plaintiffs. If so, the parties will attempt to reach anagreement as to the approach used to collect, review andproduce responsive and non-privileged documents.

f. Shared Folders -- Individual employees, groups ofemployees and entire departments at MSL are givenaccess to shared network storage locations to save andshare files. As it relates to the Human Resources [**46]

related shared folders (i.e., North America HR Drive(10.2 GB), Corporate HR Drive (440 MB), NY HR Drive(1.9 GB), Chicago HR Drive (1.16 GB), Boston HRDrive (43.3 MB), and Atlanta HR Drive (6.64 GB)),MSL will judgmentally review and produce responsiveand non-privileged documents from the North AmericaHR Drive, Corporate HR Drive, and NY HR Drive. MSLwill produce to Plaintiffs general information regardingthe content of other Shared Folders. The parties will meetand confer regarding the information gathered concerningthe other Shared Folders and discuss whether anyadditional Shared Folders should be moved to Phase I.

g. Database Servers -- MSL has indicated that it doesnot utilize any database servers, other than those thatpertain to the sources outlined above in C, which arelikely to contain information relevant to Plaintiffs' claims.

h. Halogen Software -- MSL [**47] utilizes a thirdparty product, Halogen, for performance managementand employee evaluations. The parties will meet andconfer in order to exchange additional information andattempt to reach an agreement as to the scope of data andthe approach used to collect, review and produceresponsive and non-privileged documents.

i. Noovoo -- MSL maintains a corporate Intranet sitecalled "Noovoo" where employees are able to accessCompany-related information. [*197] MSL will providePlaintiffs with any employment-related policiesmaintained within Noovoo.

j. Corporate Feedback -- MSL has maintainedvarious e-mail addresses that employees may utilize toprovide the company with comments, suggestions andoverall feedback. These e-mail addresses include"[email protected]","[email protected]","[email protected]" and"[email protected]". The partieshave agreed that all responsive and non-privileged ESIwill be produced from these e-mail accounts and anyother e-mail accounts that fall under this category ofinformation. At present, MSL intends to manually reviewthe contents of each of these e-mail accounts. However, ifafter collecting the contents [**48] of each of the e-mailaccounts MSL determines that a manual review would beimpractical, the parties will meet and confer as to theapproach used to collect, review and produce responsiveand non-privileged documents.



k. Hyperion Financial Management ("HFM") -- MSLuses an Oracle application called HFM that offers globalfinancial consolidation, reporting and analysiscapabilities.

l. Vurv/Taleo -- [**49] Since approximately 2006,MSL used an application known as Vury as its talentrecruitment software. As of August 31, 2011, as a resultof Vury being purchased by Taleo, MSL has been using asimilar application by Taleo as its talent recruitmentsoftware. The application, which is accessed throughMSL's public website, allows users to search for openpositions as well as input information about themselves.To the extent Plaintiffs contend they were denied anyspecific positions, they will identify same and the Partieswill meet and confer to discuss what, if any, informationexists within Vurv/Taleo regarding the identifiedposition. If information exists in Vurv/Taleo or anothersource regarding these positions, MSL will produce thisinformation, to the extent such information isdiscoverable.

m. ServiceNow -- MSL utilizes ServiceNow as itsHelp Desk application. This system covers a wide varietyof requests by employees for computer-related assistance(e.g., troubleshoot incidents, install software, etc.).

n. PeopleSoft -- MSL utilizes PeopleSoft, anOracle-based software product, to record employee datasuch as date of hire, date of termination, promotions,salary increases, transfers, [**50] etc. MSL has produceddata from this source and will consider producingadditional data in response to a specific inquiry fromPlaintiffs.

o. PRISM -- MSL utilizes PRISM for tracking timeand billing. It is used primarily to track an employee'sbillable time. MSL will consider producing additionaldata in response to a specific inquiry from Plaintiffs.

p. Portal -- MSL maintains a portal provided throughOracle/BEA Systems. The portal is web-based and isused for light workflow activities (such as reviewing draftdocuments).

q. Desktops/Laptops -- MSL provided employeeswith desktop and/or laptop computers to assist in workrelated activities. MSL will collect the desktop/laptophard drive data for 2 custodians and analyze the data todetermine the level of duplication of documents in thisdata source against the data contained in the EMC

SourceOne archive for the same custodians. (The partieswill meet and confer regarding the selection of the twocustodians.) The results of the analysis will be providedto Plaintiffs so that a determination can be made by theparties as to whether MSL will include this data source inits production of ESI to Plaintiffs. If so, the Parties willattempt [**51] to reach an agreement as to the approachused to collect, review and produce responsive andnon-privileged documents.

r. Publicis Benefits Connection -- Plaintiffsunderstand that MSL provides employees with access to acentralized web based site that provides access tocorporate benefits information and other related content.

s. GEARS -- MSL maintains a centralized web-basedexpense tracking and reporting system called "GEARS"where users are able to enter expenses and generatereports.

[*198] t. MS&L City -- MSL maintained acorporate web-based Intranet prior to migrating toNoovoo.

u. Adium -- This is a free and open source instantmessaging client for Mac OS X users.

v. Pidgin -- Pidgin is a chat program which lets userslog into accounts on multiple chat networkssimultaneously. However, the data resides with a thirdparty messaging provider (e.g. AIM, Yahoo!, GoogleTalk, MSN Messenger, etc.).

w. IBM Lotus Traveler and MobileIron -- MSLmaintains these systems for e-mail device sync andsecurity features for employees' mobile devices,including Blackberry devices, iPhones, iPads, Androidphones, and Android tablets.

x. Mobile Communication Devices - MSL providesmobile devices and/or connectivity [**52] includingBlackberry devices, iPhones, iPads, Android phones, andAndroid tablets to designated employees.

y. Yammer -- This is an instant messagingapplication hosted externally, used for approximately oneyear in or around 2008 through 2009.

z. SalesForce.com -- This is a web-based customerrelationship management application but it was notwidely used.



aa. Removable Storage Devices -- MSL does not restrictauthorized employees from using removable storagedevices.

D. Custodians

1. The Parties agree that MSL will search the e-mailaccounts of the following individuals as they exist onMSL's EMC SourceOne archive. (Except where a daterange is noted, the custodian's entire e-mail account wascollected from the archive.)

Custodian Name Title

1. Lund, Wendy Executive VP of Global Client

and Business Development

2. Fite, Vicki Managing Director, MSL Los Angeles

3. Wilson, Renee President, NE Region,

Managing Director NY

4. Brennan, Nancy (1/1/08 to 5/31/08) SVP/Director Corporate Branding

5. Lilien (Lillien, Kashanian), Tara SVP, North America Human Resources

6. Miller, Peter Executive Vice President, CFO

7. Masini, Rita Chief Talent Officer

8. Tsokanos, Jim President of the Americas

9. Da Silva Moore, Monique Director Healthcare Practice, Global

10. O'Kane, Jeanine (2/8/10 to 2/24/11) Director of Healthcare North America

11. Perlman, Carol Senior VP

12. Mayers, Laurie SVP MS&L Digital

13. Wilkinson, Kate Account Executive

14. Curran, Joel (5/1/08 to 5/31/10) Managing Director MSL Chicago

15. Shapiro, Maury North American CFO

16. Baskin, Rob (1/1/08 to 12/31/08) Managing Director

17. Pierce, Heather VP

18. Branam, Jud (1/1/08 to 1/31/10) Managing Director, MS&L Digital

19. McDonough, Jenni (1/1/08 VP, Director of Human Resources

to 12/31/08)

20. Hannaford, Donald (1/1/08 to 3/1/08) Managing Director

21. On, Bill (1/1/08 to 2/24/11) Managing Director

22. Dhillon, Neil (9/8/08 to 5/31/10) Managing Director MSL Washington DC

23. Hubbard, Zaneta Account Supervisor

24. Morgan, Valerie (1/1/08 to 2/24/11) HR Director

25. Daversa, Kristin (1/1/08 to 2/24/11) HR Director

26. Vosk, Lindsey (1/1/08 to 2/24/11) HR Manager

27. Carberry, Joe (1/1/08 to 2/24/11) President, Western Region

28. Sheffield, Julie (1/1/08 to 2/24/11) HR/Recruiting Associate



29. MaryEllen O'Donohue SVP (2010)

30. Hass, Mark CEO (former)

31. Morsman, Michael Managing Director, Ann Arbor (former)

[*199] E. [**53] Search Methodology1

1 As noted in Paragraphs A(1) and J of thisProtocol, Plaintiffs object to the predictive codingmethodology proposed by MSL.

1. General. The Parties have discussed themethodologies or protocols for the search and review ofESI collected from the EMC SourceOne archive and thefollowing is a summary of the Parties' agreement on theuse of Predictive Coding. This section relates solely tothe EMC SourceOne data source (hereinafter referred toas the "e-mail collection").

2. General Overview of Predictive Coding Process.MSL will utilize the Axcelerate software by Recommindto search and review the e-mail collection for productionin this case.

The process begins with Jackson Lewis attorneysdeveloping an understanding of the entire e-mailcollection while identifying a small number ofdocuments, the initial seed set, that is representative ofthe categories to be reviewed and coded (relevance,privilege, issue-relation). It is the step when the first seedsets are generated which is done by use of search andanalytical tools, including keyword, Boolean and conceptsearch, concept grouping, and, as needed, up to 40 otherautomatically populated filters available within [**54]the Axcelerate system. This assists in the attorneys'identification of probative documents for each categoryto be reviewed and coded.

Plaintiffs' counsel will be provided with preliminaryresults of MSL's hit counts using keyword searches tocreate a high priority relevant seed set, and will be invitedto contribute their own proposed keywords. Thereafter,Plaintiffs' counsel will be provided with thenon-privileged keyword hits -- both from MSL's keywordlist and Plaintiffs' keyword list -- which were reviewedand coded by MSL. Plaintiffs' counsel will review thedocuments produced and promptly provide defensecounsel with their own evaluation of the initial coding

applied to the documents, including identification of anydocuments it believes were incorrectly coded. To theextent the parties disagree regarding the coding of aparticular document, they will meet and confer in aneffort to resolve the dispute prior to contacting the Courtfor resolution. The irrelevant documents so producedshall be promptly returned after review and analysis byPlaintiffs' counsel and/or resolution of any disputes bythe Court.

The seed sets are then used to begin the PredictiveCoding process. Each [**55] seed set of documents isapplied to its relevant category and starts the software"training" process. The software uses each seed set toidentify and prioritize all substantively similar documentsover the complete corpus of the e-mail collection. Theattorneys then review and code a judgmental sample of atleast 500 of the "computer suggested" documents toensure their proper categorization and to further calibratethe system by recoding documents into their propercategories. Axcelerate learns from the new correctedcoding and the Predictive Coding process is repeated.

Attorneys representing MSL will have access to theentire e-mail collection to be searched and will lead thecomputer training, but they will obtain input fromPlaintiffs' counsel during the iterative seed selection andquality control processes and will share the informationused to craft the search protocol as further describedherein. All non-privileged documents reviewed by MSLduring each round of the iterative process (i.e., [*200]both documents coded as relevant and irrelevant) will beproduced to Plaintiffs' counsel during the iterative seedset selection process. Plaintiffs' counsel will review thedocuments produced [**56] and promptly providedefense counsel with its own evaluation of the initialcoding applied to the documents, including identificationof any documents it believes were incorrectly coded. Tothe extent the Parties disagree regarding the coding of aparticular document, they will meet and confer in aneffort to resolve the dispute prior to contacting the Courtfor resolution. Again, the irrelevant documents soproduced shall be promptly returned after review andanalysis by Plaintiffs' counsel and/or resolution of any



disputes by the Court.

At the conclusion of the iterative review process, alldocument predicted by Axcelerate to be relevant will bemanually reviewed for production. However, dependingon the number of documents returned, the relevancyrating of those documents, and the costs incurred duringthe development of the seed set and iterative reviews,MSL reserves the right to seek appropriate relief from theCourt prior to commencing the final manual review.

The accuracy of the search processes, both thesystems' functions and the attorney judgments to train thecomputer, will be tested and quality controlled by bothjudgmental and statistical sampling. In statisticalsampling, [**57] a small set of documents is randomlyselected from the total corpus of the documents to betested. The small set is then reviewed and an error ratecalculated therefrom. The error rates can then be reliablyprojected on the total corpus, having a margin of errordirectly related to the sample size.

3. Issue Tags. The parties agree that, to the extentapplicable, as part of the seed set training describedabove, as well as during the iterative review process, alldocuments categorized as relevant and not privileged, tothe extent applicable, also shall be coded with one ormore of the following agreed-upon issue tags:

a. Reorganization.

b. Promotion/Assignments.

c. Work/Life Balance.

d. Termination.

e. Compensation.

f. Maternity/Pregnancy.

g. Complaints/HR.

h. Publicis Groupe/Jurisdiction.

This issue coding will take place during the initialrandom sample, creation of the seed set and initial anditerative training (see paragraphs 4, 5 and 6 below). Thisinput shall be provided to Plaintiffs' counsel along withthe initial document productions. Plaintiffs' counsel shallpromptly report any disagreements on classification, andthe parties shall discuss these issues in good faith, so that

the seed [**58] set training may be improvedaccordingly. This issue-tagging and disclosure shall takeplace during the described collaborative seed set trainingprocess. The disclosures here made by MSL on its issuecoding are not required in the final production set.

4. Initial Random Sample. Using the Axceleratesoftware to generate a random sample of the entire corpusof documents uploaded to the Axcelerate search andreview platform, MSL's attorneys will conduct a reviewof the random sample for relevance and to develop abaseline for calculating recall and precision. To the extentapplicable, any relevant documents also will be codedwith one or more of the issue tags referenced inparagraph E.3 above. The random sample consists of2,399 documents, which represents a 95% confidencelevel with a confidence estimation of plus or minus 2%.The Parties agree to utilize the random sample generatedprior to the finalization of this protocol. However, duringPlaintiffs' counsel's review of the random sample, theymay advise as to whether they believe any of thedocuments should be coded with one or more of thesubsequently added issue codes (i.e., Complaints/HR andPublicis Groupe/Jurisdiction) and [**59] will, asdiscussed above, indicate any disagreement with MSL'sclassifications.

5. Seed Set.

a. Defendant MSL. To create the initial seed set ofdocuments that will be used to "train" the Axceleratesoftware as described [*201] generally above, MSLprimarily utilized keywords listed on Exhibits A and B tothis protocol, but also utilized other judgmental analysisand search techniques designed to locate highly relevantdocuments, including the Boolean, concept search andother features of Axcelerate. Given the volume of hits foreach keyword (Exhibit A), MSL reviewed a sampling ofthe hits and coded them for relevance as well as for thefollowing eight preliminary issues: (i) Reorganization;(ii) Promotion; (iii) Work/Life Balance; (iv) Termination;(v) Compensation; and (vi) Maternity. Specifically,except for key words that were proper names, MSLperformed several searches within each set of key wordhits and reviewed a sample of the hits. The Axceleratesoftware ranked the hits in order of relevance based onthe software's analytical capabilities and the documentswere reviewed in decreasing order of relevance (i.e., eachreview of the sample of supplemental searches startedwith the highest [**60] ranked documents). Exhibit B



identifies the supplemental searches conducted, thenumber of hits, the number of documents reviewed, thenumber of documents coded as potentially responsive andgeneral comments regarding the results. In addition, tothe extent applicable, documents coded as responsive alsowere coded with one or more issue tags. MSL will repeatthe process outlined above and will include the newlydefined issues and newly added custodians. MSL willprovide Plaintiffs' counsel with all of the non-privilegeddocuments and will provide, to the extent applicable, theissue tag(s) coded for each document, as described above.Plaintiffs' counsel shall promptly review and providenotice as to any documents with which they disagreewhere they do not understand the coding. If necessary,counsel will meet and confer to attempt to resolve anydisagreements regarding the coding applied to thedocuments in this seed set.

b. Plaintiffs. To help create the initial seed set ofdocuments that will be used to "train" the Axceleratesoftware, Plaintiffs provided a list of potential key wordsto MSL. MSL provided Plaintiffs with a hit list for theirproposed key words. This process was repeated [**61]twice with the hit list for Plaintiffs' most recent set ofkeywords attached as Exhibit C. MSL will review 4,000randomly sampled documents from Plaintiffs'supplemental list of key words to be coded for relevanceand issue tags. MSL will provide Plaintiffs' counsel withall non-privileged documents and will provide, to theextent applicable, the issue tag(s) coded for eachdocument. Plaintiffs' counsel shall promptly review andprovide notice as to any documents with which theydisagree with or where they do not understand the coding.If necessary, the Parties' counsel will meet and confer toattempt to resolve any disagreements regarding thecoding applied to the documents in this seed set.

c. Judgmental Sampling. In addition to the above, anumber of targeted searches were conducted by MSL inan effort to locate documents responsive to several ofPlaintiffs' specific discovery requests. Approximately 578documents have already been coded as responsive andproduced to Plaintiffs. In addition, several judgmentalsearches were conducted which resulted in approximately300 documents initially being coded as responsive andseveral thousand additional documents coded asirrelevant. The [**62] documents coded as relevant andnon-privileged also will be reviewed by Plaintiffs'counsel and, subject to their feedback, included in theseed set. An explanation shall be provided by MSL's

attorneys for the basis of the bulk tagging of irrelevantdocuments (primarily electronic periodicals andnewsletters that were excluded in the same manner asspam junk mail is excluded). The explanation shallinclude the types of documents bulk tagged as irrelevantas well as the process used to identify those types ofdocuments and other similar documents that were bulktagged as irrelevant.

6. Initial And Iterative Training. Following thecreation of the first seed set, the Axcelerate software willreview the entire data set to identify other potentiallyrelevant documents. MSL will then review and tag ajudgmental based sample, consisting of a minimum of500 documents, including all documents ranked as highlyrelevant or hot, of the new "Computer Suggested"documents, [*202] which were suggested by theAxcelerate software. MSL's attorneys shall act inconsultation with the Axcelerate software experts tomake a reasonable, good faith effort to select documentsin the judgmental sample that will serve [**63] toenhance and increase the accuracy of the predictivecoding functions. The results of this first iteration, boththe documents newly coded as relevant and not relevantfor particular issue code or codes, will be provided toPlaintiffs' counsel for review and comment. (Alldocuments produced by the parties herein to each other,including, without limitation, these small seed setdevelopment productions, shall be made under theConfidentiality Stipulation in this matter as well as anyclawback agreement that shall be reduced to an orderacceptable to the Court. Any documents marked asirrelevant shall be returned to counsel for MSL at theconclusion of the iterative training phase, unless therelevancy of any documents are disputed, in which casethey may be submitted to the Court for review.)

Upon completion of the initial review, and anyrelated meet and confer sessions and agreed upon codingcorrections, the Axcelerate software will be run againover the entire data set for suggestions on otherpotentially relevant documents following the sameprocedures as the first iteration. The purpose of thissecond and any subsequent iterations of the PredictiveCoding process will be to further [**64] refine andimprove the accuracy of the predictions on relevance andvarious other codes. The results of the second iterationshall be reviewed and new coding shared with Plaintiffs'counsel as described for the first iteration. This processshall be repeated five more times, for a total of seven



iterations, unless the change in the total number ofrelevant documents predicted by the system as a result ofa new iteration, as compared to the last iteration, is lessthan five percent (5%), and no new documents are foundthat are predicted to be hot (aka highly relevant), at whichpoint MSL shall have the discretion to stop the iterativeprocess and begin the final review as next described. Ifmore than 40,000 documents are returned in the finaliteration, then MSL reserves the right to apply to theCourt for relief and limitations in its review obligationshereunder. Plaintiffs reserve the right, at all times, tochallenge the accuracy and reliability of the predictivecoding process and the right to apply to the Court for areview of the process.

7. Final Search and Production. All of the documentspredicted to be relevant in the final iteration described inparagraph six above will be [**65] reviewed by MSL,unless it applies to the court for relief hereunder. Alldocuments found by MSL's review to be relevant andnon-privileged documents will be promptly produced toPlaintiffs. If more than 40,000 documents are included inthe final iteration, then MSL reserves its right to seekpayment from Plaintiffs for all reasonable costs and feesMSL incurred related to the attorney review andproduction of more 40,000 documents.

8. Quality Control by Random Sample of IrrelevantDocuments. In addition, at the conclusion of this searchprotocol development process described above, andbefore the final search and production described inParagraph 7 above, MSL will review a random sample of2,399 documents contained in the remainder of thedatabase that were excluded as irrelevant. The results ofthis review, both the documents coded as relevant and notrelevant, but not privileged, will be provided to Plaintiffs'counsel for review. (Any documents initially coded as"not relevant" will be provided subject to theConfidentiality Stipulation and any clawback agreementsentered in this matter will be returned to counsel for MSLwithin 60 days of their production.) The purpose for this[**66] review is to allow calculation of the approximatedegree of recall and precision of the search and reviewprocess used. If Plaintiffs object to the proposed reviewbased on the random sample quality control results, orany other valid objection, they shall provide MSL withwritten notice thereof within five days of the receipt ofthe random sample. The parties shall then meet andconfer in good faith to resolve any difficulties, and failingthat shall apply to the Court for relief. MSL shall not be

required to proceed with the final search and reviewdescribed in Paragraph [*203] 7 above unless and untilobjections raised by Plaintiffs have been adjudicated bythe Court or resolved by written agreement of the Parties.

F. Costs

1. MSL proposes to limit the costs of its final reviewand production of responsive ESI from the MSL emailcollection to an additional $200,000, above and beyondthe approximately $350,000 it has already paid or isanticipated to pay in e-discovery related activities aspreviously described and disclosed to Plaintiffs.

2. Plaintiffs agree to bear all of the costs associatedwith their compliance with the terms of this protocol andwith the receipt and review of ESI produced [**67]hereunder including the costs associated with its ESIexperts at DOAR Litigation Consulting who will beinvolved with Plaintiffs in all aspects of this ESIprotocol. Plaintiffs propose that MSL bear all of the costsassociated with its obligations under the terms of thisprotocol and do not agree to limit the amount ofinformation subject to the review and production of ESIby MSL.

G. Format of Production For Documents Produced FromAxcelerate

1. TIFF/Native File Format Production. Documentswill be produced as single-page TIFF images withcorresponding multi-page text and necessary load files.The load files will include an image load file as well as ametadata (.DAT) file with the metadata fields identifiedon Exhibit D. Defendant MSL will produce spreadsheets(.xls files) and PowerPoint presentations (.ppt files) innative form as well as any documents that cannot beconverted to TIFF format (e.g., audio or video files, suchas mp3s, ways, megs, etc.). In addition, for any redacteddocuments that are produced, the documents' metadatafields will be redacted where required. For the productionof ESI from non-email sources, the parties will meet andconfer to attempt to reach an agreement [**68] of theformat of production.

2. Appearance. Subject to appropriate redaction, eachdocument's electronic image will convey the sameinformation and image as the original document.Documents that present imaging or formatting problemswill be promptly identified and the parties will meet andconfer in an attempt to resolve the problems.



3. Document Numbering. Each page of a produceddocument will have a legible, unique page identifier"Bates Number" electronically "burned" onto the imageat a location that does not obliterate, conceal or interferewith any information from the source document. TheBates Number for each page of each document will becreated so as to identify the producing party and thedocument number. In the case of materials redacted inaccordance with applicable law or confidential materialscontemplated in any Confidentiality Stipulation enteredinto by the parties, a designation may be "burned" ontothe document's image at a location that does not obliterateor obscure any information from the source document.

4. Production Media. The producing party willproduce documents on readily accessible, computer orelectronic media as the parties may hereafter agree upon,[**69] including CD-ROM, DVD, external hard drive(with standard PC compatible interface), (the "ProductionMedia"). Each piece of Production Media will beassigned a production number or other unique identifyinglabel corresponding to the date of the production ofdocuments on the Production Media (e.g., "DefendantMSL Production April 1, 2012") as well as the sequenceof the material in that production (e.g. "-001", "-002").For example, if the production comprises documentimages on three DVDs, the producing party may labeleach DVD in the following manner "Defendant MSLProduction April 1, 2012", "Defendant MSL ProductionApril 1, 2012-002", "Defendant MSL Production April 1,2012-003." Additional information that will be identifiedon the physical Production Media includes: (1) textreferencing that it was produced in da Silva Moore v.Publicis Groupe SA, et al.; and (2) the Bates Numberrange of the materials contained on the ProductionMedia. Further, any replacement [*204] ProductionMedia will cross-reference the original Production Mediaand clearly identify that it is a replacement andcross-reference the Bates Number range that is beingreplaced.

5. Write Protection and Preservation. [**70] Allcomputer media that is capable of write-protection shouldbe write-protected before production.

6. Inadvertent Disclosures. The terms of the Parties'Clawback Agreement and Court Order shall apply to thisprotocol.

7. Duplicate Production Not Required. A partyproducing data in electronic form need not produce the

same document in paper format.

H. Timing.

I. To the extent a timeframe is not specificallyoutlined herein, the parties will use their reasonableefforts to produce ESI in a timely manner consistent withthe Court's discovery schedule.

2. The parties will produce ESI on a rolling basis.

I. General Provisions.

1. Any practice or procedure set forth herein may bevaried by agreement of the parties, and first will beconfirmed in writing, where such variance is deemedappropriate to facilitate the timely and economicalexchange of electronic data.

2. Should any party subsequently determine it cannotin good faith proceed as required by this protocol, theparties will meet and confer to resolve any dispute beforeseeking Court intervention.

3. The Parties agree that e-discovery will beconducted in phases and, at the conclusion of the searchprocess described in Section E above, the [**71] Partieswill meet and confer regarding whether further searchesof additional custodians and/or the Phase II sources iswarranted and/or reasonable. If agreement cannot bereached, either party may seek relief from the Court.

J. Plaintiffs' Objection

1. Plaintiffs object to this ESI Protocol in its entirety.Plaintiffs submitted their own proposed ES! Protocol tothe Court, but it was largely rejected. The Court thenordered the parties to submit a joint ES! Protocolreflecting the Court's rulings. Accordingly, Plaintiffsjointly submit this ESI Protocol with MSL, but reservethe right to object to its use in this case.

This protocol may be executed in counterparts. Eachcounterpart, when so executed, will be deemed andoriginal, and will constitute the same instrument.

By:

JANETTE WIPPER, ESQ.

DEEPIKA BAINS, ESQ.

SIHAM NURHUSSEIN, ESQ.



SANFORD WITTELS & HEISLER, LLP

Attorneys for Plaintiffs and Class

555 Montgomery Street, Ste. 1206

San Francisco, CA 94111

Telephone: (415) 391-6900

Date: , 2012

By:

BRETT M. ANDERS, ESQ.

VICTORIA WOODIN CHAVEY, ESQ.

JEFFREY W. BRECHER, ESQ.

JACKSON LEWIS LLP

Attorneys for Defendant MSLGROUP

58 South Service Road, Suite 410

Melville, NY 11747

Telephone: (631) [**72] 247-0404

Date: , 2012



ROBOCAST, INC., Plaintiff, v. APPLE, INC., Defendant. ROBOCAST, INC.,Plaintiff, v. MICROSOFT CORPORATION, Defendant.

Civil Action No. 11-235-RGA,Civil Action No. 10-1055-RGA

UNITED STATES DISTRICT COURT FOR THE DISTRICT OF DELAWARE

2012 U.S. Dist. LEXIS 24879

February 24, 2012, Decided

SUBSEQUENT HISTORY: Related proceeding atRobocast, Inc. v. Microsoft Corp., 2013 U.S. Dist.LEXIS 52619 (D. Del., Apr. 12, 2013)

COUNSEL: [*1] Thomas C. Grimm, Esq., Wilmington,Delaware; Steven J. Rizzi, Esq. (argued), New York,New York, Attorneys for Plaintiff.

Richard L. Horwitz, Esq., Wilmington, Delaware;Harrison J. Frahn IV, Esq. (argued), Palo Alto,California, Attorneys for Defendant Apple, Inc.

Thomas Lee Halkowski, Esq., Wilmington, Delaware;Adam J. Kessel, Esq. (argued), Boston, Massachusetts,Attorneys for Defendant Microsoft Corporation.

JUDGES: Richard G. Andrews, U.S. DISTRICTJUDGE.

OPINION BY: Richard G. Andrews

OPINION

MEMORANDUM OPINION

/s/ Richard G. AndrewsANDREWS, U.S. DISTRICT JUDGE:

Before the Court are motions to transfer these twoseparate cases to the Northern District of California.

On December 6, 2010, Robocast filed a Complaintagainst Microsoft. Microsoft duly filed an Answer. OnMarch 21, 2011, Robocast sued Apple. Apple respondedby filing a motion to transfer. (D.I. 16). 1 Microsoftsubsequently filed a similar motion to transfer. (No.10-1055, D.I. 17). Microsoft states forthrightly (and, inthe Court's opinion, correctly) that its transfer motionrises or falls with Apple's. (No. 10-1055, D.I. 27, pp.1-2).

1 Unless otherwise indicated, Docket Item("D.I.") citations are to filings in Robocast v.Apple, Civil Act. No. 11-235-RGA. [*2]Citations to the record in Robocast v. Microsoftinclude the case number ("10-1055").

The Complaints at issue are similar. They are eachbased on Robocast's ownership of U.S. patent No.7,155,451, which concerns an "Automated BrowsingSystem for Publishers and Users on Networks ServingInternet and Remote Devices." It was invented byRobocast's President, Damon Torres. Apple productssuch as AppleTV, Front Row, and iTunes are said toinfringe the patent. Microsoft products such as Bing andWindows Vista are also said to infringe the patent.

The transfer motions have been fully briefed andorally argued.

The statutory authority for transferring the case is §1404(a) of Title 28, which provides: "For the

Page 1

convenience of parties and witnesses, in the interest ofjustice, a district court may transfer any civil action toany other district or division where it might have beenbrought." The burden of establishing the need for transferis the movant's, see Jumara v. State Farm Ins. Co., 55F.3d 873, 879 (3d Cir. 1995), which in this case is Apple.2 The Third Circuit has set forth the framework foranalysis:

"[I]n ruling on defendants' motion theplaintiff's choice of venue should not belightly [*3] disturbed."

In ruling on § 1404(a) motions, courtshave not limited their consideration to thethree enumerated factors in § 1404(a)(convenience of parties, convenience ofwitnesses, or interests of justice), and,indeed, commentators have called on thecourts to "consider all relevant factors todetermine whether on balance thelitigation would more convenientlyproceed and the interests of justice bebetter served by transfer to a differentforum." While there is no definitiveformula or list of the factors to consider,courts have considered many variants ofthe private and public interests protectedby the language of § 1404(a).

The private interests have included:(1) plaintiff's forum preference asmanifested in the original choice; (2) thedefendant's preference; (3) whether theclaim arose elsewhere; (4) theconvenience of the parties as indicated bytheir relative physical and financialcondition; (5) the convenience of thewitnesses-but only to the extent that thewitnesses may actually be unavailable fortrial in one of the fora; and (6) the locationof books and records (similarly limited tothe extent that the files could not beproduced in the alternative forum).

The public [*4] interests haveincluded: (7) the enforceability of thejudgment; (8) practical considerations thatcould make the trial easy, expeditious, orinexpensive; (9) the relative administrative

difficulty in the two fora resulting fromcourt congestion; (10) the local interest indeciding local controversies at home; (11)the public policies of the fora; and (12) thefamiliarity of the trial judge with theapplicable state law in diversity cases.

Id. at 879-80 (citations omitted and numbering added).

2 Given Microsoft's acknowledgement that itsmotion is dependent upon Apple's, the analysiswill be limited to Apple's motion.

There is no dispute that the patent infringementaction against Apple could have been brought in theNorthern District of California, as it is a Californiacorporation with its principal place of business in theNorthern District of California.

In my view, interests (1) and (4) support theplaintiff's position that the case should not be transferred.Interests (2), (5), (6), and (8) support the defendant'srequest to transfer the case. Interests (3), (7), (10), (11)and (12) do not add much to the balancing, as they areeither inapplicable or marginally applicable to this [*5]case. Interest (9) is applicable but does not favor one sideor the other.

Plaintiff has chosen Delaware as a forum. Thatchoice weighs strongly in the plaintiff's favor, althoughnot as strongly as it would if the plaintiff had its principalplace of business (or, indeed, any place of business) inDelaware. See Shutte v. Armco Steel Corp., 431 F.2d 22,25 (3d Cir. 1970) ("plaintiff's choice of a proper forum isa paramount consideration in any determination of atransfer request"); Pennwalt Corp. v. Purex Industries,Inc., 659 F.Supp. 287, 289 (D.Del. 1986) (plaintiff'schoice of forum not as compelling if it is not plaintiff's"'home turf'"). 3 Robocast's principal place of business isin the State of New York. It is a Delaware corporation,and, while I think that gives Robocast a legitimate reasonto sue in Delaware, 4 I think that is also a reason that addsto the weight given its choice to sue in Delaware.

3 I assume, without deciding, that Delaware isnot Robocast's "home turf." Cf. IntellectualVentures I LLC v. Altera Corp., 2012 U.S. Dist.LEXIS 70653, 2012 WL 297720, *6-7 (D.Del.Jan. 23, 2012) (holding state of incorporation is"home turf" but also analyzing in the alternative).4 At oral argument, Robocast [*6] also offered

Page 22012 U.S. Dist. LEXIS 24879, *2

as a legitimate reason for suing in Delaware theexpertise that it would expect to encounter from aCourt that has a heavy patent caseload.

Defendant's preference is the Northern District ofCalifornia, where it has its principal place of business.Clearly, Defendant's decision to seek to litigate in theNorthern District of California is rational and legitimatealso. 5

5 At oral argument, I stated that I did not viewApple's decisions to sue in Delaware in othercases, or not to seek transfer from Delaware inother cases in which it was named as a defendant,as having any relevance to the instant decision.(D.I. 40, pp. 7-8). If there were a question aboutApple's capability of litigating a case in Delaware,such information might be relevant, but capabilityis not at issue.

Although the defendant has argued in its briefing thatthe claim did not arise in Delaware (D.I. 17, p.9), and thatthe Apple products that are alleged to infringe weredesigned, developed and marketed in the NorthernDistrict of California, I think the claim that is relevanthere is the plaintiff's claim that Apple's products, whichare sold and offered for sale all over the United States,including [*7] Delaware (D.I. 18, ¶ 3), infringed itspatent. Thus, I think the claims arise in every judicialdistrict. See In re Acer America Corp., 626 F.3d 1252,1256 (Fed. Cir. 2010). Consequently, this factor has noweight in the balancing. 6

6 Of course, the fact that Apple's engineers andmarketers for these products are in NorthernCalifornia is not irrelevant. It merely is analyzedas part of a different factor.

Apple is omnipresent in everyday life. It is a largeand powerful corporation. Robocast consists of itsPresident and two employees. (D.I. 24, ¶ 10). Its founderwas at the oral argument on the present motion. Itsfinancial condition pales in comparison to that of Apple.Ten years ago, it had twenty-two employees. (D.I. 24, ¶5). While Robocast's primary enterprise today may belitigation, there is little reason to believe that its pocketsare deep. (D.I. 24, ¶¶ 8, 10). "We could not even afford tofile this lawsuit, except on a contingency basis." (D.I. 24,¶ 13). There is no reason to doubt that if this litigationturns into a war of attrition, Apple will have the upperhand. I think this factor significantly disfavors transfer.

At this juncture, it is hard to tell who the witnesses[*8] might be. It seems likely that a significant number ofthe non-expert witnesses will be employees of Apple. SeeIn re Genentech, 566 F.3d 1338, 1345 (Fed. Cir. 2009)("In patent infringement cases, the bulk of the relevantevidence usually comes from the accused infringer.").Two current Apple employees (who live in California)with personal knowledge of relevant events are identifiedfrom the Complaint. (D.I. 1, ¶¶ 18-19; D.I. 19, ¶ 12).Apple has submitted the last known locations for 75people involved in the prior art references cited duringthe prosecution of the '451 patent. (D.I. 19, ¶ 7). Of thosepeople, 25 are in California and subject to the NorthernDistrict of California's subpoena power; 4 are or might be7 subject to this Court's subpoena power, and 46 aresubject to neither Court's subpoena power. 8 (D.I. 19-1,pp. 14-17). Robocast has identified two former Robocastemployees with purportedly relevant information (D.I.24, ¶¶ 14 - 16), and while they live near Delaware, theydo not appear to be within this Court's subpoena power.Even assuming their information is relevant, and I am notconvinced of that, their existence does not aid Robocast'sargument. Unfortunately, [*9] a patent case is not like acar accident, where the fact witnesses -- or likely trialwitnesses -- can be identified from day one. All that canbe concluded about likely trial witnesses is that a yearand a half to two years from now, more of them are morelikely to be in California than in Delaware or within 100miles of Wilmington. If there is a trial, 9 and if there areone or more necessary witnesses who will only testify ifsubpoenaed, there is a statistically greater likelihood thatsuch witnesses would be within the subpoena power ofthe Northern District of California than within thesubpoena power of the District of Delaware. No likelywitnesses who would be unavailable in one location butnot the other have been identified. The odds, however,are that such people exist, and that they are in California.Thus, this factor slightly favors transfer.

7 Two are listed as being in Washington, D.C.,and, depending upon their exact location, mightbe within or without this Court's subpoena power.Two are listed as being in New Jersey and wouldappear to be within this Court's subpoena power.8 A dozen of them are listed as having lastknown addresses in the State of Washington, butwhether [*10] some or all of them are current orformer employees of Microsoft is not described.9 As noted in Intellectual Ventures I LLC v.Altera Corp., 2012 U.S. Dist. LEXIS 70653, 2012


WL 297720, *10-11 (D.Del. Jan. 23, 2012), a trialis an unlikely event.

Similarly, it is hard to tell where most of the booksand records would be (other than Plaintiff's and Apple's).The only identified holders of records are the parties, andtheir records will be able to be produced in whicheverforum has the case. Robocast's President has seventyboxes of hard copy documents and ten boxes of computerhardware material in New York City. (D.I. 24, ¶ 12). Therecords of Apple relating to its accused products, whichare most likely the most important records for thislitigation, see In re Genentech, 566 F.3d at 1345, are inthe Northern District of California (D.I. 18, ¶ 2), and notin Delaware. Apple also suggests that some of theevidence in this case is sufficiently old that there mayneed to be some effort to retrieve it, and that the materialsto do so might be in the Computer History Museum orother locations in Northern California. (D.I. 29, p. 8).Such an argument is fairly speculative. In any event,there are no records identified [*11] as only beingavailable for trial in one of the two locations. 10 Thus,this factor favors transfer, but is of marginal weight giventhat it is likely that Apple's records can be produced attrial wherever trial is held.

10 While there is a paragraph in In reLink_A_Media Devices Corp., 662 F.3d 1221,1224 (Fed. Cir. 2011), about "the convenience ofthe witnesses and the location of the books andrecords," I do not understand the Federal Circuitto have altered the Third Circuit's focus on theissue being not so much where the witnesses andevidence are, but whether they can be produced incourt.

Enforceability of the judgment is not an issue.

Practical considerations that could make the trialeasy, expeditious, or inexpensive favor the NorthernDistrict of California. A trial in Delaware is likely to bemarginally less easy and more expensive for Apple than atrial in the Northern District of California, because oftravel considerations and expenses for witnesses. A trialin the Northern District of California is likely to bemarginally less easy and more expensive for Robocastthan a trial in Delaware. On balance, the overall cost andinconvenience of a trial in the Northern District [*12] ofCalifornia is likely to be less than it would be inDelaware, but not by a significant amount. Expedition,however, does not seem to depend upon the location of

the trial. Thus, this factor slightly favors transfer.

The relative administrative difficulty due to courtcongestion is difficult to assess. Plaintiff and Defendantboth cite statistics suggesting greater court congestion inthe court in which it does not want to try the case. I donot find any of the cited statistics to be meaningful inpredicting what would happen with this case, either inDelaware or the Northern District of California. I wouldalso note that, in my brief experience, "court congestion"has not caused any delay in assigning trial dates inDelaware. I therefore believe this factor is neutral.

The "local controversy" consideration is inapplicablehere. Apple is a major employer in the Northern Districtof California, but I do not think that makes the claim thatApple is infringing the patent of a Delaware companywith a principal place of business in New York into alocal California controversy. Thus, I think this factor isneutral.

The public policy of Delaware encourages the use byDelaware corporations -- [*13] such as Robocast -- ofDelaware as a forum for the resolution of businessdisputes. 11 Typically, the forum is the Delaware Court ofChancery. More generally, though, Delaware wantscorporations to incorporate in Delaware, and the taxesthat corporations pay are important to Delaware'sfinancial health. While it is not unusual that a Delawarecorporation should sue another corporation in a Delawarecourt, and is a factor that has some impact on thebalancing, cf. Micron Technology v. Rambus, 645 F.3d1311, 1332 (Fed. Cir. 2011)("Given that both partieswere incorporated in Delaware, they had both willinglysubmitted to suit there, which weighs in favor of keepingthe litigation in Delaware."), this factor significantlyoverlaps with why Robocast would have chosenDelaware as a venue in the first place, and is lesssignificant than in Micron since Apple is not a Delawarecorporation. I do not attribute it independent weight as a"public policy" factor. I believe it is taken into account inthe first factor.

11 The State of Delaware Division ofCorporations, on its website, prominently asks,"Why Choose Delaware as Your CorporateHome?" The answer follows: "More than 900,000business entities [*14] have their legal home inDelaware including more than 50% of all U.S.publicly-traded companies and 63% of theFortune 500. Businesses choose Delaware


because we provide a complete package ofincorporation services including modern andflexible corporate laws, our highly-respectedCourt of Chancery, a business-friendly StateGovernment, and the customer service-orientedStaff of the Delaware Division of Corporations."

This is not a diversity case, and thus knowledge ofstate law is irrelevant here.

Apple states that Robocast's connection withDelaware is entitled to less weight because it is a"non-practicing entit[y]." (D.I. 17, p.8). 12 Robocast wasincorporated under the laws of the State of Delaware onJanuary 10, 1997. (D.I. 24, ¶ 4). While it has no physicalconnection to Delaware, its corporate citizenship is oflong standing. There might be cases in which anon-practicing entity's connections to Delaware suggestreasons to discount that connection, cf. In re ZimmerHoldings, Inc., 609 F.3d 1378 (Fed. Cir. 2010) (accordingno weight to the plaintiff's choice of venue when its"presence in [that venue] appears to be recent, ephemeral,and an artifact of litigation."), but this [*15] is not one ofthem.

12 Robocast disputes whether it is a"non-practicing" entity, pointing to products it hasdeveloped in the past. (D.I. 24, ¶¶ 5-6).

I have considered In re Link_A_Media DevicesCorp., 662 F.3d 1221 (Fed. Cir. 2011), 13 but I do notthink it is particularly helpful in assessing the transferrequest in this case, as its facts were very different. Iwould characterize that case generally as standing for theproposition that when the parties, all the witnesses, andall the evidence are in one distant jurisdiction, and theonly connection to Delaware is that it is the state ofincorporation of the defendant, and there is no otherreason for the suit to be in Delaware, the suit must betransferred, upon timely request, to the distantjurisdiction. In the present case there is a greaterconnection to Delaware, as the plaintiff is a Delawarecorporation with its principal place of business close to

Delaware but not to California, and, also of considerablesignificance, the plaintiff would be inconvenienced bytransfer.

13 The Federal Circuit's numerous transfer casesarising from the Fifth Circuit are not controllingas the Federal Circuit interprets the law of theCircuit in [*16] which the District Court sits. SeeIn re Link_A_Media Devices Corp., 662 F.3d at1223. The law of the two Circuits in regard tohow to conduct a transfer analysis is different in anumber of regards. Of greatest relevance, "FifthCircuit precedent clearly forbids treating theplaintiff's choice of venue as a distinct factor inthe [transfer] analysis." In re TS Tech USA Corp.,551 F.3d 1315, 1320 (Fed. Cir. 2008). As notedearlier, the Third Circuit treats the plaintiff'schoice as a factor of "paramount importance."Additional differences are identified inIntellectual Ventures I LLC v. Altera Corp., 2012U.S. Dist. LEXIS 70653, 2012 WL 297720, *13n.7 (D.Del. Jan. 23, 2012).

Under Third Circuit law, considerable deference isgiven to the plaintiffs' choice of forum. I think that whenthe plaintiff is a three-person corporation with Delawareas its longstanding corporate home, and the defendant isApple, see Intellectual Ventures I LLC v. Altera Corp.,2012 U.S. Dist. LEXIS 70653, 2012 WL 297720, *3(D.Del. Jan. 23, 2012) (multi-billion dollar companiesdoing business on an international scale have a greaterburden to meet in seeking transfer), there ought to be acompelling reason to overcome plaintiff's choice offorum. Apple has offered [*17] valid reasons, but I donot think Apple has shown that the balance ofconvenience tips strongly enough in Apple's favor so thattransfer should be ordered. I will therefore deny Apple'smotion to transfer. Consequently, I will also denyMicrosoft's motion to transfer.

An appropriate order will be entered.


In Re: Actos (Pioglitazone) Products Liability Litigation. This Document Applies to:All Cases

MDL NO. 6:11-md-2299

UNITED STATES DISTRICT COURT FOR THE WESTERN DISTRICT OFLOUISIANA

2012 U.S. Dist. LEXIS 187519

July 27, 2012, DecidedJuly 27, 2012, Filed

COUNSEL: [*1] For Gary J Russo, Special Master:Gary J Russo, LEAD ATTORNEY, Carmen MRodriguez, Jones Walker et al (LAF), Lafayette, LA;Kenneth W DeJean, Office of Kenneth W DeJean,Lafayette, LA.

For Executive Committee, Plaintiff: Paul J Pennock,LEAD ATTORNEY, Weitz & Luxenberg (NY), NewYork, NY; Richard J Arsenault, LEAD ATTORNEY,Neblett Beard & Arsenault, Alexandria, LA; DawnBarrios, Barrios Kingsdorf & Casteix, New Orleans, LA;Mark P Robinson, Jr, Robinson Calcagnie et al, NewportBeach, CA; Patrick C Morrow, Morrow Morrow et al,Opelousas, LA; Stephanie O'Connor, Douglas & London,New York, NY.

For Plaintiffs Steering Committee, Plaintiff: J Pennock,LEAD ATTORNEY, Weitz & Luxenberg (NY), NewYork, NY; Richard J Arsenault, LEAD ATTORNEY,Neblett Beard & Arsenault, Alexandria, LA; Andy DBirchfield, Beasley Allen et al, Montgomery, AL;Christopher A Seeger, Seeger Weiss, New York, NY;Dianne M Nast, Nast Law, Philadelphia, PA; JayneConroy, Hanly Conroy et al, New York, NY; Jerrold SParker, Parker Waichman (FL), Bonita Springs, FL; NeilD Overholtz, Aylstock Witkin & Sasser (Pensacola),Pensacola, FL; Nicholas J Drakulich, Drakulich Firm,San Diego, CA; Stephen B Murray, Murray Law Firm,[*2] New Orleans, LA; Troy A Rafferty, Levin

Papantonio et al, Pensacola, FL; Vance R Andrus,Andrus Hood & Wagstaff, Denver, CO; W Mark Lanier,Law Offices of W Mark Lanier, Houston, TX; W JamesSingleton, Singleton Law Firm, Shreveport, LA.

For Takeda Pharmaceutical Co Ltd, Defendant: Sara JGourley, LEAD ATTORNEY, Jennifer A Foster, NathanA Huey, Sherry Ann Knutson, Sidley Austin (Chicago),Chicago, IL; Amy Sherry Fischer, Foliart Huff et al,Oklahoma City, OK; Bruce R Parker, Venable (MD),Baltimore, MD; D'Lesli M Davis, H Douglas Wabner,Jonathan B Skidmore, Katherine P Lett, Krtistin Turner,Fulbright & Jaworski (DAL), Dallas, TX; Dell PChappell, John D Martin, Nelson Mullins et al (COL),Columbia, SC; Donna L Boyce, Adams & Reese (MEM),Memphis, TN; E Paige Sensenbrenner, Jaimme AngelleCollins, Lawton Cole Callihan, Megan Haggerty Guy,Kathleen Fontenot Drew, Adams & Reese (NO), NewOrleans, LA; Jennifer Blakely D Kiefer, Stephen DMartin, Nelson Mullins et al (NC), Raleigh, NC; John EMcElligott, Jr, Davidson Meaux et al, Lafayette, LA;Jude Philip Calabrese, Squire Sanders & Dempsey (OH),Cleveland, OH; Randall Shirres Richardson, Fulbright &Jaworski (HOU), Houston, TX; Richard [*3] C Harlan,Alycia Degen, Sidley Austin (LA), Los Angeles, CA;Gerald J Corcoran, Montgomery McCracken et al (NJ),Linwood, NJ; James M Campbell, Campbell Campbell etal (MA), Boston, MA; Robert A Limbacher, GoodellDeVries et al (PA), Philadelphia, PA.

Page 1

For Takeda Pharmaceuticals America Inc, Defendant:Sara J Gourley, LEAD ATTORNEY, Jennifer A Foster,Nathan A Huey, Sherry Ann Knutson, Sidley Austin(Chicago), Chicago, IL; Alycia Degen, Richard C Harlan,Catherine Valerio Barrad, Sidley Austin (LA), LosAngeles, CA; Bruce R Parker, Venable (MD), Baltimore,MD; Carol Dan Browning, Stites & Harbison, Lexington,KY; Colin J Garry, James D Arden, Sidley Austin (NY),New York, NY; D'Lesli M Davis, H Douglas Wabner,Jonathan B Skidmore, Katherine P Lett, Krtistin Turner,Fulbright & Jaworski (DAL), Dallas, TX; David J Walz,Edward Walter Gerecke, Carlton Fields (FL), Tampa, FL;Dell P Chappell, John D Martin, Nelson Mullins et al(COL), Columbia, SC; Donna L Boyce, Adams & Reese(MEM), Memphis, TN; E Paige Sensenbrenner, JaimmeAngelle Collins, Lawton Cole Callihan, Megan HaggertyGuy, Kathleen Fontenot Drew, Adams & Reese (NO),New Orleans, LA; Jennifer Y Dukart, Joseph M Price,Faegre & [*4] Benson (MN), Minneapolis, MN; JenniferBlakely D Kiefer, Stephen D Martin, Nelson Mullins et al(NC), Raleigh, NC; John E McElligott, Jr, DavidsonMeaux et al, Lafayette, LA; Jude Philip Calabrese, SquireSanders & Dempsey (OH), Cleveland, OH; Mary NoldLarimore, Nancy Menard Riddle, Ice Miller et al,Indianapolis, IN; Randall Shirres Richardson, Fulbright& Jaworski (HOU), Houston, TX; Stephen M Brooks,Nelson Mullins et al (Atlanta), Atlanta, GA; Camille NJohnson, Snow Christensen & Martineau, Salt Lake City,UT; Gerald J Corcoran, Montgomery McCracken et al(NJ), Linwood, NJ; J Bennett Fox, Jr, Adams & Reese,Memphis, TN; James M Campbell, Campbell Campbellet al (MA), Boston, MA; John E Galvin, III, Jonathan HGarside, Fox Galvin, St Louis, MO; Robert A Limbacher,Goodell DeVries et al (PA), Philadelphia, PA; Scott WSayler, Shook Hardy & Bacon (KC), Kansas City, MO.

For Eli Lilly & Co, Defendant: J Bennett Fox, Jr, LEADATTORNEY, Adams & Reese, Memphis, TN; Sara JGourley, LEAD ATTORNEY, Jennifer A Foster, NathanA Huey, Sherry Ann Knutson, Sidley Austin (Chicago),Chicago, IL; Alan Daniel Mathis, James C Barton, Jr,Johnston Barton et al (AL), Birmingham, AL; Amy BAlderfer, [*5] Lori Chang, Greenberg Traurig (CA),Santa Monica, CA; Amy Sherry Fischer, Andrew MBowman, Foliart Huff et al, Oklahoma City, OK;Christiana C Jacxsens, PRO HAC VICE, GreenbergTraurig (GA), Atlanta, GA; Lori G Cohen, PRO HACVICE, Greenberg Traurig (GA), Atlanta, GA; Colin JGarry, James D Arden, Sidley Austin (NY), New York,NY; D'Lesli M Davis, Jonathan B Skidmore, Katherine P

Lett, H Douglas Wabner, Fulbright & Jaworski (DAL),Dallas, TX; David J Walz, Edward Walter Gerecke,Carlton Fields (FL), Tampa, FL; Dell P Chappell, John DMartin, Nelson Mullins et al (COL), Columbia, SC;Donna L Boyce, Adams & Reese (MEM), Memphis, TN;E Paige Sensenbrenner, Jaimme Angelle Collins, LawtonCole Callihan, Megan Haggerty Guy, Kathleen FontenotDrew, Adams & Reese (NO), New Orleans, LA; JenniferY Dukart, Joseph M Price, Faegre & Benson (MN),Minneapolis, MN; Jennifer Blakely D Kiefer, Stephen DMartin, Nelson Mullins et al (NC), Raleigh, NC; John EMcElligott, Jr, Davidson Meaux et al, Lafayette, LA; JoseAntonio Isasi, II, Greenberg Traurig (Chicago), Chicago,IL; Jude Philip Calabrese, Squire Sanders & Dempsey(OH), Cleveland, OH; Lauren P Buford, GreenbergTraurig (Chicago), Chicago, IL; [*6] Mary NoldLarimore, Nancy Menard Riddle, Ice Miller et al,Indianapolis, IN; Randall Shirres Richardson, Fulbright& Jaworski (HOU), Houston, TX; Stephen M Brooks,Nelson Mullins et al (Atlanta), Atlanta, GA; AlyciaDegen, Catherine Valerio Barrad, Sidley Austin (LA),Los Angeles, CA; Camille N Johnson, Snow Christensen& Martineau, Salt Lake City, UT; Gerald J Corcoran,Montgomery McCracken et al (NJ), Linwood, NJ; JamesM Campbell, Campbell Campbell et al (MA), Boston,MA; Lindsey N Oswalt, Sharon F Bridges, Adams &Reese (RL), Ridgeland, MS; Richard C Harlan, SidleyAustin (LA); Robert A Limbacher, Goodell DeVries et al(PA), Philadelphia, PA; Scott W Sayler, Shook Hardy &Bacon (KC), Kansas City, MO.

For Takeda Global Research & Development Center Inc,Defendant: Sara J Gourley, LEAD ATTORNEY, JenniferA Foster, Nathan A Huey, Sherry Ann Knutson, SidleyAustin (Chicago), Chicago, IL; Alan Daniel Mathis,James C Barton, Jr, Johnston Barton et al (AL),Birmingham, AL; Alycia Degen, Amanda VictoriaLopez, Richard C Harlan, Sidley Austin (LA), LosAngeles, CA; Amy Sherry Fischer, Foliart Huff et al,Oklahoma City, OK; Carol Dan Browning, Stites &Harbison, Lexington, KY; Colin J Garry, [*7] James DArden, Sidley Austin (NY), New York, NY; D'Lesli MDavis, H Douglas Wabner, Jonathan B Skidmore,Katherine P Lett, Fulbright & Jaworski (DAL), Dallas,TX; David J Walz, Edward Walter Gerecke, CarltonFields (FL), Tampa, FL; Dell P Chappell, John D Martin,Nelson Mullins et al (COL), Columbia, SC; Donna LBoyce, Adams & Reese (MEM), Memphis, TN; E PaigeSensenbrenner, Jaimme Angelle Collins, MeganHaggerty Guy, Kathleen Fontenot Drew, Adams & Reese


(NO), New Orleans, LA; Jane Thompson Davis, NelsonMullins et al (SC), Charleston, SC; Jennifer Blakely DKiefer, Stephen D Martin, Nelson Mullins et al (NC),Raleigh, NC; John E McElligott, Jr, Davidson Meaux etal, Lafayette, LA; Jude Philip Calabrese, Squire Sanders& Dempsey (OH), Cleveland, OH; Mary Nold Larimore,Nancy Menard Riddle, Ice Miller et al, Indianapolis, IN;Michael P Morrissey, Sidley Austin (NY), New York,NY; Randall Shirres Richardson, Fulbright & Jaworski(HOU), Houston, TX; Stephen M Bressler, Lewis &Roca, Phoenix, AZ; Stephen M Brooks, Nelson Mullinset al (Atlanta), Atlanta, GA; William Stevens Brown,Nelson Mullins et al (GV), Greenville, SC; Gerald JCorcoran, Montgomery McCracken et al (NJ), Linwood,NJ; [*8] J Bennett Fox, Jr, Adams & Reese, Memphis,TN; John E Galvin, III, Jonathan H Garside, Fox Galvin,St Louis, MO; Lindsey N Oswalt, Sharon F Bridges,Adams & Reese (RL), Ridgeland, MS; Scott W Sayler,Shook Hardy & Bacon (KC), Kansas City, MO.

For Takeda Pharmaceuticals International Inc, Defendant:Sara J Gourley, LEAD ATTORNEY, Jennifer A Foster,Nathan A Huey, Sherry Ann Knutson, Sidley Austin(Chicago), Chicago, IL; Alan Daniel Mathis, James CBarton, Jr, Johnston Barton et al (AL), Birmingham, AL;Alycia Degen, Amanda Victoria Lopez, Richard CHarlan, Sidley Austin (LA), Los Angeles, CA; AmySherry Fischer, Foliart Huff et al, Oklahoma City, OK;Carol Dan Browning, Stites & Harbison, Lexington, KY;Colin J Garry, James D Arden, Sidley Austin (NY), NewYork, NY; D'Lesli M Davis, H Douglas Wabner,Jonathan B Skidmore, Katherine P Lett, Fulbright &Jaworski (DAL), Dallas, TX; David J Walz, EdwardWalter Gerecke, Carlton Fields (FL), Tampa, FL; Dell PChappell, John D Martin, Nelson Mullins et al (COL),Columbia, SC; Donna L Boyce, Adams & Reese (MEM),Memphis, TN; E Paige Sensenbrenner, Jaimme AngelleCollins, Megan Haggerty Guy, Kathleen Fontenot Drew,Adams & Reese (NO), New Orleans, [*9] LA; JaneThompson Davis, Nelson Mullins et al (SC), Charleston,SC; Jennifer Blakely D Kiefer, Stephen D Martin, NelsonMullins et al (NC), Raleigh, NC; John E McElligott, Jr,Davidson Meaux et al, Lafayette, LA; Michael PMorrissey, Sidley Austin (NY), New York, NY; RandallShirres Richardson, Fulbright & Jaworski (HOU),Houston, TX; Stephen M Bressler, Lewis & Roca,Phoenix, AZ; William Stevens Brown, Nelson Mullins etal (GV), Greenville, SC; Gerald J Corcoran, MontgomeryMcCracken et al (NJ), Linwood, NJ; J Bennett Fox, Jr,Adams & Reese, Memphis, TN; John E Galvin, III,

Jonathan H Garside, Fox Galvin, St Louis, MO; JudePhilip Calabrese, Squire Sanders & Dempsey (OH),Cleveland, OH; Lindsey N Oswalt, Sharon F Bridges,Adams & Reese (RL), Ridgeland, MS; Scott W Sayler,Shook Hardy & Bacon (KC), Kansas City, MO.

For Takeda Pharmaceuticals L L C, Defendant: Sara JGourley, LEAD ATTORNEY, Jennifer A Foster, NathanA Huey, Sherry Ann Knutson, Sidley Austin (Chicago),Chicago, IL; Alan Daniel Mathis, James C Barton, Jr,Johnston Barton et al (AL), Birmingham, AL; AlyciaDegen, Amanda Victoria Lopez, Richard C Harlan,Sidley Austin (LA), Los Angeles, CA; Amy SherryFischer, Foliart Huff [*10] et al, Oklahoma City, OK;Carol Dan Browning, Stites & Harbison, Lexington, KY;Colin J Garry, James D Arden, Sidley Austin (NY), NewYork, NY; Dell P Chappell, John D Martin, NelsonMullins et al (COL), Columbia, SC; Donna L Boyce,Adams & Reese (MEM), Memphis, TN; E PaigeSensenbrenner, Jaimme Angelle Collins, MeganHaggerty Guy, Kathleen Fontenot Drew, Adams & Reese(NO), New Orleans, LA; Jane Thompson Davis, NelsonMullins et al (SC), Charleston, SC; John E McElligott, Jr,Davidson Meaux et al, Lafayette, LA; Michael PMorrissey, Sidley Austin (NY), New York, NY; StephenM Bressler, Lewis & Roca, Phoenix, AZ; WilliamStevens Brown, Nelson Mullins et al (GV), Greenville,SC; Gerald J Corcoran, Montgomery McCracken et al(NJ), Linwood, NJ; J Bennett Fox, Jr, Adams & Reese,Memphis, TN; John E Galvin, III, Jonathan H Garside,Fox Galvin, St Louis, MO; Jude Philip Calabrese, SquireSanders & Dempsey (OH), Cleveland, OH.

For Takeda San Francisco Inc, Defendant: Sara JGourley, LEAD ATTORNEY, Nathan A Huey, SherryAnn Knutson, Sidley Austin (Chicago), Chicago, IL;Alycia Degen, Richard C Harlan, Sidley Austin (LA),Los Angeles, CA; Jude Philip Calabrese, Squire Sanders& Dempsey (OH), [*11] Cleveland, OH.

For Takeda Ventures Inc, Defendant: Sara J Gourley,LEAD ATTORNEY, Nathan A Huey, Sherry AnnKnutson, Sidley Austin (Chicago), Chicago, IL; AlyciaDegen, Richard C Harlan, Sidley Austin (LA), LosAngeles, CA; Dell P Chappell, John D Martin, NelsonMullins et al (COL), Columbia, SC; Megan HaggertyGuy, Adams & Reese (NO), New Orleans, LA; JudePhilip Calabrese, Squire Sanders & Dempsey (OH),Cleveland, OH.

For Takeda Pharmaceuticals International L L C,


Defendant: Megan Haggerty Guy, Adams & Reese (NO),New Orleans, LA.

For Takeda Pharmaceuticals U S A Inc, formerly knownas Takeda Pharmaceuticals North America Inc,Defendant: K Michele Anderson, Sandy M Eloranto,LEAD ATTORNEYS, Tucker Ellis, Denver, CO; Sara JGourley, LEAD ATTORNEY, Jennifer A Foster, NathanA Huey, Sherry Ann Knutson, Sidley Austin (Chicago),Chicago, IL; D'Lesli M Davis, Jonathan B Skidmore,Katherine P Lett, H Douglas Wabner, Fulbright &Jaworski (DAL), Dallas, TX; Dell P Chappell, John DMartin, Nelson Mullins et al (COL), Columbia, SC;Donna L Boyce, Adams & Reese (MEM), Memphis, TN;Jaimme Angelle Collins, Lawton Cole Callihan, MeganHaggerty Guy, Kathleen Fontenot Drew, Adams & Reese(NO), New [*12] Orleans, LA; John E McElligott, Jr,Davidson Meaux et al, Lafayette, LA; Mary NoldLarimore, Nancy Menard Riddle, Ice Miller et al,Indianapolis, IN; Amanda Victoria Lopez, Sidley Austin(LA); Catherine Valerio Barrad, Sidley Austin (LA), LosAngeles, CA; E Paige Sensenbrenner, Adams & Reese(NO); Gerald J Corcoran, Montgomery McCracken et al(NJ), Linwood, NJ; J Bennett Fox, Jr, Adams & Reese,Memphis, TN; Jude Philip Calabrese, Squire Sanders &Dempsey (OH), Cleveland, OH; Robert A Limbacher,Goodell DeVries et al (PA), Philadelphia, PA; Scott WSayler, Shook Hardy & Bacon (KC), Kansas City, MO.

For Takeda Industries Inc, Defendant: Jennifer A Foster,Sidley Austin (Chicago), Chicago, IL.

For Springer Clinic Inc, Defendant: F Will DeMier,Jeffrey Lee Wilson, Barkley Law Firm, Tulsa, OK;Parker Huntington Foster, Barkley Law Firm.

For Takeda California Inc, formerly known as TakedaSan Diego Inc, Defendant: Sara J Gourley, LEADATTORNEY, Jennifer A Foster, Nathan A Huey, SherryAnn Knutson, Sidley Austin (Chicago), Chicago, IL;D'Lesli M Davis, Jonathan B Skidmore, Katherine P Lett,Fulbright & Jaworski (DAL), Dallas, TX; Dell PChappell, John D Martin, Nelson Mullins et al (COL),Columbia, [*13] SC; Donna L Boyce, Adams & Reese(MEM), Memphis, TN; Jaimme Angelle Collins, MeganHaggerty Guy, Kathleen Fontenot Drew, Adams & Reese(NO), New Orleans, LA; John E McElligott, Jr, DavidsonMeaux et al, Lafayette, LA; Amanda Victoria Lopez,Sidley Austin (LA); E Paige Sensenbrenner, Adams &Reese (NO); Gerald J Corcoran, MontgomeryMcCracken et al (NJ), Linwood, NJ; Jude Philip

Calabrese, Squire Sanders & Dempsey (OH), Cleveland,OH.

For Takeda America Holdings Inc, Defendant: Sara JGourley, LEAD ATTORNEY, Nathan A Huey, SherryAnn Knutson, Sidley Austin (Chicago), Chicago, IL; DellP Chappell, John D Martin, Nelson Mullins et al (COL),Columbia, SC; Jude Philip Calabrese, Squire Sanders &Dempsey (OH), Cleveland, OH; Megan Haggerty Guy,Adams & Reese (NO), New Orleans, LA.

For Alabama C V S Pharmacy L L C, Defendant: J AllenSydnor, Jr, LEAD ATTORNEY, Huie Fernambucq &Stewart, Birmingham, AL.

For Warren Clinic Inc, Defendant: Jeffrey Lee Wilson,Barkley Law Firm, Tulsa, OK; Parker Huntington Foster,Barkley Law Firm.

JUDGES: HONORABLE REBECCA F. DOHERTY,UNITED STATES DISTRICT JUDGE. MAGISTRATEJUDGE HANNA.

OPINION BY: REBECCA F. DOHERTY

OPINION

CASE MANAGEMENT ORDER: PROTOCOLRELATING TO THE PRODUCTION OFELECTRONICALLY [*14] STOREDINFORMATION ("ESI")

Pursuant to the agreement reached between thePlaintiffs and Defendants herein, this Court enters thefollowing Order concerning the production ofelectronically stored information in these proceedings:

A. Scope

1. General. The procedures and protocols outlinedherein govern the production of electronically storedinformation ("ESI") by the Parties. Section E titled"Search Methodology Proof of Concept" applies only tothe predictive coding and advanced analytics samplingprocedure as outlined in that Section. Sections A throughD and Sections F through J apply throughout thependency of this litigation. This Order governs all partiesto these proceedings, whether they currently are involvedor become so in the future. The Parties to this protocol("Protocol") will take reasonable steps to comply with


this agreed-upon Protocol for the production ofdocuments and information existing in electronic format.All disclosures and productions made pursuant to thisProtocol are subject to the Privilege Protocol andProtective Order entered in this matter.

2. Limitations and No-Waiver. The Parties and theirattorneys do not intend by this Protocol to waive theirrights [*15] to the attorney work-product privilege,except as specifically required herein, and any suchwaiver shall be strictly and narrowly construed and shallnot extend to other matters or information not specificallydescribed herein. All Parties preserve their attorney clientprivileges and other privileges and there is no intent bythe protocol, or the production of documents pursuant tothe protocol, to in any way waive or weaken theseprivileges. All documents produced hereunder are fullyprotected and covered by the Parties' confidentialityagreements, and order(s) of the United States DistrictCourt, as well as any clawback agreements, andprotective order(s) of the United States District Courteffectuating same.

B. ESI Preservation

1. The Parties have issued litigation notices to thoseidentified as most likely to have discoverableinformation.

C. Sources

1. While Defendants' fact gathering is ongoing, thefollowing are data sources identified to date that are mostlikely to contain discoverable information. Defendantsagree to provide additional discovered data sources likelyto contain relevant information. Defendants agree toprovide information about the data sources to the extent[*16] applicable and known in addition to that found inthe subparagraphs below, including the date range ofinformation contained in the data source, thedepartment(s) utilizing the data source, whether the datasource is hosted internally or externally, and the databasetype.

a ARISg Adverse Event Databaseb BLUE Labeling and promotional materials management systemc Galaxy Regulatory document management systemd MEDIsource Product information request databasee T-Rx Field sales call databasef TSARS (or "S Drive") Takeda Statistical Analysis and Repository Systemg T-Track Clinical Science Liaison databaseh IRIS Research grant management systemi LARC Clinical Science Liaison education resources databasej Sample Guardian Product sample management databasek TEG Takeda Educational Grant management systeml PubBase Publications management systemm Records Management System Records Operation Center ("ROC") information system

a. ARISg: ARISg is an adverse eventdatabase. It contains information that thePharmacovigilance department at TRGDU.S. receives regarding adverse eventsrelated to Takeda drugs, including adverseevent reports ("AERs") received from,without limitation, physicians, patients,clinical trials, medical literature, [*17]

and foreign entities. ARISg is the softwareused for this database, which is sometimescalled T-Gaea within Takeda. It has beenin effect since 1999.

b. BLUE: This database is used by theMarketing department in the approvalprocess for promotional materials. Itcontains a labeling module and a modulefor promotional pieces and marketingcampaigns. BLUE has been active from


April 2008 to present. The vendor isSchawk Blue.

c. Galaxy: Galaxy is a documentrepository system used by the Regulatorydepartment containing components ofregulatory submissions to the Food andDrug Administration. It went intoproduction in 2009.

d. MEDIsource: This data system isused by the Medical Information andQuality Assurance departments to captureand respond to product informationrequests and non-medical productcomplaints. It has a Siebel component thatdocuments the intake of requests forinformation from physicians and providesa response; a Documentum system withstandard response and customer responseletters; and Info Maestro which pullsinformation from the standard responseletter and from the Sieble system to createthe response letter to an individualphysician.

e. T-Rx: This database contains [*18]information regarding U.S. commercialfield sales calls.

f. TSARS (or "S Drive"): This isTakeda's Statistical Analysis andRepository System and is a Unixcentralized repository used to manageClinical and research data. It is used by theAnalytical Science department. It containsclinical SAS data sets and programs usedto analyze those data sets for purposes offinal submission reports -- tables, listings,and graphs.

g. T-Track: This database is acustomized application of Seibel'sCustomer Relationship Managementsystem for use by Takeda's field basedClinical Science Liaisons.

h. IRIS: This system is used byTakeda for the intake and processing ofexternal research grant requests. It is avendor hosted system (SteepRock is thevendor). It was implemented within thelast five years.

i. LARC: This database includesarticles, presentations, and publicationsrelated to Takeda products and thetherapeutic areas they address. Quosa isthe vender for this database. It isaccessible by Clinical Science Liaisons intheir respective therapeutic areas.

j. Sample Guardian: This databasecontains product sample management dataregarding sample transactions andinventory reconciliations.

k. TEG: [*19] Takeda EducationalGrant database is used for education grantrequest management.

l. PubBase: PubBase is aDocumentum-based system used for themanagement and storage of publicationdocuments.

m. Records Management System:This data source is used by the RecordsOperations Center ("ROC"), wherephysical records are maintained.

D. Custodians

1. The following are custodians who have beenidentified as most likely to have information relevant tothis litigation. For these custodians, data is being pulledfrom e-mail, computer hard drives, and physical files thatare in the possession, custody, and control of Takeda.Investigation is ongoing by both Parties as to potentialadditional custodians at Takeda (including potentialJapanese custodians) and Eli Lilly and Company. Currentkey custodians include:

1. Baron, David Vice President, NonClinical Safety/Efficacy

2. Spanheimer, Robert Vice President, Medical and Scientific Affairs


3. Greeby, Jennifer Director, Marketing (Diabetes)

4. Recker, David Senior Vice President, Clinical Science

5. Paris, Maria Former Vice President, Pharmacovigilance

6. Gerrits, Charles Former Senior Director, Pharmacoepidemiology

7. Johnston, Janet Associate Director, Safety Surveillance

8. Thom, Claire Former Vice President, Research and Development

9. Daly, Rich Former Vice President, Marketing

10. Perez, Alfonso Vice President, Clinical Science Strategy

11. Ortell, Una Director, Promotion and Advertising

12. Orlando, Dan Former Vice President, Sales

13. Lee, Jessie Manager, Regulatory Affairs Strategy

14. Cuomo, Maryann Associate Director, Regulatory Labeling

15. Weisbrich, Shay Vice President, Franchise Leader (Former Director, Marketing)

16. Kupfer, Stuart Vice President, Clinical Science

17. Ramstack, Mary Sr. Director, Strategic Project Planning and Management

18. Roebel, Mick Sr. Director, Regulatory Affairs

19. Lorenz, Janet Associate Director, Regulatory Affairs, Promotion and Advertising

20. Pritza, Mary Jo Former Associate Director, Regulatory Affairs

21. Caracci, Mike Former Director, Marketing

22. Tynan, Julie Assistant Project Director, Strategic Project and Planning Management

23. Hull, Andy Vice President, Alliance Management (former Vice President, Marketing)

24. Fusco, Gregory Sr. Medical Director, Pharmacoepidemiology and Analysis

25. Caggiano, Christopher Sr. Product Manager, Diabetes Marketing

26. Ryan, D'Arcy Former Director, Marketing

27. Khan, Mehmood Former Sr. Vice President, Medical and Scientific Affairs

28. Harris, Thomas Vice President, Regulatory Affairs

29. Trochanov, Anton Associate Medical Director, Pharmacovigilance

E. [*20] Search Methodology Proof of Concept

1. General. The Parties have discussed themethodologies or protocols for the search and review ofESI collected from Takeda sources, including but notlimited to e-mail, and the following is a summary of theParties' agreement on the use of a search methodologyproof of concept to evaluate the potential utility ofadvanced analytics as a document identificationmechanism for the review and production of this data.

The Parties agree to meet and confer regarding the use ofadvanced analytics for other data sources. While theParties agree to explore the use of advanced analytics as atechnique to ensure appropriate responses to discoveryrequests, the Parties agree that Defendants retain the rightto review documents after predictive coding but prior toproduction for relevance, confidentiality, and privilege. Asampling of documents withheld after such review willtake place pursuant to Section E.10.

2. General Overview of AdvancedAnalytics/Predictive Coding Process. Takeda utilizes


software provided by Epiq Systems ("Epiq") to searchand review ESI for production in this case. Epiq usesEquivio's Relevance software for advanced analytics andpredictive [*21] coding.

Epiq will collect e-mail documents from four keyTakeda custodians, which will be combined to create the"sample collection population." The Parties will meet andconfer to determine the names of the four custodians.Additionally, Takeda will add a set of regulatorydocuments which have already been collected to the"sample collection population." Takeda and Plaintiffswill each nominate three individuals ("the experts") towork collaboratively at the offices of Nelson Mullins,1320 Main Street, Columbia, SC 29201 to train theEquivio Relevance system. Plaintiffs' experts will executea Nondisclosure and Confidentiality Agreement in theform attached as Exhibit A hereto. To the extent thatPlaintiffs' experts are exposed to information that wouldbe subject to withholding or redaction under theProtective Order in this matter, Plaintiffs' experts agreenot to disclose such information to co-counsel, client, anyParty, or any third party without obtaining prior writtenconsent of the other Party regarding the particular pieceof information sought to be disclosed. Before themeeting, the Parties shall be provided a copy of theapplicable Equivio training documents, handbook, [*22]or manual. The Parties' experts will receive technicaltraining on the Equivio Relevance software and codingprocess and will work together to make one relevancedecision for documents in the Control and Training sets,as described in more detail below.

The Parties will review a number of documentsrequired by the Equivio Relevance system for the data toreach Stability as described below. Once Stability isreached, the Control and Training sets are then used tobegin the predictive coding process. Using the Controland Training documents, the system calculates relevancescores for the entire sample collection population, witheach document in the sample collection populationreceiving a relevance score of 0 through 100.

Attorneys representing Takeda will have access tothe entire sample collection population to be searched andwill lead the computer training, but they will workcollaboratively with Plaintiffs' counsel during theAssessment and Training phases. Takeda's experts willconduct an initial review of documents presented by theEquivio Relevance system for privilege. The privileged

documents will be either entirely withheld from viewingby Plaintiffs' experts or printed and [*23] redacted. Aprivilege log for such documents will be provided. TheParties, after review of the privilege log, reserve the rightto require that such documents be deemed as "skip"(same as designation used for technical problemdocuments). Otherwise, these documents may still beused to train the system. Both Parties will then review allof the non-privileged documents during the trainingprocess (i.e., both documents coded as relevant andirrelevant). The Parties' experts will review thedocuments in collaboration and determine the coding tobe applied to the documents. To the extent the Partiesdisagree regarding the coding of a particular document ordesignation of privilege, they will meet and confer in aneffort to resolve the dispute prior to contacting the Courtfor resolution.

At the conclusion of the training process and uponcalculation of relevance scores, the Parties will meet andconfer regarding which relevance score will provide acutoff for documents to be manually reviewed by defensecounsel for production. However, the Parties reserve theright to seek relief from the Court prior to thecommencement of the final manual review.

At the recommendation of Epiq, no seeding [*24]will take place at this time. The Parties may meet andconfer if it is determined that seeding may be applicableat a later date.

Plaintiffs' experts and counsel shall not remove anyof the Control or Training documents from the offices ofNelson Mullins, nor shall they be allowed to copy suchdocuments. The Parties agree that Defendants do notwaive protection of trade secret or confidentialinformation in allowing Plaintiffs to review documentsunder this sampling mechanism. All documents reviewedpursuant to this sampling protocol shall be done under theProtective Order in this matter as well as any PrivilegeProtocol or clawback agreement that shall be reduced toan order acceptable to the Court.

3. Relevance Tags. The Parties agree that as part ofthe Assessment and Training phases, all of thenon-privileged and privilege-redacted documentsreviewed by both parties' experts will be categorized asrelevant, not relevant, or skip (to be used for documentswith technical problems). The privileged-withhelddocuments will be categorized by Defendants' experts asrelevant, not relevant, or skip, subject to the Parties' right


to have any privileged-withheld documents categorizedas a "skip." [*25] The Parties shall immediately discussany disagreements on coding in good faith, so that thetraining may be improved accordingly, and may seekguidance from the Court or the Court appointed specialmasters if necessary.

4. Collection & Data Preparation. The Parties willmeet and confer to agree upon the four custodians thatwill be selected for the sampling. E-mail and attachmentdocuments will be collected from the four custodians andadded to the collected regulatory documents, togethercomprising the sample collection population. Documentsmay be removed from the sample collection population ifthey are:

a. Spam,

b. Commercial e-mail,

c. Files without text,

d. Exact duplicates within thecustodians (see Section G.6 regardingproduction of information for duplicatedocuments), and

e. System files, etc. (i.e., thedocuments that the samples will beselected from will be de-NISTED)

Epiq will extract the sample collection populationdocuments' text and build an index.

5. Assessment Phase. The Equivio Relevancesoftware generates an initial simple random sample of500 documents from the sample collection population.Takeda's experts will initially review the documents forprivilege. Any documents [*26] deemed privileged byTakeda's experts will be either entirely withheld fromviewing by Plaintiffs' experts or printed and redactedprior to viewing by Plaintiffs' experts, and logged on aprivilege log consistent with the Privilege Protocol in thismatter. These documents may still be used to train thesystem. To the extent the Parties disagree regarding theprivilege decision for a particular document, they willmeet and confer in an effort to resolve the dispute prior tocontacting the Court for resolution. The Parties' expertswill then work collaboratively to determine the relevanceof the non-privileged and privilege-redacted documents.

The relevance of the privileged-withheld documents willbe determined by Defendants' experts. The documentsreviewed in the Assessment Phase make up the ControlSet. The Control Set is used for estimating richness(percentage of relevant documents in a population), andalso serves as a reference point for calculating recall andprecision.

a. The application's estimates of richnessuse a confidence level of 95%. The initialControl Set of 500 documents yields aconfidence estimation of richness with anerror margin of plus or minus 4.3%. Thisis [*27] a worst-case error marginassuming richness of 50%. For lowerlevels of richness, the error margin willalso be lower. For example, for richness of10%, the error margin would be plus orminus 2.6%, while for 5%, the errormargin would be plus or minus 1.9%.

b. The Control Set also creates a basisfor calculating recall and precision, whichare then used for monitoring trainingprogress and calculating results.

c. Equivio Relevance tracks theprogress of the Assessment Phase toachieve the appropriate level of statisticalvalidation. These levels of validation arereferred to in the Equivio system as"Baseline," at the lowest level, through"Statistical," at the highest level. Theterms "Baseline" and "Statistical" are usedby Equivio Relevance as indicators to theuser as to the progress of the AssessmentPhase. The validation level achieveddepends on the number of relevantdocuments found by the user in theControl Set. At the "Baseline" level, thenumber of relevant documents in thecontrol set is too low to allow statisticallyvalid estimates of recall and precision. TheParties will ensure that the number ofControl Set documents reviewed willreach the "Statistical" level.

d. For [*28] informational purposes,the "Statistical" level of validation inEquivio requires the presence of at least


70 relevant documents in the Control Set.For document collections with richness of14% and above, a Control Set of 500documents is sufficient to reach the"Statistical" level of validation. For lowerlevels of richness, additional documentswill need to be reviewed in theAssessment Phase in order to reach the"Statistical" level.

e. Based on a confidence level of95%, the Statistical level of validationyields an error margin on recall estimatesof plus or minus 11.7%. This is aworst-case error margin assuming recall of50%. The Parties will continue theAssessment Phase, beyond the"Statistical" level, until the Control Setcontains at least 385 relevant documents.This sample will yield an error margin onrecall estimates of plus or minus 5%.

6. Iterative Training Phase. Following the creation ofthe Control Set at the Statistical validation level, theEquivio Relevance system selects a random sample offorty documents. Takeda's experts will initially reviewthe forty documents for privilege. Any documentsdeemed privileged by Takeda's experts will be eitherentirely withheld [*29] from viewing by Plaintiffs'experts or printed and redacted prior to viewing byPlaintiffs' experts, and logged on a privilege logconsistent with the Privilege Protocol in this matter.These documents may still be used to train the system.The Parties' experts will then work collaboratively todetermine the relevance of the non-privileged andprivilege-redacted documents. The relevance of theprivileged-withheld documents will be determined byDefendants' experts, subject to the Parties' right to haveany privileged-withheld documents categorized as a"skip" and not included in the training. To the extent theParties disagree regarding the relevance or privilegedecision for a particular document, they will meet andconfer in an effort to resolve the dispute prior tocontacting the Court for resolution.

a. Once the experts have completed thefirst Training Set, the Equivio Relevancesystem calculates the Training Status. Thethree possible states are "Not Stable,"

"Nearly Stable," or "Stable."

b. The experts continue to reviewsamples of forty documents each, usingthe process outlined in paragraph 6 above,until the Stable Training Status is reached.

c. The subsequent samples of forty[*30] documents are selected using anActive Learning approach. ActiveLearning means that each training sampleis selected based on what has been learnedfrom previous samples. The object is tomaximize the sample's contribution to thetraining process. Therefore, the systemchooses samples that providecomprehensive coverage of the population(reducing under-inclusiveness), whilefine-tuning the concept of relevance thatthe Classifier is developing (reducingover-inclusiveness). The system reachesStability when the marginal contributionof additional samples to the enhancementof the Classifier approaches zero, asdetermined by the Equivio software andwhich determination (Stability) is notconfigurable.

7. Calculation of Relevance Scores. Uponcompletion of the Training Phase once Stability isreached, and any related meet and confer sessions andagreed upon coding corrections, the Equivio Relevancesystem will run over the sample collection population andcalculate relevance scores for each document in thesample collection population. Each document in thesample collection population receives a relevance scoreof 0 through 100, with 0 being least likely to be relevantand 100 being [*31] most likely.

8. Final Search, Review, and Production of SampleCollection Population Documents. The Parties will meetand confer regarding which relevance score will providea cutoff that will yield a proportionate set of documentsthat will be manually reviewed by Takeda for production.All of the documents above the agreed upon relevancescore in the sample collection population will bereviewed by Takeda. Documents found by Takeda'sreview to be relevant and non-privileged documents willbe produced to Plaintiffs.


9. Quality Control by Random Sample of IrrelevantDocuments. In addition, at the conclusion of the processdescribed above, and prior to generating the review set,the Parties will collaboratively review at the offices ofNelson Mullins in Columbia, SC a random sample ofdocuments in the sample collection population withrelevance scores below the cut-off score set forestablishing the review set (aka the "Rest"). Thesedocuments are flagged for culling, and will not beincluded in the review set. In Equivio Relevance, this testis referred to as "Test the Rest." The purpose for thisphase is to verify that the Rest contains a low prevalenceof relevant documents and that [*32] the proportionalityassumptions underlying the cut-off decision are valid.

a. The Test the Rest sample is designedto provide a confidence level of 95%. Thedefault sample size is 500 documents. Themargin of error depends on the percentageof relevant documents in the Rest. Forexample, if 5% of the Rest documents arefound to be relevant, the margin of error is1.9%. If 1% are relevant, the margin oferror is 0.8%.

b. Takeda's experts will initiallyreview the Rest sample documents forprivilege. Any documents deemedprivileged by Takeda's experts will beeither entirely withheld from viewing byPlaintiffs' experts or printed and redactedprior to viewing by Plaintiffs' experts, andlogged on a privilege log consistent withthe Privilege Protocol in this matter. TheParties' experts will then workcollaboratively to determine the relevanceof the non-privileged andprivilege-redacted documents. Therelevance of the privileged-withhelddocuments will be determined byDefendants' experts, subject to the Parties'rights to have any privilege-withhelddocument categorized as a "skip" forpurposes of the Test the Rest sample. Tothe extent the Parties disagree regardingthe relevance or [*33] privilege decisionfor a particular document, they will meetand confer in an effort to resolve thedispute prior to contacting the Court forresolution.

10. Sampling of Documents Not Produced AfterPredictive Coding. After the predictive coding processcompletes, and Takeda's counsel reviews and producesdocuments from the sample collection populationconsistent with paragraph 8, the Parties willcollaboratively review at the offices of Nelson Mullins inColumbia, SC a random sample of documents above theagreed-upon cutoff relevance score that were withheldfrom production on relevance grounds. The Parties agreeto meet and confer regarding an appropriate sample size.

a. Takeda's experts will initially reviewthe sample documents for privilege. Anydocuments deemed privileged by Takeda'sexperts will be either entirely withheldfrom viewing by Plaintiffs' experts orprinted and redacted prior to viewing byPlaintiffs' experts, and logged on aprivilege log consistent with the PrivilegeProtocol in this matter. The Parties'experts will then work collaboratively todetermine the relevance of thenon-privileged and privilege-redacteddocuments. The relevance of theprivileged-withheld [*34] documents willbe determined by Defendants' experts,subject to the Parties' rights to have anyprivilege-withheld document categorizedas a "skip" for this purpose. To the extentthe Parties disagree regarding therelevance or privilege decision for aparticular document, they will meet andconfer in an effort to resolve the disputeprior to contacting the Court forresolution.

11. Post-Predictive Coding Sampling Meet andConfer. The Parties shall meet and confer in good faith toresolve any difficulties and finalize the method forsearching documents on a going forward basis. To theextent that the Parties cannot agree, they shall apply tothe Court for relief. Defendant shall not be required toproceed with the final search and review unless and untilobjections raised by either Party have been adjudicatedby the Court or resolved by written agreement of theParties. The Parties reserve the right to request a meetand confer regarding the designation of any document as


a "skip" for purposes of the control sample, training, orTest the Rest, if agreement cannot be reached.

F. Costs

1. Takeda reserves its right to seek relief from theCourt (e.g., a cost shifting award and pursuant to [*35]the principles of proportionality). See Fed. R. Civ. P. 1,26(b)(2)(C), 26(b)(2)(B), & 26(g); Electronic Discovery,11 Sedona Conf. J. 289 (2010); see also Fed. R. Evid.403 (inadmissibility of cumulative evidence).

2. Plaintiffs agree to bear all of the costs associatedwith their compliance with the terms of this protocol.Plaintiffs agree to bear all of the costs associated with thereceipt and review of ESI produced hereunder includingthe costs associated with its ESI experts who will be

involved with Plaintiffs in all aspects of this ESIprotocol.

G. Format of Production For Documents Produced byDefendants

1. TIFF/Native File Format Production. Documentswill be produced as single-page TIFF images withcorresponding multi-page text, native file formatdocument if applicable under paragraph G.2, andnecessary load files. Native files, along with allcorresponding metadata, will be preserved. TIFF imageswill be of 300 dpi quality or better. The load files willinclude an image load file as well as a metadata (.DAT)file with the metadata fields identified below on thedocument level to the extent available.

Field Summation Field (Florida) Definition Doc Type

1 SOURCE SOURCE Name of party producing the document All

2 CUSTODIAN CUSTODIAN Name of person or non- human datasource from where documents/files areproduced. **Where redundant namesoccur, individuals should be distin-guished by an initial which is kept con-stant throughout productions (e.g.,Smith, John A. and Smith, John B.Where data is collected from anarchive, the archive will be listed ascustodian.

All

3 CUSTODIANAPPEND-MULTI

CUSTODIANAPPENDMULTI Name of Takeda person or non-humandata source from where duplicate doc-uments/files were suppressed.**Where redundant names occur, indi-viduals should be distinguished by aninitial which is kept constantthroughout productions (e.g., Smith,John A. and Smith, John B. Where datais collected from an archive, thearchive will be listed as custodian.

All

4 CUSTODIAN ID CUSTODIAN ID Each CUSTODIAN from #2 or 3above will be assigned a unique nu-meric identifier that will be maintainedthroughout productions. Where data iscollected from an archive, the archivewill be listed as custodian.

All

5 BEGBATES BEGDOC# Beginning Bates Number (productionnumber)

All


6 ENDBATES ENDDOC# End Bates Number (production num-ber)

All

7 PGCOUNT PGCOUNT Number of pages in the document All

8 FILESIZE FILESIZE File Size All

9 APPLICAT APPLICAT Commonly associated application forthe specified file type.

All

10 FILEPATH FILEPATH (for Edocs) File source path for electronically col-lected documents other than emails,which includes location, file name, andfile source extension.

Edocs

11 RELATIVE PATH AP-PEND

RELATIVE PATH APPEND(for Edocs)

File source path for duplicate electron-ically collected documents other thanemails, which includes location, filename, and file source extension.

Edocs

12 NATIVEFILELINK DOCLINK For documents provided in nativeformat only

All

13 TEXTPATH LOGFILE or FULLTEXT File path for OCR or Extracted Textfiles

All

14 MSGID MSGID Value extracted from parent messageduring processing

Email

15 FROM FROM Sender Email

16 TO TO Recipient Email

17 cc cc Additional Recipients Email

18 BCC BCC Blind Additional Recipients Email

19 SUBJECT SUBJECT Subject line of email Email

20 PARENTBATES PARENTID BeginBates number for the parentemail of a family (will not be popu-lated for documents that are not part ofa family)

Email

21 ATTACHBATES ATTACHID Bates number from the first page ofeach attachment

Email

22 BEGATTACH (will be provided from AT-TRANGE)

First Bates number of family range(i.e. Bates number of the first page ofthe parent email)

Email

23 ENDATTACH (will be provided from AT-TRANGE)

Last Bates number of family range (i.e.Bates number of the last page of thelast attachment)

Email

24 ATTACHCOUNT ATTACHMENT COUNT Number of attachments to an email Email

25 ATTACHNAME ATTACHMENT LIST Name of each individual attachment Email

26 DATESENT (mm/dd/yyyyhh:mm:ss AM)

DATESENT Date Sent Email

27 DATERCVD(mm/dd/yyyy hh:mm:ss

DATERCVD Date Received Email


AM)

28 EMAILDATSORT(mm/dd/yyyy hh:mm:ssAM)

DATESENT Sent Date of the parent email(physically top email in a chain, I.e.immediate/direct parent email)

Email

29 Email Outlook Type Email Outlook Type Type of Outlook item, e.g.email, cal-endar item, contact, note, task

Email

30 HASHVALUE MD5HASH MD5 Hash Value All

31 TITLE DOCTITLE Title provided by user within the docu-ment

Edocs

32 AUTHOR AUTHOR Creator of a document Edocs

33 DATECRTD DATECRTD Creation Date Edocs

34 MODIFIED BY LAST EDITED BY Person who has modified a document Edocs

35 LASTMODD(mm/dd/yyyy

LASTMODD (mm/dd/yyyyhh:mm:ss

Last Modified Date Edocs

36 DocumentType DocumentType Descriptor for the type of document:"E-document" for electronic docu-ments not attached to emails;"Emails" for all emails; "E-attachments" for files that were at-tachments to emails; and "Physicals"for hard copy physical documents thathave been scanned and converted to anelectronic image.

All

37 Importance Importance High Importance - indicates PriorityEmail message.

Email

38 Redacted Redacted Descriptor for documents that havebeen redacted. Yes for redacted docu-ments; "No" for unredacted docu-ments.

All

39 ProdVol ProdVol Name of media that data was producedon.

All

Wave 00 I - Hard Drive

40 Confidentiality Confidentiality Indicates if the document has been des-ignated as Confidential pursuant to anyapplicable Protective Order. "Yes" forConfidential documents; No for docu-ments that are not so designated.

All

41 Email folder Email folder Folder in which non- archive collectedemail is stored within the custodiansmailbox, such as "inbox", "sent", de-leted, "draft", or any custom folder.

Email

42 Relevance score Relevance score Relevance score assigned by Equiviofor documents that have been throughthe predictive coding process

All


a. [*36] This list of fields does notcreate any obligation to create or manuallycode fields that are not automaticallygenerated by the processing of the ESI;that do not exist as part of the originalMetadata of the document; or that wouldbe burdensome or costly to obtain.

2. Defendants will produce spreadsheets (.xls/.xlsxfiles) and PowerPoint presentations (.ppt/.pptx files) innative form as well as audio and video files (e.g., mp3s,wavs, mpegs, etc.), except that spreadsheets andPowerPoint documents will be produced in TIFF formatif redactions are applied. Audio and video files shall beedited if redactions are required, subject to appropriateidentification of any such audio or video files havingbeen edited. In addition, for any redacted documents thatare produced, the documents' metadata fields will beredacted where required. The Parties will meet and conferregarding a request for the production of any othermaterials including documents in native file format.

3. The Parties agree to meet and confer regarding theformat of production for structured databases.

4. Appearance. Subject to appropriate redaction, eachdocument's electronic image will convey the sameinformation and [*37] image as the original document,including formatting, such as bolding, highlighting, fontsize, italics. Documents will be produced in black andwhite. After production, a Party may request that adocument be produced in color at which time the Partiesmay meet and confer about such production. Documentsthat present imaging or formatting problems will beidentified and the Parties will meet and confer in anattempt to resolve the problems.

5. Document Numbering. Each page of a produceddocument will have a legible, unique page identifier"Bates Number" electronically "burned" onto the imageat a location that does not obliterate, conceal or interferewith any information from the source document. TheBates Number for each page of each document will becreated so as to identify the producing Party and thedocument number. In the case of materials redacted inaccordance with applicable law or confidential materialscontemplated in any Protective Order or ConfidentialityStipulation entered into by the Parties, a designation may

be "burned" onto the document's image at a location thatdoes not obliterate or obscure any information from thesource document.

6. De-NISTing and Deduplication. [*38] Electronicfile collections will be De-NISTed, removingcommercially available operating system and applicationfile contained on the current NIST file list. Defendantswill globally deduplicate identical ESI as follows:

a. Electronic Files: Duplicated electronicfiles will be identified based uponcalculated MD5 Hash values for binaryfile content. File contents only will beused for MD5 Hash value calculation andwill not include operating system metadata(filename, file dates) values. All filesbearing an identical MD5 hash value are aduplicate group. The document reviewedby Defendants for privilege, relevance, orconfidentiality shall be deemed theprimary duplicate document within thegroup. Generally, the Defendants shall notremove any of the objective coding fieldslisted in paragraph G.1 above, in eitherprimary or duplicate documents. Ifredactions are applied to the subject and/ortext fields, however, Defendants mayapply the same redactions to all otherdocuments within the duplicate group.Defendants shall only produce onedocument image or native file forduplicate ESI documents within the group.For Takeda sources, the followingmetadata fields as described in Section[*39] G.1 associated with the produceddocument will provide information forduplicate documents not produced:CustodianAppendMulti andRelativePathAppend.

b. Messaging Files: Duplicatemessaging files will be identified basedupon MD5 Hash values for the messagefamily, including parent object andattachments. The following fields will beused to create the unique value for eachmessage: To; From; CC; BCC; Date Sent;


Subject; Body; and, MD5 Hash values forall attachments, in attachment order.Duplicate messaging materials will beidentified at a family level, includingmessage and attachment(s). All filesbearing an identical MD5 Hash value are aduplicate group. The documents reviewedby Defendants for privilege, relevance, orconfidentiality shall be deemed theprimary duplicate document within thegroup. For identified duplicate ESI, theDefendants shall not remove any of theobjective coding fields listed in paragraphG.1 above. If redactions have been appliedto such fields, Defendants may substituteand replace the subject and text fields withthose reviewed by Defendants' counsel forthe primary duplicate ESI document forthe other documents within the duplicategroup. Defendants [*40] shall onlyproduce one document image or native filefor duplicate ESI documents within thegroup. For Takeda sources, the followingmetadata field as described in Section G.1associated with the produced documentwill provide information for duplicatedocuments not produced:CustodianAppendMulti.

c. E-mail Threading: The producingParty may identify e-mail threads whereall previous emails which make up thethread are present in the body of the finale-mail in the thread. Any party electing touse this procedure must notify allreceiving parties that e-mail threadsuppression has been proposed to beperformed on a specified production andthe Parties agree to meet and conferregarding the format of this production,and reserve the right to seek Courtguidance on the issue should agreementnot be reached.

7. Production Media. The producing Party mayproduce documents via a secure file transfer mechanismand/or on readily accessible, computer or electronicmedia as the Parties may hereafter agree upon, includingCD-ROM, DVD, external hard drive (with standard PC

compatible interface), (the "Production Media"). Eachpiece of Production Media will be assigned a productionnumber or other [*41] unique identifying labelcorresponding to the date of the production of documentson the Production Media (e.g., "Defendant TakedaProduction April 1, 2012") as well as the sequence of thematerial in that production (e.g. "-001", "-002"). Forexample, if the production comprises document imageson three DVDs, the producing Party may label each DVDin the following manner "Defendant Takeda ProductionApril 1, 2012", "Defendant MSL Production April 1,2012-002", "Defendant Takeda Production April 1,2012-003." Additional information that will be identifiedon the physical Production Media includes: (1) textreferencing that it was produced in In re: Actos(Pioglitazone) Products Liability Litigation; and (2) theBates Number range of the materials contained on theProduction Media. Further, any replacement ProductionMedia will cross-reference the original Production Mediaand clearly identify that it is a replacement andcross-reference the Bates Number range that is beingreplaced.

8. Write Protection and Preservation. All computermedia that is capable of write protection should bewrite-protected before production.

9. Inadvertent Disclosures. The terms of the CaseManagement Order: [*42] Assertions of Attorney-ClientPrivilege and Work Product Doctrine shall apply to thisprotocol.

10. Duplicate Production Not Required. The Partiesshall meet and confer regarding any Party's request toproduce identical paper copies of data already producedin electronic form.

H. Timing.

1. The Parties will use their reasonable efforts toproduce ESI in a timely manner consistent with theCourt's discovery schedule.

2. The Parties will produce ESI on a rolling basis.

I. General Provisions.

1. Any practice or procedure set forth herein may bevaried by agreement of the Parties, and first will beconfirmed in writing, where such variance is deemedappropriate to facilitate the timely and economical


exchange of electronic data.

2. Should any Party subsequently determine it cannotin good faith proceed as required by this protocol; theParties will meet and confer to resolve any dispute beforeseeking Court intervention.

3. The Parties agree that e-discovery will beconducted in phases and the Parties will meet and conferregarding discovery of data sources not listed herein.

4. Regardless of the foregoing, the Parties are undera continuing obligation to produce identified responsive,non-privileged [*43] documents and to identify sourcesof potentially discoverable materials consistent with theirobligations under Federal Rules of Civil Procedure.

J. Items Requiring Meet and Confer.

1. The Parties agree to meet and confer regarding thefollowing items in advance of impacted productions:

a. Whether the E-mail Property metadatafield is able to be produced

b. Certain technical specifications forproductions:

(1) Hard copy documentunitization

(2) Microsoft "Auto"features or macros

(3) Embedded objects

(4) Compressed Files

(5) Load fileorganization

IT IS SO ORDERED.

THUS DONE AND SIGNED in Lafayette,Louisiana, this 27 day of July, 2012.

/s/ Rebecca F. Doherty

HONORABLE REBECCA F. DOHERTY

UNITED STATES DISTRICT JUDGE


KLEEN PRODUCTS LLC, et al., Plaintiffs, v. PACKAGING CORPORATION OFAMERICA, et al., Defendants.

No. 10 C 5711

UNITED STATES DISTRICT COURT FOR THE NORTHERN DISTRICT OFILLINOIS, EASTERN DIVISION

2012 U.S. Dist. LEXIS 139632

September 28, 2012, DecidedSeptember 28, 2012, Filed

PRIOR HISTORY: Kleen Prods., LLC v. PackagingCorp. of Am., 775 F. Supp. 2d 1071, 2011 U.S. Dist.LEXIS 38546 (N.D. Ill., 2011)

COUNSEL: [*1] For Kleen Products LLC, Plaintiff:Heidi M. Silton, W. Joseph Bruckner, LEADATTORNEYS, Lockridge Grindal Nauen P.L.L.P.,Minneapolis, MN; Trent M. Johnson, LEADATTORNEY, PRO HAC VICE, Foley & Lardner Llp,Milwaukee, WI; Anthony D. Shapiro, Hagens BermanSobol Shapiro, Seattle, WA; Charles P Goodwin, PROHAC VICE, H. laddie Montague, PRO HAC VICE,Berger & Montague, P.c., Philadelphia, PA; ChristopherM. Burke, PRO HAC VICE, Kristen M. Anderson, PROHAC VICE, Walter W. Noss, PRO HAC VICE,ScottScott LLP, San Diego, CA; Daniel A. Bushell,Berman DeValerio, Palm Beach Gardens, FL; Daniel EGustafson, Daniel C Hedlund, Gustafson Gluek PLLC,Minneapolis, MN; Daniel Jay Mogin, PRO HAC VICE,The Mogin Law Firm, San Diego, CA; Dianne M Nast,PRO HAC VICE, Roda & Nast, P.C., Lancaster, PA;Donald Lewis Sawyer, Freed Kanner London & MillenLLC, Bannockburn, IL; Edward A. Diver, PRO HACVICE, Howard Langer, Peter E. Leckman, PRO HACVICE, Langer Grogan & Diver, P.c., Philadelphia, PA;Erin C. Burns, PRO HAC VICE, RodaNast, P.C.,Lancaster, PA; Geoffrey C Rushing, PRO HAC VICE,Richard Alexander Saveri, PRO HAC VICE, Saveri &

Saveri, Inc., San Francisco, CA; Jeffrey Sprung, PROHAC VICE, Hagens [*2] Berman Sobol Shapiro Llp,Seattle, WA; Joseph Goldberg, PRO HAC VICE,Freedman Boyd Hollander, Goldberg Ives & Duncan,PA, Albuquerque, NM; Manuel Juan Dominguez, CohenMilstein Sellers & Toll, Palm Beach Gardens, FL; MartinI. Twersky, PRO HAC VICE, Berger & Montaque, P.C.,Philadelphia, PA; Matthew Sinnott, PRO HAC VICE,The Mogin Law Firm, P.c., San Diego, CA; MichaelJerry Freed, Michael E. Moskovitz, Robert J. Wozniak,Steven A Kanner, Freed Kanner London & Millen, LLC,Bannockburn, IL; Steven J. Greenfogel, Lite DePalmaGreenberg, LLC, Philadelphia, PA; William J Heye, PROHAC VICE, Saveri & Saevri, Inc., San Francisco, CA.

For Ferraro Foods of North Carolina, LLC., FerraroFoods, Inc., Plaintiffs: Heidi M. Silton, W. JosephBruckner, LEAD ATTORNEYS, Lockridge GrindalNauen P.L.L.P., Minneapolis, MN; Daniel Jay Mogin,The Mogin Law Firm, San Diego, CA; Donald LewisSawyer, Robert J. Wozniak, Freed Kanner London &Millen LLC, Bannockburn, IL; Manuel Juan Dominguez,Cohen Milstein Sellers & Toll, Palm Beach Gardens, FL;Robert G. Eisler, PRO HAC VICE, Grant & EisenhoferP.A., Wilmington, DE; Vincent J Esades, PRO HACVICE, Heins Mills & Olson, P.L.C., Minneapolis, MN.

For Distributors [*3] Packaging Group, LLC, RHEHatco, Inc., Plaintiffs: Heidi M. Silton, W. Joseph

Page 1

Bruckner, LEAD ATTORNEYS, Lockridge GrindalNauen P.L.L.P., Minneapolis, MN; Daniel Jay Mogin,The Mogin Law Firm, San Diego, CA; Donald LewisSawyer, Robert J. Wozniak, Freed Kanner London &Millen LLC, Bannockburn, IL; Manuel Juan Dominguez,Cohen Milstein Sellers & Toll, Palm Beach Gardens, FL.

For R.P.R. Enterprises, Inc., Plaintiff: W. JosephBruckner, LEAD ATTORNEY, Brian D. Clark, PROHAC VICE, Heidi M. Silton, Lockridge Grindal NauenP.L.L.P., Minneapolis, MN; Daniel Jay Mogin, TheMogin Law Firm, San Diego, CA; Donald Lewis Sawyer,Robert J. Wozniak, Freed Kanner London & Millen LLC,Bannockburn, IL; Manuel Juan Dominguez, CohenMilstein Sellers & Toll, Palm Beach Gardens, FL;Richard Frank Lombardo, PRO HAC VICE, ShafferLombardo Shurin, Kansas City, MO.

For Chandler Packaging, Inc., Plaintiff: Heidi M. Silton,W. Joseph Bruckner, LEAD ATTORNEYS, LockridgeGrindal Nauen P.L.L.P., Minneapolis, MN; Daniel JayMogin, The Mogin Law Firm, San Diego, CA; DonaldLewis Sawyer, Robert J. Wozniak, Freed Kanner London& Millen LLC, Bannockburn, IL; Lee Albert, PRO HACVICE, Murray Frank LLP, New York, NY; Manuel [*4]Juan Dominguez, Cohen Milstein Sellers & Toll, PalmBeach Gardens, FL.

For Mighty Pac, Inc., Plaintiff: Heidi M. Silton, W.Joseph Bruckner, LEAD ATTORNEYS, LockridgeGrindal Nauen P.L.L.P., Minneapolis, MN; Daniel J.Kurowski, Hagens Berman Sobol Shapiro LLP, OakPark, IL; Daniel Jay Mogin, The Mogin Law Firm, SanDiego, CA; Donald Lewis Sawyer, Robert J. Wozniak,Freed Kanner London & Millen LLC, Bannockburn, IL;Manuel Juan Dominguez, Cohen Milstein Sellers & Toll,Palm Beach Gardens, FL.

For Packaging Corporation of America, Defendant:Barack S. Echols, Daniel E. Laytin, Douglas JamesKurtenbach, Jeannie Young Evans, Leonid Feller, LukeChristian Ruse, Kirkland & Ellis LLP, Chicago, IL.

For International Paper, Defendant: James T. Mckeown,LEAD ATTORNEY, Foley & Lardner Llp, Milwaukee,WI; Michael M. Conway, LEAD ATTORNEY, JoanneLee, Foley & Lardner, Chicago, IL; Gerald MichaelHalfenger, James T. McKeown, PRO HAC VICE, Foley& Lardner, Milwaukee, WI; Nathan P. Eimer, Susan M.Razzano, Eimer Stahl LLP, Chicago, IL.

For Norampac Industries Inc., Cascades, Inc.,Defendants: Scott M. Mendel, LEAD ATTORNEY, JohnEdward Susoreny, Lauren Nicole Norris, K&L GatesLLP, Chicago, IL.

For Weyerhaeuser [*5] Company, Defendant: DavidMarx, Michelle S. Lowery, LEAD ATTORNEYS,McDermott, Will & Emery LLP (Chicago), Chicago, IL;Jennifer A Smulin Diver, Rachael V Lewis, PRO HACVICE, McDermott Will & Emery LLP, Chicago, IL.

For Georgia Pacific LLC, Defendant: Deborah KayBrown, PRO HAC VICE, Quinn Emanuel, New York,NY; James R. Figliulo, Figliulo & Silverman, Chicago,IL; Kyle R Taylor, PRO HAC VICE, Marc L.Greenwald, PRO HAC VICE, Sami Husayn Rashid, PROHAC VICE, Stephen R Neuwirth, PRO HAC VICE,Quinn Emanuel Urquhart & Sullivan, Llp, New York,NY; Michael B Carlinsky, PRO HAC VICE, QuinnEmanuel Urquhart Oliver & Hedges LLP, New York,NY; Stephanie D. Jones, Figliulo & Silverman, P.C.,Chicago, IL.

For Temple-Inland, Inc., Defendant: Andrew StanleyMarovitz, Britt Marie Miller, Courtney Lynn Anderson,Mayer Brown LLP, Chicago, IL.

For RockTenn CP, LLC, Defendant: R. Mark McCareins,LEAD ATTORNEY, James Franklin Herbison, JosephLawrence Siders, Michael P Mayer, Winston & StrawnLLP, Chicago, IL.

For Cascades Canada Inc., Norampac Holdings U.S. Inc.,Defendants: Scott M. Mendel, LEAD ATTORNEY,K&L Gates LLP, Chicago, IL.

JUDGES: Nan R. Nolan, United States MagistrateJudge.

OPINION BY: Nan R. Nolan

OPINION

MEMORANDUM [*6] OPINION AND ORDER

Lawyers have twin duties of loyalty:While they are retained to be zealousadvocates for their clients, they bear aprofessional obligation to conductdiscovery in a diligent and candid manner.


Their combined duty is to strive in the bestinterests of their clients to achieve the bestresults at a reasonable cost, with integrityand candor as officers of the court.Cooperation does not conflict with theadvancement of their clients' interests--itenhances it. Only when lawyers confuseadvocacy with adversarial conduct arethese twin duties in conflict.

The Sedona Conference, The Sedona ConferenceCooperation Proclamation, 10 Se-dona Conf. J. 331, 331(2009).

This matter is before the Court on Plaintiffs' Motionto Compel Defendants to Produce Documents and Datafrom All Reasonable Accessible Sources [Doc. 347],Plaintiffs' Motion to Compel Temple-Inland to IncludeAdditional Document Custodians [Doc. 366], Plaintiffs'Motion to Compel International Paper Company toInclude Additional Document Custodians [Doc. 382], andDefendant Georgia-Pacific LLC's Motion for ProtectiveOrder [Doc. 373]. 1 However, this is a story as muchabout cooperation as dispute.

1 Plaintiffs have [*7] also filed a Motion toCompel Defendants to Produce Documents andData for the Time Periods Proposed by Plaintiffs[Doc. 345], which the Court will address in aseparate order.

I. BACKGROUND

A. Procedural

This action is the outgrowth of a number of classactions that had been brought against DefendantsGeorgia-Pacific LLC ("GP"), Packaging Corporation ofAmerica ("PCA"), International Paper Company ("IP"),Cascades Canada, Inc./Norampac Holdings U.S., Inc.("Norampac"), Weyerhaeuser Company("Weyerhaeuser"), Temple-Inland, Inc. ("TIN") andSmurfit-Stone Container Corporation ("Smurfit") 2

charging violations of Sherman Act §1. The instantaction, which is the first-filed case in these consolidatedactions, was filed on September 9, 2010, and the relatedlater-filed cases were subsequently reassigned to theDistrict Judge. On November 8, 2010, Plaintiffs filed aConsolidated and Amended Complaint ("Complaint").

2 On May 27, 2011, Smurfit was acquired byRock-Tenn Company. To effect the acquisition,Smurfit was merged into a subsidiary ofRock-Tenn Company. The surviving entity fromthe merger became RockTenn CP, LLC("RockTenn"), a limited liability company andwholly owned subsidiary of [*8] Rock-TennCompany. On June 16, 2011, the Court allowedRockTenn to be substituted as a defendant inplace of Smurfit.

B. The Complaint3

3 This summary of the Complaint is adoptedfrom the Court's opinion denying Defendants'motions to dismiss.

Countless industrial and consumer products aremanufactured from container-board, the principal rawmaterial used to manufacture corrugated products such aslinerboard and corrugated boxes. (Compl. ¶ 36.) Thus,the prices of those corrugated products are tied directly tothe price of containerboard. (Id. ¶ 40.) From the 1930sonward, the containerboard industry has been subject toextensive antitrust litigation and other charges of unfaircompetition. (Id. ¶¶ 57-62.) In this instance, Plaintiffsallege the existence of anticompetitive behaviorbeginning in August 2005 and continuing through thepresent. (Id. ¶ 1.)

During the class period, the containerboard industrywas heavily consolidated. (Compl. ¶ 39.) Significantbarriers to entry in the form of capital-intensive startupcosts and high transportation costs make that industrysusceptible to an oligopolistic structure. (Id. ¶ 41.)Because of those barriers to entry and the degree ofconsolidation [*9] in the industry, containerboardindustry firms share a similar cost structure. (Id. ¶¶47-48.) While the containerboard industry isconsolidated, no single firm has sufficient market powerto control the supply and price of the product. (Id. ¶ 52.)Because there are no close substitutes for containerboard,the demand for the product is inelastic. (Id. ¶¶ 50-51.)

In August 2005, the containerboard industry facedcomplex environmental factors, including declining profitmargins, rising demand and a promising economicenvironment. (Compl. ¶ 64.) Between 2003 and 2005,individual producers had tried and failed to institute priceincreases at least twice. (Id.) In 2005, many of theDefendants, including Smurfit, PCA, IP, TIN, GP and


Norampac, significantly reduced their productioncapacity through plant closures, capacity idling orscheduled production downtime. (Id. ¶¶ 71-72, 75, 78-81,84-87.) Those capacity reductions coincided in time withthe existence of a high demand for containerboard. (Id. ¶¶73- 75.)

There are a number of important industry groups andtrade associations in the containerboard industry. (Compl.¶¶ 53-56.) Most Defendants belong to one or both of twoprominent organizations: [*10] the Fibre BoxAssociation ("FBA") and the American Forest and PaperAssociation. (Id. ¶¶ 53-54.) In June 2005, industryleaders, including many representatives of Defendants,attended an industry conference where pricing strategieswere discussed. (Id. ¶¶ 76-77.) Just over three monthsafter the conference, Smurfit, PCA and GP eachannounced a $30 per ton price increase effective October1, 2005. (Id. ¶¶ 83-84.) After an FBA conference onSeptember 27, 2005, the remaining Defendants followedsuit. (Id. ¶ 89.) Only a few months later, on November28, the FBA Board of Directors met. (Id. ¶ 91.) At thatmeeting, both Weyerhaeuser and PCA announced $40 perton price increases effective January 1, 2006, which wassoon matched by all other Defendants. (Id. ¶¶ 91, 98.)Then on March 14, 2006, the FBA Executive Committeemet again, and Defendants raised their prices by $50 perton only a few weeks later. (Id. ¶ 104.)

Even though prices were increasing throughout 2005and into 2006, during that period many of the Defendantsreduced capacity. (Compl. ¶¶ 115, 117-18.) Such capacitydecreases continued through 2007. (Id. ¶¶123, 125, 132.)Shortly after a large industry conference in June 2007,[*11] both PCA and Smurfit announced $40 per ton priceincreases effective August 1. (Id. ¶¶ 126-27.) On or aboutAugust 1, all other Defendants followed suit. (Id. ¶ 129.)

In late March 2008, as the economy was beginningto decline, the industry conducted a series of conferences.(Compl. ¶¶ 138-39.) Less than two months later, in earlyMay 2008, both GP and Smurfit announced $55 per tonprice increases to take effect July 1. (Id. ¶ 141.) Most ofthe remaining Defendants instituted identical priceincrease in the same time period. (Id.) About a monthlater, IP announced its intention to raise prices by anadditional $60 per ton on October 1, 2008. (Id. ¶ 145.)IP's price increase was followed by the majority of theindustry soon thereafter. (Id.) Over the course of theremainder of the year, the industry saw a continued

decrease in production. (Id. ¶¶ 150, 152.)

Despite a drastic economic slowdown in 2009,containerboard prices were inelastic and, for the mostpart, remained at the levels achieved by earlier priceincreases. (Compl. ¶¶ 154-58.) In the face of economicweakness and normal seasonal weakness, Defendantsraised prices by $50 per ton on January 1, 2010.Defendants continued to [*12] raise prices through thesummer of 2010, raising prices by $60 per ton on April 1and August 1, 2010. (Id. ¶¶ 163, 167.)

II. DISCOVERY PROCESS

A. Timeline

On January 14, 2011, Defendants each filed a motionto dismiss. On April 8, 2011, the Court entered amemorandum opinion and order denying Defendants'motions. Kleen Prods., LLC v. Packaging Corp. ofAmerica, 775 F. Supp. 2d 1071 (N.D. Ill. 2011). On May2, 2011, Defendants filed their answers.

Following the Court's denial of the motions todismiss, the parties began their initial discovery. On May3, 2011, Plaintiffs served Defendants with their FirstRequest for Production of Documents ("RPD"). The RPDhad 94 requests, which Plaintiffs divided into threegeneral categories: (1) conduct requests, (2) datarequests, and (3) inquiries, investigations and priorlitigation requests. Defendants each filed responses andobjections to the RPD on June 6, 2011.

Defendants began producing responsive hard copydocuments in August 2011. Shortly thereafter, the partiesbegan meet-and-confer discussions in an attempt toresolve issues raised in Defendants' responses andobjections and to draft a protocol for the production ofESI. On October 26, 2011, [*13] the Court entered astipulated Order governing the production format ofe-mail, e-files and paper documents and other relatedmatters ("ESI Production Order") [Doc. 245].

Defendants served their First Set of Requests toPlaintiffs for the Production of Documents on May 6,2011. Plaintiffs served their responses and objections onJune 10, 2011. Plaintiffs began producing paperdocuments in August 2011 and completed their paperproduction in November 2011.

In November 2011, Plaintiffs served Rule 30(b)(6)


notices on each Defendant relating to documentpreservation and information systems issues. Afterconducting meet-and-confer sessions, the parties agreedthat Defendants would provide written responses to theinformation sought in the Rule 30(b)(6) notices and thedepositions would be deferred pending Plaintiffs' reviewof the responses. 4

4 The parties ended up supplementing thewritten responses with Rule 30(b)(6) depositions.Nevertheless, the Court commends the parties'efforts to address the 30(b)(6) issues withoutlengthy depositions.

While the parties were able to resolve many of theirdiscovery issues through meet-and-confer sessions, byDecember 2011, four issues remained unresolved: [*14](1) reconciling Defendants' production to the individualRPD requests; (2) ESI search methods, (3) scope ofdocument searches, and (4) relevant time periods.

As to the first issue, Plaintiffs requested that eachDefendant provide an index associating its produceddocuments with the 29 categories Plaintiffs included inthe RPD. (Pls.' Statement [Doc. 266] 1.) Defendantsasserted that complying with Plaintiffs' request wouldimpose great expense and is not required by Rule 34.Defendants argued that the metadata required by theparties' agreed ESI Production Order provides Plaintiffswith sufficient identifying information. (Defs.' Statement[Doc. 267] 10.) Plaintiffs responded that metadata is notthe issue and that the coding fields they are requesting arenot addressed by the ESI Production Order. (Pls.'Statement 3.)

With regard to the ESI search dispute, Plaintiffscriticized Defendants' use of a Boolean search method toidentify responsive documents. (Pls.' Statement 4-16.)Plaintiffs argued that the Boolean keyword process is perse "subject to the inadequacies and flaws inherent whenkeywords are used to identify responsive documents."(Id. 8.) They requested that Defendants use [*15]"content-based advanced analytics ('CBAA') technologyanalytics to conduct natural language, subject mattersearches across corporate departments or corporatefunctions, using content-based search technology ratherthan keywords." (Id. 5.)

Defendants defended their use of Boolean keywords,arguing that their testing and validations processes "willhave a degree of accuracy that meets or surpasses not

only industry standards but also the likely accuracy ofany other available methodology." (Defs.' Statement 3.)Further, given that by December 2011, Defendants hadalready collected and produced a significant amount ofresponsive information, they contend that Plaintiffs'approach would involve additional costs and burdens notcontemplated by the Federal Rules, the Seventh CircuitElectronic Discovery Pilot Program, 5 or case law. (Id.8-10.)

5 See http://www.discoverypilot.com.

As to the third issue, Plaintiffs criticized Defendantsfor limiting their search to "main active servers." (Pls.'Statement 16.) Defendants stated they were willing toconsider searching for responsive archived documents ifPlaintiffs both identify a "particular and compellingneed" and agree to bear the costs of [*16] restoring thearchived information. (Defs.' Statement 12-13.)

Finally, with regard to the temporal scope dispute,Plaintiffs broke down their document requests into thefollowing time periods:

Conduct Requests: January 1,2002-December 31, 2010;

Data Requests: January 1,2000-December 31, 2010;

Inquiries, Investigations and PriorLitigation Requests: January 1,1996-December 31, 2010; and

The prior antitrust litigation identifiedin the Complaint: no time limitation.

(Pls.' Statement 17.) In response, Defendants proposed toproduce conduct request documents only back to January1, 2004, and data request documents back to January 1,2003. (Defs.' Statement 14.) They argued that those timeperiods were appropriate based on the Complaint'sallegations that the conspiracy began in mid-2005. (Id.)

On January 10, 2012, the case was referred to theundersigned Magistrate Judge for discovery supervision.In February and March 2012, the Court conducted twofull days of evidentiary hearings on the ESI searchdispute issue. Over the next six months, the Court held 11status hearings and Rule 16 conferences with all parties,many of which lasted a half day or longer. The Court also


facilitated three Rule 16 [*17] conferences betweenPlaintiffs and individual Defendants. A clientrepresentative attended many of these conferences. Dueto counsel's hard work, the parties and the Court wereable to resolve a number of the issues throughmeet-and-confer discussions. These cooperativeendeavors are described in the next section.

B. Issues on Which Parties Have Reached Agreement

1. Search Methodology

As discussed above, the parties had a fundamentaldispute over what search methodology Defendants shouldutilize to identify documents responsive to Plaintiffs'RPD. Defendants argued that in order to best identifypotentially responsive ESI, they engaged leadingconsulting companies to develop Boolean search terms.(Pls.' Evidentiary Hr'g Br. [Doc. 288] 3.) During aniterative process, Defendants and their consultantsrevised and refined the search terms over the course ofseveral months. (Id. 4.) Sampling procedures were usedthroughout the process to evaluate the effectiveness andreliability of the search terms. (Id.)

On the other hand, Plaintiffs argued that Defendants'search methodology is likely to find less than 25% ofresponsive documents. (Defs.' Evidentiary Hr'g Mem.[Doc. 290] 1.) They asserted [*18] that their proposedcontent-based advance analytics ("CBAA") search wouldfind more than 70% of responsive documents at nogreater cost. (Id. 1-2.) Plaintiffs contended that becauseCBAA searches do not focus on matching words butinstead on identifying relevant concepts out of thedocuments, CBAA searches provide a richer,substantially more accurate return than Boolean searches.(Id. 5.) Thus, they concluded that "[a] CBAA search ofdocuments collected by corporate or department functionis the best and most practical method for locatingresponsive ESO in this case." (Id. 6.)

In an attempt to resolve this impasse, an evidentiaryhearing was held on February 21 and March 28, 2012,with both sides presenting witnesses in support of theirpositions. At the conclusion of the second day, the Courtobserved that under Sedona Principle 6, "[r]espondingparties are best situated to evaluate the procedures,methodologies, and techniques appropriate for preservingand producing their own electronically storedinformation." See The Sedona Conference, The Sedo-naConference Best Practices Commentary on the Use of

Search and Information Retrieval Methods inE-Discovery, 8 Sedona Conf. J. 189, 193 (Fall 2007).[*19] Accordingly, the Court urged the parties toconsider whether there was a method to refine orsupplement Defendants' Boolean search so that Plaintiffshad a reasonable assurance that they were receiving ahigh percentage of responsive documents withoutcompletely scrapping Defendants' search methodology.

Over the course of the next five months, the partiesand the Court continued to meet and confer on this issue.Finally, in August 2012, the parties came to anunderstanding on the ESI search methodology issue, andon August 21, 2012, the Court entered a stipulated orderrelating to ESI search ("ESI Search Order") [Doc. 385]. 6

In the ESI Search Order, Plaintiffs agreed to withdrawtheir demand that Defendants apply CBAA to documentsthat have been or will be collected in response to the RPD("First Request Corpus"). As to any documents or ESIbeyond the First Request Corpus, Plaintiffs agreed not toargue or contend that Defendants should be required toapply CBAA or "predictive coding" methodology withrespect to any requests for productions served on anyDefendant prior to October 1, 2013. With respect to anyrequests for production served on any Defendant on orafter October 1, 2013 [*20] that requires the collection ofdocuments beyond the First Request Corpus, the partiesagreed to meet and confer regarding the appropriatesearch methodology to be used for such newly collecteddocuments.

6 A copy of the ESI Search Order is attached asExhibit A.

2. Custodians

After Defendants filed their responses and objectionsto the RPD in June 2011, the parties beganmeet-and-confer sessions in an attempt to work out theirdifferences. In August 2011, Defendants identified 109individuals for which they intended to collectcustodian-specific documents responsive to the RPD.Because the Complaint alleged that "Defendants and theirco-conspirators conducted their conspiracy in secret andkept it mostly within the confines of their higher-levelexecutives" (Compl. ¶ 192; see id. ¶ 6), Defendantsincluded as custodians their higher-level executives withdecision-making responsibility for the pricing and sale ofcontainer-board and corrugated products during therelevant period, along with some of their subordinates.Plaintiffs complained that more custodians were needed,


especially within the sales and marketing departmentsand at the plant level. Subsequent meet-and-conferdiscussions, [*21] some of which the Court supervised,resulted in a few additional custodians.

During the Court's meet-and-confer sessions with theparties, the Court encouraged Defendants to expand theirlists of custodians and include at least a sample oflower-level and plant-level employees. At the Court'surging, Defendants produced litigation hold lists to aidPlaintiffs in identifying appropriate custodians. 7 (See,e.g., Docs. 325, 350-52.) Thereafter, to move the processalong, the Court urged Plaintiffs to make formal requestsfor specific custodians from each Defendant. On July 27,2012, after reviewing the litigation hold lists anddocuments produced by Defendants to date, Plaintiffsmade formal requests for additional custodians.Thereafter, the parties promptly engaged inmeet-and-confer discussions. Those discussionsultimately led to agreements with five of the sevenDefendants. 8 (See Docs. 409, 410.) In those agreements,the parties agreed on specific additional custodians and aprotocol for how they would handle future requests foradditional custodians.

7 A copy of the order is attached as Exhibit B.8 A copy of one such agreement is attached asExhibit C.

3. Document Requests

Plaintiffs [*22] have objected to the form andsubstance of many of Defendants' objections andresponses to the RPD. They contend that the responsesmake it difficult to determine what was produced inresponse to each document request. Duringmeet-and-confer sessions facilitated by the Court,Defendant IP prepared a chart that described the specificESI searches it made to respond to each of the separateRPD requests. Plaintiffs agreed that this chart was helpfuland could alleviate some of their concerns. Thereafter, IPagreed to revise its RPD responses to incorporate theinformation in the chart. IP served its revised RPDresponses on July 23, 2012.

The Court determined that the most efficient use ofthe Court's and the parties' time is to address the RFPissue in stages. During phase one, Plaintiffs are meetingand conferring solely with IP in an effort to addressPlaintiffs' objections to IP's revised responses. AfterPlaintiffs finish their review of the revised responses,

they will meet and confer with IP to see if they can workout any remaining disputes. After the meet-and-conferprocess is completed, Plaintiffs may file a motion tocompel against IP regarding any unresolved RFPresponses. [*23] Phase two will include the remainingDefendants and will commence after the Court has ruledon Plaintiffs' motion to compel against IP. To narrow anyremaining disputes, the Court expects the parties to usethe IP process as a guide.

4. Sources of Data

Plaintiffs have asserted that Defendants shouldsearch and produce responsive information regardless ofwhere it is stored. Defendants countered that they willsearch all active files for each of their custodians.Defendants are also prepared to search for categories ofdata that are stored in centralized corporate systems andexist on media or servers which can be searched usingDefendants' current system capabilities and withoutincurring undue costs. With regard to archived media,Defendants will consider searching for responsivedocuments where Plaintiffs identify a specific need andagree to bear the costs of restoring the archiveddocuments.

The parties and the Court have conductedmeet-and-confer sessions in an effort to find somecommon ground on these issues. While the partiescontinue to dispute both the cost and the need to restoreinformation from archived media, Defendants haveagreed to preserve all such media during the pendency[*24] of the case. Further, each Defendant has agreed thatto the extent it has not already done so, it will determineif any of its custodians maintained potentially responsiveinformation on any personal device not synced to itsservers. If any potentially responsive information isidentified, it will be reviewed and responsive,nonduplicative, nonprivi-leged information will beproduced. 9

9 The parties also continue to meet and confer inan effort to agree on the information Defendantswill provide in response to Plaintiffs' transactionaldata requests.

III. DISCUSSION

The following issues remain in dispute.

A. Georgia-Pacific Motion for Protective Order


On August 14, 2012, Defendant Georgia-Pacificfiled a Motion for Protective Order. In its Motion, GPrequests that the Court quash Plaintiffs' SixthInterrogatory.

1. Background

The Court facilitated a Rule 16 Conference betweenGP and Plaintiffs on May 31, 2012. Prior to theConference, the parties engaged in numerous discussionsconcerning GP's responses to Plaintiffs' initial documentrequests. (Resp. 6.) Plaintiffs complained that "GPrefuses to identify which of its personnel were involvedin Con-tainerboard Products business other than [*25]'primary decision makers' or reveal its corporatestructure." (Mot. Ex. 7 at 11.) During the Conference, GPexplained that since 2005, it has been a privately-heldcompany and does not maintain organizational charts orjob descriptions. (Hr'g Tr., May 31, 2012, at 36, 46.)However, in the interests of cooperation and compromise,GP agreed to produce a list of the individuals whoreceived the litigation-hold notice in connection with thislitigation, along with their job titles. (Id. 36-41, 48-50.)

On June 26, 2012, GP produced the list oflitigation-hold recipients. (Mot. Ex. 1, Attach. A.) Thelist included not only the names and titles of thelitigation-hold recipients, but also the GP division inwhich each recipient works. (Id.)

Three days later, on June 29, 2012, Plaintiffs servedtheir Sixth Interrogatory request, demanding variousbackground information over an eight-year period foreach of the approximately 400 persons on thelitigation-hold list. (Mot. Ex. 1.) GP informally objectedto the request, contending that it was unfair andburdensome. (Mot. 5.) Thereafter, the parties met andconferred in an effort to address GP's concerns. (Id. 5-6,Ex. 9.) During a telephonic status [*26] conference onJuly 25, 2012, GP indicated to the Court that it may haveto move for protection from the Sixth Interrogatory. (Id.Ex. 10 at 27-29.) In response, Plaintiffs stated that theyexpected the upcoming Rule 30(b)(6) deposition,scheduled for August 1, to largely eliminate Plaintiffs'need for the Sixth Interrogatory. (Id. Ex. 10 at 29-30.)

Plaintiffs took the Rule 30(b)(6) deposition onAugust 1, 2012. GP's designee, George Ragsdale, wasprepared to answer questions about the structure of, andpersonnel in, GP's containerboard and packagingbusinesses, including specific questions about the

individuals and job titles on GP's list of litigation-holdrecipients. (Ragsdale Dep. 166-68, 192-99.) 10 WhenPlaintiffs' counsel announced, well before the expirationof seven hours, that he had completed questioning thewitness, GP's counsel reminded Plaintiffs' counsel thatthe witness was available to answer questions about thelitigation-hold list. (Id. 205-08.) Thereafter, Plaintiffscontinued the deposition, asking the witness furtherquestions, but ultimately ended the deposition before theexpiration of seven hours. (Mot. 8.)

10 The transcript of the Rule 30(b)(6) depositionis available [*27] at Exhibit 11 to GP's Motionand Exhibit C to Plaintiffs' Response.

The next day, GP requested that Plaintiffs withdrawthe Sixth Interrogatory, believing that Plaintiffs hadobtained all of their needed information during the Rule30(b)(6) deposition. (Mot. 8.) Plaintiffs refused, statingthat the Sixth Interrogatory "is hardly burdensome" andcan be "answered by a small production of paper." (Id.Ex. 12 at 13.)

2. Analysis

GP contends that the Sixth Interrogatory (a)"imposes undue and disproportionate burdens, andconstitutes an abusive response to GP's agreementvoluntarily to provide its list of litigation hold recipientsand their job titles;" (b) "with its multiple subparts, bothon its own and when combined with prior interrogatories,exceeds the number of interrogatories permitted by theFederal Rules;" and (c) "violates Plaintiffs' expresscommitment not to seek further discovery about GP'sorganizational structure after GP responded to previousinterrogatories on these subjects." (Mot. 3.) Plaintiffsdisagree, contending that the Sixth Interrogatory "can beanswered through the production of documents pursuantto Fed. R. Civ. P. 33(d), if GP would produce the jobdescriptions [*28] that it maintains in the ordinary courseof its business." (Resp. 1-2.) After carefully reviewing theissue, the Court finds that a protective order is warranted.

First, issuing the Sixth Interrogatory within days ofreceiving the list of litigation-hold recipients violated thespirit of cooperation that this Court has encouraged theparties to pursue. The Rule 16 conference was held tofacilitate cooperative discussions between the parties onissues related to Defendants' document production.During the conference, the discussion largely focused onPlaintiffs' need to understand GP's key personnel and


who some of the individuals are whose names wereappearing in the document production. GP explained thatit is a privately-held company and does not maintainorganizational charts or job descriptions. (Hr'g Tr., May31, 2012, at 36, 46.) The Court, acting as a neutral,facilitated an informal brainstorming discussion with theparties.

The Court observed that without a formalorganizational chart, Plaintiffs need a betterunderstanding of GP's organizational structure. In thiscontext, the Court suggested that one possible solutionmight be to provide Plaintiffs with a list of GP'slitigation-hold [*29] recipients. (See, e.g., Hr'g Tr. 10,11, 14, 20, 31, 33, 35, 36.) However, the Court'ssuggestion was never intended to generate additionaldiscovery obligations. Accordingly, the Court expectedthat Plaintiffs would use the list of litigation-holdrecipients, which included the recipients' job titles and thedivision in which they worked, in conjunction with theirreview of GP's documents to create their ownorganizational charts. (Id. 50.) Instead, Plaintiffs violatedthe spirit of cooperation and used the list oflitigation-hold recipients to request the additionaldiscovery. Such a response could have a chilling effect onboth litigants and courts to engage in candid discussions.

Second, GP has established an undue burden inresponding to the Sixth Interrogatory. "All discovery issubject to the limitations imposed by Rule 26(b)(2)(C)."Fed. R. Civ. P. 26(b)(1). The Rule 26 proportionality testallows the Court to "limit discovery if it determines thatthe burden of the discovery outweighs its benefit." In reIKB Deutsche Industriebank AG, No. 09 CV 7582, 2010U.S. Dist. LEXIS 35924, 2010 WL 1526070, at *5 (N.D.Ill. Apr. 8, 2010). Rule 26(b)(2)(C)(iii) requires a court tolimit discovery if it determines that [*30] "the burden orexpense of the proposed discovery outweighs its likelybenefit, considering the needs of the case, the amount incontroversy, the parties' resources, the importance of theissues at stake in the action, and the importance of thediscovery in resolving the issues." In other words, "Rule26(b)(2)(C)(iii) empowers a court to limit the frequencyor extent of discovery if it determines that the burden orexpense of the proposed discovery outweighs its likelybenefit or that it is unreasonably cumulative orduplicative." Sommerfield v. City of Chicago, 613 F.Supp. 2d 1004, 1017 (N.D. Ill. 2009) objectionsoverruled, 06 C 3132, 2010 U.S. Dist. LEXIS 19440,2010 WL 780390 (N.D. Ill. Mar. 3, 2010). "The 'metrics'

set forth in Rule 26(b)(2)(C)(iii) provide courtssignificant flexibility and discretion to assess thecircumstances of the case and limit discovery accordinglyto ensure that the scope and duration of discovery isreasonably proportional to the value of the requestedinformation, the needs of the case, and the parties'resources." The Sedona Conference, The SedonaConference Commentary on Proportionality in ElectronicDiscovery, 11 Sedona Conf. J. 289, 294 (2010); seeSommerfield, 613 F. Supp. 2d at 1017 [*31] ("Theapplication of Rule 26(b)(2)(C)(iii) involves a highlydiscretionary determination based upon an assessment ofa number of competing considerations.").

The Sixth Interrogatory requests for each of theapproximately 400 persons on the litigation-hold list, andfor an eight-year period, that GP:

(a) describe their job functions;

(b) identify their employer;

(c) identify all the persons to whomthey reported, the period during whichthey so reported, and the job titles,employer and division of the person towhom they reported; and

(d) identify all the persons whoreported to them, the period during whichthey so reported, and the job titles,employer and division of the reportingperson.

(Mot. Ex. 1.) GP has demonstrated that it "does notmaintain any single report in the ordinary course ofbusiness from which this information could be readilyobtained." (Mary K. McLemore Decl. ¶ 4.) Indeed, giventhe temporal scope of the request, some of thisinformation would be difficult, or perhaps impossible, toobtain. (Id. ¶¶ 4-8.) To attempt to respond to the SixthInterrogatory, GP would need to (a) conduct a multi-stepprocess to compile the information that is available in itsSAP Data Warehouse, [*32] and then (b) interview eachof the litigation-hold recipients-- along with theirsupervisors and direct reports--to confirm theinformation. (Id. ¶¶ 4-6.) GP estimates that the processcould take as many as 800 hours to accomplish. (Id. ¶ 7.)Even after these hours were devoted, the response wouldlikely be incomplete and could be impossible to verify.(Id. ¶ 8.)


In responding to the Motion, Plaintiffs do not addressthe Rule 26 proportionality principle. In other words, theydo not explain how the value of the proposed discoveryoutweighs its burden. Instead, Plaintiffs contend that GPcan answer the Sixth Interrogatory by producing the jobdescriptions that it maintains. (Resp. 13) ("GP can readilyanswer Plaintiffs' Sixth Interrogatory through theproduction of a defined set of documents that GP canreadily assemble and produce.") However, GP hasconsistently stated that it does not maintain formalorganizational charts or job descriptions. (See, e.g., Hr'gTr. 36 ("Since 2005, [GP] has not been a public company[and] there are not in existence organizational charts."),46 ("There are no written job descriptions in [GP's]system."); Resp. Ex. E at 2 (GP "has searched and doesnot believe [*33] such formal written job descriptionsexist.").)

Plaintiffs assert that at the Rule 30(b)(6) deposition,"Plaintiffs learned for the first time that as anorganization, GP's practice was to generate a Roles,Responsibilities and Expectations ("RRE") document foreach employee that sets forth that employee's duties, aswell as providing pertinent reporting information." (Resp.Ex. E at 2.) Thus, Plaintiffs assert that "[w]ere GP toproduce its RREs for the individuals identifies on GP'slitigation hold list, Plaintiffs would accept their SixthInterrogatory as answered." (Id. 2-3.)

But Plaintiffs misapprehend the testimony of GP'srepresentative. The phrase "job description" does notappear in the transcript, and Plaintiffs' counsel did noteven inquire if RREs were job descriptions. On thecontrary, the witness testified that job descriptions wouldbe meaningless because an individual employee'sresponsibilities are defined annually on an individualbasis by the employee and his or her supervisor and aresubject to change at any time. (Ragsdale Dep. 144-45.)Thus, the RRE is retained locally "as a living documentbetween the boss and the subordinate[;] it's updatedannually, or at least [*34] it's reviewed annually forupdate, and, theoretically, once it's updated, the old onegoes away." (Id. 147-48.) Furthermore, the RREs do notnecessarily provide all of the information required by theSixth Interrogatory. "For example, rather than providing afixed job description for a title, a RRE is intended to be anegotiated framework between each employee and his orher supervisor that is used to focus an employee onplanned outcomes for each year." (McLemore Decl. ¶11.) And, while the RREs identify the employee's

supervisor, they do not provide the additional reportingrelationships sought by Plaintiffs. (Id.)

The Court agrees. The RRE is a document used aspart of an individual employee's job performance review,not a document used by GP to describe a particular job.Significantly, job descriptions are generic documentsmeant to apply to all individuals at all times, whileperformance evaluations are highly personal documentsthat apply to a single employee at a particular point intime. See, e.g., Hooper v. Total System Services, Inc., 799F. Supp. 2d 1350, 1362 (M.D. Ga. 2011) (describing "jobdescriptions" as "generic and meant to be used acrossbusiness units"); Loeb v. Best Buy Co, Inc., No. 05-720,2007 U.S. Dist. LEXIS 58039, 2007 WL 2264729, at *1,*15 (D. Minn. Aug. 6, 2007) [*35] (describing "jobdescriptions" as "fairly generic and broad" and"performance reviews" as means to determine ifemployee is meeting current expectations).

In any event, the Rule 26 proportionality testcautions against producing the RREs. Indeed, producingthe RREs as an alternative to responding to the SixthInterrogatory would be no less burdensome. "RREs arenot centrally maintained in any location and can beupdated more than once each year." (McLemore Decl. ¶12.) Thus, "GP would have to interview each of the[litigation-hold recipients] to determine whether theymaintain any current or historical RREs." (Id.) "GPwould also be required to interview each of theirsupervisors to determine if they maintain any RREs forthe identified individuals." (Id.) Plaintiffs contend thatGP could simply request all RREs by emailing each ofthe litigation-hold recipients. (Resp. 13-14 & n.8.) ButPlaintiffs fail to discuss how this process would workwith former employees. Furthermore, GP could notprovide a complete, accurate and verified response to theSixth Interrogatory without also determining who theemployees' supervisors were during the eight-year periodand contacting them to collect [*36] any RREs in theirpossession. (See McLemore Decl. ¶¶ 5, 12.) Moreover,Plaintiffs do not explain how any benefit from securingmore RREs outweighs the burden of producing them.Despite having identified some 25 RREs in GP'sproduction (Resp. 13), Plaintiffs do not describe howthese documents have aided their document review orcould lead to relevant information.

Finally, much of the information sought in the SixthInterrogatory has already been obtained through a more


convenient, less burdensome, and less expensive method."[T]he court must limit the frequency or extent ofdiscovery otherwise allowed by these rules or by localrule if it determines that . . . the discovery sought isunreasonably cumulative or duplicative, or can beobtained from some other source that is more convenient,less burdensome, or less expensive . . . ." Fed. R. Civ. P.26(b)(2)(C)(i). "In other words, where relevantinformation is available from multiple sources, the Rulesgive courts the authority to limit discovery to the leastburdensome source, thus empowering courts to controllitigation costs and promote efficiency in accordance withRule 1." The Sedona Conference, The Sedona ConferenceCommentary on Proportionality in Electronic Discovery,11 Sedona Conf. J. 289, 296-97 (2010); [*37] see Fed. R.Civ. P. 1 (The Rules "should be construed andadministered to secure the just, speedy, and inexpensivedetermination of every action and proceeding.").

Here, GP provided a Rule 30(b)(6) witness toprovide testimony about the litigation-hold recipients,specifically "their responsibilities at GP, the identities ofthose reporting to him or her, and the identities of thoseto whom he or she reported, whether directly or in a'dotted line' relationship." (Resp. Ex. R.) The witnessanswered almost all of the questions asked of himregarding GP's organizational structure and job titles, andPlaintiffs concluded the deposition early rather thanasking any additional questions of the deponent.(Ragsdale Dep. 166-68, 192-99, 205-08.) In fact, inresponse to GP's request that Plaintiffs use the remainingtime to ask additional questions, Plaintiffs' counselresponded that while "[t]here may be small areas thatremain where we need some further information," "Ibelieve, on my recollection without review of thetranscript, that I have accomplished what Plaintiffs, at aminimum to satisfy [the Sixth Interrogatory], need toaccomplish." (Id. 206.) Thereafter, Plaintiffs continuedthe [*38] deposition, asking the witness furtherquestions, but ultimately ending the deposition before theexpiration of seven hours. (Id. 206-30.)

Plaintiffs do not dispute that GP's corporate witnesswas qualified and able to provide significant informationon the above topic. Instead, they argue that "there weremore than a half dozen instances where Mr. Ragsdale waseither unable to testify concerning the nature of a positionor had to speculate about job functions." (Resp. 12 &n.7.) In response, GP has provided verified writtenanswers to each of the specific questions referenced in

footnote 7 to Plaintiffs' Response. (Doc. 407.) Theseadditional answers, together with the Rule 30(b)(6)deposition, provide Plaintiffs with a proportionalresponse to their Sixth Interrogatory.

In sum, the Court finds that the Sixth Interrogatoryviolated the spirit of cooperation. Further, the SixthInterrogatory does not pass the Rule 26 proportionalitytest. GP's burden of responding to the Sixth Interrogatoryoutweighs any benefits, and Plaintiffs were able to getmuch of the information in a less burdensome way.Defendant Georgia-Pacific LLC's Motion for ProtectiveOrder is granted. Plaintiffs' Sixth [*39] Interrogatory isquashed.

B. Custodians

As discussed above, the Court facilitatedmeet-and-confer discussions with the parties in an effortto find a cooperative solution to the parties' disputes overwho is an appropriate custodian. Those discussionsultimately led to agreements with GP, Norampac,Weyerhaeuser, PCA and RockTenn. (Docs. 409, 410.)However, Plaintiffs were unable to come to agreementswith IP and TIN and have filed motions to compeladditional custodians from them.

Plaintiffs contend that "[i]n this price-fixingconspiracy, [IP and TIN] should not have the unilateralability to select their own 'priority' document custodians,limit such custodians to high level executives with'primary decision making authority,' and reject Plaintiffs'reasonable requests for additional custodians that arebelieved likely to possess responsive documents."(Custodian Mem. 7.) In their motions, Plaintiffs request19 additional custodians from TIN (TIN Mem. 6) and 16additional custodians from IP (IP Mem. 8).

IP and TIN oppose the motions. IP and TIN wereeach willing to include 16 additional custodians but onlyif Plaintiffs agreed to limit the document sources for theadditional custodians [*40] and a written assurance thatany future custodian requests be accompanied withsubstantive support. (IP Resp. 1-5, 15; TIN Resp. 1-4.)Plaintiffs are unwilling to make this compromise. Theyassert that limiting sources to a subset of ESI documentswould preclude paper documents, which may the onlyviable source of discovery for the earlier time periods andfor former employees. (TIN Reply [Doc. 392] 14; see IPReply [Doc. 402] 6.) Plaintiffs also contend that IP's andTIN's "efforts to tie any additional custodians to specific


allegations in the [Complaint] or documents alreadyproduced" is inconsistent with the Federal Rules,especially Rule 26. (TIN Reply 4-5; see IP Reply 2-3.)

In antitrust cases, courts generally take an expansiveview of relevance and permit broad discovery. In re Auto.Refinishing Paint Antitrust Litig., MDL No. 1426, 2004U.S. Dist. LEXIS 29160, at *7-8 (E.D. Penn. Oct. 29,2004); see United States v. International BusinessMachines Corp., 66 F.R.D. 186, 189 (S.D.N.Y. 1974)(observing that "discovery in antitrust litigation is mostbroadly permitted and the burden or cost of providing theinformation sought is less weighty a consideration than inother cases") (citation omitted). "Broad [*41] discoveryis permitted because direct evidence of ananticompetitive conspiracy is often difficult to obtain, andthe existence of a conspiracy frequently can beestablished only through circumstantial evidence, such asbusiness documents and other records." In re Auto.Refinishing Paint Antitrust Litig., 2004 U.S. Dist. LEXIS29160, at *8; see Hosp. Bldg. Co. v. Trs. of Rex Hosp.,425 U.S. 738, 746, 96 S. Ct. 1848, 48 L. Ed. 2d 338(1976) (In antitrust cases, relevant evidence is "largely inthe hands of the alleged conspirators.") (citation omitted);Callahan v. A.E.V. Inc., 947 F. Supp. 175, 179 (W.D.Penn. 1996) ("Discovery in an antitrust case isnecessarily broad because allegations involve improperbusiness conduct. Such conduct is generally covert andmust be gleaned from records, conduct, and businessrelationships.") (citation omitted). Courts also note "thepublic importance of the decision, the need of largecorporate defendants to know which of their manyactivities are attacked, [and] the issue narrowing functionof discovery." Bass v. Gulf Oil Corp., 304 F. Supp. 1041,1046 (S.D. Miss. 1969). These factors create apredisposition among courts to allow broad discovery ofantitrust defendants.

However, "[a]ll [*42] discovery, even if otherwisepermitted by the Federal Rules of Civil Procedurebecause it is likely to yield relevant evidence, is subjectto the court's obligation to balance its utility against itscost." U.S. ex rel. McBride v. Halliburton Co., 272F.R.D. 235, 240 (D.D.C. 2011) (Facciola, M.J.); see Fed.R. Civ. P. 26(b)(2)(C). More specifically, Rule26(b)(2)(C) requires the court to consider whether "(1)the discovery sought is unreasonably cumulative orduplicative, or obtainable from a cheaper and moreconvenient source; (2) the party seeking the discovery hashad ample opportunity to obtain the sought information

by earlier discovery; or (3) the burden of the discoveryoutweighs its utility." McBride, 272 F.R.D. at 240-41; seeWillnerd v. Sybase, Inc., No. 09 C 500, 2010 U.S. Dist.LEXIS 121658, 2010 WL 4736295, at *3 (D. Idaho Nov.16, 2010) ("In employing the proportionality standard ofRule 26(b)(2)(C) . . . , the Court balances [the requestingparty's] interest in the documents requested, against thenot-inconsequential burden of searching for andproducing documents."). The third factor requires thecourt to consider (a) the needs of the case; (b) the amountin controversy; (c) the parties' resources; [*43] (d) theimportance of the issues at stake in the action; and (e) theimportance of the discovery in resolving the issues. Fed.R. Civ. P. 26(b)(2)(C)(iii). Nevertheless, "[t]he partyopposing a motion to compel carries a 'heavy' burden ofpersuasion." U.S. v. AT&T Inc., No. 11-1560, 2011 U.S.Dist. LEXIS 156303, 2011 WL 534178, at *5 (D.D.C.Nov. 6, 2011).

While the record does not afford a precisecalculation, the Court can presume, given the nature ofthe antitrust claims and the size of the companiesinvolved, that the amount in controversy is very large andthat Defendants' resources are greater than Plaintiffs'.Further, claims of collusion in the containerboard andcorrugated box industries raise important, vital issues ofpublic importance. Thus, these factors weigh in favor ofthe discovery sought by Plaintiffs. See McBride, 272F.R.D. at 241.

On the other hand, Defendants protest that Plaintiffshave not demonstrated that the proposed additionalcustodians will have important, nonduplicativeinformation. (IP Resp. 9; TIN Resp. 11-14.) Indeed,Plaintiffs do not point to any specific, noncumulativeevidence they expect to find with the additionalcustodians. Instead, they selected the additionalcustodians [*44] by examining each Defendant'sorganizational charts and the list and titles of personswho received a litigation hold notice 11 and usingmetadata to analyze which individuals were sending andreceiving emails from the sales and marketing people thatwere already identified as custodians. (Hr'g Tr. 18-21,Aug. 21, 2012.) Thus, Plaintiffs contend that theproposed individuals should be included as custodiansbecause they are senior executives with responsibilities incontainerboard, boxes, pricing, strategic planning,marketing and sales, and who "exchanged an unusuallylarge" number of emails with top sales and marketingexecutives already named as custodians. (IP Reply 3-4;


see TIN Reply 13-15.)

11 Because Temple-Inland sent a litigation-holdnotice to all employees, it did not produce alitigation-hold list.

But just because a proposed custodian exchanged alarge number of emails with a current custodian does notmean that the proposed custodians will have a significantnumber of important, non-cumulative information.Further, until Plaintiffs have had an opportunity to reviewthe huge quantity of information already produced fromthe existing custodians, it is difficult for the Court [*45]to determine the utility of the proposed discovery. SeeMcBride, 272 F.R.D. at 241 ("Without any showing ofthe significance of the non-produced e-mails, let alonethe likelihood of finding the 'smoking gun,' the [party's]demands [for additional custodians] cannot possibly bejustified when one balances its cost against its utility.");Jones v. Nat'l Council of Young Men's Christian Ass'ns ofthe United States, No. 09 C 6437, 2011 U.S. Dist. LEXIS123008, 2011 WL 7568591, at *2 (N.D. Ill. Oct. 21,2011) ("The Court finds that Plaintiffs' untargeted,all-encompassing request fails to focus on key individualsand the likelihood of receiving relevant information.");Garcia v. Tyson Foods, Inc., No. 06-2198, 2010 U.S.Dist. LEXIS 135678 , 2010 WL 5392660, at *14 (D.Kan. Dec. 21, 2010) (Waxse, M.J.) ("Plaintiffs present noevidence that a search of e-mail repositories of the 11employees at issue is likely to reveal any additionalresponsive e-mails. . . . Plaintiffs must present somethingmore than mere speculation that responsive e-mails mightexist in order for this Court to compel the searches andproductions requested.").

The Court also notes that IP already has 75custodians--by far the most of any Defendant--and hasengaged in good faith meet-and-confer [*46] discussionswith Plaintiffs that enlarged the scope of documentcollection and production that IP initially agreed toundertake. (See IP Resp. 1, 6-7.) Similarly, TIN alreadyhas 28 custodians, which is more than most of the otherDefendants.

However, the selection of custodians is more than amathematical count. The selection of custodians must bedesigned to respond fully to document requests and toproduce responsive, nonduplicative documents during therelevant period. See, generally, Eisai Inc. v.Sanofi-Aventis U.S., LLC, No. 08-4168, 2012 U.S. Dist.LEXIS 52885, 2012 WL 1299379, at *9 (D.N.J. April 16,

2012). First of all, Plaintiffs are entitled to discoveryabout Defendants' box businesses. The Complaintadequately alleges a conspiracy both of containerboardand of corrugated products, including corrugated boxes.See Kleen Prods., 775 F. Supp. 2d at 1082. Further, whilethe Complaint alleges a conspiracy mostly amonghigher-level executives (Compl. ¶ 192), it does notexclude lower level employees. More importantly, even ifthe conspiracy is among higher-level executives,lower-level employees may possess important, relevantinformation which could reasonably lead to admissibleevidence. Fed. R. Civ. P. 26(b)(1); [*47] see In reCoordinated Pretrial Proceedings in Petroleum Prods.Antitrust Litig., 906 F.2d 432, 453 (9th Cir. 1990) ("Withregard to the appellees' contention that Rogers was toolow-level an employee to be of significance, we see noreason for concluding that such information gatheringcannot be delegated to subordinates. Accordingly, thefact that Rogers did not himself have authority to makeARCO pricing decisions is not dispositive."); In re HighFructose Corn Syrup Antitrust Litig., 295 F.3d 651, 662(7th Cir. 2002) ("One of Staley's HFCS plant managerswas heard to say: 'We have an understanding within theindustry not to undercut each other's prices.' (He wascommenting on a matter within the scope of hisemployment and his comment was therefore admissibleas an admission by a party. Fed. R. Evid.801(d)(2)(D).)"); see also In re SRAM Antitrust Litig.,No. 07-1819, 2010 U.S. Dist. LEXIS 132172, at *40,*45-48, *51 (N.D. Cal. Dec. 10, 2010). Thus, in anantitrust case such as this, Plaintiffs are at least entitled toa sample of lower-level and plant-level employees todetermine if they possess significant and nonduplicativeinformation.

Further, IP and TIN have not established [*48] anundue burden to producing information from theadditional custodians. IP contends that its burden isundue because it has "already produced more than 4million pages, with more to come based upon prioragreements with Plaintiffs." (IP Resp. 7 (emphasisadded); see id. 11.) Similarly, TIN requests "anacknowledgment of the considerable burdens it alreadyhas been subjected to and some assistance . . . incontrolling this burden going forward." (TIN Resp. 11.)But a party must articulate and provide evidence of itsburden. While a discovery request can be denied if the"burden or expense of the proposed discovery outweighsits likely benefit," Fed. R. Civ. P. 26(b)(2)(C)(iii), a partyobjecting to discovery must specifically demonstrate how


the request is burdensome. See Heraeus Kulzer, GmbH v.Biomet, Inc., 633 F.3d 591, 598 (7th Cir. 2011); Sauer v.Exelon Generation Co., No. 10 C 3258, 2011 U.S. Dist.LEXIS 90511, 2011 WL 3584780, at *5 (N.D. Ill. Aug.15, 2011). This specific showing can include "an estimateof the number of documents that it would be required toprovide . . . , the number of hours of work by lawyers andparalegals required, [or] the expense." Heraeus Kulzer,633 F.3d at 598. Here, TIN's [*49] and IP's conclusorystatements do not provide evidence in support of theirburdensome arguments.

In sum, in this situation, the Rule 26(b)(2)(C) factorsdo not overwhelmingly favor either Plaintiffs orDefendants. However, because Plaintiffs had no input onthe initial custodian determinations and the case is still inthe early stages of discovery, the Court finds thatPlaintiffs should be allowed a small number of additionalcustodians. Accordingly, Plaintiffs may select eightadditional custodians from the prioritized list it sent to IP.(IP Mot. Ex. 9.) Similarly, Plaintiffs may select eightadditional custodians from the proposed list it sent toTIN, two of whom should be a random sample of theindividuals identified as mill managers. (See TIN Mot.Ex. A.)

TIN and IP argue that they should not have to searchall sources of information for any new custodians. (TINResp. 9-11; IP Resp. 10-11.) They generally propose tosearch only ESI from particular servers. (Id.) Theycontend that "the burden and expense of searching [other]sources . . . would certainly outweigh its likely benefit."(IP Resp. 11; see TIN Resp. 11.)

The Court disagrees. While Plaintiffs have focusedtheir review on [*50] email and other ESI documents,they have not asserted that non-ESI documents areunimportant. Given the dearth of emails produced in theearly time periods, hard copy documents for thoseperiods may prove valuable. Similarly, for formeremployees, hard copy documents may be the onlyinformation available.

TIN and IP also seek a written assurance that anyfuture custodian request would be subject to a good faithbelief that the individual possesses information tending toprove the alleged conspiracy. (See IP Resp. 12; TINResp. 4.) Specifically, IP requests assurance that "[a]nyfuture request for custodians would be based on a goodfaith belief, arising from a review of the documentsproduced in this case and taking depositions, than an

individual possesses information tending the prove theallege conspiracy, and Plaintiffs agree to provide thebasis for their good faith belief to IP." (IP Resp. 12)(emphasis omitted). Similarly, TIN seeks confirmationthat "if Plaintiffs seek to add any additional custodians, itwould have to be based on specific evidence from therecord indicating that the proposed custodian would haveresponsive evidence tending to prove the allegedconspiracy." [*51] (TIN Resp. 4.) Plaintiffs counter thatthe agreement they reached with PCA provides sufficientprotection for all parties and is consistent with Rule 26.(IP Reply 4-5.)

The Court declines to set any restrictions on futurecustodian requests. The Court finds that IP's and TIN'srequests are too restrictive and run contrary to the FederalRules. For example, discoverable information "need notbe admissible at the trial if the discovery appearsreasonably calculated to lead to the discovery ofadmissible information." Fed. R. Civ. P. 26(b)(1).Nevertheless, the parties are free to enter into anystipulations or agreements that fit their needs. Absent anyagreements, however, the Court will be guided in anyfuture custodian disputes by the Federal Rules andapplicable case law. 12 See, e.g., Fed. R. Civ. P. 1,26(b)-(c).

12 Plaintiffs are cautioned, however, that theCourt expects them to support any futurecustodian requests with specific evidence of theexpected utility of the additional custodiansderived from their review of existing documents.See Fed. R. Civ. P. 26(b)(2)(C)(iii).

Plaintiffs' motions to compel IP and TIN to includeadditional custodians are granted in part.

C. [*52] Data Sources

In Plaintiffs' Motion to Compel Defendants toProduce Documents and Data from All ReasonablyAccessible Sources, they request that Defendants becompelled to search all reasonably accessible sources thatpotentially contain nonduplicative responsive documentsor data, including backup tapes. (Mot. 10.) Defendantsrespond that their backup tapes are not reasonablyaccessible and Plaintiffs have not shown good cause forrequiring their production. (Resp. 16-27.)

The resolution of Plaintiffs' motion is dependent onthe application of Rule 26(b)(2)(B). This Rule provides:


Specific Limitations on ElectronicallyStored Information. A party need notprovide discovery of electronically storedinformation from sources that the partyidentifies as not reasonably accessiblebecause of undue burden or cost. Onmotion to compel discovery or for aprotective order, the party from whomdiscovery is sought must show that theinformation is not reasonably accessiblebecause of undue burden or cost. If thatshowing is made, the court maynonetheless order discovery from suchsources if the requesting party shows goodcause, considering the limitations of Rule26(b)(2)(C). The court may specify [*53]conditions for the discovery.

"Pursuant to this Rule, defendants must produceelectronically stored information that is relevant, notprivileged, and reasonably accessible, subject to thediscovery limitations in Rule 26(b)(2)(C)." Major Tours,Inc. v. Colorel, No. 05-3091, 2009 U.S. Dist. LEXIS97554, 2009 WL 3446761, at *2 (D.N.J. Oct. 20, 2009).If Defendants establish that the requested backup tapesare "inaccessible" within the meaning of Rule26(b)(2)(B), the information must still be produced ifPlaintiffs establish good cause considering the limitationsin Rule 26(b)(2)(C). "The decision whether to require aresponding party to search for and produce informationthat is not reasonably accessible depends not only on theburdens and costs of doing so, but also on whether thoseburdens and costs can be justified in the circumstances ofthe case." Fed. R. Civ. P. 26(b)(2), advisory committee'snote (2006). Factors to examine in this analysis include:

(1) the specificity of the discoveryrequest; (2) the quantity of informationavailable from other and more easilyaccessed sources; (3) the failure toproduce relevant information that seemslikely to have existed but is no longeravailable on more easily accessed [*54]sources; (4) the likelihood of findingrelevant, responsive information thatcannot be obtained from other, more easilyaccessed sources; (5) predictions as to theimportance and usefulness of furtherinformation; (6) the importance of theissues at stake in the litigation; and (7) the

parties' resources.

Id.

"Under this framework, a court does not reach thetwo-fold question of whether inaccessible sources ofelectronically stored information should be searched and,if so, which party should bear the associated costs unlessit is first satisfied that the request seeks relevantinformation that is not available from accessible sources."Baker v. Gerould, 03-CV-6558L, 2008 U.S. Dist. LEXIS28628, 2008 WL 850236, at *2 (W.D.N.Y. Mar. 27,2008) This is because relevant considerations indetermining whether to order a search of inaccessiblesources include "the quantity of information availablefrom other and more easily accessed sources" and "thelikelihood of finding relevant information that seemslikely to have existed but is no longer available on moreeasily accessed sources." Fed. R. Civ. P. 26(b)(2),advisory committee's note (2006); see Zubulake v. UBSWarburg LLC, 217 F.R.D. 309, 323 (S.D.N.Y. 2003)(Scheindlin, [*55] J.) (one of the two most importantconsiderations is "the availability of such informationfrom other sources").

Courts generally agree that backup tapes arepresumptively inaccessible. See, e.g., Zubulake, 217F.R.D. at 319-20 (" 'Inaccessible' data . . . is not readilyusable. Backup tapes must be restored using a processsimilar to that previously described, fragmented datamust be de-fragmented, and erased data must bereconstructed, all before the data is usable. That makessuch data inaccessible."); Major Tours, 2009 U.S. Dist.LEXIS 97554, 2009 WL 3446761, at *3 (Backup tapesare "typically classified as inaccessible."); Go v.Rockefeller Univ., 280 F.R.D. 165, 175-76 (S.D.N.Y.2012) ("Information stored on backup tapes is generallyconsidered 'not reasonably accessible.' ") (citationomitted); Clean Harbors Envtl. Servs., Inc. v. ESIS, Inc.,No. 09 CV 3789, 2011 U.S. Dist. LEXIS 53212, 2011WL 1897213, at *2 (N.D. Ill. May 17, 2011) (Courts[*56] have already agreed that when information is storedon backup tapes, it is 'likened to paper records lockedinside a sophisticated safe to which no one has the key orcombination.' ESIS has given us no reason to believe thatthe information on the backup tapes in this case would bemore easily accessible.") (quoting Zubulake v. UBSWarburg LLC, 216 F.R.D. 280, 291 (S.D.N.Y. 2003)).

In addition, Defendants have demonstrated a cost


burden to restoring the backup media. They providedaffidavits indicating that to restore the backup tapeswould cost each Defendant at least $200,000, with someestimates well over $1,000,000. (See Resp. Exs. D, H.)Plaintiffs dispute the cost to restore Defendants' backupmedia. They contend that sampling the media todetermine whether they contain responsivenonduplicative information could reduce costs. (Reply6-7; Han-ners Decl. ¶ 5.)

In any event, the Court finds that Plaintiffs' requestto produce the backup tapes is premature. There is nodiscovery cutoff date in this case, and Plaintiffs are only20% complete with their first level review of Defendants'documents. Thus, Plaintiffs should complete their reviewof Defendants' ESI, including the information [*57]produced from the additional custodians, before seekingto have archived backup tapes restored. "The volumeof--and the ability to search--much electronically storedinformation means that in many cases the respondingparty will be able to produce information from reasonablyaccessible sources that will fully satisfy the parties'discovery needs." Fed. R. Civ. P. 26(b)(2), advisorycommittee's note (2006). Accordingly, Plaintiffs "shouldobtain and evaluate the information from such sourcesbefore insisting that the responding party search andproduce information contained on sources that are notreasonably accessible." Id.

If, at the appropriate time, Plaintiffs decide to pursuethe backup tapes, the parties and their experts are urgedto work together in a cooperative manner to determinethe actual cost of restoring the archived media. If feasibleand cost efficient, sampling methods should be pursued.However, because each Defendant's ESI storageprotocoal is unique, it may be difficult or impossible toextrapolate any sampling results from one Defendant tothe others.

Plaintiffs' Motion to Compel Defendants to ProduceDocuments and Data from All Reasonably AccessibleSources is denied [*58] without prejudice.

IV. CONCLUSION

Since its publication in 2009, more than 100 federaljudges have endorsed the Cooperation Proclamation. Inan effort to aid courts and counsel, The SedonaConference has published guides and toolkits to facilitateproportionality and cooperation in discovery. 13

Moreover, a number of courts have instituted model

orders to assist counsel in transitioning to the cooperativediscovery approach. 14

13 See The Sedona Conference, The SedonaConference Cooperation Proclamation: Resourcesfor the Judiciary (2011), available athttp://www.thesedonaconference.org; TheSe-dona Conference, The Sedona ConferenceCooperation Guidance for Litigators & In-HouseCounsel (2011), available athttp://www.thesedonaconference.org.14 See, e.g., Seventh Circuit ElectronicDiscovery Pilot Program, Model Standing Order,available at http://www.discoverypilot.com;Southern District of New York Pilot Program,available at http://www.nysd.uscourts.gov;District of Delaware, Default Standard forDiscovery, Including Discovery of ElectronicallyStored Information ("ESI"), available athttp://www.ded.uscourts.gov; see also David J.Waxse, Cooperation--What Is It and Why Do It?,XVIII Rich. [*59] J. L. & Tech. 8 (2012),available at jolt.richmond.edu/v18i3/article8.pdf.

In pursuing a collaborative approach, some lessonshave been learned. First, the approach should be startedearly in the case. It is difficult or impossible to unwindprocedures that have already been implemented. Second,in multiple party cases represented by separate counsel, itmay be beneficial for liaisons to be assigned to eachparty. Finally, to the extent possible, discovery phasesshould be discussed and agreed to at the onset ofdiscovery.

The Cooperation Proclamation calls for a "paradigmshift" in how parties engage in the discovery process. TheSedona Conference, The Sedona Conference CooperationProclamation, 10 Sedona Conf. J. 331, 332-33 (2009). Insome small way, it is hoped that this Opinion can be ofsome help to others interested in pursuing a cooperativeapproach. 15 The Court commends the lawyers and theirclients for conducting their discovery obligations in acollaborative manner.

15 The Court notes that there are very few modelagreements available for parties and courts tofollow. Accordingly, the Court suggests that TheSedona Conference and the Seventh CircuitElectronic Discovery Pilot [*60] Program serveas repositories for gathering ESI discoveryagreements.


For the foregoing reasons, Plaintiffs' Motion toCompel Defendants to Produce Documents and Datafrom All Reasonable Accessible Sources [347] isDENIED WITHOUT PREJUDICE; Plaintiffs' Motionto Compel Temple-Inland to Include AdditionalDocument Custodians [366] is GRANTED IN PARTAND DENIED IN PART; Plaintiffs' Motion to CompelInternational Paper Company to Include AdditionalDocument Custodians [382] is GRANTED IN PARTAND DENIED IN PART; and DefendantGeorgia-Pacific LLC's Motion for Protective Order [373]is GRANTED.

Dated: September 28, 2012

ENTER:

/s/ Nan R. Nolan

Nan R. Nolan


EXHIBIT A

UNITED STATES DISTRICT COURT NORTHERNDISTRICT OF ILLINOIS EASTERN DIVISION

KLEEN PRODUCTS LLC, et al., individually andon behalf of all those similarly situated, v. Plaintiff,PACKAGING CORPORATION OF AMERICA, et al,Defendants.

Civil Case No. 1:10-cv-05711

Hon. Milton I. Shadur

Hon. Nan R. Nolan

STIPULATION AND ORDER RELATING TO ESISEARCH

WHEREAS, in response to plaintiffs' May 3, 2011Request for Production of Documents (the "FirstRequest") in this matter, defendants have collectedelectronic and hard [*61] copy documents;

WHEREAS, defendants will continue to collectcertain documents in response to the First Request,including without limitation such documents that may becollected for review in response to discovery conferencesor pursuant to judicial orders arising out of motionsbrought on the First Request (e.g., any documents that the

Court orders included for review and production based onthe motions to be filed in July and August, 2012)(collectively all of the documents that have been or willbe collected in response to the First Request shall bereferred to in this Stipulation as the "First RequestCorpus");

WHEREAS, defendants have employed ESI vendorsto process the electronic documents contained within theFirst Request Corpus, and those ESI vendors have doneso and continue to do so;

WHEREAS, defendants have applied and continue toapply their ESI search methodology (hereafter"Defendants' ESI Methodology"), which was describedduring the evidentiary hearings conducted on February21, 2012 and on March 28, 2012 (the "EvidentiaryHearings"), to those processed electronic documentswithin the First Request Corpus;

WHEREAS, plaintiffs have challenged Defendants'ESI Methodology [*62] for the identification ofdocuments responsive to the First Requests and asked theCourt to order defendants to employ a "Content BasedAdvanced Analytics" ("CBAA") approach, as defined byplaintiffs, instead of Defendants' ESI Methodology;

WHEREAS, defendants have opposed thischallenge;

WHEREAS, the parties have made a substantialnumber of written submissions and oral presentations tothe Court with their views on this issue, and the Courtheld the Evidentiary Hearings to address this dispute; and

WHEREAS, the parties continue to have a number ofdisputes, but in order to narrow the issues, the partieshave reached an agreement that will obviate the need foradditional evidentiary hearings on the issue of thetechnology to be used to search for documents responsiveto the First Requests.

THEREFORE, based upon and incorporating theforegoing, the parties, through their respective counsel ofrecord, hereby stipulate to and the Court hereby orders:

1. Plaintiffs withdraw their demand that defendantsapply CBAA to documents contained within the FirstRequest Corpus. Plaintiffs will not claim that defendantsmust use an electronic search process other thanDefendants' ESI Methodology to locate [*63] relevant


documents contained in the First Request Corpus.

2. As to any documents or ESI beyond the FirstRequest Corpus, plaintiffs will not argue or contend thatdefendants should be required to use or apply the types ofCBAA or "predictive coding" methodology andtechnology that were proposed by plaintiffs in connectionwith the Evidentiary Hearings with respect to anyrequests for production served on any defendant prior toOctober 1, 2013. With respect to any requests forproduction served on any defendant on or after October 1,2013, that requires the collection of documents beyondthe First Request Corpus, the parties will meet and conferregarding the appropriate search methodology to be usedfor such newly collected documents. If the parties fail toagree on a search methodology, either party may file amotion with the Court seeking resolution.

3. Plaintiffs do not waive any additional objectionsthey may have to defendants' search methodology for theFirst Requests, including any additional objectionsrelating to defendants' identification, collection,custodians, data sources, search terms, statisticalvalidation, review or production of documents, and thatdefendants' objections [*64] to the First Request undulynarrowed the scope of responsive documents, anddefendants will not argue or contend that plaintiffs, inwhole or in part, have waived or otherwise failed to fullyreserve such additional objections by entering into thisStipulation. The Court has established briefing schedulesand other processes to resolve some of these issues by theend of September 2012.

4. Defendants reserve all rights they currently havewith respect to their position that their documentcollection and production efforts met or exceededrelevant legal standards.

5. In light of this agreement by the parties, theEvidentiary Hearings are discontinued.

Stipulated and agreed:

By:Daniel J. MoginMatthew T. SinnottTHE MOGIN LAW FIRM, P.C.707 Broadway, Suite 1000San Diego, CA 92101(619) [email protected]

[email protected]

INTERIM CO-LEAD COUNSEL FOR THEPROPOSED CLASS

By:

Michael J. Freed

Steven A. Kanner

FREED KANNER LONDON & MILLEN LLC

2201 Waukegan Road, Suite 130

Bannockburn, IL 60015

(224) 632-4500

[email protected]

[email protected]


By:Stephen R. NeuwirthDeborah BrownSami H. Rashid

QUINN EMANUEL URQUHART & SULLIVAN,LLP

51 [*65] Madison Avenue, 22nd FloorNew York, NY 10010(212) [email protected]@[email protected]

James R. FigliuloStephanie D. JonesFIGLIULO & SILVERMAN, P.C.10 South LaSalle Street, Suite 3600Chicago, IL 60603(312) [email protected]@fslegal.com

COUNSEL FOR DEFENDANTGEORGIA-PACIFIC LLC

By:Douglas J. Kurtenbach, P.C.


Daniel E. LaytinBarack S. EcholsLeonid FellerKIRKLAND & ELLIS LLP300 North LaSalleChicago, IL 60654(312) [email protected]@[email protected]@kirkland.com

COUNSEL FOR DEFENDANT PACKAGINGCORPORATION OF AMERICA

By:

Nathan P. Eimer

EIMER STAHL LLP

224 South Michigan Avenue, Suite 1100

Chicago, IL 60604-2516

(312) 660-7600

[email protected]

James T. McKeown

FOLEY & LARDNER LLP

777 East Wisconsin Avenue

Milwaukee, WI 53202-5306

(414) 297-5530

[email protected]

COUNSEL FOR DEFENDANT INTERNATIONALPAPER COMPANY

By:

R. Mark McCareins

Michael P. Mayer

James F. Herbison

WINSTON & STRAWN LLP

35 West Wacker Drive

Chicago, IL 60601

(312) 558-5600

[email protected]

>[email protected]

>[email protected]

COUNSEL FOR DEFENDANT ROCKTENN CP,LLC

By:Andrew [*66] S. MarovitzBritt M. MillerMAYER BROWN LLP71 S. Wacker DriveChicago, IL 60606(312) [email protected]@mayerbrown.com

COUNSEL FOR DEFENDANT TEMPLE-INLANDINC.

By:David Marx Jr.Jennifer S. DiverRachel LewisMcDERMOTT WILL & EMERY LLP227 W. Monroe StreetChicago, IL 60606(312) [email protected]@[email protected]

COUNSEL FOR DEFENDANTWEYERHAEUSER COMPANY

By:

Scott M. Mendel

John E. Susoreny

Lauren N. Norris

K&L GATES LLP


70 W. MADISON ST.

SUITE 3100

CHICAGO, IL 60602

(312) 372-1121

[email protected]

>[email protected]

>[email protected]

COUNSEL FOR DEFENDANTS CASCADES,INC. AND NORAMPAC HOLDING U.S. INC.

Dated: August 21, 2012

ENTER:

/s/ Nan R. Nolan

NAN R. NOLAN


EXHIBIT B

Name of Assigned Judge or Magistrate Judge

Nan R. Nolan

Sitting Judge if Other than Assigned Judge

CASE NUMBER

10 C 5711

DATE

6/25/2012

CASE TITLE

Kleen Products, et al. vs. Packaging Corporation ofAmerica, et al.

DOCKET ENTRY TEXT

Rule 16 conference between Plaintiffs and DefendantGeorgia-Pacific held on 5/31/2012. As stated below andon the record, various agreements were reached.

STATEMENT

At the 5/31/2012 Rule 16 conference, Plaintiffs andDefendant [*67] Georgia-Pacific each agreed to producethe names and titles of all persons who have received alitigation hold notice related to this action, along with thedate that the notice was made. If the exact date of thenotice cannot be readily determined, an approximaterange of dates will be provided. The production of thenames of persons receiving the litigation hold notice shallnot constitute a waiver of the work-product doctrine orany other privilege. In addition, the inclusion of a personon the list of those receiving the litigation hold noticeshall not create any presumption, or change theapplicable standards for determining, that the person is anappropriate document custodian for purposes of ESI anddocument searches, or is otherwise subject to discovery.

Defendant Georgia-Pacific also agreed to includeJames Hannan, Chief Executive Officer and President, asa document custodian. Defendant Georgia-Pacific'swillingness to compromise, in the context of a meet andconfer supervised by the Court, on the foregoing issues oflitigation hold recipients and making Hannan a documentcustodian shall not be a factor in determining whetherHannan shall be subject to any further discovery [*68] inthis litigation.

EXHIBIT C

UNITED STATES DISTRICT COURT NORTHERNDISTRICT OF ILLINOIS EASTERN DIVISION

KLEEN PRODUCTS LLC, et al. individually and onbehalf of all those similarly situated, Plaintiff, v.PACKAGING CORPORATION OF AMERICA, et al,Defendants.

Civil Case No. 1:10-cv-05711

Hon. Milton I. Shadur

Hon. Nan R. Nolan

STIPULATION AND ORDER RELATING TOCUSTODIANS

In order to fully and finally resolve the parties'custodian disputes as outlined in their submissions to theCourt, Plaintiffs and the undersigned defendants("Defendants") agree as follows:

1. Defendants have agreed, in connection with thisStipulation and Order and their respective agreements


with Plaintiffs, to add certain additional custodians to thedocument review. Plaintiffs have no present intention ofseeking additional custodians from any of theDefendants.

2. Plaintiffs shall not make a future request forcustodians from Defendants unless Plaintiffs' counselhave a good faith belief, arising from a review of thedocuments produced in this case, from depositions, orfrom some other identifiable source, that each requestedindividual custodian has information, not unreasonablycumulative or duplicative, that is relevant [*69] to theconduct alleged in Plaintiffs' Complaint.

3. Plaintiffs will provide the Defendant from whomthey are seeking any additional custodians with a writtensummary explanation of their basis for requesting eachadditional custodian, including a specific articulation ofthe basis for the request, such as Plaintiffs' review ofdocuments and deposition testimony. In providing thisinformation, Plaintiffs do not waive any work productprotection that might apply.

4. If, after receiving the written explanationdescribed in paragraph 3, an agreement cannot be reachedamong the relevant parties, then either party may take theissue up with the Court.

5. Defendants reserve the right to oppose the additionof future custodians.

6. Nothing in this agreement is intended to modifythe applicability of the standard set forth in Fed. R. Civ.P. 26(b)(2)(C) in the event that a future dispute ariseswith respect to custodians.

STIPULATED AND AGREED:

By: Daniel J. MoginDaniel J. MoginMatthew T. SinnottTHE MOGIN LAW FIRM, P.C.707 Broadway, Suite 1000San Diego, CA 92101(619) [email protected]>[email protected]


By: Michael J. Freed

Michael J. Freed

Steven [*70] A. Kanner

FREED KANNER LONDON & MILLEN LLC

2201 Waukegan Road, Suite 130

Bannockburn, IL 60015

(224) 632-4500

[email protected]

[email protected]


By: R. Mark McCareinsR. Mark McCareinsMichael P. MayerJames F. HerbisonWINSTON & STRAWN LLP35 West Wacker DriveChicago, IL 60601(312) [email protected]>[email protected]>[email protected]

COUNSEL FOR DEFENDANT ROCKTENN CP,LLC

By: Jennifer S. DiverDavid Marx Jr.Jennifer S. DiverRachel LewisMcDERMOTT WILL & EMERY LLP227 W. Monroe StreetChicago, IL 60606(312) [email protected]>[email protected]>[email protected]

COUNSEL FOR DEFENDANTWEYERHAEUSER COMPANY

By: Scott M. Mendel

Scott M. Mendel


John E. Susoreny

Lauren N. Norris

K&L GATES LLP

70 W. MADISON ST.

SUITE 3100

CHICAGO, IL 60602

(312) 372-1121

[email protected]

[email protected]

[email protected]

COUNSEL FOR DEFENDANTS CASCADES,INC. AND NORAMPAC HOLDING U.S. INC.

Dated: September 17, 2012

ENTER:

/s/ Nan R. Nolan

NAN R. NOLAN



IN THE COURT OF CHANCERY OF THE STATE OF DELAWARE

EORHB, INC., a Georgia corporation, : and COBY G. BROOKS, EDWARD J. GREENE, : JAMES P. CREEL, CARTER B. WRENN : and GLENN G. BROOKS, each as personal : representatives and trustees of the : estate of Robert H. Brooks, : : Plaintiffs, : : vs. : Civil Action : No. 7409-VCL HOA HOLDINGS LLC, a Delaware limited : liability company, and HOA RESTAURANT : GROUP, LLC, a Delaware limited : liability company, : :

Defendants and : Counterclaim Plaintiffs,:

: vs. :

: EORHB, INC., a Georgia corporation, : and COBY G. BROOKS, EDWARD J. GREENE, : JAMES P. CREEL, CARTER B. WRENN : and GLENN G. BROOKS, each as personal : representatives and trustees of the : estate of Robert H. Brooks, :

: Counterclaim Defendants.:

--------------------------------------: Chancery Courtroom 12C

New Castle County Courthouse Wilmington, Delaware Monday, October 15, 2012 2:00 p.m.

- - -

BEFORE: HON. J. TRAVIS LASTER, VICE CHANCELLOR

- - - MOTION FOR PARTIAL SUMMARY JUDGMENT

MOTION TO DISMISS COUNTERCLAIM AND RULING OF THE COURT

CHANCERY COURT REPORTERS

EFiled: Oct 19 2012 11:36AM EDT Transaction ID 47152282 Case No. 7409VCL

2

1

2 APPEARANCES:

3

4 RICHARD P. ROLLO, ESQ. JOHN MARK ZEBERKIEWICZ, ESQ.

5 Richards, Layton & Finger, P.A. for Plaintiffs/Counterclaim Defendants

6

7 A. THOMPSON BAYLISS, ESQ. Abrams & Bayliss LLP

8 -and- DOUGLAS H. HALLWARD-DRIEMEIER, ESQ.

9 CHRISTOPHER G. GREEN, ESQ. AMY D. ROY, ESQ.

10 of the Massachusetts bar Ropes & Gray LLP

11 for Defendants/Counterclaim Plaintiffs

12

13

14 - - -

15

16

17

18

19

20

21

22

23

24


3

1 THE COURT: Welcome, everyone.

2 Mr. Bayliss, how are you?

3 MR. BAYLISS: Good afternoon, Your

4 Honor. Tom Bayliss on behalf of HOA Restaurant Group.

5 I rise to introduce Mr. Douglas Driemeier, Chris Green

6 and Amy Roy from Ropes & Gray.

7 THE COURT: Great. Welcome to all of

8 you. Mr. Rollo, how are you doing?

9 MR. ROLLO: I'm doing well, Your

10 Honor. Yourself?

11 THE COURT: Great.

12 MR. ROLLO: For the record, this is

13 Rich Rollo of Richards, Layton & Finger, and I

14 represent the plaintiffs in this action. With me at

15 counsel table is my colleague, John Mark Zeberkiewicz.

16 THE COURT: Mr. Zeberkiewicz, how are

17 you. Mr. Zeberkiewicz has not often darkened the

18 doors of the courtroom.

19 MR. ZEBERKIEWICZ: Absolutely correct.

20 THE COURT: It's good to see you, a

21 transactional lawyer learning how to make his way down

22 to 500 King Street. You're always welcome here,

23 Mr. Zeberkiewicz.

24 MR. ZEBERKIEWICZ: Thank you, Your


4

1 Honor.

2 MR. ROLLO: Your Honor, we're before

3 the Court today on our motion for partial summary

4 judgment and also a motion to dismiss. I know Your

5 Honor is familiar with the papers, so I'll briefly

6 summarize our position and then address any questions

7 Your Honor may have.

8 We believe this case is relatively

9 simple. Sophisticated parties negotiated a settlement

10 that included the exchange of a specific release.

11 Neither side claims they were tricked or defrauded or

12 some mistake occurred that resulted in the release.

13 Rather, we simply disagree over what

14 the words on the page mean. The release was executed

15 in May 2011 which was about four months after the

16 parties engaged in a merger transaction pursuant to

17 which the buyers purchased the Hooters restaurant

18 chain from the sellers.

19 A release was exchanged as part of a

20 global settlement of a prior litigation before Your

21 Honor involving, among other things, who had the right

22 to purchase the restaurant chain.

23 At the time in May of 2011, the

24 transaction had closed, but the parties had continuing


5

1 obligations under the merger agreement which we wanted

2 to preserve. So we included a carveout. It's the

3 scope and operation of that carveout that is the

4 dispute before the Court today.

5 Now, the release is attached as

6 Exhibit G to our opening brief. On the third page is

7 the language that's disputed, the beginning of it. It

8 starts, and it's a single sentence that begins with,

9 I'll say, your typical laundry list identifying the

10 types of claims and rights released, including the

11 conflicting adjectives to include everything under the

12 sun and make clear that there are no limitations.

13 Now, because this is a specific

14 release, toward the bottom of the page, about five

15 lines from the bottom, there is a limitation on that

16 laundry list, and I'm paraphrasing, arising from or

17 related to indirectly or directly any of nine

18 enumerated categories. In clause five, it says the

19 sale of Hooters. Now, I'm paraphrasing that as well,

20 but it's the sale of Hooters pursuant to the merger

21 transaction.

22 Those rights and any obligations under

23 the merger agreement would have been eliminated

24 without a carveout. So following the list of nine


6

1 categories, there are two provisos separated by

2 semicolons. The first, about halfway down the page on

3 the third page, begins "provided however," and it

4 creates a specific carveout for enforcement of the

5 settlement agreement. It doesn't purport to modify

6 the definition of released claims. It says "provided

7 however, nothing in this release."

8 Now, the second proviso begins four

9 lines later and starts after the second semicolon with

10 "provided further, however," and it creates a carveout

11 for the merger and related transactions. That's

12 followed by a carveback where it says "except that"

13 and the carveback says released claims cannot be the

14 basis for a breach of the merger agreement and won't

15 result in a purchase price adjustment pursuant to the

16 operative contract.

17 We think that the only reasonable

18 read -- in fact, the only read of this language -- is

19 that the parties agreed that they could, on a

20 forward-looking basis, enforce the merger agreement

21 and that any breach of contract claims that existed

22 from that date back in time were released.

23 Now, the indemnifications we're here

24 today about are predicated upon purported breaches of


7

1 the merger agreement that occurred before the release

2 date; in fact, in or around January 2011 when the reps

3 and warranties were supposed to have been true, and

4 there's also one claim with respect to a pre-closing

5 operation of the company, the pre-closing operations

6 covenant. All of those were released, we say, under

7 the operative agreement.

8 My friends make several arguments, and

9 I'll respond to most of them on rebuttal, but there is

10 one I'd like to address briefly. That's the argument

11 that some temporal ambiguity exists in the document

12 that precludes summary judgment. Briefly, and they'll

13 state it for themselves, in the beginning part of the

14 contract, there is a phrase that says "which now

15 exists or heretofore after existed or may hereafter

16 exist." I think I mangled that. I could say it

17 again, but Your Honor understands the concept. Seven

18 lines later, there's another phrase that says "in

19 existence from the beginning of time to the date of

20 this agreement."

21 Now, that alleged conflict -- and we,

22 in our reply brief, say we don't believe it's a

23 conflict, but let's assume that is a conflict. It

24 doesn't matter in this case. We're not arguing about


8

1 breaches that supposedly occurred after execution of

2 the release. We're arguing about breaches that

3 existed on the date of the release under either

4 definition. Under either temporal phrase, they fall

5 within the definition of released claims. Because

6 they fall within the definition of released claims,

7 they were barred under the second part of the second

8 proviso.

9 Now, I'm happy to address any of the

10 arguments or questions Your Honor may have. I think

11 it may be more efficient if my friends, since they

12 have several arguments, would state the arguments and

13 I can reply on rebuttal.

14 THE COURT: Let me ask you one thing

15 before you sit down.

16 MR. ROLLO: Yes, Your Honor.

17 THE COURT: How do you pronounce the

18 entity that is now successor to the estate, EORHB?

19 MR. ROLLO: EORHB is how I say it.

20 It's the Estate of Robert H. Brooks.

21 THE COURT: The estate, as the

22 predecessor to EORHB, signed a comparable release that

23 is one of the ones that you collected, that is

24 collected behind Exhibit E; true?


9


2 THE COURT: EORHB's claim to

3 additional monies from the indemnification agreement

4 arose at closing, right?

5 MR. ROLLO: I don't believe I agree

6 with that, Your Honor. A certain amount of the

7 consideration payable to us was set aside in an escrow

8 account. And if valid claims were not presented under

9 that, we received it when the escrow expired. That is

10 a contractual obligation under the merger agreement

11 and separately under the escrow agreement. And under

12 the releases, there is a carveout for future

13 enforcement of the merger agreement.

14 THE COURT: You think that claim for

15 the access of the indemnification agreement falls

16 under the enforcement of the merger agreement proviso?


18 THE COURT: You don't think that a

19 right against the escrow agreement is a right that

20 would fall under the definition of released claims?

21 MR. ROLLO: I don't, Your Honor, for

22 two reasons. First, it is a contractual obligation

23 under the escrow, and that if valid claims are not

24 made against the escrow, the escrow agent is directed


10

1 to tender the rest of that money to us last summer.

2 So if, at that point in time, $11 million remained,

3 it's either given to us or it stays there presumably

4 in perpetuity because the buyers wouldn't have a claim

5 against it.

6 THE COURT: I hear you. Released

7 claims means any and all claims. It means rights,

8 blah, blah, blah, of any kind whatsoever, whether

9 known or unknown. Well, this one was known. Fixed or

10 contingent. Well, this is a little bit of both. As

11 you say, it's a fixed contractual right of which the

12 amount the contingent.

13 The definition of released claims, if

14 one were to read it broadly to include everything

15 related to the sale of HOA, it seems to me is

16 sufficiently capacious to cover, in the first

17 instance, EORHB's claim against the escrow fund.

18 MR. ROLLO: Perhaps I misunderstood

19 Your Honor's earlier question. I believe the initial

20 definition does. I think then it is saved by the

21 second proviso that says "provided further, however,

22 the foregoing shall not include any claims to enforce

23 the terms and conditions of the merger agreement or

24 directly related to the transaction contemplated


11

1 thereby."

2 THE COURT: Why doesn't it then

3 founder again on the idea that receipt of that money

4 would result in an adjustment to the merger

5 consideration?

6 MR. ROLLO: Because receipt of that

7 money, by definition on the bottom of the second

8 proviso, does not result in a modification of the

9 merger agreement. There is a --

10 THE COURT: It doesn't fall under the

11 modification of the merger agreement, but --

12 MR. ROLLO: Purchase price adjustment.

13 We expressly excluded that in the bottom of the second

14 proviso. No release claim and no liability or payment

15 under the accompanying settlement agreement --

16 THE COURT: The problem with that is

17 Section 9.7 which says, "All payments made pursuant to

18 this Article 9 shall be treated as adjustments to the

19 merger consideration for tax purposes." I agree

20 that's for tax purposes, but it would seem to say that

21 the release of funds from the escrow would be an

22 adjustment to the merger consideration.

23 MR. ROLLO: It would be an adjustment

24 to the merger consideration and --


12

1 THE COURT: That would take it into

2 the second "except" clause which would mean you'd be

3 back to the idea that really what I ought to be doing

4 is granting summary judgment for the other side saying

5 that they released their claims to everything in the

6 escrow fund.

7 MR. ROLLO: I disagree, because I

8 think the second portion, Your Honor, is predicated on

9 breaches of the merger agreement. The first is

10 "obligations under," and the second is "breaches

11 under." Future enforcement of contractual provisions

12 is not a breach of the merger. If tomorrow we

13 breached, it would fall within that provision, so I

14 think there's a distinction in the first part between

15 "obligations under" and "breaches under."

16 THE COURT: Walk me through that

17 again.

18 MR. ROLLO: The indemnification

19 obligation that my friends are seeking to pursue is

20 predicated upon purported breaches of the merger

21 agreement.

22 THE COURT: That's a problem for them

23 under Romanette "i".

24 MR. ROLLO: Correct. But, for


13

1 example, I'll use an example --

2 THE COURT: I'm focused on Romanette

3 "ii". I think your problem is the breadth of your

4 argument as to released claims is so powerful that it

5 seems to me that it runs into exception two. I agree

6 with you your issue isn't that it's a breach under

7 exception one.

8 What seems to me to be your problem is

9 that you agree contractually that any additional money

10 you got was going to be an adjustment to the merger

11 consideration, and you're now telling me that you

12 released anything that fit within the definition of

13 released claims that could result in an adjustment to

14 the merger consideration, and so by the force of your

15 powerful interpretation of released claims, hasn't

16 your fellow given up his adjustment to the merger

17 consideration?

18 MR. ROLLO: I don't believe so,

19 because as I read the second Romanette, it says "no

20 released claims shall result in adjustment to the

21 merger consideration."

22 Now, the merger consideration is a

23 defined term under the merger agreement. The escrow

24 didn't result in a lower amount. It was simply an


14

1 amount of money that was set aside and the merger

2 consideration itself remained constant. That payment

3 was just delayed. Had it resulted in a reduction,

4 which I believe Your Honor is positing, that escrow

5 amount would then be deducted from the overall merger

6 consideration. That would then be barred by Romanette

7 "ii". I believe Romanette "ii" operates to prevent

8 the exact issue that Your Honor is raising.

9 There's a provision, I believe, at

10 Section 9.3 that talks about how purchase price

11 adjustments are made under the merger agreement. It's

12 Section 2.3 of the merger agreement. Basically, it

13 provides a process post-closing where either side can

14 identify certain issues that would result in a change

15 in the merger consideration.

16 Those adjustments aren't based upon a

17 claim that we breached the merger agreement or that

18 the other side breached the merger agreement. It was

19 simply the accounting process in place. Romanette

20 "ii" in the release is focused on that process.

21 In order for I believe what Your Honor

22 is positing that the escrow functionally is waived,

23 that would presuppose that by putting the escrow funds

24 in escrow, it somehow impacts the merger


15

1 consideration, and I would submit to Your Honor it

2 does not. The consideration is the consideration.

3 11.5 of that consideration was simply held for a

4 period of time subject to the rights under 9.3 for

5 indemnification.

6 THE COURT: It does create an

7 interesting interpretive question because 2.1(b) says

8 that your company membership interests that EORHB

9 owned were converted into the right to receive

10 Romanette "i", net merger consideration plus various

11 other amounts which include amounts released from the

12 escrow agreement, and it defines those latter amounts

13 as contingent merger payments.

14 So, again, I look at that, and I

15 think, okay, well, are contingent merger payments part

16 of the merger consideration such that they represent

17 an adjustment? They seem to be something other than

18 net merger consideration. Then I get back here to

19 9.7, treatment of indemnity payments, and it says,

20 "All payments pursuant to this Article 9 shall be

21 treated as adjustments to the merger consideration for

22 tax purposes."

23 I guess what you're telling me is that

24 Section 9.7 is only intended to be payments to HOA and


16

1 folks, not releases from the escrow agreement.

2 MR. ROLLO: I think functionally

3 that's correct, Your Honor, because the escrow

4 agreement itself is a separate document that says if

5 you don't have indemnification claims validly made

6 against it, then, at the end of its term, those funds

7 are released, and that in terms of merger

8 consideration, whether that consideration is

9 contingent or absolute, it's still merger

10 consideration.

11 So that Your Honor's definition of

12 2.1(b), contingent merger payments, those are still

13 part of the merger consideration. Then if you go back

14 to the release Romanette "ii" it says that the

15 settlement agreement and whatnot will not result in a

16 modification of the merger consideration,

17 paraphrasing, of course.

18 THE COURT: You obviously resist any

19 broad construction of the release that gave up the

20 escrow.

21 MR. ROLLO: Absolutely, Your Honor.

22 THE COURT: All right. Let me hear

23 from the other side.

24 MR. ROLLO: Thank you, Your Honor.


17

1 MR. DRIEMEIER: Good afternoon, Your

2 Honor. I think that it's important to begin with the

3 context in which this release was entered. It was

4 entered as part of a settlement of the ROFR

5 litigation, the litigation about who was going to be

6 able to purchase Hooters of America or whether it

7 would be the Wellspring, NRI parties or our clients

8 that would be able to spend the $223 million to

9 acquire a very complicated enterprise of global reach.

10 It is really beyond logic to think

11 that as part of a release that was executed in the

12 context of tying up the loose ends of that litigation

13 over those threshold questions of who would get to

14 enter into the substantive agreement to purchase HOA,

15 that our clients released all of their substantive

16 rights under that $223 million agreement.

17 In fact, not only does the context

18 suggest that that is highly unlikely; the text of the

19 agreement provides a specific preservation of the

20 claims under the agreement. Because after

21 delineating -- of course, it's not a general release.

22 After delineating categories of claims, specific

23 categories of claims that are released, the agreement

24 provides specifically that the foregoing, that list of


18

1 claims, shall not include any claims to enforce the

2 terms and conditions of the amended and restated

3 Holdings merger agreement.

4 Now, my friend would have the Court

5 read into that clause a temporal limitation, a

6 temporal scope; only those claims to enforce the terms

7 and conditions going forward. He also wants to

8 rewrite that language to eliminate the words "terms

9 and conditions of the amended and restated merger

10 Holdings agreement" as obligations because that

11 comports with his temporal-only forward-looking view.

12 THE COURT: I don't think that's where

13 he gets the temporal issue. Maybe I'm misunder-

14 standing it, but I think that he agrees with you that

15 there is a broad preservation of claims to enforce the

16 merger agreement. I think the temporal issue comes

17 into play because under the first and second

18 exceptions, there are carveouts from that broad

19 preservation, and the temporal issue comes from your

20 inability to release future claims. So, therefore,

21 those carveouts are limited to the date of the

22 execution of the release.

23 Again, he can correct me, but I don't

24 think the temporal limitation comes from the


19

1 preservation of the merger agreement enforcement

2 right.

3 MR. DRIEMEIER: But based on that

4 understanding, that the temporal limitation as

5 provided by Delaware common law as they argue in the

6 reply, then the proviso and the "except that" clause

7 are wholly unnecessary because solely on the force of

8 the release itself in the Romanettes, they would not

9 have encompassed suits to enforce the agreement going

10 forward and would only have released claims for past

11 breach.

12 So the two clauses that the parties

13 spent a lot of time negotiating, there's a lot of

14 exchanges about how the "specifically included"

15 becomes surplusage. But I would also point out that

16 it's inconsistent, because under the agreement, it

17 explicitly provides, among the scope of types of

18 things that are released, the word "obligations," so,

19 again, the argument that they advance really turns

20 this into a circular exercise because obligations are

21 released, because the word -- one of the first words

22 in terms of the scope of the release is "any and all

23 claims, demands, rights, actions, potential actions,

24 causes of action, liability, damages, lawsuits,


20

1 obligations," so any extant obligation would be

2 released.

3 Then there's the proviso, and they say

4 the "except that" clause means that all that was

5 released is still released so that "all obligations"

6 which would include all obligations extant at that

7 date, all obligations under the merger agreement are

8 released.

9 And there's nothing to, on that view,

10 the proviso which is clearly intended to do something.

11 Both common sense and rules of construction of

12 contracts provides that the Court should not adopt a

13 construction that would render the language

14 surplusage. On their view, that proviso which the

15 parties spent so much time about, so many exchanges

16 about, becomes a nullity.

17 And if there were any question about

18 that, the fact that on the day before the agreement

19 was signed we wrote to counsel for sellers and

20 specifically said that it was our understanding that

21 this proviso preserved our ability to bring typical

22 buyer/seller claims, there was no rejection of that.

23 There was no even question about what that term meant.

24 The only response was one of agreement.


21

1 They recognized that the proviso

2 preserved the right to enforce the terms and

3 conditions of the merger agreement. Of course, one

4 enforces the representations and warranties when the

5 party does not make good on them by suing for breach,

6 and that's what we're here today about.

7 I think that Your Honor's focus in

8 your questions of my friend on the trouble that their

9 interpretation has with making sense of the second

10 Romanette of the "except that" clause is right.

11 Because, really, they trip themselves up. If the

12 release was as broad as they now contend, they would

13 have asked for the escrow to have been returned to

14 them because there really would have been virtually

15 nothing for us to have asked for them that would have

16 implicated the escrow.

17 More importantly perhaps, they would

18 not have paid a purchase price adjustment in late May

19 of that year based on facts that were alerted to them

20 in March of 2011 and that involve a lot of the same

21 types of issues that gave rise to the representation

22 of the warranty claims.

23 Exhibit 24 of the Schulman affidavit

24 that we submitted is the document relating to the


22

1 purchase price adjustment. If you look at the last

2 few pages of that exhibit, it goes through a list of

3 the types of items that were subsumed in the

4 adjustment, and they include things like the North

5 Carolina Dram Shop litigation. They include things

6 like taxes. They include things like the fact that we

7 were not getting the kinds of royalties that we had

8 expected from Wings Over Germany. All of those issues

9 were made part of a purchase price adjustment.

10 Then they would today say that -- I

11 don't actually understand quite their explanation

12 about why, under their theory of the release, they

13 were not also released from paying that purchase price

14 adjustment, and yet they did, just weeks after having

15 signed the release.

16 So we have the contemporaneous

17 communications between the parties. We have the

18 conduct of the parties subsequent to the signing of

19 the release. But more importantly from our view,

20 because we don't think you even need to get to those,

21 we have the context in which it was negotiated, and

22 the fact that a very specific proviso was inserted to

23 preserve precisely this type of claim, and that

24 proviso is rendered meaningless on their


23

1 interpretation.

2 I think it's interesting to contrast

3 the proviso and "except that" clause in the release

4 that we're discussing now with the release that was

5 included in the January 24 amended and restated merger

6 agreement, because, there, there are two things that

7 are notable. One is that the proviso preserves claims

8 to enforce obligations, and it specifies obligations.

9 It doesn't use the broader "terms and conditions"

10 under the loan agreement that relate to conduct that

11 was to occur after December 24th.

12 THE COURT: I thought I had flagged

13 that provision in the first amendment. I remember it

14 being in the seven's. 7.14. Okay. I got it now.

15 Get back on your horse now.

16 MR. DRIEMEIER: So what we have is we

17 have a distinction both because the temporal

18 limitations are spelled out explicitly. It's limited

19 to obligations, and it says obligations based on

20 conduct after December 24th. So all of those features

21 of the January 24th release are absent from the

22 release that we see on May 3rd.

23 So it really, I think, highlights the

24 extent to which one has to write into the May 3rd


24

1 release the kinds of limitations that the sellers are

2 advocating today, and why it is that we had no

3 expectation that that would be the interpretation.

4 In fact, it's interesting that the

5 language that the sellers have seized upon, the sale

6 of HOA, in the context of this list of Romanettes,

7 it's really the emphasis that is on the sale of HOA to

8 the private equity parties. Now, why is that included

9 in our release with the sellers? Well, you do have to

10 kind of go back to the history of the development of

11 the release.

12 THE COURT: Go back and explain to me

13 what is the business reason you think for -- what is

14 the inference I should draw from the plain meaning of

15 the distinction that you pointed out between 7.14(b)

16 and the release found in Exhibit G? In other words,

17 why in 7.14(b) do people go to the trouble of

18 including the dates and not include them in the

19 release? What was on peoples' minds at the time that

20 7.14 made them do that?

21 MR. DRIEMEIER: Well, I think with

22 respect to 7.14, it was that there was a release of

23 the kind of delay -- that was one of the big issues,

24 the disputes between the parties, the delay on the


25

1 part of the sellers in signing the merger agreement

2 and beginning to cooperate. They were willing to

3 release -- the buyers were -- it was the sellers delay

4 up until December 24th but not their delay thereafter.

5 So it was a temporal distinction that the parties were

6 drawing.

7 But with respect to the release on

8 May 3rd, it's a substantive distinction that the

9 parties are drawing. We are releasing, and we mean

10 it, that's what the "except that" clause basically

11 means, all of the claims relating to this dispute

12 about who would buy HOA, would it be NRI, and you have

13 the first of the Romanettes, the NRI merger agreement.

14 Now, mind you, one of the significant

15 features of this list of Romanettes is that there is

16 no Romanette that simply says "the amended and

17 restated Holdings merger agreement." That is, I

18 think, very significant. Then you have the separate

19 and apart from the NRI merger agreement, the entering

20 into or termination of the NRI merger agreement.

21 So already we think, okay, these

22 clauses are pretty specifically crafted because

23 there's some distinction that the drafters see between

24 the NRI merger agreement and the entering into,


26

1 meaning like the fact of the entering into and the

2 fact of the termination of.

3 THE COURT: Really, in your view of

4 the Romanettes, you don't even need the second

5 proviso.

6 MR. DRIEMEIER: True, Your Honor, but

7 I think it's fair to say that we anticipated that we

8 might be here without the proviso. We never

9 anticipated that we would be here with the proviso.

10 We thought the proviso was belt and suspenders and it

11 provided the clarity that we needed.

12 When the other side said, well, wait a

13 minute, you can't then turn around and urge that the

14 delay and all those various things that we're

15 resolving here as part of this release and settlement,

16 you're not going to be able to come back and

17 recharacterize those as the basis of a breach of

18 contract claim based on the merger agreement.

19 Now, in their reply, the sellers said,

20 well, that's meaningless because you couldn't breach

21 an agreement that wasn't in force. But, in fact, it's

22 a little too quick, a little too easy. If you look at

23 the reps and warranties and other provisions of the

24 amended and restated merger agreement, many of those


27

1 provisions have effective dates that predate

2 December 1st, 2010, even October 29, 2010.

3 So I'm not suggesting that we would

4 have done this, but the other side -- we could

5 understand why the other side would be concerned that

6 we would try to recharacterize these types of claims

7 as a breach of the merger agreement which, of course,

8 we thought was effective as of December 1.

9 We thought that when we sent them a

10 signed merger agreement that was the equivalent of the

11 agreement that they had entered into with NRI; that

12 that constituted a binding contract as of that point.

13 Of course, that too was an issue of dispute between

14 the parties, but we understood why they wanted that

15 clarification, and we gave it to them.

16 THE COURT: Remind me when the closing

17 actually was.

18 MR. DRIEMEIER: January 24th, 2011.

19 THE COURT: The same day as the

20 amended and restated agreement.

21 MR. DRIEMEIER: Yes.

22 I do want to, if I could, just go back

23 again to the Romanettes, and as we were saying,

24 they're kind of narrowly and specifically drawn,


28

1 "entering into" or "termination of" being different

2 from the NRI merger agreement itself.

3 We then have "entering into" the

4 Holdings merger agreement, our agreement, and of

5 course, there was no termination of it, so there's no

6 parallel there. But the effective equivalent of that,

7 or an analogue would be the consummation of it. So we

8 say the sale of HOA to our clients.

9 Now, these Romanettes were initially

10 drafted as part of a release that was to be included

11 in the agreement with NRI, and of course, that was

12 critical that we have that language in a release that

13 NRI was granting to us or granting to the sellers

14 because that was, of course, the whole basis of their

15 claim against the sellers, or against us; was that

16 they sold HOA to us instead of selling it to NRI, so

17 we had to have that included there to protect

18 ourselves.

19 But there was never a thought that

20 that would prevent us from enforcing the terms and

21 conditions of the merger agreement itself, the merger

22 agreement never being a Romanette by itself. And the

23 proviso so specifies.

24 So, again, as I said, and I think Your


29

1 Honor's question was well put, we think that the

2 Romanettes themselves, properly construed, protected

3 us. But understanding how, months later, years later,

4 the parties' views can evolve, we added that proviso

5 for clarity, and we don't think that it fairly can be

6 construed -- and certainly that contemporaneous

7 correspondence when we made clear our understanding

8 that it preserved traditional buyer/seller claims with

9 no dispute from the other side, we think there is no

10 question but that we have preserved our rights.

11 If Your Honor has no further

12 questions.

13 THE COURT: I don't, thank you.

14 Mr. Rollo.

15 MR. ROLLO: Thank you, Your Honor.

16 I tried to take notes on the various

17 points, so I'll try and respond as best I can to each

18 of them. I heard a lot about parol evidence, and I

19 don't need to recite all the case law that says it

20 can't be used to create an ambiguity.

21 In our reply brief, we walk through

22 the history. My friend suggested, I think twice,

23 there was a lot of negotiation around the second

24 proviso. Where is that in the papers? He also


30

1 testified, he testified, that they had no

2 understanding or belief.

3 Let me go back to kind of the

4 fundamental concept. It really doesn't matter what

5 they thought. It really doesn't matter if they

6 believed or didn't believe that it operated this way,

7 because the words on the page say what they say.

8 We talk about the spirit or intent

9 or -- I forget the phrase they use as to how to

10 rewrite the nine categories. But I didn't once hear

11 that the plain words on the page for category five do

12 not include the obligations under the merger

13 agreement.

14 Now, there were several loose

15 statements made during the argument, and I want to

16 correct them because they are important. There is the

17 suggestion that we are here contending that all of the

18 obligations under the merger agreement were

19 eliminated. That's not what we said. That's not

20 what's in our papers.

21 Yes, the initial definition was

22 everything, and then there is a carveback, and it says

23 "your obligations." Those obligations are saved. But

24 you can't sue us for a breach, a claim of breach based


31

1 upon what happened beforehand.

2 Now, my friends say that the second

3 proviso, I think it is conceded, that under their

4 interpretation, the second proviso is surplusage.

5 They said that it is unnecessary. Well, that's a

6 concession that their interpretation is inconsistent

7 with Delaware law.

8 We give meaning to the second proviso.

9 We give meaning to the second proviso in part if you

10 look at the negotiating history because we're the ones

11 who put it in. If you look at pages two through I

12 think it's seven of our reply brief, we walk through

13 the iterations. They proposed -- and I'm not

14 suggesting Your Honor should use any of this

15 information to create an ambiguity, but you can use it

16 secondarily in order to confirm the conclusion that

17 there is no ambiguity.

18 With that said, the first draft said

19 the merger agreement and everything related to it is

20 preserved in its entirety. I think it's an April 18th

21 draft from Mr. Bueker. Paragraph eight, it breaks

22 into two parts, 8-A, the lawsuit, or 8-B, a list of a

23 bunch of different things, including the transaction.

24 Any suggestion that they really meant


32

1 there that, oh, everything after 8-B is really

2 qualified by the lawsuit, it would have been worded

3 differently. They wouldn't have said "or." They

4 would have said "including but not limited to." They

5 didn't do that.

6 We rejected that carveout. We then

7 counter-proposed a carveout that contains functionally

8 this structure. Now, there's some modifications that

9 happened over time, but it didn't materially change.

10 The draft that we circulated back -- I

11 believe it has a paragraph 8-E, and it's cited in our

12 reply brief that makes clear how the ultimate

13 definition of "released claims" was intended to

14 operate because it didn't incorporate the definition

15 of released claims initially. That's consistent with

16 what we said.

17 I don't need to drag Your Honor

18 through the negotiating history because I don't think

19 Your Honor needs to get there, but I will say one

20 thing about this whole forthright negotiator. We sent

21 two emails. Both of them say -- and you can debate

22 how it's worded -- that the operation of the second

23 proviso prevented a warranty claim. Consistent with

24 this side. Not even addressed.


33

1 Had they had a piece of paper where

2 they said, "You're absolutely wrong, Mr. Rollo, you're

3 crazy, you're lying to the Court," I imagine Your

4 Honor would have it. This was a voicemail left near

5 the close of business on a Friday when this deal was

6 about to get done. And on a Monday morning, we get an

7 email that's vaguely worded about typical buyer and

8 seller claims.

9 I'll submit to Your Honor if I pass

10 out a bunch of pieces of paper and pens and said,

11 without talking to anybody, everyone write down in the

12 room what a typical buyer and seller claim is, we

13 wouldn't have agreement, not complete. We may have a

14 general notion of what it is, but those words, they

15 are critical to their position, do not show up in the

16 settlement agreement. They don't show up in the

17 release. They don't show up anywhere.

18 So while maybe you make a vague

19 statement at the end of a negotiation to preserve an

20 argument later that you want to create an ambiguity,

21 that doesn't modify the agreement. They proposed a

22 change. We rejected it. We counter-proposed. They

23 accepted it. Done.

24 Now, we add a footnote about a comma,


34

1 and I know we haven't talked about punctuation today,

2 and I don't think Your Honor needs to even address

3 whether or not that comma is necessary. I haven't

4 heard an argument today about the semicolons or lack

5 of semicolons in the definitions, and I think all of

6 that's laid out in the papers. I think Your Honor

7 basically understands our position.

8 But let me address one or two other

9 arguments. Your Honor asked about 7.14(b). I can

10 tell you exactly what was on everyone's mind at the

11 time. There were counterclaims, and those

12 counterclaims said we refused to sign the merger

13 agreement and as a result of it, we'd harmed them.

14 All of that conduct occurred beforehand.

15 Now, Your Honor asked what were we

16 thinking at the time. We had a deal with a different

17 bidder. Same reps and warranties. Same restrictions.

18 Now, things had happened between when we signed that

19 deal and this time.

20 For example, they consented to certain

21 things that violated the reps and warranties and

22 covenants. Now, those concessions were binding on the

23 prior bidder but not on my friends. So if we signed a

24 document that says something has to be true and where


35

1 the other earlier counter party had breached the

2 agreement, because as soon as we signed it, we

3 breached the agreement, we needed something that said

4 you couldn't sue us for that.

5 In their papers, they suggest that the

6 dispute in the earlier lawsuit were additional claims

7 against us. That's just not consistent with the

8 record. The counterclaims say what they say. The

9 counterclaims were filed before this release.

10 This release says basically it's the

11 cooperation covenants and we had to do everything in

12 our power to make sure this deal goes forward. They

13 can suggest that they believe maybe there were some

14 claims, but that's not what was at issue. Nor is

15 there a single document that suggests that the second

16 proviso was animated or included based upon those

17 concerns.

18 I'm generally pretty simple when it

19 comes to contract construction. There's a

20 straightforward answer, and we think it's the one we

21 put forward, and to reach an alternative conclusion,

22 Your Honor has to go through machinations, some

23 contortion of the language to reach the conclusion,

24 and the ultimate conclusion my friends concede does


36

1 not give any meaning to the second proviso.

2 Now, there was one additional point

3 concerning I'll call it the circularity in the second

4 proviso, and it's included in the briefs, but I think

5 today there was the suggestion that it was

6 unnecessary. I want to address that point because I

7 think it's just wrong.

8 The first part of the proviso says

9 notwithstanding this broad definition, which would

10 include the merger agreement and any obligations or

11 breaches, there's a savings clause. Obligations under

12 the merger agreement, and broader than that,

13 transactions directly related thereto. It's a broad

14 transaction.

15 The next one simply refers to claims

16 for breach. That is a smaller subset. So even if you

17 agree that that first clause, the savings clause, is

18 broad, as broad as you'd like it to be, ultimately

19 because there is a difference in the wording between

20 the first clause and the second clause, they're not

21 circular.

22 We submit that the reason they were

23 worded the way they are is the merger agreement has

24 obligations just like the escrow agreement, just like


37

1 everything else, and we wanted to insure that those

2 were not eliminated, because they would have been.

3 Without a proviso, our obligation to not use the

4 Hooters trade name would have been eliminated.

5 Now, with respect to the escrow

6 agreement which Your Honor raised initially about did

7 we release that, there's money in escrow. What right

8 would the buyers have to have the funds flow to them

9 in Your Honor's hypothetical. None.

10 So while I understand Your Honor's

11 hypothetical in terms of how you get there, it doesn't

12 lead to the conclusion that the buyers would have an

13 entitlement to that $11.5 million. So I don't think

14 ultimately that example creates a flaw in our

15 analysis, or alternatively, any ambiguity that would

16 render summary judgment inappropriate.

17 I guess the final point, unless Your

18 Honor has questions, one thing we haven't heard today

19 and anywhere in the brief, is an alternative

20 reasonable interpretation of the contract as a whole.

21 The buyer's position is predicated, the whole thing,

22 on the conclusion that the initial definition of

23 "released claims" does not include the merger

24 agreement obligations. They spend several pages on


38

1 that and that argument.

2 Now, I think that's flatly

3 inconsistent with the negotiating history and the

4 language in the contract. But if you eliminate that

5 assumption, every other argument in the answering

6 brief falls apart because there isn't a cohesive

7 explanation of the contract.

8 How was it intended to operate? What

9 did they think or what did they propose these

10 conflicting temporal clauses mean? Nothing. They

11 simply are trying to create an excuse to get to parol

12 evidence. Then when you get to the parol evidence

13 that they want to look at, it falls apart.

14 Now, one piece -- because Your Honor

15 had questions on it on the purchase price adjustment.

16 Back to what we said in our brief. The purchase price

17 obligation -- the adjustment is an obligation under

18 the contract. My friend pointed to I think it's

19 Exhibit 24. I don't believe the word "breach" is in

20 that document.

21 There is no claim for breach of the

22 merger agreement. We had a contract that says here's

23 how you adjust it. That is an obligation of the

24 merger agreement that continued to be enforceable. It


39

1 is not within the second part of the proviso that says

2 "except no release can be the basis for a breach of

3 the contract." That's the distinction.

4 THE COURT: So you beat that one, but

5 why wouldn't that then result in a released claim

6 affecting an adjustment to the merger consideration?

7 MR. ROLLO: I would submit to Your

8 Honor that release of the escrow is not a

9 modification.

10 THE COURT: Purchase price adjustment.

11 Purchase price adjustment is a modification to the

12 merger agreement consideration because you're

13 following one of these standard mechanisms. Yours was

14 a three-day, pre-closing, you submit your estimates,

15 and then within 60 days post-closing you do the

16 true-up, and you were modifying the merger

17 consideration to reflect what was really on the books

18 at the time of closing as opposed to what was in the

19 estimates.

20 So if the first half of the release

21 encompasses claims for the merger agreement such that

22 those are released claims, didn't you all accept that

23 a released claim was used to make an adjustment to the

24 merger consideration?


40

1 MR. ROLLO: I don't know if I agree

2 with that assertion, Your Honor, since this is the

3 first time I've kind of thought through it, that

4 particular argument, because it's not raised in the

5 papers. I think we talked about it earlier.

6 THE COURT: That's what I thought they

7 were saying.

8 MR. ROLLO: I didn't get that from the

9 papers, let me put it that way. That may be where

10 Your Honor got it from; the papers. As I think I said

11 in my reply brief, we didn't quite follow the

12 argument.

13 So let me respond in two ways. First,

14 even if we did make a payment we weren't obligated to

15 make --

16 THE COURT: You're just good guys.

17 MR. ROLLO: Let's say we did. That's

18 parol evidence. Post-execution conduct cannot be used

19 to modify the terms of the agreement. Let's say we

20 made that, and we had a right to say, "No, never

21 mind," at best, that's a waiver argument, and there's

22 the provision that says a waiver of one provision is

23 not a waiver of anything else.

24 So while I think in terms of structure


41

1 we can walk through whether or not we could have said

2 no around the same time we were agreeing to a

3 settlement with our friends, I don't know if it

4 ultimately impacts this question because it doesn't

5 modify the text.

6 But I'm happy to address any other

7 arguments or questions Your Honor has.

8 THE COURT: No, I don't have any other

9 questions.

10 All right. Well, thank you both for

11 your presentation. Let me give you a couple of

12 thoughts. First, purely non-substantively, two things

13 for the Delawareans to note for the future, and I give

14 this same advice to some of the other outstanding

15 firms, so don't take this personally. I really hate

16 this footnote structure in briefs. I'll tell you why:

17 Because I actually want to know what you're citing for

18 these statements. To figure out what you're citing

19 for these statements, I have to jump down to the

20 footnotes. Perhaps somebody with better visual acuity

21 than I is readily able to jump down from text to

22 footnote to text to footnote to text to footnote, but

23 I can't. So I get a quote that sounds really good. A

24 good example is Footnote 70. That is one that was


42

1 such a good example that I wrote it down. So I'm

2 reading along and this is -- let's get there. This is

3 a great statement. "Indemnity, in its most basic

4 sense, means reimbursement may lie when one party

5 discharges" blah, blah, blah, Footnote 70."

6 I have to look down to the footnote to

7 find out that that is an unreported Delaware Superior

8 Court case from 2008. Now, I'm not saying there's

9 anything wrong with an unreported Delaware Superior

10 Court case. Certainly we like unreported cases here

11 in Delaware, but I think it would be undisputed that

12 that case would have more heft were it a reported

13 Supreme Court case.

14 So it may be that my colleagues like

15 this footnote style, and it eases their minds when

16 they're reading because they don't have to actually

17 look at where the sources are from or that type of

18 thing. I find it very difficult to deal with. So

19 just in terms of submissions to me, if it's going to

20 be some lengthy string cite, you can put it down in

21 the footnote, but otherwise put these things in the

22 text, because, particularly in a section of the brief

23 where there's carpet bombing of footnotes after every

24 sentence, I'm bouncing down and back after every


43

1 sentence.

2 You'll notice that that's why I don't

3 use the Garner footnote style in my opinions. It's

4 because, again, I think it's ineffective for someone

5 who is immersed in the law of the particular

6 jurisdiction and therefore cares about what case

7 you're actually citing for a proposition. If you were

8 a lawyer who is itinerant and wanders from circuit to

9 circuit and is not as concerned with particular case

10 names or particular authorities, Garner is fantastic.

11 Why bother? I mean, that footnote method is great.

12 But I actually care about what cases you're citing for

13 these principles and where things are coming from.

14 Nobody should take it personally or anything like

15 that. But that's a little constructive point on that.

16 The other thing is I do not understand

17 why I got dueling transmittal affidavits where 80

18 percent of the documents were the same. Now, it

19 turned out to be helpful because I read everything out

20 of the Rollo affidavit. One of my cats decided that

21 the Rollo affidavit was an appropriate litter box

22 substitute. Because of that, I was glad to be able to

23 resort to the Hannigan affidavit. Generally speaking

24 though, I do not want to lug two things like this. I


44

1 mean, look, I've gotten to the point -- and the reason

2 why, when Mr. Bayliss worked for me, he had to carry

3 my bags was not because I was some crazy corner office

4 partner who wanted somebody to carry my bags, but

5 because I got tennis elbow. I didn't get tennis elbow

6 from playing tennis. I like to play tennis. I got

7 tennis elbow from carrying around a big heavy lit bag.

8 The lit bag is twice as heavy when I have the Schulman

9 affidavit and the Hannigan affidavit and the Rollo

10 affidavit.

11 Now, the last affidavit did have a

12 bunch of emails that weren't in the first two, but the

13 first two, I would say 80 percent of those documents

14 were the same. I got two copies of the merger

15 agreement. I got two copies of the escrow agreement.

16 I got two copies of the back and forth. I got two

17 copies of the blooming pleadings from the December

18 case. I don't need that.

19 So those are two practice points for

20 you all going forward, idiosyncratic though they may

21 be. Perhaps other members of the Court would like you

22 to handle their documents differently. But if you

23 would like me to be happy and smiling when I read your

24 papers, those are two things to remember.


45

1 I am going to give you my ruling now.

2 First of all, I want to start by giving you the

3 factual background because I think the time line is

4 important. I'll give you the punch line up front. I

5 am denying the plaintiff's motion for summary

6 judgment, and under the authority of Stroud V. Grace

7 and XO Communications LLC versus Level 3

8 Communications, because I think the language of the

9 contract is plain, I am granting summary judgment in

10 favor of the defendants on the interpretation of the

11 release.

12 So what that will leave, as far as I

13 understand it, is a case about the counterclaims. I

14 do believe that the counterclaims state a claim. I

15 think even this odd claim for the airplane usage --

16 the airplane judgment -- is something that, frankly, I

17 don't understand what's going on there. It's bizarre.

18 It seems to me that it's the type of thing where

19 conceivably one guy was in control of both entities,

20 and he said, "You know what? I'd rather have this be

21 a judgment against this one rather than that one."

22 If that pans out, and all I have is

23 the pleadings right now, but if that pans out, it's

24 conceivable to me that that could be a situation where


46

1 there would be grounds for indemnification. We'd have

2 to run it through the merger agreement. We'd have to

3 run it through the reps. We'd have to run it through

4 the disclosure schedules. But it's reasonably

5 conceivable. So I think but for the -- except for the

6 release argument -- the counterclaims state claims.

7 Now I am going to address the release

8 argument. The time line is that on October 29th of

9 2010, the plaintiffs originally signed up a deal to

10 sell the Hooters restaurant chain to Neighborhood

11 Restaurants Inc., which people refer to as NRI, and

12 which was connected with Wellspring Capital

13 Management; hence, those references in various

14 documents to the Wellspring claims and things like

15 that.

16 On December 1st, 2010, the private

17 equity group that now owns the Hooters chain through

18 HOA Holdings exercised a preexisting right of first

19 refusal that it had under an outstanding loan

20 document.

21 Six days later, on the 7th of

22 December, the plaintiffs filed a suit claiming that

23 they couldn't figure out whether the right of first

24 refusal had been validly exercised. Three days later,


47

1 on December 10th, HOA, the second bidder private

2 equity firm, filed counterclaims and cross claims.

3 In their answers to those, the

4 plaintiffs suddenly found pellucid clarity as to

5 whether the right of first refusal had been validly

6 exercised and conceded that it had been. Based on

7 that, on December 20th, 2010, I granted judgment on

8 the pleadings as to the valid exercise, the concededly

9 valid exercise of the right of first refusal. That

10 led essentially to a situation where there were two

11 merger agreements in play.

12 So, on December 22nd, 2012, there was

13 a first amendment to the merger agreement with HOA.

14 That's Exhibit C to the Rollo affidavit. That

15 agreement extended the closing deadline, it cut back

16 on the plaintiff's indemnification rights, facially

17 because they had created the mess, and in that

18 document, plaintiffs and the HOA Holdings group agreed

19 to give each other mutual releases but with carveouts

20 preserving rights under the operative transaction

21 documents.

22 Notably, paragraph nine left in place

23 the price adjustment provisions of the merger

24 agreement but provided that payment of litigation


48

1 expenses relating to the then extant litigation

2 wouldn't be treated as increasing current liabilities

3 or indebtedness or as reducing cash or cash

4 equivalents or otherwise. In other words, it wouldn't

5 be treated as having an effect on the price adjustment

6 provisions that were preserved.

7 On January 24, 2011, there was an

8 amended and restated merger agreement between

9 plaintiffs and HOA. That agreement superseded the

10 first amendment to the original merger agreement and

11 picked up and incorporated its provisions. A couple

12 things are important about this merger agreement.

13 First, it had an extensive section on indemnification

14 for breaches and inaccuracies of reps and warrantees.

15 The vast majority of the representations were extended

16 for a year plus 180 days post-closing. Fundamental

17 representations were extended forever. Tax

18 representations were based on the expiration of the

19 related tax oriented statute of limitations. A total

20 of $61.5 million was put into escrows for various

21 buckets of payments, and to govern that escrow

22 arrangement and to provide for the indemnification,

23 there was an escrow agreement dated January 24, 2011.

24 As I indicated in comments with


49

1 counsel, Section 2.1(b) of the merger agreement

2 provided that the LLC interests -- to facilitate that

3 transaction, the Hooters entity which originally was a

4 corporation had been converted into an LLC -- the LLC

5 interests were converted into the right to receive the

6 net merger consideration plus amounts received from

7 the shareholder escrow plus tax reimbursements.

8 Section 2.2 provided for a price true-up mechanism.

9 As I described earlier, it called for delivery three

10 days pre-closing of estimates of cash, cash

11 equivalents and indebtedness with a post- closing

12 purchase price adjustment to be completed 60 days

13 after closing based on actual figures.

14 Of importance to me at least, and to

15 an understanding what was going on, is what the

16 parties agreed to in the introductory paragraph of

17 Article 4. All of the representations and warrantees

18 that were set forth in the amended and restated merger

19 agreement were qualified, recognizing the existence of

20 the then still extant litigation, the Chancery

21 litigation.

22 The obvious purpose of that was

23 because the outcome of that litigation -- indeed, the

24 existence of that litigation -- could have had


50

1 significant implications for various representations

2 and warranties. Immediately jumping to mind are

3 Section 4.5(b), the absence of any undisclosed

4 liabilities other than those on the schedules. One

5 could envision being tripped up by Section 4.6, the

6 absence of certain changes; effectively, a "no MAE"

7 clause. And Section 4.7, no other litigation.

8 There were also potential issues for

9 the parties in terms of the conduct of business

10 between signing and closing. So Section 6.1 listed a

11 pretty extensive, very extensive, list of closing

12 covenants in terms of the operation of the business,

13 generally limiting the business to ordinary course of

14 business activities. A lot of the items that are

15 listed in there could have been affected by and

16 breached by, created problems for by, the outcome of

17 the then extant Chancery litigation.

18 In Section 7.4, what I interpret that

19 to be, that's the release that I discussed with

20 counsel, and it seems to be a release designed to

21 resolve any of the disputes that had been generated by

22 the Chancery litigation up until the date specified in

23 that provision, but which otherwise preserved the

24 right to enforce the transaction documents.


51

1 The merger closed promptly after the

2 signing of the merger agreement. I understand from

3 counsel it was, in fact, the same day. Now, four

4 months later, just under four months later, on

5 May 3rd, 2011, the Chancery litigation was settled.

6 NRI got $9 million. Everybody else got releases.

7 There was a side agreement among the folks in this

8 room about how to divvy up the $9 million payment.

9 Now, this was one of these settlements

10 where nobody actually wanted to have to ultimately

11 show the full agreement whenever they wanted to invoke

12 their releases, so rather than there just being an

13 agreement containing the releases, people signed a

14 settlement agreement and then signed a list of

15 releases from each party to the other parties. That's

16 what gets us to Exhibit G which is the specific

17 release that HOA gave to the plaintiffs.

18 Then, finally, October 4th, 2011,

19 within the schedule contemplated by the

20 indemnification provisions of the merger agreement,

21 HOA served the first of several notices for losses.

22 That's a defined term, "Losses," for indemnification

23 claims against the escrow fund based on alleged

24 breaches of reps and warranties.


52

1 Now, we're here today because in

2 response to those notices, the plaintiffs cited

3 Exhibit G and said, "Sorry, HOA, even though we set

4 aside this money for escrow, even though we had this

5 expansive procedure in the merger agreement to handle

6 post-closing indemnification claims, and even though

7 there's nothing specifically addressing the giving of

8 those up in the settlement agreement, when you granted

9 your broad release found at Exhibit G, that language

10 can actually be read to release any claim you might

11 have for anything you might bring up in the next year

12 plus 180 days for most representations, anything you

13 might bring up forever for fundamental

14 representations, and anything you might bring up for

15 taxes at any point during the period before the

16 expiration of the statute of limitations. You guys

17 gave all that away."

18 Well, that was a claim with which HOA

19 disagreed. It is also one with which I disagree.

20 The release has three pertinent parts.

21 I decide this entirely based on plain language. The

22 release has three pertinent parts. The first is an

23 expansive definition of what qualifies as a Released

24 Claim, and that is with a capital R, capital C.


53

1 The second important part has two

2 provisos where, after the definition of Released

3 Claim, there are two sections where the parties said

4 "provided that."

5 Then finally there are two exceptions

6 to the second proviso.

7 Let's start with the definition of

8 released claims. It is typically expansive. It is

9 plainly attempting to give broad and global releases

10 as to everything related to the enumerated items. In

11 other words, it has the type of language that one

12 would see in a general broad universal release

13 covering everything from the beginning of time with

14 every adjective that any lawyer who ever touched the

15 form language could find in the Thesaurus. But then

16 rather than simply stopping with that broad language,

17 it lists items that the claims have to relate to.

18 It's, therefore, a specific release.

19 In my view, the plain language of the

20 enumerated items facially reflect what actually was

21 being settled; namely, the dispute over the then

22 extant Chancery litigation about which the merger

23 agreement had been validly entered into, how those

24 events occurred, and what the consequences of those


54

1 events would be.

2 The plain language of the Exhibit G

3 release, as well as the plain language of all of the

4 mix-and-match releases that the parties entered into

5 to implement this settlement agreement, drew

6 distinctions between the original NRI merger agreement

7 and the HOA merger agreement.

8 For example, Romanette "i" releases

9 all claims relating to the NRI merger agreement. It's

10 gone, done, over. Romanette "iii" releases all claims

11 relating to the entering into or termination of the

12 NRI merger agreement. Just in case you didn't realize

13 from Romanette "i" that it was gone, over and done,

14 Romanette "iii" is clear that anything related to the

15 entering into of that transaction or the termination

16 of that transaction is gone, over and done.

17 Contrast that with what it says about

18 the Holdings merger agreement. Romanette "iv" only

19 releases matters relating to the entering into of the

20 Holdings merger agreement. What that distinction is

21 plainly attempting to capture is the idea that the

22 Chancery litigation focused on the events leading up

23 to the entering into of the Holdings merger agreement.

24 Thus, while the parties were getting


55

1 rid of, in its entirety, the NRI merger agreement,

2 everything relating to the entering into of that

3 agreement, and everything relating to the termination

4 of that agreement, all people were focused on, as

5 shown by the plain language of Romanette "iv" was the

6 entering into of the Holdings merger agreement.

7 Likewise, in Romanette "v" they were

8 worried about the sale of HOA to a specific set of

9 buyers; namely, the exercisers of the ROFR. Read in

10 context, the plain language of that phrase

11 distinguishes between the sale of HOA to the second

12 set of buyers; namely, the exercisers of the ROFR, as

13 contrasted to the first folks in the door, NRI.

14 Now, the problem with that is that

15 although that is the plainest reading of what the

16 romanettes said, the global release language that

17 precedes the specific items that make it a specific

18 release, is quite expansive. There could be

19 uncertainty, particularly litigation-driven

20 uncertainty, as to the interpretation of what would

21 happen to the existing Holdings merger agreement.

22 Prudent transactional attorneys might worry that if

23 the romanettes were read too broadly, someone would

24 argue that the Holdings merger agreement had been, in


56

1 fact, itself released.

2 So, hence, you have two provisos: The

3 first proviso which was in all of the myriad

4 mix-and-match release documents is that "Provided,

5 however, that nothing in this release shall bar any

6 party from taking any action necessary to enforce the

7 terms of the accompanying settlement agreement," and

8 that's the settlement agreement relating to the

9 Chancery litigation. This first proviso was included

10 because otherwise the release is so blooming broad

11 that even the settlement agreement itself could be

12 released. What that first proviso demonstrates is

13 that the settlement agreement was not part of the

14 defined term "Released Claims," so if you start out --

15 imagine a circle. Think Venn diagrams encompassing

16 released claims. We are then taking a bite out of

17 that circle relating to the enforcement of the

18 settlement agreement.

19 We then get to the second proviso

20 where it says, "Provided further, however, that the

21 foregoing shall not include any claims to enforce the

22 terms and conditions of the amended and restated

23 Holdings merger agreement or directly relating to the

24 transactions contemplated thereby." Again, otherwise


57

1 some litigation-minded parties potentially could argue

2 that the amended and restated Holdings merger

3 agreement itself had been covered and released.

4 Now, I agree with Mr. Rollo that in an

5 ideal world perhaps, instead of "provided further

6 however" someone truly channeling Brian Garner would

7 have said, "for the avoidance of doubt," but I think

8 that it is sufficiently plain from the "provided

9 further," particularly when combined with the first

10 "provided further" clause, that the intent of this

11 construction was to carveout from the definition of

12 Released Claims any claim for enforcement of the

13 amended and restated Holdings merger agreement.

14 That is another bite out of the circle

15 that otherwise would be Released Claims. In other

16 words, if you had any doubt at all based on the

17 structure of the nine romanettes, we are now

18 confirming through this "provided further" clause that

19 claims to enforce the merger agreement are not part of

20 the Released Claims.

21 Now, this, however, created a problem

22 for the sellers. Why? Because there were matters

23 that were the subject of the Chancery litigation that

24 could be used to claim indemnifiable breaches of the


58

1 reps and warranties or breaches of the covenants as to

2 how the business was to be operated between signing

3 and closing.

4 I discussed in my factual exposition

5 why there are some relatively clear items that might

6 jump out. Just to put a finer point on it, the

7 $9 million payment to NRI, how would that fit into the

8 liabilities and the need to disclose liabilities? It

9 would be hard to call that a MAC, but we have seen

10 weaker MAC claims. Would that have been -- would the

11 agreement to do that, or the exposure of the company

12 to that claim, be something that could be shoe horned

13 into one of the closing covenants if not into one of

14 the representations?

15 We, therefore, have the "except that"

16 provisions. The first "except that" provision says

17 "No released claim shall be the basis for any claim of

18 breach of the amended and restated Holdings merger

19 agreement." What that is saying is if you had any

20 doubt that we were releasing the Released Claims, we

21 really are giving them up, and you, HOA, as buyer, or

22 this is a release from HOA, so it's we, as HOA, as

23 buyer, will not claim that anything that happened and

24 was at issue in that Chancery litigation about the


59

1 ROFR or the delay in payment or the incurrence of

2 contingent liabilities that might thereby have been

3 not adequately disclosed Liabilities, we're not going

4 to claim that any of those are breaches of the merger

5 agreement. We're letting those go.

6 You also have the second proviso which

7 says that except that no Released Claim and no

8 liability or payment under the accompanying settlement

9 agreement shall result in any adjustment to the merger

10 consideration due under the merger agreement. This is

11 exactly the same concept spelled out slightly

12 differently to make sure that no clever transactional

13 lawyer, or clever private equity guy at the Karp firm,

14 could try to get back some of his expenses incurred in

15 the litigation or his piece of that $9 million payment

16 as part of the price adjustment or as part of an

17 indemnification claim.

18 This is saying, "No, we're not going

19 to try to re-trade the settlement by saying that the

20 expenditure of cash to pay that $9 million actually is

21 something that we can then assert as an

22 indemnification claim to come back on you."

23 It doesn't give up the entire

24 indemnification framework because it would only give


60

1 up the entire indemnification framework if all of that

2 article was part of the definition of Released Claim.

3 As I have already said, for two independent reasons,

4 that is not the case. That is not the case, first,

5 because the plain language of the series of romanettes

6 makes clear that they are not releasing claims for

7 enforcement of the merger agreement, and to again

8 avoid any litigation-oriented reinterpretation of the

9 romanettes, the second proviso makes clear that claims

10 to enforce the amended and restated Holdings merger

11 agreement, including things like your indemnification

12 rights, don't fall within the definition of Released

13 Claims.

14 So, given all this, it is clear to me

15 that the release does not mean what the plaintiffs are

16 now arguing; namely, that claims for indemnification

17 under the merger agreement are Released Claims and

18 therefore can't be part of the indemnification process

19 which includes the need to assert that there was some

20 breach of a representation and warranty, thereby

21 falling afoul of romanette "i" of the first exception,

22 in plaintiff's view. Nor is it a Released Claim that

23 would lead to an adjustment of the merger

24 consideration and therefore running afoul of the


61

1 second exception, in the plaintiff's view.

2 Contrary to plaintiff's argument, the

3 plain language of the settlement agreement carves out

4 enforcement of the merger agreement from the

5 definition of Released Claims. Contrary to the

6 plaintiff's argument, the structure of the release as

7 a whole was clearly intended to address the then

8 extant Chancery litigation and the potential

9 re-cycling of those claims through either a breach of

10 the merger agreement assertion, or, more importantly,

11 through the indemnification process.

12 Frankly, it is facially implausible

13 and absurd, given the detailed indemnification

14 provisions, given the sequence of events that led to

15 this settlement, given the nature of the settlement

16 payment, that in agreeing to these releases, the

17 buying parties gave up an otherwise quite detailed

18 indemnification article that entitled them to assert

19 breaches of most reps and warranties for 545 days and

20 other representations longer.

21 Now, I need not reach extrinsic

22 evidence, but were I to do so, I think it's consistent

23 with the plain meaning, and most importantly, I look

24 at the parties' post-contracting behavior. I should


62

1 say post-contracting behavior prior to the buyers

2 actually asserting a meaningful claim for

3 indemnification at which point the sellers suddenly

4 raised this release argument.

5 Prior to those events in October, the

6 parties' post-contracting behavior was consistent with

7 the plain meaning of the release and contrary to the

8 plaintiff's position. So, first of all, there was no

9 effort to shut down, or more likely, modify this

10 indemnification escrow. Again, there was a lot of

11 money in this thing, and even though there's a

12 provision saying that it's capped for tax purposes at

13 the amount that goes to EORHB at 20 million,

14 20 million is still a chunk of change.

15 The escrow agreement is a document

16 that limits the type of things in which the escrow

17 agent can invest. If EORHB and the clever fellows on

18 that side of the deal really thought that they had

19 gotten a release essentially giving up the ability, in

20 which the buyers gave up their ability to raise

21 breaches of reps and warranties under the theory that

22 all of those had to occur or not occur pre-closing, I

23 guarantee you that that event would have been followed

24 quite promptly by a demand for some portion of the


63

1 escrow agreement based on the idea that there is no

2 way that 20 million ought to be sitting in there for

3 545 days when virtually every rep and warranty claim

4 and breach of covenant claim had been given up.

5 Separately and independently, the

6 plaintiffs went forward with a price adjustment. The

7 price adjustment provision is the short-term

8 adjustment for which the indemnification section is

9 the long-term adjustment. All of the same arguments

10 that are being raised now about the indemnification

11 issue could have been raised about the price

12 adjustment with the exception -- I agree with

13 Mr. Rollo on this -- that the argument would not be

14 based on a breach. The argument would be based on

15 romanette "ii" that said "all released claims were

16 given up and shall have no effect as a price

17 adjustment."

18 People went forward with the price

19 adjustment blissfully -- perhaps not blissfully but

20 blithely. Certainly blithely with respect to the idea

21 that there had been some type of release of this

22 mechanism and anything pre-closing that might have

23 been a deviation from what the actual results were.

24 That confirms, in my mind, what the


64

1 plain language says and that this has been a --

2 perhaps it wasn't a late-adopted strategy. Perhaps it

3 was an anticipated strategy. I don't know. I don't

4 need to make that decision. But certainly nobody

5 acted as if this was really a release until the big

6 dollar indemnification claims came in, and one can

7 almost imagine people saying, "Whoa, we got to figure

8 out some reason why these aren't valid. How about

9 those releases."

10 Lastly, although I do think that there

11 are perhaps some contractual gymnastics that one can

12 go through to preserve a claim, the full import of the

13 plaintiff's theory in terms of the capaciousness of

14 the release language could be read to give up their

15 right to the escrow. The original language of the

16 released claims is just so darn broad, and the link of

17 that to price adjustments in the second exception is

18 so problematic for price adjustments that are paid out

19 of the escrow, particularly given the language of the

20 merger agreement that defines those payments as price

21 adjustments for tax purposes that, again, it renders,

22 in my mind, highly implausible the argument that the

23 plaintiffs are now advancing as the plain meaning of

24 this release.


65

1 So, to come full circle, I am granting

2 summary judgment for the defendants on Count I

3 regarding the nature of the release. I'm not saying

4 that it's ambiguous. I'm saying that, plainly read,

5 it doesn't do what the plaintiffs say it does, and

6 plainly read, it preserves the defendants' right to

7 seek this type of indemnification claim that they have

8 asserted.

9 Again, I am denying the motion to

10 dismiss as to the counterclaims. Having reviewed the

11 counterclaims, I think that but for the release

12 argument, it is reasonably conceivable that they state

13 a claim. It's also reasonably conceivable to me that

14 there could be, depending on how the facts pan out,

15 something relating to this litigation over the plane.

16 So, as far as my view of the matter, and people can

17 discuss this, but it seems to me the case is going

18 forward only as to counterclaims.

19 Now, before I say that so

20 definitively, Mr. Rollo, is there something other than

21 the counterclaims that I am missing that would still

22 be live given those rulings? I understand you

23 disagree with those rulings. I'm not asking you to

24 agree with them, but stuck as you are at least for the


66

1 present with those rulings, is there anything that you

2 think would go forward other than the counterclaims?

3 MR. ROLLO: Not that I can recall.

4 THE COURT: Rise up.

5 MR. ROLLO: I apologize, Your Honor.

6 Not that I recall.

7 THE COURT: Just the speaker.

8 MR. ROLLO: No, Your Honor; not that I

9 am aware of at this point.

10 THE COURT: Thank you. Why don't you

11 all talk about a scheduling order for the litigation

12 on the counterclaims. This seems to me to be an ideal

13 non-expedited case in which the parties would benefit

14 from using predictive coding. I would like you all,

15 if you do not want to use predictive coding, to show

16 cause why this is not a case where predictive coding

17 is the way to go.

18 I would like you all to talk about a

19 single discovery provider that could be used to

20 warehouse both sides' documents to be your single

21 vendor. Pick one of these wonderful discovery super

22 powers that is able to maintain the integrity of both

23 side's documents and insure that no one can access the

24 other side's information. If you cannot agree on a


67

1 suitable discovery vendor, you can submit names to me

2 and I will pick one for you.

3 One thing I don't want to do -- one of

4 the nice things about most of these situations is once

5 people get to the indemnification realm, particularly

6 if you get the business guys involved, they have some

7 interest in working out a number and moving on. The

8 problem is that these types of indemnification claims

9 can generate a huge amount of documents. That's why I

10 would really encourage you all, instead of burning

11 lots of hours with people reviewing, it seems to me

12 this is the type of non-expedited case where we could

13 all benefit from some new technology use.

14 What else should we talk about today?

15 Mr. Rollo, from your side?

16 MR. ROLLO: At this point there is

17 nothing else I think that we can talk about today.

18 THE COURT: Mr. Bayliss, anything that

19 you'd like to discuss?

20 MR. BAYLISS: Nothing, Your Honor.

21 Thank you.

22 THE COURT: All right. Thank you all

23 for coming in. It was very well briefed, and I

24 appreciate you all getting me so prepared that I was


68

1 able to give you a ruling today.

2 We stand in recess.

3

4 (The Court adjourned at 3:35 p.m.)

5

6

7 -----

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24


69

CERTIFICATE

I, MAUREEN M. McCAFFERY, Official Court

Reporter of the Chancery Court, State of Delaware, do

hereby certify that the foregoing pages numbered

3 through 68 contain a true and correct transcription

of the proceedings as stenographically reported by

me at the hearing in the above cause before the Vice

Chancellor of the State of Delaware, on the date

therein indicated.

IN WITNESS WHEREOF, I have

hereunto set my hand at Dover, this 17th day of

October, 2012.

/s/Maureen M. McCaffery ----------------------------

Maureen M. McCaffery Official Court Reporter

of the Chancery Court State of Delaware


SETH HARRIS, Acting Secretary of Labor, United States Department of Labor,Petitioner, - v - SUBCONTRACTING CONCEPTS, LLC, Respondent.

Civ. No. 1:12-MC-82 (DNH/RFT)

UNITED STATES DISTRICT COURT FOR THE NORTHERN DISTRICT OFNEW YORK


March 11, 2013, DecidedMarch 11, 2013, Filed

COUNSEL: [*1] For the Petitioner: ANDREW M.KATZ, ESQ., OF COUNSEL, UNITED STATESDEPARTMENT OF LABOR OFFICER OF THESOLICITOR, New York, New York.

For the Respondent: DAVID F. JASINSKI, ESQ., OFCOUNSEL, JASINSKI, P.C., Newark, New Jersey.

JUDGES: Randolph F. Treece, U.S. Magistrate Judge.

OPINION BY: Randolph F. Treece

OPINION

RANDOLPH F. TREECEUnited States Magistrate Judge

MEMORANDUM-DECISION and ORDER

This matter was submitted to this Court by theUnited States Department of Labor's Wage and HourDivision (hereinafter "DOL") seeking to compelSubcontracting Concepts, LLC (hereinafter "SCI LLC")to comply with its Subpoena Ad Testificandum(hereinafter "Subpoena"). Dkt. No. 1, Pet'r Mot. toCompel. On February 11, 2013, this Court issued aMemorandum-Decision and Order (hereinafter "MDO")granting in part DOL's Motion. Dkt. No. 12. For the mostpart, SCI LLC was directed to testify at a deposition and

to produce requested documents, albeit with specificlimitations. Some of the production is subject torepresentative sampling, while disclosure of SCI LLC'sclient list was contingent upon "the [*2] understandingthat DOL is not to publish, disclose, nor reveal this list toany third party outside the context of any prospectivelitigation." Id. at p. 20. Lastly, the applicable statute oflimitations was tolled for a definite duration. Id. at pp.21-22.

On February 25, 2013, SCI LLC filed a Motion forReconsideration of the MDO on the ground of newlydiscovered evidence. Dkt. No. 13. Additionally, SCI LLCseeks clarification of the MDO as well as a more formaland extensive protective order. Id. Pursuant to the Court'sAmended Text Notice, the Motion for Reconsiderationwas placed on an expedited schedule requiring DOL tofile its Response in Opposition to the Motion on anabbreviated time table, Dkt. No. 14, DOL's Opp'n, datedMar. 5, 2013, and eliminating the need for replies andsurreplies.

MOTION FOR RECONSIDERATION STANDARD

Normally, prior decisions made within the same casemust be followed under the law of the case doctrine.Catanzano by Catanzano v. Wing, 103 F.3d 223, 231 n. 5(2d Cir. 1996); Shomo v. City of New York, 579 F.3d 176,186 (2d Cir. 2009); United States v. Millet, 208 F.3d 204

Page 1

(2d Cir. 2000) (noting that court should "not depart fromthis sound policy absent [*3] cogent or compellingreasons"). Generally, reconsideration of a court's priordecision is warranted only where the moving partydemonstrates (1) an intervening change of controllinglaw; (2) the availability of new evidence; and/or (3) theneed to correct a clear error or prevent manifest injustice.Crucible Materials Corp. v. Certain Underwriters atLloyd's London & London Market Companies, 681 F.Supp. 2d 216, 225 (N.D.N.Y. 2010); Caidor v.Harrington, 2009 U.S. Dist. LEXIS 24343, 2009 WL799954, at *1 (N.D.N.Y. Mar. 24, 2009) (Suddaby, J.)(quoting United States v. Sanchez, 35 F.3d 673, 677 (2dCir.), cert. denied, 514 U.S. 1038, 115 S. Ct. 1404, 131L. Ed. 2d 291 (1995); see also Pescatore v. PanAmerican World Airways, Inc., 97 F.3d 1, 8 (2d. Cir.1996) (one ground for reconsideration includes anintervening change of controlling law); Delaney v. Selsky,899 F. Supp. 923, 925 (N.D.N.Y. 1995) (citing Doe v.New York City Dep't of Soc. Servs., 709 F.2d 782, 789(2d Cir.), cert. denied, 464 U.S. 864, 104 S. Ct. 195, 78L. Ed. 2d 171 (1983)). Thus, the moving party must"point to controlling decisions or data that the courtoverlooked -- matters, in other words, that mightreasonably be expected to alter the conclusion reached bythe court." Shrader v. CSX Transp., Inc., 70 F.3d 255,257 (2d Cir. 1995) [*4] (citations omitted).

"[A] motion to reconsider should not be grantedwhere the moving party seeks solely to relitigate an issuealready decided." Id. at 257. "[A]ny litigant consideringbringing a motion for reconsideration must evaluatewhether what may seem to be a clear error of law is infact simply a point of disagreement between the Courtand the litigant." Gaston v. Coughlin, 102 F. Supp. 2d 81,83 (N.D.N.Y. 2000) (citation omitted). Of significancehere, "[a] motion for reconsideration is not an opportunityfor a losing party to advance new arguments to supplantthose that failed in the prior briefing of the issue."Fredericks v. Chemipal, Ltd., 2007 U.S. Dist. LEXIS49185, 2007 WL 1975441, at *1 (S.D.N.Y. July 6, 2007).In other words, it is not an opportunity to take a "secondbite at the apple." Analytical Surveys, Inc. v. TongaPartners, L.P., 684 F.3d 36, 52 (2d Cir. 2012) (quotingSequa Corp v. GBJ Corp., 156 F.3d 136, 144 (2d Cir.1998)); see also In re Health Mgmt. Sys., Inc. Sec. Litig.,113 F. Supp. 2d 613, 614 (S.D.N.Y. 2000)("[R]econsideration of a previous order is anextraordinary remedy to be employed sparingly in theinterests of finality and conservation of scarce judicial

resources.") [*5] (quotation marks and citationsomitted); In re Bird, 222 B.R. 229, 235 (Bankr. S.D.N.Y.1998) ("A motion for reconsideration is not a forum fornew theories or for plugging the gaps of a lost motionwith additional matters.") (internal quotation marks andcitation omitted).

NEWLY DISCOVERED EVIDENCE

SCI LLC seeks reconsideration of the MDO basedupon newly discovered evidence. The Court is told thatthis new revelation became apparent to SCI LLC duringthe latter stage of the Motion to Compel discourse whenDOL filed its reply and included an especially curiousexhibit. That curious exhibit was a redacted check issuedby Subcontracting Concepts, Inc (hereinafter "SCI").1 SeeDkt. No. 8-1, Lisa Schneider Supp. Decl. & Supp. Ex. A.Relying upon Investigator Schneider's averment that shehad "obtained copies of checks issued by "'SCI' to one ofthe individuals termed an independent contractor" inorder to show that there may be a business relationshipbetween SCI LLC and SCI, this Court accepted it forwhat it portrayed - a blank check. See id. at ¶ 3 (attachinga copy of "a partially redacted [check] to protect theidentity of the payee"). However, still inexplicable to thisCourt, SCI [*6] LLC was able to discern that this Exhibitwas a check issued to Milton Greene, who had previouslychallenged his independent contractor status with theNew York State Division of Human Rights and the EqualEmployment Opportunity Commission. Dkt. Nos. 13-1,Resp't Mem. of Law at p. 1, 13-3, Peter Fidopiastis, Esq.,Decl., dated Feb. 25, 2013, at ¶¶ 3 & 7-12. Prior to thisrevelation, SCI LLC "had no idea why DOL wasinvestigating its global operation," Fidopiastis Decl. at ¶4, and now posits that "DOL's investigation and"unrestrained demands for information were prompted bya single complaint filed by Milton Greene," Resp't Mem.of Law at p. 1. Purportedly contributing to SCI LLC'ssuspicion that Mr. Greene is the sole impetus for thisinvestigation is DOL's letter to SCI LLC reminding it notto retaliate against Mr. Greene for the disclosure of hisname in this case. Dkt. No. 13-4, Pet'r Lt., dated Feb. 11,2013.2

1 Subcontracting Concepts Inc. is a separateentity that may have a business connection to SCILLC. That connection or relationship is to beexplored during the ordered deposition of SCILLC.2 Apparently, the impetus for DOL's Letter was


SCI LLC's General Counsel's Supplemental [*7]Declaration identifying Mr. Greene. Dkt. No.10-1, Peter Fidopiastis Supp. Decl., dated Feb. 7,2013, at ¶¶ 3-9. Because Mr. Greene wasidentified in this manner and concerned about itsoverarching implications, DOL felt compelled toremind SCI LLC that it would be unlawful toretaliate against "any employee." Dkt. No. 13-4,Ex. B at p. 1.

Based upon this newly unearthed revelation andDOL's Letter, SCI LLC extrapolates that Greene'scomplaint is the sole impetus for DOL's investigation,which would not justify the magnitude of DOL'sinvestigation and the unreasonable production ofdocuments demanded by DOL. SCI LLC exclaims thatDOL intentionally kept it and the Court "in the dark as tothe real basis for its investigation" and the Court was"deprived of this vital piece of evidence" which wouldhave critically impacted the ultimate decision rendered inthe MDO. Resp't Mem. of Law at pp. 5-6. With this inmind, SCI LLC argues that DOL's investigation isunreasonable and its investigative powers have beenstretched "well beyond justifiable end[s]," the orderedproduction is disproportionate to the "real" scope of theinvestigation, and that DOL has no evidence that SCILLC committed [*8] any violations of the Fair LaborStandard Act (hereinafter "FLSA"). See generally Resp'tMem. of Law. Lastly, SCI LLC seeks clarification of theMDO by asking that a more formal protective order beissued under these circumstances. Id.

In opposing the Motion for Reconsideration, DOLchides the entire premise of SCI LLC's Application,especially the propositions that this constitutes newlydiscovered evidence and that its investigation isconstrained solely to Mr. Greene or even to a singlecomplaint. See Dkt. No. 14, Pet'r Opp'n Mem. of Law,dated Mar. 5, 2013. DOL contends that the whollyredacted check, which contains no personal identifyinformation, "say[s] nothing about the origin of theSecretary's investigation" and, the proposition that thisinvestigation should be limited to only those transactionsinvolving Mr. Greene would "lead to an absurd result -the Secretary [of Labor] would be impeded frominvestigating widespread and pervasive employmentviolations where the trigger for the investigation was asingle complaint." Id. at pp. 3 & 4. For the Petitioner, thisMotion for Reconsideration constitutes nothing short of aruse to supplement SCI LLC's previously ineffectual

opposition [*9] to its Motion to Compel and poses as aveiled opportunity to re-argue relevancy, reasonableness,burden, and confidentiality, which were firmly rejectedby the Court. See generally Pet'r Mem. of Law. Thus,DOL seeks a denial of this Motion for Reconsideration.

ANALYSIS

Even if the Court was to deem Mr. Greene's identityand his role in this investigation as newly discoveredevidence, it could only serve as an abstract or obtuse facthaving no purposeful bearing on this Court's MDO. ThisCourt does not adopt SCI LLC's myopic view that asingle complaint legally impedes DOL from conducting amuch more broader investigation. To do so would requirea tremendous leap in logic. It is well established that anagency cannot conjure up an investigation and itsdemands to produce are subject to the limitations ofreasonableness, United States v. Constr. Prods. Research,Inc., 73 F.3d 464, 471 (2d Cir. 1996), but as long as theinvestigation is conducted pursuant to a legitimatepurpose, the information is not already within theagency's possession, and all required administrative stepshave been followed, a court's role in enforcing anadministrative subpoena is "extremely limited," E.E.O.C.v. United Parcel Serv., Inc., 587 F.3d 136, 139 (2d Cir.2009). [*10] Under the statute, DOL may investigate andgather information relative to wages, hours andconditions of employment in order to determine if therehas been a violation. See 29 U.S.C. §§ 209 & 211(a) &Mem.-Dec. & Order at pp. 3-4.

Even though DOL has not fully disclosed the genesisof its investigation, nor does it have to, SCI LLC knowsfull well, or should know, that this investigation does notrest solely on Mr. Greene's complaint alone. To argueotherwise is pure obfuscation. A better indicator as towhy DOL's investigation has encompassed SCI LLC isDOL's ongoing investigation regarding Zion DeliveryServices Inc., a client of SCI LLC. Dkt. No. 8, Attach. A,Peter Fidopiastis Decl., dated Dec. 31, 2012, & B, Cent.Dist. of California Case No. 12-09956, Order, dated Jan.15, 2013. SCI LLC is keenly aware that DOL brought apetition to enforce an administrative subpoena to obtaindrivers' 1099s completed by SCI LC, which wasultimately upheld by that district court. Dkt. No. 8, Ex. B,Order, dated Jan. 15, 2013. The California investigationwas not narrowly confined to a particular driver, andobviously neither is this investigation. Setting all of thataside, this Court already [*11] found that the purpose of


this investigation is to determine whether the independentowner operators are misclassified as independentcontractors rather than employees and whether SCI LLCand the logistic couriers may be joint employers of thesedrivers. Mem.-Dec. & Order at p. 9.

Newly discovered evidence must have somesignificance to conceivably change the outcome of theprevious ruling, but none can be found in thispresentation. In re Bernard L. Madoff Inv. Sec., LLC, 489F. App'x 519, 520 (2d Cir. 2012) (citing United States v.Int'l Bhd. of Teamsters, 247 F.3d 370, 392 (2d Cir. 2001).Mr. Greene's complaint and its impact as to thereasonableness of its scope is of minute importance.Hence this Court does not find that this is relevant newlydiscovered evidence nor would it require a modificationto the MDO, and accordingly there is no legitimate basisfor this Motion.

Continuing, because it deserves mentioning, thisCourt agrees with DOL that this Motion is uniquelydisguised as "a second bite of the apple" as SCI LLCreargues all of the issues that should have been fullyaddressed during the Motion to Compel. This is nothingmore than a veiled attempt to overcome previousdeficiencies [*12] and to plug in information SCI LLCfailed to reveal earlier. For example, when the Courtaddressed the matter of disclosing SCI LLC's client list, Inoted that "[o]ther than conclusory statements that itsclient list is 'carefully guarded,' SCI LLC has not outlinedin any detail how it protects or guards its customerlists[.]" Mem.-Dec. & Order at p. 19. Now, SCI LLCpresents Affidavits, especially one from Ryan Wise, theVice President of Information Technology, extolling ingraphic detail all of the steps SCI LLC takes to maintainthe confidentiality of its "proprietary" information. SeeDkt. No. 13-5, Ryan Wise Decl., dated Feb. 25, 2013.Additionally, SCI LLC's Memorandum of Law is merelya recapitulation of those same central themes it argued inopposing the Motion to Compel - relevance,reasonableness, burdensome, trade secrets, andconfidentiality.

These are belated submissions to overcome previousdeficiencies. This Court will not allow SCI LLC anotheropportunity to pitch its position, and the MDO and all ofits directions stand.

Nonetheless, this Court takes a moment to addressSCI LLC's complaint that if it has to adhere to the MDO,it will have to produce nearly 45,000 pages [*13] of

documents. Initially, the Court realized the DOL wasseeking an enormous amount of documents and, in thatrespect, I reduced the scope of the Subpoena and alsoimposed representative sampling in order to make thedemand for documents more reasonable. See generallyMem.-Dec. & Order. By directing that SCI LLC give adeposition within sixty days, it is conceivable that thesize of disclosure may be decreased. But this Court findsRyan Wise's Declaration very telling in terms of SCILLC's ability to produce these documents withoutunfairly intruding upon its twenty-two (22) employees.Both the client list and 1099 tax records are in both"physical form" and "in electronic form in [its] computerand server systems." Wise Decl. at ¶¶ 2-3. Although thisCourt may not share the technological sophistication ofMr. Wise, I certainly know that the amount of time, cost,and effort expended to produce these records from thecomputer is significantly less than by hand. With theadvent of software, predictive coding, spreadsheets, andsimilar advances, the time and cost to produce largereams of documents can be dramatically reduced.Further, suggesting to DOL to accept the production ofthese [*14] documents in either native format, or througha zip file, or some other electronic transaction shouldminimize SCI LLC's anxiety. Hence, the Court is moreconvinced than ever that SCI LLC is not subject to anoverwhelming and incomprehensible burden. See e.g.,E.E.O.C. v. Sterling Jewelers, Inc., 2011 U.S. Dist.LEXIS 126585, 2011 WL 5282622, at *5 (W.D.N.Y.Nov. 2, 2011) (noting that reviewing 54,000 personnelfiles was not burdensome nor would it disrupt thebusiness). Quite frankly, no manifest injustice has beendemonstrated.

Finally, SCI LLC's request for a more formalprotective order to ensure that all necessary measures aretaken by DOL to protect the names and addresses of itsclients is as baseless as it is superfluous. Throughout itsargument against the Motion to Compel, SCI LLC raisedthe alarm that DOL will reveal its proprietary informationand that its profitability would be impaired. To remindSCI LLC, the Court found its "fear that DOL intends toshare all of this information with SCI LLC's universe ofcompetitors [] utterly without basis." Mem.-Dec. & Orderat p. 16. Nonetheless, the Court directed DOL "not topublish, disclose nor reveal this list to any third partyoutside the context of any [*15] prospective litigation."Id. at p. 20. To suggest that this broadly stated protectiveorder is toothless because a penalty or consequence wasnot definitely and explicitly meted out should this federal


agency not abide by the Court's order is meritless. Partiesto litigation are expected to fully comprehend themagnitude of an order and the consequences to bear forfailing to abide by such an order. And to surmise that aparty will breach such an order is presumptuous andunfounded. SCI LLC has conveniently forgotten that thisCourt maintains jurisdiction over this application for theenforcement of a subpoena and clearly has the power ofcontempt when a party has failed to comply with such aclearly stated directive.3 Thus a more formal protectiveorder is not necessary.

3 The Court may turn to those inherent powers,which are innate to its creation, to impose respectfor its lawful mandates. United States v. Seltzer,227 F.3d 36, 39-42 (2d Cir. 2000); DLC Mgmt.Corp. v. Town of Hyde Park, 163 F.3d 124, 136(2d Cir. 1998). Federal courts have always had theinherent power to manage their own proceedingsand to control the conduct of those who mayappear before them, and when a party [*16] acts"in bad faith, vexatiously, wantonly, or foroppressive reasons," the courts may exercise their

discretion in fashioning a remedy. Chambers v.Nasco, Inc., 501 U.S. 32, 45-46, 111 S. Ct. 2123,115 L. Ed. 2d 27 (1991). Additionally, the Courtmay rely upon the authority granted in the FederalRules of Civil Procedure. FED. R. CIV. P.37(b)(2)(A)(vii) & 45(e).

Based upon the foregoing, it is hereby Ordered thatSCI LLC's Motion for Reconsideration, Dkt. No. 13, isdenied. To reiterate further, the entire MDO remains ineffect.

IT IS SO ORDERED.

March 11, 2013Albany, New York

/s/ Randolph F. Treece

Randolph F. Treece

U.S. Magistrate Judge.


CHEVRON CORPORATION, Plaintiff, -against- STEVEN DONZIGER et al.,Defendants.

11 Civ. 0691 (LAK)

UNITED STATES DISTRICT COURT FOR THE SOUTHERN DISTRICT OFNEW YORK


March 15, 2013, DecidedMarch 15, 2013, Filed

SUBSEQUENT HISTORY: Motion granted by, in part,Motion denied by, in part Chevron Corp. v. Donziger,2013 U.S. Dist. LEXIS 46381 (S.D.N.Y., Mar. 26, 2013)

PRIOR HISTORY: Chevron Corp. v. Donziger, 2013U.S. Dist. LEXIS 24086 (S.D.N.Y., Feb. 21, 2013)

COUNSEL: [*1] For Plaintiff: Randy M. Mastro,Andrea E. Neuman, Scott A. Edelman, Kristen L.Hendricks, William E. Thompson, GIBSON, DUNN &CRUTCHER, LLP.

For Patton Boggs LLP: James K. Leader, S. AlyssaYoung, LEADER & BERKON, LLP.

For the Donziger, Defendants: John W. Keker, Elliot R.Peters, KEKER & VAN NEST, LLP.

For Hugo Gerardo Camacho Naranjo and Javier PiaguajePayaguaje, Defendants: Julio C. Gomez, JULIO C.GOMEZ, ATTORNEY AT LAW LLC; Craig Smyser,Larry R. Veselka, Tyler G. Doyle, SMYSER KAPLAN& VESELKA, L.L.P.

JUDGES: Lewis A. Kaplan, United States DistrictJudge.

OPINION BY: Lewis A. Kaplan

OPINION

TABLE OF CONTENTS

SUMMARY

FACTS

I. THE LAGO AGRIO LITIGATION

A. THE BACKGROUND

B. THE LAWSUIT, THE TERMINATION OF JUDICIAL INSPECTIONS, THE

CABRERA APPOINTMENT, THE MAKING OF CRUDE, AND THE JUDGMENT

II. THIS LITIGATION

Page 1

III. EVIDENCE THAT THE JUDGMENT WAS OBTAINED BY FRAUD

A. ALLEGED BRIBING OF THE JUDGE

B. ADDITIONAL EVIDENCE THAT THE LAPS -- NOT THE JUDGE --

WROTE THE JUDGMENT

C. EVIDENCE OF FRAUD WITH RESPECT TO THE JUDICIAL INSPECTION PROCESS

1. DR. CALMBACHER

2. THE TERMINATION OF THE JUDICIAL INSPECTIONS, CABRERA'S

APPOINTMENT, AND THE CABRERA REPORT

3. NEW EVIDENCE OF FRAUD IN RESPECT OF THE CABRERA REPORT

4. THE "CLEANSING REPORTS"

IV. PB'S INVOLVEMENT

A. THE STRATUS SECTION 1782 PROCEEDING AND EVIDENCE OF FRAUD ON THE COURT

B. THE CLEANSING REPORTS

C. POST-TRIAL SUBMISSIONS TO THE LAGO AGRIO COURT

D. PB'S OTHER ACTIVITIES

IV. PROCEEDINGS WITH RESPECT TO THE PB SUBPOENA

A. PB'S MOTION TO QUASH

B. ARGUMENT AND THE INITIAL NARROWING OF THE SUBPOENA

V. CHEVRON'S UNSUCCESSFUL ATTEMPTS TO OBTAIN DISCOVERY FROM THE LAP

REPRESENTATIVES' AGENTS IN ECUADOR AND FROM THE DEFENDANTS

A. DEFENDANTS' REFUSAL TO PRODUCE DOCUMENTS AND EVIDENCE FROM ECUADOR

B. DEFENDANTS' REFUSAL TO COMPLY WITH THEIR OTHER DISCOVERY

OBLIGATIONS DISCUSSION

I. IN RE FRIEDMAN

A. ATTORNEY-CLIENT PRIVILEGE AND WORK PRODUCT

1. BASIC PRINCIPLES

(A) ATTORNEY-CLIENT PRIVILEGE

(B) WORK PRODUCT DOCTRINE

(C) THE CRIME-FRAUD EXCEPTION

2. APPLICATION IN THIS CASE

(A) IMPROBABILITY OF MANY RESPONSIVE ATTORNEY-CLIENT

COMMUNICATIONS, IF ANY

(B) CHEVRON'S SUBSTANTIAL NEED OVERCOMES ORDINARY WORK

PRODUCT PROTECTION

(C) THE CRIME-FRAUD EXCEPTION

3. FURTHER LIMITATION OF THE SUBPOENA

B. PB'S ROLE


(1) WHETHER DISCOVERY WOULD DISRUPT THE LITIGATION

(2) WHETHER PB LIKELY HAS RELEVANT EVIDENCE

C. THE NEED FOR DISCOVERY FROM PATTON BOGGS AND THE EXTENT OF

DISCOVERY ALREADY CONDUCTED

II. ALLEGED UNDUE BURDEN AND COST SHIFTING

CONCLUSION

LEWIS A. KAPLAN [*2] , District Judge.

An Ecuadorian court in 2011 entered an $18.2 billionjudgment (the "Judgment") against Chevron Corporation("Chevron")1 in an action brought by 47 individualsreferred to as the Lago Agrio Plaintiffs (the "LAPs"), twoof whom, the LAP Representatives, have appeared in thisaction.2 Chevron brought this action against the LAPs,their lead U.S. attorney, Steven Donziger and his lawoffices (the "Donziger Defendants"), and others involvedin the Lago Agrio Litigation,3 claiming among otherthings that the Judgment is the product of fraud and that itis a central part in a pattern of violations of the RacketeerInfluenced and Corrupt Organizations Act ("RICO") thathas included extortion, fraud, money laundering, andobstruction of justice, among other offenses.

1 DI 168 (Lago Agrio Judgment).2 The other LAPs have defaulted and are notdefending this action. DI 469.3 In addition, the amended complaint identifies anumber of alleged "co-conspirators" who are notnamed as defendants.

This matter is now before the Court on a dispute overnon-party discovery. Patton Boggs LLP ("PB") is a majorU.S. law firm that has provided services to the LAPssince early 2010 with respect [*3] to the Ecuadorianlitigation. It has represented the LAPs in much of theU.S. litigation relating to the Judgment, although notformally in the district court in this case. It has beeninvolved in the Ecuadorian litigation behind the scenes. Itis a named co-conspirator in this case. In addition, it hassued Chevron on its own behalf at least three times onclaims relating to this controversy.

Chevron served PB with a subpoena duces tecum(the "Subpoena") in this case. PB seeks to avoid entirelyany obligation to comply or, at least, to minimize thescope of any disclosure. Its position rests mainly on

claims of work product protection and attorney-clientprivilege, and on a contention that the requesteddiscovery would be unduly burdensome.

Summary

The Court begins with the proposition thatdepositions of opposing trial counsel are disfavored.4

Courts "have resisted the idea that lawyers shouldroutinely be subject to broad discovery."5 Nevertheless,"the disfavor with which the practice of seekingdiscovery from adversary counsel is regarded is not atalisman for the resolution of all controversies of thisnature."6 The Second Circuit has said that such efforts:

"require a flexible approach [*4] tolawyer depositions whereby the judicialofficer supervising discovery takes intoconsideration all of the relevant facts andcircumstances to determine whether theproposed deposition would entail aninappropriate burden or hardship. Suchconsiderations may include the need todepose the lawyer, the lawyer's role inconnection with the matter on whichdiscovery is sought and in relation to thepending litigation, the risk of encounteringprivilege and work-product issues, and theextent of discovery already conducted . . . .Under this approach, the fact that theproposed deponent is a lawyer does notautomatically insulate him or her from adeposition nor automatically require priorresort to alternative discovery devices, butit is a circumstance to be considered."7

4 See generally In re Subpoena Issued to DennisFriedman, 350 F.3d 65, 71 (2d Cir. 2003)(hereinafter In re Friedman).


5 Id. at 70.6 Id. at 71.7 Id. at 72.

These considerations again have led this Court toengage in a painstaking, step-by-step process to dealappropriately with a subpoena addressed to a LAPlawyer, a process that in this instance has consumedmonths.8 [*5] In the final analysis, it has reached theseconclusions.

8 In a decision affirmed on appeal, it previouslygranted and enforced a subpoena againstDonziger. See In re Chevron Corp., 749 F. Supp.2d 141 (S.D.N.Y. 2010) aff'd sub nom. LagoAgrio Plaintiffs v. Chevron Corp., 409 F. App'x393 (2d Cir. 2010).

First, Chevron has established at least probable causeto believe there was fraud or other criminal activity in theprocurement of the Judgment and in other respectsrelating to the Lago Agrio litigation in which thatJudgment was rendered and in certain litigations in theUnited States relating to the Ecuadorian litigation.Without alluding here to the entirety of its showing, thereis probable case to suspect, and often stronger evidence,9

that:

o Representatives of the LAPs bribed theEcuadorian judge to obtain the result theywanted and, as part of the deal, wrote theJudgment to which the judge put his name.Indeed, there is substantial evidencecorroborating this assertion, not least of itthe fact that significant parts of theJudgment match -- word-for-word --internal work product documents of theLAPs that never were publicly filed in theLago Agrio case. This latter evidencealone [*6] at least gives rise to probablecause to conclude that whoever wrote theJudgment had access to and copiednon-record materials that originated withthe LAPs.

o At an earlier stage of the LagoAgrio litigation, representatives of theLAPs coerced the then-presidingEcuadorian judge to terminate judicialinspections of alleged pollution sites, toreplace that process with a global expert

charged with making an independentevaluation, and to appoint the LAPs'candidate, Richard Stalin Cabrera Vega("Cabrera"), to that position. They did soby threatening him with a judicialmisconduct complaint if he did not accedeto their wishes.

o The report that Cabrera ultimatelysubmitted in fact was planned and written,at least in major part and quite possiblyentirely, by lawyers and consultantsretained on behalf of the LAPs though itwas signed by Cabrera and filed as if itwere his independent work. LAPrepresentatives, moreover, took a numberof steps to create or reinforce the entirelyinaccurate contention that the Cabrerareport was the unbiased work of anindependent expert when, in fact, it hadbeen the work of the LAPs' representativesthemselves and was not independent in theslightest [*7] respect.

o Once the improprieties surroundingCabrera began to come to light, the LAPsor their representatives then obstructedjustice and committed fraud in at least oneSection 1782 proceeding in the UnitedStates by submitting to a court in Coloradoa deceptive account of the LAPs'relationship with Cabrera.

o At a still earlier stage of the lawsuitin Ecuador, the LAPs filed two siteinspection reports with the trial court overthe signature of one of their experts thatthe expert neither adopted nor agreed with.The evidence readily gives rise to theinference that the LAP lawyers wrote thereports, affixed the expert's signature tothem in the knowledge that they did notreflect his views, and filed them.

9 Some of this evidence is undisputed. TheCourt already has determined on a motion forpartial summary judgment that there is no genuineissue of material fact as to much of it. SeeChevron v. Donziger, 886 F. Supp. 2d 235,


286-90 (S.D.N.Y. 2011).

The Court's second conclusion is this. Although thescope of the Subpoena has been limited dramatically inprior proceedings, described below, it will be limitedfurther to avoid any undue imposition on PB's position ascounsel, to further [*8] reduce any burden ofcompliance, and to minimize any genuine work productprotection and attorney-client privilege issues. It will belimited to documents pertaining to the subjects as towhich Chevron has established probable cause to suspectfraud or criminal activity. This limitation will greatlyreduce any legitimate claim of work product protection orattorney-client privilege because documents that relate tothe crime or fraud and would be discoverable providedonly that the documents were in furtherance of the crimeor fraud.

Third, Chevron has established substantial need forthe materials sought by the Subpoena as further narrowedhere. It cannot obtain their substantial equivalentelsewhere without undue hardship, if at all. The qualifiedprotection afforded to "ordinary" work product thereforehas been overcome. Documents that fall within theSubpoena as further narrowed here and allegedlyprotected from disclosure only by an ordinary workproduct claim are discoverable without regard to whetherthey furthered a crime or fraud.

The sum of these three conclusions thus will be thatPB must produce (1) all documents responsive to theSubpoena as finally narrowed -- that is, documents [*9]relating to the subjects as to which there is a probablecause to suspect crime or fraud -- as to which either (a)no claim of privilege or work product protection is made,or (b) the only claim of protection from disclosure is thatthe documents contain ordinary, i.e., non-opinion, workproduct, and (2) a privilege log as to responsivedocuments which PB claims did not further the suspectedcrimes or frauds.

One more preliminary comment is appropriate. It isimportant to recognize that the crime-fraud exception towork product protection and the attorney-client privilegeis established where there is "probable cause to believethat a fraud or crime has been committed [by someone]and that the communications in question were infurtherance of the fraud or crime."10 If probable causeexists as to the commission of a fraud or crime, it is notnecessary to show also that a lawyer from whomotherwise privileged or protected documents may be

sought was a culpable or knowing participant in the fraudor crime.11 It therefore is unnecessary to determine forpresent purposes whether there is probable cause tosuspect that PB or any of its personnel was a culpable orknowing participant in any [*10] alleged fraud or crime.

10 United States v. Jacobs, 117 F.3d 82, 87 (2dCir. 1997).11 See Clark v. United States, 289 U.S. 1, 15, 53S. Ct. 465, 77 L. Ed. 993 (1933) ("the loss of the[attorney client] privilege [does not] depend . . .upon proof that client and attorney are involved inequal guilt. The attorney may be innocent, andstill the guilty client must let the truth come out");In re Grand Jury Subpoena Duces Tecum DatedSept. 15, 1983, 731 F.2d 1032, 1038 (2d Cir.1984) (attorney client "communications areproperly excluded from the scope of the privilegeeven if the attorney is unaware that his advice issought in furtherance of . . . an improperpurpose.").

The balance of this opinion proceeds as follows.

Part I of the Facts is a basic description of thegeneral course of the Lago Agrio litigation, which isessential to providing the time line with respect to eventsin Ecuador and a framework for understanding therelevance of this and other litigation in the United States.Part II sets forth the essentials with respect to thislawsuit. Part III describes the evidence and makes thefinding that Chevron has established probable cause tosuspect fraud or criminal behavior in a number ofrespects and the scope [*11] of that probable causefinding. Part IV then addresses PB's role, to the extent itis known, with respect to both the Ecuadorian and U.S.litigation. This is relevant to understanding, among otherthings, the extent to which the policies that underliecourts' reluctance to subject lawyers to discovery actuallyapply here and the types, significance of, and need for,the documents that PB almost certainly has. With thatfactual predicate thus established, the Discussion sectionanalyzes the applicable law and comes to the ultimateconclusions described above.

Facts

I. The Lago Agrio Litigation

A. The Background


In 1993, a group of Ecuadorians brought a classaction in the Southern District of New York againstTexaco seeking billions of dollars in damages for harm asubsidiary caused during its oil explorations in Ecuador'sOriente region in the 1960s-1990s (the "AguindaAction").12 The case ultimately was dismissed on forumnon conveniens grounds.13

12 Republic of Ecuador v. ChevronTexacoCorp., 376 F. Supp. 2d 334 (S.D.N.Y. 2005).13 Aguinda v. Texaco, Inc., 142 F. Supp. 2d 534(S.D.N.Y. 2001).

On October 9, 2001, while the Aguinda Action waspending, a wholly owned subsidiary of Chevron merged[*12] with and into Texaco. Texaco was the survivingentity. Chevron became the owner of all of Texaco'scommon stock. Chevron did not acquire or assume any ofTexaco's assets or liabilities by merger.14

14 Chevron v. Donziger, 886 F. Supp. 2d at 243.

B. The Lawsuit, the Termination of Judicial Inspections,the Cabrera Appointment, the Making of Crude, and theJudgment

The Lago Agrio Litigation began in 2003 when theLAPs, represented by defendant Steven Donziger (albeitnot of record in the Ecuadorian courts) and other lawyers,sued Chevron in Ecuador under that country'sEnvironmental Management Act of 1999.15 The LAPsasserted, inter alia, claims for damages for allegedenvironmental harm said to have been caused by Texaco.

15 "In 1999 the [Republic of Ecuador] enactedthe Environmental Management Act of 1999("EMA"), which created a private right of actionfor Ecuadorians who have been individuallyaffected to seek damages related to environmentalharms to the community." Id. at 242.

It is unnecessary to detail every aspect of the longhistory of the Lago Agrio litigation. It suffices tohighlight only the aspects critical to this dispute.Beginning in 2004, the Lago Agrio court ordered judicial[*13] site inspections to "assess the approximately 122wells and production installations in the formerconcession granted by the Ecuadorian government towhat was called the PETROECUADOR-TEXACOConsortium."16 Each party selected experts to be presentduring the judicial inspections and to submit their

findings to a panel of "settling experts" that would"provide decisive opinions . . . [and] comment solely onthe reports presented by the experts appointed by theparties."17 Some of the inspections were completed,including two reports, which the LAPs submitted underthe signature of one of their experts, Dr. CharlesCalmbacher.18 By 2006, however, the LAPs asked theLago Agrio court to end the judicial inspection processand petitioned for the appointment of an independentexpert to perform a global assessment of the allegedenvironmental effects.19 The Lago Agrio court adoptedthat proposal, appointed Richard Stalin Cabrera Vega("Cabrera") to serve as the independent global expert, andcancelled most of the remaining inspections.20 As willappear below, these events and subsequent activities areamong several foci of this case.

16 Id. at 244 (internal quotation marks omitted).17 Id. (quoting [*14] Hendricks Decl. [DI 31] ¶198 & Ex. 121 pt.1 (Aug. 7, 2004 oral hearingsummary), at 1).18 Id.19 Id.20 Id.

While all of this was going on, Steven Donziger,then the key lawyer on the LAP side, approached filmmaker Joe Berlinger to create a documentary depictingthe Lago Agrio case from the perspective of his clients.For the next three years, Berlinger and his crewshadowed the plaintiffs' lawyers and filmed "the eventsand people surrounding the trial," compiling six hundredhours of raw footage.21

21 In re Application of Chevron Corp., 709 F.Supp. 2d 283, 287 (S.D.N.Y. 2010) (internalquotation marks omitted), aff'd sub nom. ChevronCorp. v. Berlinger, 629 F.3d 297 (2d Cir. 2011).

On April 1, 2008, Cabrera submitted what purportedto be his independent report which attributed $16.3billion in damages to Chevron.22 A supplement to thereport increased the damage assessment to $27.3billion.23

22 Chevron Corp. v. Donziger, 886 F. Supp. 2dat 244.23 Id.

In 2009, Berlinger released two versions of hispurported documentary, which is called Crude, one on


DVD and the other on Netflix. The Netflix versioncontained a scene or scenes not included on the DVDversion. As has been discussed in prior [*15] opinions, itrevealed or at least suggested that the LAPs and Donzigerhad collaborated with Cabrera. It suggested also that hisreport had been prepared by the LAPs and was notCabrera's independent work product.

This led to the filing by Chevron and others in April2010 of a Section 1782 action against Berlinger thatsought the video that Berlinger had shot that did notappear in the released films, which is referred to asouttakes.24 This Court granted that discovery in May2010 and the Second Circuit affirmed in January 2011.25

The pendency of the appeal, however, did not long delaythe release of most of the outtakes, as the Court ofAppeals in July 2010 directed that Berlinger comply inmajor part with this Court's disclosure order.26

24 The actions were filed on the Court'smiscellaneous docket, which at that time was notelectronic. The relevant docket entries appear inBerlinger's brief and appendix in the Court ofAppeals, Chevron Corp. v. Berlinger, 2d Cir. Nos.10-1918, 10-1919, at DI 196-98.25 See In re Application of Chevron Corp., 709F. Supp.2d 283; Chevron Corp. v. Berlinger, 629F.3d 297 (2d Cir. 2011).26 Chevron Corp. v. Berlinger, [*16] 2d Cir.Nos. 10-1918, 10-1919, DI 277.

Following the release of Crude and, in some cases,the outtakes, Chevron sought other discovery in theUnited States under 28 U.S.C. § 1782 relating to the LagoAgrio litigation and the Cabrera report.27 Based in parton evidence gathered through the Section 1782proceedings, Chevron argued that the Calmbacherreports, Cabrera's appointment, and the Cabrera reportwere fraudulent.28 In an effort to meet the mountingevidence of fraud, the LAP team hired new experts toaddress Cabrera's findings and submit new reports --referred to as "cleansing reports"-- to the Lago Agriocourt. On September 10, 2010, seven reports weresubmitted.

27 Chevron Corp. v. Donziger, 886 F. Supp. 2dat 245.28 Id.

The Lago Agrio court issued the Judgment onFebruary 14, 2011.29 It awarded $8.646 billion in

remediation damages and another $8.646 billion to bepaid unless Chevron issued a public apology within 15days. Chevron issued no apology.30

29 Id.30 Id. at 246.

Both the LAPs and Chevron appealed. The LAPssought additional damages, and Chevron sought to havethe Judgment reversed or declared a nullity on multiplegrounds, including fraud.31 The appellate court affirmedthe [*17] Judgment on January 3, 2012.32 It declined toaddress many of Chevron's allegations of fraud. Inaffirming the damages award, the appellate courtspecified that two trusts were to be set up and managedby defendant Amazon Defense Front ("ADF") -- one forthe $8.646 billion in remediation damages and the otherfor the $8.646 in punitive damages.33

31 Id.32 Id.33 Id. at 247.

Chevron sought clarification on several aspects ofthe appellate decision, including whether the appellatecourt had considered Chevron's claims that the Judgment"had been based on information foreign to the record"and that the Lago Agrio court "ha[d] received 'secretassistance' in drafting it."34 The court rejected what itdescribed as Chevron's "accusations" that "'the [J]ugmentha[d] been based on information foreign to the record'and that the Lago Agrio court 'had received "secretassistance" in drafting' it."35

34 Id.35 Id. (quoting Lago Agrio Judgment).

II. This Litigation

Chevron brought this action on February 1, 2011against the LAPs, the Donziger Defendants,36 the StratusDefendants,37 and a number of other individuals andentities. Chevron alleges among other things that theJudgment is the product of fraud and [*18] violations ofthe Racketeer Influenced and Corrupt Organizations Act("RICO"). The RICO and, to some extent, the fraudclaims rest on allegations that Donziger, a New Yorklawyer, and others based in the United States, hereconceived, substantially executed, largely funded, andsignificantly directed a scheme to extort38 and defraudChevron by, among other things, (1) bringing the Lago


Agrio case;39 (2) fabricating (principally in the UnitedStates) evidence for use in that lawsuit and corruptingand intimidating the Ecuadorian judiciary in order toobtain a tainted judgment;40 (3) exerting pressure onChevron to coerce it to pay money not only by means ofthe Lago Agrio litigation and judgment, but also bysubjecting Chevron to public attacks in the United Statesand elsewhere based on false and misleadingstatements;41 (4) inducing U.S. public officials toinvestigate Chevron;42 and (5) making false statements toU.S. courts and intimidating and tampering withwitnesses in U.S. court proceedings to cover up theirimproper activities.43 PB was named as a co-conspiratorbut not a defendant.44

36 The Donziger Defendants are StevenDonziger and his law firm, variously referred toas the [*19] Law Offices of Steven Donziger andDonziger & Associates, PLLC.37 The Stratus Defendants are StratusConsulting, Inc. ("Stratus"), the consulting firm,that allegedly wrote all or most of the Cabrerareport, and two of its personnel, Douglas Beltmanand Ann Maest.38 Amended complaint ("Cpt.") ¶ 1 (allegingthat defendants "sought to extort, defraud, andotherwise tortiously injure plaintiff Chevron bymeans of a plan they conceived and substantiallyexecuted in the United States."); id. ¶ 2 ("Theenterprise's ultimate aim is to create enoughpressure on Chevron in the United States to extortit into paying to stop the campaign against it.").39 Id. ¶ 3.40 E.g., id. ¶ 145 ("Back in the United States,preparations were well underway for draftingCabrera's report [the report of a supposedlyindependent, court-appointed expert]."); id. ¶ 151("While Stratus [LAP environmental consultant]was the primary coordinator of the . . . CabreraReport, other members of the [LAP's] U.S.-basedteam of experts . . . also contributed to the reportwithout attribution in the report or disclosure toChevron."); id. ¶¶ 353--56; Mastro Decl. [DI 746]Ex. C (hereinafter "Guerra Decl.").41 Cpt. ¶ 214.42 Id. ("And [*20] they have taken this pressurecampaign to U.S. state and federal agencies,seeking their falsely induced assistance in thisracketeering scheme."); id. ¶ 216.43 Id. ¶¶ 273-77, 291-300, 311-16.

44 Id. ¶ 18(s).

Several counts of Chevron's complaint have beendismissed by this Court on motions by the defendants.45

One count, which had been severed and became aseparate action, was dismissed at the direction of theCourt of Appeals (the "Count 9 Action").46 The principalcounts remaining are the RICO and state law fraudcounts.47

45 Chevron Corp. v. Donziger, 871 F. Supp. 2d229 (S.D.N.Y. 2012); DI 472 (May 24, 2012Decision on Stratus Defendants' Motion toDismiss).46 Count 9 of Chevron's complaint sought adeclaration that the Judgment was unenforceableand unrecognizable because it was tainted byfraud in a judicial system (Ecuador's) which couldnot afford a fair and impartial proceeding. TheCourt issued a preliminary injunction barringenforcement of the Judgment pendente lite inMarch 2011. Chevron Corp. v. Donziger, 768 F.Supp. 2d. 581 (S.D.N.Y. 2011). The SecondCircuit vacated the preliminary injunction andremanded with instructions to dismiss Count 9entirely on the ground that, [*21] in its view, "theprocedural device [Chevron] has chosen topresent those claims [in Count 9] is simplyunavailable: The [New York Recognition ofForeign Country money Judgments Act] nowhereauthorizes a court to declare a foreign judgmentunenforceable on the preemptive suit of a putativejudgment creditor." Chevron Corp. v. Naranjo,667 F.3d 232, 240 (2d Cir. 2012), cert. denied,133 S. Ct. 423, 184 L. Ed. 2d 288 (U.S. 2012).47 Also remaining are claims for tortiousinterference with contract, trespass to chattels, andconspiracy against defendants, and a claim thatthe Donziger Defendants violated Section 487 ofthe New York Judiciary Law.

III. Evidence that the Judgment Was Obtained by Fraud

Chevron has contended since the inception of thislawsuit that the Cabrera report was not written byCabrera but by the LAPs, who then foisted it upon theworld and a perhaps unsuspecting Ecuadorian judge asthe work of an independent, neutral expert. Soon after theJudgment came down, Chevron pointed to evidence that,it argued, suggested that portions of the Cabrera reportwere copied from the LAPs' internal documents. Most


recently, Chevron has come forward with evidence that,if credited, [*22] would establish that the Judgment waswritten by the LAPs, who bribed the Lago Agrio judge tosubmit it under his name.

A. Alleged Bribing of the Judge

On January 28, 2013, Chevron submitted adeclaration of Alberto Guerra Bastidas, a former judge ofthe Provincial Court of Sucumbios, Ecuador,48 the courtthat rendered the Judgment, and at one time the judgeassigned to the case against Chevron. In summary,Guerra states that the LAPs' counsel -- Pablo Fajardo andDonziger -- bribed the judge who signed the EcuadorianJudgment to obtain their desired result and, in fact,supplied him with the decision, which the LAPs hadwritten in all important respects.

48 Mr. Guerra was dismissed as a judge in 2008,ostensibly for private statements to the effect thatthe case against Chevron should be dismissed. Heavers, however, that the real reason for hisdismissal was his confrontation of two judgeswho later served on the Chevron case "regardingseveral dubious and illegal rulings that had issuedin the proceedings, and regarding their practice ofasking the settling experts for 25 percent of theirfees in consideration for having them appointed assuch." Guerra Decl. ¶ 6.

Turning to the details, [*23] Guerra's declaration,which is corroborated in some particulars by otherpublicly filed declarations,49 states that when JudgeNicolás Zambrano was assigned the Chevron case on thefirst of two occasions,50 he asked Guerra "to get in touchwith the attorneys for Chevron in order to negotiate anagreement by which the company would pay Mr.Zambrano and [Guerra] for issuing the final judgment inChevron's favor."51 Guerra obliged, but Chevron rejectedthe overture. So Judge Zambrano, who had told Guerrathat he already had reached an agreement with the LAPs'representatives "to quickly move the case along in theirfavor," suggested that Guerra meet with the LAPs'Ecuadorian lawyer, Fajardo.52 Guerra then met withFajardo. They discussed the fact that Guerra already hadan arrangement with Zambrano pursuant to which Guerrawrote Zambrano's decisions in civil cases. Guerra andFajardo then agreed that (1) Guerra would make theChevron case move quickly, (2) "Chevron's proceduraloptions would be limited by not granting their motions onalleged essential errors in rulings [Guerra] was to write,"

and (3) the LAPs' "representatives would pay [Guerra]approximately USD $1,000 per month for [*24] writingthe court rulings Mr. Zambrano was supposed to write."53

49 Mastro Decl. [DI 746] Ex. E (hereinafter"Callejas Decl."); id. Ex. F (hereinafter "RacinesDecl."); id. Ex. G (hereinafter "CampuzanoDecl."), id. Ex. H (hereinafter "Carvajal Decl.").50 Judge Zambrano first began presiding overthe case in September -- October 2009 when hereplaced Judge Nuáez. Stavers Decl. [DI 754] Ex.3102, at 4-6. He was replaced by Judge Ordoáezon March 12, 2010. Id. at 15. As will appear,Judge Zambrano later replaced Judge Ordoáez inOctober 2010. Id. at 18-19.51 Guerra Decl. ¶ 12.52 Id. ¶ 13.53 Id.

This arrangement is said to have continued untilJudge Zambrano was replaced on the Chevron case by aJudge Ordoáez.54 In time, however, a motion to recuseJudge Ordoáez in the Chevron case came before JudgeZambrano, which the latter allegedly "saw . . . as anopportunity to once again take control of the Chevroncase, and asked [Guerra] to help him write the courtruling sustaining Judge Ordoáez's disqualification fromthe case."55 Judge Zambrano allegedly "saw this as anopportunity to once again approach Chevron's attorneysto see if they were willing to pay to have the case decidedin their favor."56 [*25] He dispatched Guerra to explorethat possibility, but Chevron again rejected theoverture.57

54 Id. ¶ 20.55 Id. ¶ 21.56 Id.¶ 22.

As noted, Judge Zambranoreplaced Judge Ordoáez on theChevron case in the fall of 2010.Supra note 50; see also CarvajalDecl. ¶ 3.

57 Guerra Decl. ¶ 22.

At that point, Judge Zambrano, according to Guerra,"suggested and authorized [Guerra] to seek an agreementwith the Plaintiffs' representatives so that they couldobtain a verdict in their favor, in exchange for a payment


of at least USD $500,000 to Mr. Zambrano; and whateveramount [Guerra] could negotiate or agree to for [himself].The proposal entailed Plaintiffs writing a draft of thejudgment and Judge Zambrano signing it and issuing it ashis own."58 Guerra took the offer to Fajardo, whoexpressed interest and said that he would discuss it withDonziger. Later, Guerra received a call from Fajardo whoasked him to a meeting with himself, Donziger, and LuisYanza.59 At that meeting, Guerra summarized theproposal. Donziger replied that they, the LAPs, did notthen have that sum of money. Subsequently, however,Zambrano told Guerra that he had been in direct contactwith Fajardo and that "the Plaintiffs' representatives[*26] had agreed to pay him USD $500,000 fromwhatever money they were to collect on the judgment, inexchange for allowing them to write the judgment inPlaintiffs' favor."60 Zambrano told Guerra that he wouldshare the money with Guerra.61

58 Id. ¶ 23.59 Yanza is the co-founder of the AmazonDefense Front ("ADF"), a non-profit organizationthat purports to represent the LAPs, and thegeneral manager of Selva Viva, an Ecuadoriancompany that administers funds for the LagoAgrio litigation.60 Guerra Decl.¶ 23.61 Id.

Guerra then resumed his role as Judge Zambrano'sghostwriter. When it came to the final judgment,however, Guerra relates that his role changed somewhat.About two weeks before the Judgment was issued,"Zambrano gave [Guerra] a draft of the judgment [thathad been written by the attorneys for the plaintiffs anddelivered to Zambrano] so that [Guerra] could revise it."Zambrano asked Guerra "to work on the document tofine-tune and polish it so it would have a more legalframework."62 He did so at Zambrano's residence usingFajardo's computer.63 He made few changes, making "itseem more like a judgment issued by the Sucumbios [i.e.,Lago Agrio] court."64

62 Id. ¶ 25.63 Id.64 Id. ¶¶ 26--27.

When [*27] Guerra was through, he returned thedocument to Zambrano, which "was not too differentfrom the one the Plaintiffs had given him."65 Zambranotold him that the LAPs' lawyers "made changes to the

judgment up to the very last minute before it waspublished."66

65 Id. ¶ 27.66 Id. ¶ 28.

As noted, Guerra's account is corroborated in anumber of respects by other declarations that recentlywere filed, some publicly and two under seal. AndChevron has submitted further corroborating evidence,including drafts of nine of the twelve court orders thatJudge Zambrano67 issued during his tenure on theChevron case that were found on Guerra's computer68

and emails among Donziger and Fajardo and others thatappear to confirm the arrangement the LAPs had withGuerra.69

67 Judge Zambrano reportedly was removedfrom the bench after rendering the decisionagainst Chevron "for releasing a suspected drugtrafficker in an act of 'obvious negligence or aninexcusable mistake.'" Eduardo Garcia andAlexandria Valencia, Chevron Hopes to Benefitfrom Ecuador's Judge Dismissal, REUTERS(Mar. 9, 2012) (available athttp://www.reuters.com/article/2012/03/09/ecuador-chevron-idUSL2E8E9ETI20120309)(last visited Feb. 17, [*28] 2013).68 Mastro Decl. [DI 746] Ex. C (Guerra Decl.)& Attachment O, P, Q, R, S, T, U, V, W.69 Two emails appear strongly to corroborateGuerra's assertion that during Judge Zambrano'sfirst tenure as the presiding judge of the LagoAgrio case (September 2009 through March 12,2010), Judge Zambrano had an arrangement withFajardo "to quickly move the case along in [theLAPs'] favor," that Judge Zambrano suggestedthat Guerra -- who wrote his decisions in civilcases -- meet with Fajardo, and that Guerra andFajardo the agreed that Guerra would move thecase quickly, that he would limit Chevron'sprocedural options, and that the LAPs'representatives would pay Guerra $1,000 permonth.

On September 15, 2009, Fajardo wrote toDonziger and others: "The puppeteer [allegedlyreferring to Guerra] is pulling the string and thepuppet [Judge Zambrano] is returning the package. . . . By now it's pretty safe that there won't beanything to worry about . . . ." Stavers Decl. [DI


754] Ex. 3140. A month later, Fajardo wrote toDonziger and Yanza that "[t]he puppeteer won'tmove his puppet until the audience doesn't payhim something . . . ." Id. Ex. 3154. Deposit slipsin Guerra's bank records show that [*29] $1,000was deposited into Guerra's account a month laterby someone whom Chevron identifies as a LAPemployee. Guerra Decl. ¶ 14 & Attachment K, L,M, & N; DI 752, Rule 56.1 Statement ¶ 205.Chevron has submitted other bank recordsshowing further $1,000 deposits made intoGuerra's account by the same employee.

The Guerra affidavit, if it ultimately is credited,establishes that the Judgment was fraudulently obtainedby the LAPs. Even before the Guerra affidavit was filed,however, Chevron had presented substantial evidence offraud in the procurement of the Judgment. More recently,it has submitted still more.

B. Additional Evidence that the LAPs -- Not the Judge --Wrote the Judgment

On July 31, 2012, the Court granted in part anddenied in part Chevron's motion for partial summaryjudgment seeking dismissal of the defenses of collateralestoppel and res judicata.70 In that opinion, the Courtheld that there was no genuine issue of material fact thatfraud tainted various aspects of the Lago Agrio litigation.Chevron subsequently has submitted additional evidencethat further suggests that the Judgment was obtained byfraud.

70 Chevron Corp. v. Donziger, 886 F. Supp. 2d235.

In its motion [*30] for partial summary judgment,Chevron's "ghostwriting" allegations related principallyto three internal LAP documents -- that is, documents thatwere not part of the court record but parts of whichwound up in the Judgment:

o A document entitled The Merger ofChevron Inc. And Texaco Inc. (the"Unfiled Fusion Memo") that was writtenby one or more members of the LAP teamand that addresses, among other things,successor liability.71

o The Index Summaries --spreadsheets prepared by the LAP team

that list and summarize documents filed inthe Lago Agrio court.

o The Selva Viva Data Compilation,which "consists of spreadsheets containingenvironmental sampling data."72

Chevron submitted analyses by experts that establishedthat portions of each of these documents appearedverbatim in the Judgment.73 Thus, Chevron contendedthat (1) relevant portions of the Judgment either werewritten by the LAPs, who had generated the documents,and passed to the judge ex parte or (2) these documentsthemselves were provided ex parte to the Lago Agriocourt, which copied at least parts of them word-for-wordinto the Judgment.

71 Id. at 253.72 Id.73 Id. at 253-54.

There was no evidence disputing the fact that parts[*31] of the Judgment were virtually identical to theUnfiled Fusion Memo. This Court therefore held that thisvirtual identity "demonstrate[d] as a matter of law thatwhoever wrote the Judgment had access to and copiedportions of the [Unfiled Fusion] Memo."74 The Court,however, concluded that "[t]he identity of languagebetween parts of the Unfiled Fusion Memo and parts ofthe Judgment -- troublesome as it is -- d[id] not alonewarrant the conclusion that Chevron has established thatthere is no genuine issue of material fact on this issue."75

With respect to the Index Summaries and Selva VivaData Compilation, the Court held that "[w]hile thesimilarities between those documents and aspects of theJudgment supports the premise that the author or authorsof the Judgment had access to and copied them, thenature of their contents makes it more difficult toconclude that any ex parte submission of them materiallyimpacted Chevron's ability to present its defense."76

74 Id. at 286.75 Id. at 288.76 Id. at 287 n.319.

Chevron now has submitted additional evidence that,it claims, shows that at least three other portions of theJudgment are identical to other internal LAP documentsthat never were [*32] part of the Ecuadorian courtrecord:


o A memorandum by a Donziger intern(the "Moodie Memorandum"), whichanalyzes causation under a doctrineapplied in California asbestos litigation.

o An analysis entitled La ExplotacionDe Petroleo En La Zona Conesionada ATexaco Y Sus Impactos En La Salud DeLas Personas, which was written by LAPconsultant Richard Clapp (the "ClappReport").

o An email regarding trusts sent byFajardo to alleged co-conspirators on June18, 2009(the "Fajardo Trust Email").

Chevron supports its contention with respect to eachdocument by expert analyses that conclude that (1) thelanguage or analysis in the relevant LAP internaldocument and the Judgment is nearly identical, and (2)the analysis or language is found nowhere else in thecourt record.77 In each case, Chevron's evidence issufficient to warrant the conclusion that the items inquestion further support the existence of probable causeto suspect that the LAPs, not the judge, wrote the relevantpart of the Judgment.78

77 See, e.g., Mastro Decl. [DI 658], Ex. 3003(Green Decl.), at 2, ¶ 12 ("it is highly unlikely,perhaps even more so than highly, that thepassages in the [judgment regarding causation]were [*33] prepared independently of the Moodiememorandum.").78 Moodie Memorandum. Chevron's expertexplains that both the Moodie Memorandum andthe Judgment analyze "causation law of theUnited States and Australia, not Ecuador. . . .Absent some explanation, [he] would consider ithighly unlikely that an Ecuadorian court wouldindependently choose United States andAustralian causation law, rather than Ecuadoriancausation law." Mastro Decl. [DI 658], Ex. 3003(Green Decl.), at 4, ¶ 13. Moreover, the expertopines that the causation analysis in both theJudgment and the memorandum was incorrectunder U.S. law and, in any case, inapplicable tothe Lago Agrio case. While the LAPs argue thatthe expert's work was not thorough and would notbe admissible in its present form at trial, it affordsa sufficient factual basis to support the finding of

probable cause given the standard applicable here-- probable cause to suspect a crime or fraud.

Clapp Report. Chevron has submittedevidence showing that portions of the ClappReport were included in an annex to the CabreraReport. Mastro Decl. [DI 658], Ex. 3005 (LeonardReport) at 9-10, 13. Emails between Donziger anda Stratus employee make clear that [*34] theysought to keep the Clapp Report's true authorshipsecret, presumably to create the appearance thatCabrera had written it. See Hendricks Decl. [DI34], Ex. 188 (Email from Doug Beltman atStratus to Donziger stating "We have to talk toClapp about that 5-pager, and how we have tolimit its distribution. It CANNOT go into theCongressional Record as being authored byhim."). The annex to the Cabrera report, however,"did not contain the entirety of the Clapp Report .. . , and portions that were not included appear inthe judgment." DI 657, at 7; Mastro Decl. [DI658], Ex. 3005 (Leonard Report) at 33-34.Chevron's expert notes that this includes a34-word string which appears both in the ClappReport and the Judgment, as well as twoadditional 16-word overlaps. Ex. 3006 (JuolaReport) at ¶¶ 23, 24. The portions of the ClappReport that appear in the Judgment appearnowhere else in the Lago Agrio court record. Id.at ¶ 29.

The LAPs do not dispute that sections of theClapp Report appear verbatim in the Judgmentand are nowhere else in the Lago Agrio courtrecord. They argue only that "Chevron has notproved that the materials [including the ClappReport] were not provided to the [*35] Judge insome informal manner such that the materials didnot get signed and numbered as part of therecord." DI 712, at 5. But the LAPs point to noevidence that the Clapp Report ever wassubmitted to the judge in a manner that did notresult in its being part of the court record muchless that any such submission would not havebeen improper. Given the identity of language inparts of the Clapp Report and the Judgment andthe absence of that language anywhere in thecourt record, this too supports the conclusion thatthere is probable cause to suspect fraud.


Fajardo Trust Email. The same can be saidfor the Fajardo Trust email. On June 18, 2009,Fajardo sent an email to Donziger and othersconcerning Ecuadorian trust law and the Andradev. Conolec case. DI 401, Ex. 2174. Portions ofthat email -- including a misquotation of theAndrade case -- appear in the Judgment. TheCourt noted in its opinion on Chevron's motionfor partial summary judgment that "Chevron hassubmitted no expert reports documenting allegedplagiarism in the Judgment from the Fajardoemail, or indicating whether or not the Fajardoemail was or was not a part of the Lago Agriocourt record." Chevron v. Donziger, 886 F. Supp.2d at 254 n.116. [*36] It recounted the LAPs'contention that any language that was common tothe Fajardo Trust Email and the Judgment wasmerely "stock language" that could have beenfound4independently in the court record. Id. TheCourt therefore declined to find that there was nogenuine issue of material fact that the FajardoTrust Email had been submitted fraudulently tothe court.

Chevron's new evidence overcomes thesedeficiencies, at least for present purposes.Chevron's expert now concludes that "parts of the[Judgment] must likely have had their origin inthe unfiled Fajardo Trust email." Mastro Decl. [DI658] Ex. 3005(Leonard Decl.), at 33. Thisincludes identical word strings which are foundnowhere else in the court record, id. at 30, andimproper reliance on Andrade, which both theFajardo Trust email and the Judgment incorrectlycite as dealing with "the legal basis of the trust,"although the case itself says nothing about trusts.Id. at 31. Furthermore, following a completereview of the Lago Agrio court record, Chevron'sexpert now has concluded that the Fajardo TrustEmail is found nowhere within it. Hernandez Aff.[DI 548] ¶ 26. The LAPs do not attempt toexplain how the language from [*37] the FajardoTrust email ended up in the Judgment despite thefact that it was never part of the Lago Agrio courtrecord. Thus, the Fajordo Trust Email furthersupports the probable cause finding.

C. Evidence of Fraud With Respect to the JudicialInspection Process

1. Dr. Calmbacher

As indicated above, Dr. Charles Calmbacher was theLAPs' expert for some of the early judicial inspections.Indeed, two reports were filed with the Lago Agrio courtunder his name and with his signature. Those two reports,however, contained conclusions and findings that he latertestified he did not reach.79 The Court held on summaryjudgment that "the evidence is uncontradicted" that"[w]hile his signatures on the reports were genuine, thetext associated with them was not."80 The Court thereforeconcluded that "[i]t . . . is at least arguable that Chevronhas established from Dr. Calmbacher's uncontradictedtestimony . . . that the LAP lawyers wrote the reportssubmitted over Dr. Calmbacher's signatures and thatCalmbacher did not in fact hold the views there stated."81

But the Court declined to decide on summary judgmentwhether the submission of the purported Calmbacherreport was fraudulent. The evidence, [*38] however,warrants a finding of probable cause, which the Courtmakes.

79 Chevron Corp. v. Donziger, 886 F. Supp. 2dat 288.80 Id.81 Id.

2. The Termination of the Judicial Inspections, Cabrera'sAppointment, and the Cabrera Report

As noted, the LAPs applied to the Lago Agrio courtto terminate the judicial inspections that it had ordered.While that application was pending, Donziger drafted amisconduct complaint against the judge to whom the casethen was assigned. The complaint accused the judge of"trading jobs for sex in the court."82 Pablo Fajardo, theLAPs' principal Ecuadorian lawyer, then had ex partemeetings with the judge, during which he (1) attempted topersuade the judge to appoint Cabrera as the globalexpert, which ultimately occurred on March 19, 2007,and (2) "let [the judge] know [the LAPs] might file it[i.e., the misconduct complaint] if he does not adhere tothe law and what we need."83

82 Id. (internal quotation marks omitted).83 Id. (internal quotation marks omitted).

The Court held in its summary judgment opinion thatthere was no genuine issue of material fact that "thedecisions to terminate judicial inspections, to pursue theglobal assessment, and to select Cabrera [*39] as the


global expert were tainted by the duress and coercionapplied to [the judge] by Donziger, Fajardo, and perhapsother."84 The Court held also that there was no genuineissue of fact that Cabrera's

"report was not entirely or evenpredominantly his own work or that of anyassistants or consultants working only forhim. There is no genuine issue withrespect to the facts that the LAP teamsecretly prepared his [i.e., Cabrera's] workplan, worked closely with him in carryingit out, and drafted most of the report andits annexes. Nor is there any genuine issueregarding the fact that the LAP team thenpublicly objected to the very report thatthey, in large part, secretly had drafted . . .. The answers filed by Cabrera in responseto the LAPs' (and Chevron's) objections --like the report itself -- were drafted at leastin substantial part by the LAP team andwritten to read as if Cabrera had writtenthem."85

It concluded as well that, while the Lago Agrio court"disclaimed reliance" on the Cabrera report, Chevron hadpresented evidence from which it could be concluded thatthe Lago Agrio court had in fact relied on the Cabrerareport at least for certain determinations.86

84 Id. at 289.85 Id.86 Id. at 289-90.

3. [*40] New Evidence of Fraud in Respect of theCabrera Report

Subsequent to the partial summary judgmentdecision, Chevron submitted a declaration of RamiroFernando Reyes Cisneros ("Reyes"),87 an Ecuadorianpetroleum and environmental engineer,88 that supportsthe conclusion that the object of terminating the originaljudicial inspections and appointing a single global expertwas to obtain the appointment of someone chosen by theLAPs who would "play ball" with them.

87 Id. ¶ 3.88 See Mastro Decl. [DI 658] Ex. 3014 (ReyesDecl.).

Reyes' declaration states that he was asked by the

LAPs' attorneys, before the termination of the judicialinspections and the appointment of Cabrera as the globalexpert, to serve as an independent expert to "monitor" thesettling experts in the Lago Agrio case. Donzigerexplained to Reyes at the time that he was unhappy withthe reports plaintiffs' then-experts had submitted andwanted Reyes to submit to the Lago Agrio court anotherreport that established "that the findings of the settlingexperts' report . . . were wrong."89 Reyes makes clear thatDonziger and the other attorneys aimed to keep theirrelationship with Reyes secret so that his report wouldappear to [*41] be "independent."90 He claims that,despite Donziger's wishes, he believed that "the evidencedid not support Mr. Donziger's position and [he] couldnot twist [his] professional assessments to make them fitthe plaintiffs' interests."91 Accordingly, in the reportReyes prepared and showed to Donziger, he concludedthat, while "the settling experts had failed to strictlyfollow their judicial mandate . . . their report containedenough information for the Court to make its ownruling."92 Donziger was dissatisfied with Reyes' reportand did not ask him to submit it to the court.93

89 Id. ¶ 20.90 Id. ¶ 15.91 Id. ¶ 20.92 Id.93 Id.

In 2006, after the Lago Agrio court halted thejudicial inspections, Donziger and the other LAP lawyersinformed Reyes that they wanted him to serve as theglobal "court-appointed" expert94 and that he would"need . . . to state that Chevron was the only partyresponsible for environmental damages and the harm tothe local community."95 When Donziger later told Reyesthat the "judge . . . was putting up hurdles to [Reyes']appointment as expert," Reyes and Donziger discussedappointing Cabrera instead.96 At a later meeting withCabrera and certain LAP attorneys which [*42] Reyesattended, "Mr. Fajado, Mr. Yanza and Mr. Donzigerdropped any pretense that Mr. Cabrera would actindependently in writing an expert report that would betechnically sound and executed according to professionalstandards. On the contrary, it was obvious that theplaintiffs had already predetermined the findings of theglobal assessment, that they themselves would write areport that would support their claim . . . and wouldsimply put Mr. Cabrera's name on it."97


94 Id. ¶ 22.95 Id. ¶ 25.96 Id. ¶¶ 29-30.97 Id. ¶ 35.

4. The "Cleansing Reports"

As the evidence of infirmities in and affecting theCabrera fraud were coming to light, the LAP teamprivately began to "acknowledge[] problems associatedwith Cabrera's lack of independence."98 The LAPs'lawyers' planned to hire a new expert to address Cabrera'sfindings and "submit an additional expert report to theLago Agrio court that would appear to be independent ofbut, in fact, would rely on the data and conclusionsreached in the Cabrera report."99 One of the LAPs'lawyers explained that, with the cleansing report:

"The path for an Ecuadorian decisionwill be simple. We would hope the judgewould say/rule: There has been muchcontroversy [*43] surrounding theCabrera report, and objections to it.[Perhaps: The court did not anticipate thatthere was the degree of collaborationbetween plaintiffs' counsel and Cabrera,that they may have been. Given theseissues, the court is not relying on Cabrerafor its ruling.] However, the Court nowhas additional submissions from theparties . . . The court finds the new report(demonstrating damages of -- billion) to bepersuasive, reliable and accurate andtherefore rules . . . ."100

98 Chevron Corp. v. Donziger, 886 F. Supp. 2dat 261.99 Id.100 Champion Decl. [DI 400] Ex. 2060 (June14, 2010 email).

Ultimately, the LAPs petitioned the Lago Agriocourt to allow the parties to submit "supplementaryinformation to aid th[e] Court in the process of assessingthe global damages."101 The court granted the LAPs'request and the LAPs submitted seven new reports to theLago Agrio court from newly hired experts.102 The newexperts had been instructed to rely on the Cabrera report,but were not told that it had been written by Stratus and

the LAP team rather than Cabrera,103 even though, in thewords of Donziger, the LAP team's "general idea" hadbeen that "Stratus would draft the report in a form [*44]that could be submitted directly to the Ecuadorian courtby Mr. Cabrera."104 "[T]he known [new] experts thatsubmitted reports later admitted that they never hadtraveled to Ecuador for the purpose of gathering data tosupport their reports. At least four relied on the data andconclusions in the Cabrera report."105 For example, oneof the purportedly independent cleansing experts lateradmitted that he had "relied on parts of the Cabrerareport" and "made no efforts to independently verifyunderlying data."106 Another expert said he relied on"data series and cost figures" from the Cabrera reportwithout "know[ing] one way or the other whether they'recorrect or not."107

101 Stavers Decl. [DI 549] Ex. 2420 (Fajardodeclaration submitted to Lago Agrio court (the"Fajardo Lago Agrio Decl.")), at 2.102 Chevron Corp. v. Donziger, 886 F. Supp. 2dat 261.103 See e.g., Stavers Decl. [DI 549] Ex. 2417(Shefftz deposition transcript), at 68:14-24(cleansing expert was told "that the Cabrera reportwas prepared by an independent [and neutral]expert").104 Hendricks Decl. [DI 8] Ex. 6 (Donzigerdeposition transcript), at 2253:5-11.105 Chevron Corp. v. Donziger, 886 F. Supp. 2dat 261.106 Champion Decl. [*45] [DI 400] Ex. 2073(Allen deposition transcript), at 171:18-172:3.107 Champion Decl. [DI 400] Ex. 2074 (Shefftzdeposition transcript), at 63:3-21.

In the Judgment, the Lago Agrio court disclaimedreliance on the Cabrera report. The Judgment, however,appears to reveal that the court in fact relied upon it by itsreliance on the cleansing reports. For example, inassessing damages, the Lago Agrio court cited acleansing report that "contain[ed] no damage assessmentindependent of that in the Cabrera report."108 Thus, thisCourt concluded on the partial summary judgmentmotion that "[t]he uncontradicted evidence . . . shows thatthe Cabrera report was tainted and that the Lago Agriocourt relied to some extent on that report, both directlyand via its reliance on the [cleansing] report."109

108 Chevron Corp. v. Donziger, 886 F. Supp. 2d


at 290.109 Id.

* * *

Although this Court concluded that summaryjudgment was not warranted with respect to certain of theincidents described above -- e.g., the Calmbacher reportand the preparation of the Judgment -- i.e., that thestandard relevant to the crime-fraud exception to thework product doctrine and the attorney client privilege isless demanding. [*46] The privilege is overcome, inrelevant part, by a showing of probable cause -- "aprudent person ha[s] a reasonable basis to suspect theperpetration or attempted perpetration of a crime orfraud."110

110 In re Grand Jury Subpoena Duces TecumDated Sept. 15, 1934, 731 F.2d at 1039.

This record establishes probable cause to suspect,taking the matters essentially in chronological order, that(1) the LAPs wrote the Calmbacher reports that werefiled with the Lago Agrio court and attachedCalmbacher's signatures to them, knowing that thereports did not reflect his views, (2) the judicialinspection process was terminated, the global expertproposal adopted, and Cabrera selected as the globalexpert as a result of the LAPs' threat that they would filea misconduct complaint against the judge if he did notaccede to their wishes that he take these actions, (3) theLAPs secretly planned and wrote all or at least the greatmajority of Cabrera's report, were complicit in itspresentation to the Lago Agrio court as Cabrera'sindependent work, and took other steps to bolster thefalse pretense that the report had been independent, (5)the LAPs entered into an improper relationship withJudge Zambrano [*47] during his first tenure as thepresiding judge pursuant to which Judge Zambranoagreed "to quickly move the case along in their favor,"(6) and the LAPs then entered into a supplementary andequally improper relationship with Guerra pursuant towhich Guerra agreed to move the case quickly and limitChevron's procedural options "by not granting theirmotions on alleged essential errors in rulings [Guerra]was to write, in exchange for payment by the LAPs'representatives of "approximately USD $1,000 per monthfor writing the court rulings Mr. Zambrano was supposedto write." In addition, there is probable cause also tosuspect that LAP lawyers and other representatives laterbribed Judge Zambrano to obtain the result they wanted

and, pursuant to the arrangement they struck with him,actually wrote the decision to which he signed his namedafter some cosmetic and inconsequential editing byGuerra.

IV. PB's Involvement

PB argues that requiring it to produce documentswould be inappropriate because, among other things, itwas not retained until February 2010 and it neitherparticipated in nor has much if any direct knowledge ofthe preceding events. But that is too facile and misleading[*48] a contention. Careful consideration of PB's rolewith respect to the Lago Agrio litigation, the Judgment,and certain other events, shows that PB participatedheavily in certain critical activities that make it likely thatit is an important and, in many respects, unique source ofevidence of the alleged fraud that is available nowhereelse and that at least some of the materials in itspossession or control were in furtherance of crimes orfrauds regardless of whether PB was aware of them.

The period from PB's initial involvement, whichapparently began in February 2010, through the filing ofthe Lago Agrio court's Judgment in February 2011 isparticularly telling. It is important to understand wherematters stood when PB came on the scene and then tofocus on three of PB's activities that went on, in varyingdegrees, at the same time during this period: (1) PB'sefforts to assist in preventing Chevron from obtainingdiscovery from Stratus in a Section 1782 proceeding ithad begun in December 2009 -- discovery that eventuallyoccurred and made clear that Stratus had had extensivecontact with Cabrera and substantially written theCabrera report, thus destroying or at least badlyundermining [*49] the pretense that he had been anindependent expert; (2) PB's role in recruiting andorchestrating the work of the so-called "cleansingexperts," whose reports were submitted to the Lago Agriocourt in order to provide a basis for a decision favorableto the LAPs; and (3) PB's role in drafting the finalalegato -- the closing argument or closing briefs --submitted on behalf of the LAPs to the Lago Agrio court.

A. The Stratus Section 1782 Proceeding and Evidence ofFraud on the Court

In December 2009, Chevron brought a Section 1782proceeding against Stratus and related individuals in theDistrict of Colorado.111 Chevron argued that discoverywas appropriate because similarities between the Cabrera


report and documents created by Stratus and itsemployees suggested that Stratus had written all or atleast part of the Cabrera report. It contended that it wasentitled to discovery to determine the degree to whichthat in fact was so as well as the LAPs' involvement inthe process.112 That application was pending when PBarrived on the LAP scene in February 2010.

111 Chevron v. Stratus Consulting, Inc., No. 10Civ. 00047 (D. Colo.).112 Id. DI 2 (Chevron Mem. of Points andAuthorities in Support [*50] of Petition).

The District of Colorado granted Chevron's Section1782 application on March 4, 2010.113 The LAPs,realizing that production from Stratus was virtuallyinevitable, were anxious to "minimize the effects" of thecourt-ordered production of Stratus' documents.114 In anemail to Donziger, Fajardo, and others, Julio Prieto, oneof the LAPs' Ecuadorian lawyers wrote:

"Today Pablo [Fajardo] and Luis[Yanza] were kind enough to tell us whatwas going on in Denver, and the fact thatALL will be made public, includingcorrespondence . . . . Apparently this isnormal in the U.S. and there is no riskthere, but the problem, my friend, is thatthe effects are potentially devastating inEcuador (apart from destroying theproceeding, all of us, your attorneys,might go to jail), and we are not willing tominimize our concern and to sit to wait forwhatever happens. For us it is NOTacceptable for the correspondence, thee-mails, between Stratus and Juanpa[Saenz] and myself to be divulged."115

A month later, the LAPs filed a motion for a protectiveorder with the District of Colorado, claiming that thedocuments and testimony that Stratus had been ordered toproduce were protected from [*51] disclosure by theattorney-client privilege and work product protection.116

The motion was later supported by a declaration of PabloFajardo (the "Fajardo Declaration"), the LAPs' leadEcuadorian counsel.117

113 Id. DI 22.114 Hendricks Decl. [DI 9], Ex. 11 (March 30,2010 Email from Julio Prieto to Donziger,

Fajardo and Others).115 Id. (all emphasis added).116 Chevron Corp. v. Stratus Consulting, Inc.,No. 10 Civ. 00047, DI 68.117 Id. DI 99 (filed May 5, 2010).

PB was heavily involved in drafting the FajardoDeclaration118 which, in the words of one PB partner,Eric Westenberger, was intended to "'cleanse anyperceived impropriety related to the Cabrera Report."119

The LAPs' American lawyers debated what the affidavitshould reveal and whether Fajardo should be the one tosign it. When PB circulated a draft of the affidavit onMay 3, 2010, one lawyer from Emery Celli Brincherhoff& Abady, LLC, which also represented the LAPs,responded:

"I don't quite get the purpose of thisaffidavit. Pablo mentions one documentsubmission but not the other. If he'ssubmitting an affidavit about whathappened, why omit the most importantpart? It seems misleading at best. I justdon't see how he can sign [*52] an aff.that documents his submissions toCabrera without mentioning that he sentdocuments that originated from Stratuswhich is the one thing the judge is going towant to know about. . . . [And] I wouldn'temphasize too much that Cabrera wasindependent and court-appointed. Once[Fajardo] says that in an American court,we'll never be able to back off from it."120

Westenberger expressed his concern that Fajardo might"be subject to deposition[.] This is why we struggled withwho would sign the declaration. If Steve [Donziger]signs, he will most certainly be deposed. Same for anyother counsel in the US. We figured that with [Fajardo],they likely would not slow down the process by deposinghim."121

118 See Stavers Decl. [DI 549] Ex. 2407 (May 3,2010 email from PB attorney to others attaching"a draft of Pablo Fajardo Mendoza's Declarationin support of our motion to be filed in Denver.").119 Hendricks Decl. [DI 9] Ex. 13 (email fromWestenberger to others).120 Stavers Decl. [DI 549] Ex. 2407 (May 3,2010 email from Ilann Maazel).


121 Id. (May 3, 2010 email from Westenberger).

The Fajardo Declaration that ultimately was filedgave a bland description of the process by which thejudicial [*53] inspections had been terminated, theglobal expert proposal adopted, and Cabrera in particularselected. But it failed to mention that Fajardo andDonziger had threatened the judge with a misconductcomplaint unless the judge agreed to their demands andappointed Cabrera. And while it acknowledged that theLAPs had "delivered materials to Mr. Cabrera,"122 it didnot mention the March 3, 2007 meeting at which theLAPs laid out the plan for Cabrera's report and indicatedthat the work would be done by them.123 Nor did it revealthat Stratus and the LAPs' counsel in fact had written allor most of Cabrera's report.124 In other words, it omittedwhat the Emery Celli lawyer said was "the mostimportant part" -- that Fajardo "sent documents thatoriginated from Stratus." The declaration similarlyneglected to report that the LAPS "chang[ed] the focus of[Cabrera's] data at [their] offices."125

122 Chevron Corp. v. Stratus Consulting, Inc.,No. 10 Civ. 00047, DI 99, ¶ 17.123 Chevron Corp. v. Donziger, 886 F. Supp. 2dat 258.124 Id. at 258-260.125 Hendricks Decl. [DI 32], Ex. 156 (July 17,2007 email from Donziger) ("we think thatRichard [Cabrera] should suspend his work in thefield and we should [*54] not pay the team untilafter the recess. We just need him to tell the teamand Texaco that he's going to start all over afterthe recess so there is nothing strange, everythingappears normal.")

Notwithstanding the Fajardo Declaration, the Districtof Colorado denied the LAPs' motion for a protectiveorder and ultimately ordered Stratus to turn over itsdocuments.126 Following that ruling, the LAP teambrainstormed ways to delay further the production ofStratus' documents and, realizing that production wasinevitable, to mitigate its effects. One of the LAPs'lawyers sent an email to the LAP team emphasizing that"Stratus will be under a court order to produce allmaterials it gave Cabrera. Stratus will not risk a contemptmotion, it will comply. Unless we want theStratus/Cabrera revelation to come out in CO, whichseems like the worst possible place, we need to make oursubmission in Ecuador and fast."127 PB's Westenberger

responded, "[w]hat about the following? Appeal; movefor stay; if we win with [the District of Colorado] great; ifwe lose, we produce whatever we want (narrow read);[Gibson Dunn] complains and then we move forclarification. If we lose again, we think about another[*55] appeal."128

126 Chevron Corp. v. Stratus Consulting, Inc.,No. 10 Civ. 00047 (MSK) (MEH), 2010 U.S.Dist. LEXIS 55049, 2010 WL 2135217 (D. Colo.May 25, 2010) order clarified, No. 10 Civ. 00047(MSK) (MEH), 2010 U.S. Dist. LEXIS 61821,2010 WL 2232371 (D. Colo. June 1, 2010).127 Hendricks Decl. [DI 47] Ex. 292 (May 27,2010 email).128 Id.

Nearly five months after the court denied the LAPs'motion for a protective order, Chevron had yet to receivethe majority of responsive documents from Stratus andfiled a motion to compel production. The District ofColorado granted Chevron's motion on October 1,2011;129 the LAPs filed an emergency motion to staypending appeal three days later. The District of Coloradodenied the LAPs' motion the next day,130 and Stratusbegan producing documents.

129 Chevron Corp. v. Stratus Consulting, Inc.,2010 U.S. Dist. LEXIS 110023, 2010 WL3923092, at *12.130 Chevron Corp. v. Stratus Consulting, Inc.,No. 10 Civ. 00047, DI 268.

The Prieto email131 makes it clear, or at least highlylikely, that the LAPs' Ecuadorian counsel knew what hadtaken place among themselves, Stratus and Cabrera -- theCabrera report was not the work of an independent expertbut of Stratus, and the LAPs and the various statementsand other tactics designed to portray the report [*56] asindependent were, and were intended to be, highlymisleading. They made at least a good part of that clear tothe American lawyers. PB and Emery Celli, amongothers, then collaborated on the drafting of the FajardoDeclaration, which acknowledged having "deliveredmaterials to Mr. Cabrera" without revealing that theLAPs had prepared the plan for the report he had filed,had extensive contacts with Cabrera, had had Stratuswrite all or most of the report itself. That declaration thenwas submitted to the court in Denver for the purpose ofpersuading it that nothing amiss had occurred withrespect to the Cabrera report in an effort to prevent the


disclosure of Stratus' files, including "the e-mailsbetween Stratus and Juanpa [Saenz] and myself" thatPrieto was determined to keep concealed lest the case inEcuador be destroyed and the LAPs' Ecuadorianattorneys "go to jail."132

131 See supra pp. 30-31.132 The LAPs argue that "a close[] reading" ofPrieto's email suggests that the email betrayedconcern that disclosure of the emails betweenStratus, on the one hand, and Prieto and Saenz, onthe other was not acceptable because it wouldhave violated duties they owed to their clients as[*57] Ecuadorian lawyers, not, as Chevronmaintains, that he was concerned that disclosureof what had occurred vis-a-vis Cabrera amountedto misconduct such that the lawyers might go tojail. DI 712, 10-11. Perhaps. But that contention,whatever impact it someday may have before atrier of fact, is neither material at this stage nor, inall the circumstances, persuasive.

First, the issue for present purposes iswhether "a prudent person [would have] areasonable basis to suspect the perpetration orattempted perpetration of a crime or fraud." In reGrand Jury Subpoena Duces Tecum Dated Sept.15, 1983, 731 F.2d at 1039. Regardless of thePrieto email, the ultimate question with respect tothe proceedings in Colorado is whether there isprobable cause to suspect that the Fajardodeclaration that was submitted to the district courtthere was fraudulent because it suggested thatnothing was amiss with respect to Cabrera and hisreport but failed to disclose a host of highlymaterial facts quite plainly suggesting thecontrary. As discussed above, probable causeexists to suspect fraud in that respect independentof the interpretation of the Prieto email. ThePrieto email goes to the motive for [*58] thatnondisclosure, not to its deceptive nature.Moreover, even if the Prieto email shed light onthe latter issue, and even if PB's interpretationwere plausible, "a finding of probable cause," asdiscussed below, "is not negated by 'an innocentexplanation which may be consistent with thefacts alleged." United States v. McDonald,01-CR-1168JSWDW, 2002 U.S. Dist. LEXIS9866, 2002 WL 31956106, at *5 (E.D.N.Y. May9, 2002) (quoting A.I.A. Holdings, S.A. v. Lehman

Bros. Inc., 97 Civ. 4978 (LMM) (HBP), 1999U.S. Dist. LEXIS 1183, 1999 WL 61442, at *5(S.D.N.Y. Feb. 3, 1999) (citing United States v.Fama, 758 F.2d 834, 838 (2d Cir. 1985)).

Second, Chevron's interpretation of the emailis more plausible than PB's. Prieto's expressedconcern was not only with potential criminalexposure for the Ecuadorian lawyers. He wasconcerned also that disclosure of what had goneon between Stratus and the Ecuadorian lawyerscould or would "destroy[] the proceeding," i.e.,the Lago Agrio case. It is quite improbable that adisclosure of the Stratus documents by theEcuadorian lawyers in breach of professionalresponsibilities, even in criminal breach of suchresponsibilities, would have destroyed the LagoAgrio case, whatever consequences it might have[*59] had for the lawyers themselves. Indeed, theStratus documents were disclosed and theEcuadorian court entered the Judgment anyway.

Finally, the suggestion that Prieto wasconcerned that the disclosure of the emails wouldhave violated Article 335.1 of the EcuadorianJudicial Code (DI 712, at 11) borders on thefanciful. Although the LAPs were desperatelyseeking to stop disclosure of the Stratusdocuments in Colorado, no such argument wasmade to that Court. Nor was any such argumentmade to this Court when Donziger's files weresubpoenaed and produced in the 1782 proceedingor in the Count 9 action when the depositions ofEcuadorian lawyers were noticed. The notion thatdiscovery from the Ecuadorian lawyers wouldviolate Ecuadorian law first was raised in 2012 inthis case, more than two years after the date of thePrieto email and almost two years after thesubmission of Fajardo's declaration to theColorado district court.

In these circumstances, there is probable cause tosuspect that at least some of those involved, as allegedextensively in the amended complaint,133 committed mailand/or wire fraud134 and obstructed justice in at least theStratus 1782 proceeding in Colorado by [*60]formulating and filing the Fajardo declaration, which wasa seriously misleading account of what had happened.

133 See e.g., Am. Cpt. ¶¶ 266-303, 354-55,359-65.


134 "The elements of mail or wire fraud are (i) ascheme to defraud (ii) to get money or property,(iii) furthered by the use of interstate mail orwires." United States v. Autuori, 212 F.3d 105,115 (2d Cir. 2000). "[T]he well-settled meaningof 'fraud' require[s] a misrepresentation orconcealment of material fact." Neder v. UnitedStates, 527 U.S. 1, 22, 119 S. Ct. 1827, 144 L. Ed.2d 35 (1999) (first emphasis added, second inoriginal). Chevron alleged in its amendedcomplaint that the RICO defendants committedwire fraud by the electronic filing of court papersthat contained false and misleading statements. ¶354(e). For example, the Fajardo declaration"falsely attested that Cabrera was 'independent'and omitted from his declaration the substantialrole he and other conspirators had played insecuring Cabrera's appointment, his numerouspersonal meetings with Cabrera . . . and themassive, U.S. project to write, translate andsubmit the fraudulent Cabrera Report." ¶ 287.Chevron has submitted subsequently evidencethat shows -- at the very least -- that there [*61] isprobable cause to suspect that the Fajardodeclaration -- which was filed electronically withseveral U.S. courts -- misrepresented or concealedmaterial facts about the LAPs' relationship withCabrera.

PB's involvement in the Section 1782 proceedingswas not limited to the one in Colorado. It has representedthe Ecuadorian plaintiffs in at least six of the proceedingsacross the country135 and was involved behind the scenesin others.136 Just as in Colorado, the LAPs' employedPB's strategy of "fight[ing] hard on all fronts all the timeand conced[ing] nothing, [to] buy as much time aspossible."137 Indeed, the LAPs filed the FajardoDeclaration in at least fifteen other Section 1782proceedings across the country, including this one.138

135 See Chevron Corporation v. Allen, No.2:10-mc-00091-wks (D. Vt.); In re ChevronCorp., No. 1:10-mc-00053-SSB-KLL (S.D.Ohio); In re Chevron Corp., No. 7:10-mc-00067(W.D. Va.); In re Chevron Corp., No.2:10-mc-00208-JD (E.D. Pa.); Chevron Corp. v.Picone, No. 8:10-cv-02990-AW (D. Md.); In reChevron Corp., No. 2:10-cv-02675-KM-MCA(D.N.J.).136 See, e.g., Mastro Decl. [DI 658], Ex. 3010

(1/17/2011 text message from Westenberger toDonziger) ("I've [*62] convened a call for 1230on the 1782s."); Stavers Decl. [DI 547] Ex. 241(Dec. 13, 2010 E-mail from Eric Daleo of PattonBoggs) ("we have prepared for transferdocuments that previously appeared on privilegelogs in the Allen (Vermont), Picone (Maryland),Rourke (Maryland), and Scardina (Virginia)[Section 1782] proceedings.")137 DI 496, Ex. 5; see also Hendricks Decl. [DI48] Ex. 323 (email from Donziger explaining thatit is important to "adhere to the fundamentalprinciple of our strategy as outline by Jim" Tyrrellof Patton Boggs to "appeal everything on thetheory that we gain a greater advantage byfighting them on everything, and tying them up,than conceding any one thing").138 Stavers Decl. [DI 549] ¶ 44; see In reChevron Corp., 11-cv-24599 (S.D. Fla. June 26,2012 & Apr. 16, 2012) Dkts. 55-26 & 86-26;Chevron Corp. v. Donziger, 11-cv-0691(S.D.N.Y. Feb. 8, 2011 and Feb. 25, 2011) Dkts.66-2 & 138-13; In re Application of ChevronCorp., No. 2:10-cv-02675-SRC-MAS (D.N.J.June 7, 2010 and May 17, 2011) Dkts. 5-15 &54-10; Chevron Corp. v. Allen, No.2:10-mc-00091-WKS (D. Vt. Nov. 16, 2010) Dkt.23-9; In re Chevron Corp., No.1:10-mc-0053-SSB-KLL (S.D. Oh. Nov. 16,2010) Dkt. [*63] 20-10; Chevron Corp. v.Picone, No. 8:10-cv-02990 (D. Md. Nov. 16,2010) Dkt. 19-9; In re Chevron Corp., No.10-mc-10352 (D. Mass. Nov. 16, 2010) Dkt. 25-5;Chevron Corp. v. Scardina, No.7:10cv-00549-JCT (W.D. Va. Nov. 16, 2010)Dkt. 15-9; In re Application of Chevron Corp.,No. 10-mc-00021-JCH-LFG (D.N.M. Aug. 26,2010) Dkt. 63-2; Chevron Corp. v. Champ, No.10-mc-00027 (W.D.N.C. Aug. 24, 2010) Dkt.20-10; In re Application of Chevron Corp., No.3:10-cv-00686 (M.D. Tenn. Aug. 11, 2010) Dkt.44-9; In re Application of Chevron Corp., No.10-cv-01146-IEG-WMC (S.D. Cal. June 26,2010) Dkt. 16-4; In re Application of ChevronCorp., No. 10-mc-00371-CKK (D.D.C. June 11,2010) Dkt. 22-2; Chevron Corp. v. 3TM Int'l, Inc.,No. 4:10-mc-00134 (S.D. Tex. May 18, 2010)Dkt. 49-3; In re Chevron Corp., No. 10-cv-00047(D. Colo. May, 5, 2010).


B. The Cleansing Reports

In May 2010, the District of Colorado granteddiscovery from Stratus, although documents were notproduced until much later. Also in May 2010, this Courtgranted Chevron's application to subpoena the Crudeouttakes from the filmmaker, Joseph Berlinger.139 Theouttakes further supported Chevron's contention thatDonziger and the LAPs had [*64] orchestrated Cabrera'sappointment and written his report in whole or in majorpart. In this context, PB then led efforts -- in the wordschosen by PB's Mr. Westenberger--to "cleanse anyperceived impropriety related to the Cabrera Report."140

139 In re Chevron Corp., 709 F. Supp. 2d 283(S.D.N.Y. 2010).140 Hendricks Decl. [DI 9] Ex. 13 (email fromWestenberger to others).

First, as explained above, the LAP team -- by way ofanother Fajardo declaration -- petitioned the Lago Agriocourt to order supplemental reports on the issue ofdamages.141 The declaration, drafted by PB and otherLAP American lawyers, also was intended to argue thatthe submissions the LAPs' had made to Cabrera wereappropriate and acceptable under Ecuadorian law.Discovery has shown, however, that the LAPs' lawyersdebated whether and to what extent the declarationshould reveal the LAPs' "meetings with Cabrera"142 andthe nature and extent of their relationship with him. AJune 5, 2010 email by PB lawyer Edward Yennock said:

"The sole open issue [with thedeclaration] is the specificity with whichwe describe the 'meetings' with Cabrera . .. If we . . . are to have no more certaintythan we do today, we are inclined [*65] torevise this to include little or noinformation about the meetings, as muchas we'd like to be able to use thissubmission to fully air the facts. However,two events might occur to change thatapproach: (1) The Stratus emails areexpeditiously uploaded into a searchableformat that allows us to identify anycommunications related to the meetings byearly next week; and/or (2) we aresomehow able to review the relevantCrude outtakes."143

Yennock made clear that understanding what the Crudeouttakes and the Stratus documents would reveal wascrucial to determining whether to disclose the true natureof the LAPs' relationship with Cabrera. He wrote that"the pivotal nature of this submission, and the potentiallydevastating effect of making a representation that is laterproven to be wrong or incomplete by way of the emailsor the outtakes (assuming Chevron gets them), wouldseem to warrant a review [of the documents and theouttakes] if it can be done quickly."144

141 Stavers Decl. [DI 549] Ex. 2420 (FajardoDeclaration submitted to Lago Agrio court);Hendricks Decl. [DI 9] Ex. 13 (email fromWestenberger at PB to others attaching draftoutline of petition to be filed in Ecuadorian [*66]court.)142 Champion Decl. [DI 400] Ex. 2060 (June 5,2010 Email from PB lawyer Edward Yennock).143 Id.144 Id.

An Emery Celli lawyer responded that there was notenough time to review the outtakes and the Stratus emailsbecause "a court ruling -- relying solely on Cabrera -- ispotentially imminent [in Ecuador] if we don't getsomething on file immediately."145 He proposed thatthey:

"[T]ake out any references to alleged'contacts' between Cabrera and plaintiffs'counsel and/or with plaintiffs' consultingexperts and just confess to havingauthored specific portions of the report.We emphasize we think there was nothingwrong with this . . . . . The 'contacts' issueis too uncertain and gaining clarity willtake too long. If we cop to having writtenportions of the report, the details ofexactly how that might have beenaccomplished will be for another day,when and if the relevant people aredeposed as part of the 1782s, but hopefullyby that time, the process of having bothsides cure this with new submissions willbe under way and render the details of theCabrera report a thing of the past."146

145 Id. (June 14, 2010 email).


146 Id. (emphasis added).

Ultimately, the declaration informed the [*67] courtthat the LAPs had made submissions to Cabrera but didnot actually "confess to having authored specific portionsof the report."147 Instead, it stated that "Chevron enjoyedthe same opportunity as the Plaintiffs to provideinformation to Cabrera in support of its position in thecase,"148 and, while

"Chevron elected to ignore thisopportunity . . . [t]he Plaintiffs . . . tookadvantage of the opportunity to defendtheir own findings, conclusions andvaluations to Cabrera in order for the latterto consider potentially adopting them. Theinformation the plaintiff provided toCabrera included proposed factualfindings and economic valuations of theenvironmental and other damages Texpet'spractices and contamination caused.Naturally, Cabrera was free to adopt theplaintiff's viewpoint, proposed findingsand valuations. And of course, because heapparently considered them credible,Cabrera adopted the plaintiffs' proposals,analysis and conclusions regarding thedamages and assessment."149

147 Id.148 Stavers Decl. [DI 549] Ex. 2020 (FajardoDeclaration submitted to Lago Agrio court), at 6.149 Id. at 6-7.

This declaration went farther than that filed inDenver with its statement that [*68] the plaintiffs hadprovided Cabrera with "proposed factual findings andeconomic valuations of the . . . damages" and itscontention that Cabrera adopted them "because heapparently considered them credible." But that tooappears to have been deceptive, whether or not PB thenwas aware of the extent to which that was the case. Thereis at least probable cause to suspect that Cabrera washandpicked by the Lago Agrio plaintiffs because hewould "play ball" with them, that the entire report wasplanned and written by the LAPs and Stratus, and thatCabrera "played ball" by simply affixing his name to it,acting all the while under the pretense -- fostered by theLAPs -- that the report was Cabrera's independent work.

The Lago Agrio court granted the LAPs' request tofile additional damages assessments, and the LAPs puttogether the new team. PB "hired the Weinberg Group tomanage the [cleansing] process,"150 in August 2010.151

The two firms facilitated the hiring of new experts. PBwas the "lead"152 in developing the strategy for the newexpert reports,153 and had continuous "interaction" withthe Weinberg Group and the cleansing experts.154

150 Stavers Decl. [DI 549] Ex. 2406 (Donzigerdeposition [*69] transcript), at 1666.151 Hendricks Decl. [DI 36] Ex. 215 (Weinbergretention agreement).152 Stavers Decl. [DI 549] Ex. 2406 (Donzigerdeposition transcript), at 1666.153 See, e.g., Hendricks Decl. [DI 36] Ex. 214(Aug. 18, 2010 email from PB lawyer AdlaiSmall to others).154 Stavers Decl. [DI 549] Ex. 2406, at 1666-67.

As explained previously, the new experts wereinstructed to use Cabrera's data as a "starting point . . . todevelop [their] own valuation[s]."155 But at least one ofthe experts has testified that he was not told that Stratusand the LAPs had written all or at least most of Cabrera'sreport156 Another has admitted that he relied directly onparts of the Cabrera report and made no effortsindependently to verify Cabrera's data.157 None of thenew experts ever traveled to Ecuador to gather their owndata. Indeed, as one PB attorney wrote to Donziger, thepurpose of the cleansing reports was to "addressCabrera's findings in such a subtle way that someonereading the new expert report (the Court in Lago or anenforcement court elsewhere) might feel comfortableconcluding that certain aspects of Cabrera [were] a validbasis for damages.'"158

155 Champion Decl. [DI 400] Ex. 2073 [*70](Allen deposition transcript), at 140:25-141:11.156 Stavers Decl. [DI 549] Ex. 2417 (Shefftzdeposition transcript), at 68:14-24.157 Champion Decl. [DI 400] Ex. 2073 (Allendeposition transcript), at 140:25-141:11.158 Hendricks Decl. [DI 36] Ex. 214.

In the end, seven new reports were filed with theLago Agrio court on September 16, 2010. Six of theseven were by U.S. experts; one was anonymous.159

159 Chevron Corp. v. Donziger, 886 F. Supp. 2dat 261.


The Lago Agrio court issued the Judgment inFebruary 2011. It disclaimed reliance on the Cabrerareport, but noted that it had considered other expertassessments, including some of the reports submitted bythe new experts hired by PB, in rendering theJudgment.160

160 DI 168 (Lago Agrio Judgment) at 48-51.

C. Post-Trial Submissions to the Lago Agrio Court

PB was involved also in drafting the LAPs' post-trialsubmissions to the Lago Agrio court. Beginning no laterthan March 2010, it revised and re-wrote the alegatos --post-trial briefs--before they were filed.161 It "handle[d]the alegato assignments"162 and worked on"re-draft[ing]" and "complet[ing]" the alegatos beforethey ultimately were submitted to the Lago Agrio court inDecember 2010 and [*71] January 2011.163

161 On March 6, 2010, Aaron Marr Page, whoworked for Donziger, sent an email to Donziger,attaching the draft alegato and explaining that "Ithink the draft attached here can be sent toselected counsel, as we discussed . . . . I think thatmay be precisely what new counsel needs to seeto understand where we are in the process and tostart thinking about how they can effectivelyengage." Mastro Decl. [DI 658] Ex. 3007. OnApril 8, 2010, Laura Garr, another individualworking for Donziger, sent an email to EricWestenberger attaching "the latest working draftof the Alegato in the Ecuador trial." HendricksDecl. [DI 355] Ex. 1135.162 Mastro Decl. [DI 658] Ex. 3010 (textmessage on 1/14/11 at 1:00 p.m. fromWestenberger to Donziger saying "I'll handlealegato assignments"); DI 558 (PB Mem. Of Lawin Support of Mot. to Quash the Subpoena), at 8("Recognizing that these filings may be read byU.S. courts and abroad . . . Patton Boggs provideddrafting assistance.").163 Stavers Decl. [DI 549] Ex. 2439.

D. PB's Other Activities

PB's role has gone beyond that described above.

PB authored an undated memorandum called"Invictus," which laid out a plan to enforce the Judgment"quickly, [*72] if not immediately, on multiple

enforcement fronts -- in the United States and abroad."164

The memorandum, which recognized that enforcement ofthe Judgment may be difficult in the United States,emphasized that "Patton Boggs' current and formerrepresentation of numerous, geographically diverseforeign governments means that barriers to judgmentrecognition in a given country may not necessarilypreclude enforcement there."165 It further elaborated that"Patton Boggs will use its political connections andstrategic alliances to ascertain which nations'governments are not beholden to Chevron, so as tominimize the prospect of adverse governmentalinterference in the enforcement process."166

Subsequently, PB has been heavily involved in the LAPs'efforts to enforce of the Judgment in other countries.

164 Hendricks Decl. [DI 6] Ex. 341 (InvictusMemorandum), at 12.165 Id. at 19.166 Id.

PB also controls, at least to some degree, how theLAP team spends the money it receives to fund thelitigation. In October 2010, the LAPs entered into afunding agreement with Treca Financial pursuant towhich Treca agreed to provide for legal expenses inreturn for a stake in the Judgment.167 Not only did [*73]PB play the "primary role" in "convincing [Treca'sfinancial advisor] to invest in the Lago Agriolitigation,"168 it also had authority over how the fundsTreca provided were spent. Indeed, "fees andout-ofpocket expenses of lawyers . . . advisors, experts,witnesses and others . . . ." were subject to the approvalof "James Tyrrell (the lead partner at Patton BoggsLLP)," as well as Donziger.169 Any costs associated withcourt proceedings, arbitral tribunals, fees relating toclaims or counterclaims by or against the LAPs, and feesand expenses of LAPs' counsel were subject also toTyrrell's and Donziger's approval.170

167 11 Civ. 3718 (LAK), Champion Decl. [DI44] Ex. 46 (Treca Funding Agreement).168 Stavers Decl. [DI 549] Ex. 2406 (Donzigerdeposition transcript), at 3239:22-2340:03.169 11 Civ. 3718 (LAK), Champion Decl. [DI44] Ex. 46 (Treca Funding Agreement), at 2.170 Id. Treca has since terminated itsrelationship with the LAPs. On September 29,2011, the Burford Group, Treca's financialadvisor, sent a letter to Donziger and the LAPs,


stating that "[i]t is clear from the evidence that hascome to light subsequent to our discussions withyou and Treca's entry into the Funding Agreement[*74] that [you] have engaged in conduct andactivity that gives rise to numerous breaches ofthe Funding Agreement. In addition to breachingthe Funding Agreement -- throughmisrepresentations and other failures -- theconduct discovered amounts to fraud." ElliotDecl. [DI 714] Ex. 1.

Finally, PB itself has become an active litigant incases related to this one. PB has brought three actionsagainst Chevron in which PB itself is the plaintiff -- twoin the District Court for the District of Columbia171 andone in the District of New Jersey, which recently wastransferred to this Court.172

171 Patton Boggs, LLP v. Chevron Corp., No.10 Civ. 1975 (HHK) (D. DC); Patton Boggs LLPv. Chevron Corp., No 11-00799 (HHK) (D.DC).172 Patton Boggs, LLP v. Chevron Corp., No.12 Civ. 9176 (LAK).

IV. Proceedings With Respect to the PB Subpoena

The Subpoena was served on June 15, 2012, andcontained 58 discrete requests for the production ofdocuments.

A. PB's Motion to Quash

PB moved to quash on July 20, 2012.173 It arguedprincipally that the Subpoena would require it to discloseprivileged or other protected material and thatcompliance would subject it to undue burden.174

173 DI 522.174 See DI 527.

On the issue of [*75] privilege, the Court held that(1) "there [was] reason to doubt whether many -- if any --of the subpoenaed documents [were] protected by theattorney-client privilege,"175 (2) PB's motion "essentiallyignore[d] the fact that the protection accorded to attorneywork product in important respects is less extensive thanthat accorded to attorney client communications,"176 and(3) PB had failed to show that it was not in possession ofresponsive documents obtained from third parties thatwere not protected from disclosure.177 Moreover, theCourt noted that certain of the protected documents might

be subject to disclosure under the crime-fraud exceptionto the attorney client privilege and the work productdoctrine.178

175 DI 571 at 2.176 Id. at 3.177 Id. at 4.178 Id. at 4-5.

With respect to burden, the Court determined that itwas premature to quash the Subpoena as undulyburdensome before its scope had been resolved.179 TheCourt noted also that "it is far from clear that the burdenof complying with this subpoena . . . would be out of linewith what occurs in comparable litigation in thiselectronic age."180

179 Id. at 6.180 Id. at 7.

The Court denied the motion to quash withoutprejudice to PB's [*76] privilege and burden claims,which it would consider in connection with any Rule45(c)(2)(B) objections PB might raise. The Courtdeferred PB's obligation to produce a privilege log untilfurther notice.181

181 Id. at 8.

B. Argument and the Initial Narrowing of the Subpoena

PB then served Chevron with 186 pages ofobjections to the Subpoena. The Court held a hearing onSeptember 25 and 27, 2012 to address them, ruling on themajority during the hearings182 and the balance in aNovember 16, 2012, order (the "November Order"),which also memorialized the rulings it had made at thehearings. The Court sustained PB's objections to 15 of the58 requests and limited 10 others.183

182 The Court made clear that its rulings restednot only on relevancy, but on Fed. R. Civ. P.26(b)(2)(C), which "gives district courtsdiscretion to limit the extent of discovery, even ofrelevant matters, for several reasons. One of themis that its burden or expense outweighs its likelybenefit, considering the needs of the case, theamount in controversy, the parties' resources, theimportance of the issues at stake, and theimportance of the discovery in resolving theissues." It sustained many objections in whole


[*77] or in part in order to reduce alleged burdenon PB and intrusion into its role as litigatingcounsel rather than on the ground that theinformation sought was not relevant. See e.g., Tr.,Sept. 25, 2012, at 6-7.183 See DI 621.

Both sides agree that the Court's rulings substantiallylimited the Subpoena's scope and the extent of the effortthat would be necessary to comply with it as modified.According to Chevron's experts, the effect of the Court'srulings was to reduce by 90 to 99 percent the amount ofdata that would have to be searched and reviewed inorder to comply184 and reduced the cost of complianceby 85 percent.185 PB's experts estimate that, even usingChevron's proposed search terms, the November Orderreduced the cost of reviewing and logging responsivedocuments by over 80 percent.186

184 DI 713, at 7-8.185 Id.186 Compare DI 665, at 2-3 (To review and logthe documents resulting from Chevron's searchterms will likely . . . cost between $1,060,000 and$1,290,000.") with DI 527, at 2-3 ("At aminimum, it is estimated that the review ofelectronic documents alone will cost the firmbetween $6.35 million and $7.75 million dollars,plus roughly $550,000 for electronic document[*78] collection").

Subsequent to the November Order, the Courtreceived supplemental briefing on (1) the extent to whichthe Subpoena as modified is unduly burdensome, and (2)whether the Subpoena in its modified form soughtmaterials in areas in which "the first of the two prongs ofthe crime-fraud exception [to the attorney clientprivilege] is satisfied."187

187 Sept. 27, 2012 Tr., at 139.

PB contends that the Subpoena still is impermissiblyoverbroad. PB complains principally that Chevron seeksinformation from its "adversary's litigation counsel" that"it will likely obtain . . . from the parties or othernonparties."188 Moreover, PB claims that the Subpoenaseeks documents concerning events in which PB had noinvolvement and asks that the Court "limit its areas ofinquiry to Chevron's allegations of Patton Boggs'purported misconduct or alleged conduct witnessed bycounsel and to exclude historical information learned in

Patton Boggs' role as counsel."189 Finally, PB argues thatmost of the documents it has are privileged or attorneywork product and that the crime-fraud exception does notapply.

188 DI 665, at 3.189 Id. at 4.

Chevron argues that compliance with the Subpoena-- at least given [*79] the extent to which it has beennarrowed by the Court and the parties' agreements --would impose no undue burden.190 Moreover, Chevroncontends that it has provided sufficient evidence to satisfythe first prong of the crime-fraud exception to the workproduct doctrine and attorney client privilege.

190 See DI 713.

V. Chevron's Unsuccessful Attempts to Obtain Discoveryfrom the LAP Representatives' Agents in Ecuador andFrom the Defendants

It is important to recognize that the LAPs and theirEcuadorian lawyers and associates have refused toprovide any meaningful discovery of documents andwitnesses located in Ecuador or to include informationfrom Ecuador in their own responses to discoveryrequests.

A. Defendants' Refusal To Produce Documents andEvidence from Ecuador

In both this action and the Count 9 action,191

defendants have refused to produce any documents in thepossession, custody, or control of their attorneys andagents in Ecuador despite orders by this Court compellingthem to do so. While the bottom line has been consistent,the rationale has changed.

191 "Count 9 action" refers to 11 Civ. 3718(LAK), which came into existence by theseverance of Count 9 of Chevron's amended[*80] complaint in this action. The Count 9 actionhas been dismissed.

In the Count 9 action, defendants contended that theywere unable to produce those documents because theEcuadorian attorneys were not their agents and thereforewere not subject to their control. After the Court heldotherwise and ordered the documents produced,192

defendants refused again, arguing that collecting,reviewing, and producing documents from their


"approximately 87 lawyers" in Ecuador would be tooburdensome.193

192 11 Civ. 3718, DI 101, at 2-3.193 See 11 Civ. 3718, DI 240, at 7.

In this action, defendants' obstinance with respect todiscovery of evidence from Ecuador has reached a newlevel. Chevron moved on August 13, 2012 to compel theLAP Representatives and the Donziger Defendants toproduce documents in the hands of their Ecuadorianlawyers and other associates.194 Although this time theyconceded that Fajardo is the LAP Representatives' agent,the LAP Representatives nevertheless resisted, arguinginter alia that the motion was premature and, in any case,that the Ecuadorian lawyers were precluded fromcomplying by Ecuadorian law and that the defendantshere therefore did not control them.195 They submitted[*81] a declaration by an Ecuadorian lawyer thatpurported to support their position. Chevron replied with,inter alia, a declaration by another Ecuadorian lawyerthat disputed the conclusions of the first.196 And whilethat motion was pending, Chevron renewed its contentionin a second motion to compel on November 6, 2012.197

In due course, the Court ordered production regardless ofwhether responsive documents are in Ecuador.198 Thesedefendants, however, have refused to comply with theorder.199 Now, however, they advance a new basis.

194 DI 562.195 DI 563, DI 564.196 Ltr., Randy Mastro, Sept. 5, 2012.197 DI 608, at 4.198 DI 787.199 See DI 836, DI 841.

While Chevron's motion was pending andunbeknownst to the Court or to Chevron, AttorneySmyser, counsel for the LAP Representatives in this case,by his own admission, "suggest[ed] to the Ecuadorianlegal team that someone should consider seeking anEcuadorian court ruling on the issue of documentproduction."200 In consequence, a lawsuit was filed inEcuador in October 2012 in the name of one of thenon-appearing LAPs, Octavio Ismael Cordova Huanca,against Cordova's attorneys -- Fajardo, Saenz, and Prietoas well as the head of the Amazon Defense [*82] Front,Luis Yanza -- to bar them from turning over anyinformation in discovery in this case. The case obviouslywas collusive in the sense that, as the judgment recites,the ostensible defendants agreed with the ostensible

plaintiff.201 Indeed, section FOUR quotes Fajardo atlength in support of the arguments advanced, supposedlyagainst Fajardo, on behalf of Mr. Cordova.202 TheEcuadorian court did what all of the parties before it --the plaintiff and the defendants -- asked it to do: itentered an order barring disclosure to this Court of anyinformation possessed by or known to the defendants inthat case concerning the Chevron litigation.203

200 Ltr., Craig Smyser and Larry R. Veselka,Mar. 8, 2013, at 2.201 DI 734, Ex. A (Ecuadorian Judgment). Inaddition, Fajardo holds a broad power of attorneyon behalf of all of the LAPs. Hendricks [DI 106]Ex. 481. He thus was in a position to facilitate allof the U.S. discovery on behalf of all of them, notto mention himself, had he wished to do so. Helikewise was in a position to have resisted theCordova suit, both individually and on behalf ofthe other LAPs, had he so desired. Moreover, itbears noting that Fajardo held the power toinstitute [*83] the action against himself and hiscolleagues in Mr. Cordova's name. To this it mustbe added that Mr. Smyser claims, in an unswornletter, that "the two parties to the [Ecuadorian]proceeding were represented by separatecounsel." Supra note 200. On that last point,however, there is no evidence of record that thiswas indeed the case. The record does disclose,however, that only one attorney appearance isnoted on the decision -- that of Fajardo.202 DI 734, Ex. A.203 Id. at 4-5.

The fact that this collusive lawsuit was brought in thename of one of the LAPs against his own lawyerslikewise is troublesome, especially when one recognizesthat the only attorney appearance listed on the decision isthat of Fajardo. The fact that this was done behind theback both of Chevron and of this Court while a motion tocompel was pending is even more troublesome. TheCourt notes also the following:

Section FIVE of the Ecuadorian court's decision,which contains its reasoning, begins as follows:

"The plaintiff, in bringing a suit for theprotection of constitutional rights, mustprove that rights referred to that might beviolated will cause him serious harm,


which has been proven in this case.Therefore, [*84] it is appropriate to applySection 21 of Art. 66, and Arts. 75 and 76of the Constitution of the Republic ofEcuador referred to by the plaintiff. Onpages 51 et seq. of the record and in thepublic hearing, the plaintiff submitteddocumentation with a translation intoSpanish by [name omitted], in which itappears that the Judge of the United Statesof American Lewis A. Kaplan of thesecond district of the State of New Yorkdemands our confidential information infavor of his jurisdiction, this has beenrequested from our attorneys . . ."204

Note that the writer of the decision on line 2 spoke of theplaintiff in the third person -- as "him." Two sentenceslater, however, in lines 8 and 9, the writer referred to theconfidential information and the LAPs' attorneys as "our"information and "our" attorneys instead of the plaintiff'sinformation and the plaintiff's attorneys. For presentpurposes, however, it is unnecessary to decide exactlyhow, why, and by whom the decision came to be writtenin this curious way.

204 Id. at 4 (emphasis added).

But the bottom line here is that the LAPRepresentatives and the Donziger Defendants are defyingthis Court's order to produce information in the [*85]hands of their Ecuadorian attorneys and even theirnon-lawyer associates. Rather than attempt to facilitatethe full and fair disclosure to which any litigant isentitled, they have engaged in extraordinary efforts toraise obstacles to such disclosure. This episode lendsfurther strength to Chevron's contention that relevantdisclosure from PB -- which is at the heart of the LAPs'and Donziger's efforts both in the United States and, itappears in Ecuador and elsewhere -- is extremelyimportant and certainly unlikely to be forthcoming fromEcuador.

B. Defendants' Refusal to Comply with Their OtherDiscovery Obligations

Chevron has fared no better in getting the LAPs (ortheir agents and experts) to submit to deposition205 or toprovide adequate responses to its interrogatories andrequests for admission ("RFAs"). In both the Count 9action and this one, the LAPs have made many baseless

objections and arguments in efforts to justify their refusalto respond adequately to Chevron's RFAs.206 In theCount 9 action, the LAPs claimed not to have knowledgeof or recollect the information necessary to answer manyof Chevron's interrogatories.207 In this action, the LAPsand Donziger initially [*86] refused to respond to asingle interrogatory propounded by Chevron based on anunfounded numerosity objection.208

205 The LAP Representatives refused to submitto deposition in New York in the Count 9 action.See 11 Civ. 3718, DI 225. And in that action,counsel for one or the other group of thesedefendants repeatedly directed witnesses AaronPage, Laura Garr, and Andrew Woods, the lattertwo of whom were employed by Donziger atvarious points, not to answer questions on groundof alleged privilege despite prior rulings bymagistrate judges in both the District of Marylandand this Court that the claimed privileges hadbeen waived, vitiated by the crime-fraudexception or both. DI 797, at 1-2. Most recently inthis action, when Chevron noticed the depositionof Saenz -- one of the LAPs' Ecuadorian lawyers-- counsel for the LAPs informed Chevron thatSaenz refused to make himself available fordeposition. See Champion Decl. [Ex. 770] Ex. 4;see also DI 778.206 For example, in the Count 9 action, theLAPs refused to respond to any of Chevron'sRFAs because they claimed the term "drafted"was ambiguous. See 11 Civ. 3718, DI 71, Ex. 3.207 See 11 Civ. 3718, DI 72, Ex. 3, 4.208 DI 781 Exs. 4, [*87] 5; DI 809 (OrderGranting Chevron's Motion to Compel).

Donziger has been little more forthcoming in certainrespects. When Chevron subpoenaed him in the Section1782 proceeding, Donziger sought to avoid complyingwith the subpoena, first by moving unsuccessfully toquash it on burden and privilege grounds209 and then byproviding an untimely privilege log that was 2,000 pageslong and claimed privilege as to 8,652 documents.210 Infact, however, not one of the documents "was written byor addressed to any of the Lago Agrio plaintiffs -- theclients whose privilege supposedly [was] beingasserted."211 Moreover, Donziger's motion and theprivilege log lacked any detail or substance, which"foreclosed any serious attempt to come to grips with thespecific privilege claims," and "served the tactical


interests of Donziger and the Lago Agrio plaintiffs andsubstantially prejudiced the parties seeking discovery."212

And when Donziger was deposed in the Section 1782proceeding, he "gave many unresponsive, self-servinganswers to questions that should have been answereddirectly, with no embellishment."213 So evasive andunresponsive was Donziger at the outset that the SpecialMaster overseeing [*88] the proceeding requested thatthis Court intervene to order his compliance.214

209 In re Chevron Corp. 749 F. Supp. 2d 135,140 (S.D.N.Y. 2010).210 In re Chevron Corp., 749 F. Supp. 2d 170,173 (S.D.N.Y. 2010).211 Id.212 Id. at 183.213 Hendricks Decl. [DI 48], Ex. 336 (Feb. 6,2011 Order).214 See id. ("I have cautioned [Mr. Donziger]many times and have stricken unresponsive,evasive, self-serving testimony . . . [but] myinstructions to the witness and striking portions ofhis answers seemed to have little effect.").

Chevron's inability to obtain appropriate discoveryunderscores its need to get it from PB, to the extent itmay do so consistent with governing principles. PB hasbeen intimately involved in shaping and carrying outdefendants' strategy since early 2010 -- a role thatnecessarily involves awareness not only of what hastranspired since it came onto the matter, but of whatoccurred before. It is against this backdrop that the Courtproceeds to resolve the remaining issues concerning theSubpoena.

Discussion

I. In re Friedman

The Subpoena is directed to a law firm. Although PBhas not formally appeared before this Court in this case, ithas been involved in the Court of Appeals. It [*89] hasassisted the LAPs in the Ecuadorian litigation, and it hasappeared on their behalf in other U.S. litigation relatingto the broad controversy at issue here.

There is an obvious tension between ensuring alawyer's ability to represent the lawyer's clientsvigorously and preserving an adverse party's right toobtain evidence necessary to prosecute its case. Where, ashere, discovery is sought from attorneys, especially

attorneys actively involved in litigation against anadversary seeking discovery from such attorneys,"[c]ourts have been especially concerned about theburdens imposed on the adversary process . . . and haveresisted the idea that lawyers should routinely be subjectto broad discovery."215 Nevertheless, the Second Circuithas made clear that the fact that a party seeks discoveryfrom a lawyer "does not automatically insulate [thelawyer]. . . nor automatically require prior resort toalternative discovery devices."216 In In re SubpoenaIssued to Dennis Friedman, it instructed courts to take a"flexible approach" to discovery of lawyers "whereby the[court] . . . takes into consideration all of the relevantfacts and circumstances to determine whether theproposed [*90] [discovery] would entail an inappropriateburden or hardship."217 These include "the need to [getdiscovery from] the lawyer, the lawyer's role inconnection with the matter on which discovery is soughtand in relation to the pending litigation, the risk ofencountering privilege and work-product issues, and theextent of discovery already conducted."218

215 In re Subpoena Issued to Dennis Friedman,350 F.3d 65, 70 (2d Cir. 2003).216 Id. at 72.217 Friedman, 350 F.3d at 72.218 Id.

As a preliminary matter, the Court turns first toChevron's argument that Friedman has no bearing herebecause PB quite deliberately has not appeared in thedistrict court in this case. It argues that Friedman isinapplicable because it involved depositions of opposinglitigation counsel whereas this Subpoena seeks onlydocuments and those from a firm that is not formallyinvolved in this action.

While PB's avoidance of an appearance before thisCourt may be relevant in some respects, the Court sees noreason to disregard Friedman's wise teachings entirely onthat account. It cannot be gainsaid that PB, whatever itsformal role before this Court, is broadly involved in thiscontroversy as an adversary to Chevron. [*91] Many ofthe concerns that arise when an adverse trial counsel issubjected to deposition -- e.g., possible compromise ofthe attorney-client relationship or intrusion on anattorney's work product -- also are present when a lawfirm involved behind the scenes is required to providedocuments to an adversary. Chevron has presented nocompelling reason that the Court should not be guided by


the factors set forth in Friedman, taking account of all ofthe relevant circumstances.

That said, it must be recognized that this Subpoenapresents a very unusual situation. The Subpoena in majorpart -- but not entirely -- seeks discovery with respect tosubjects on which the LAPs' Ecuadorian lawyers andnon-lawyer associates reasonably might be expected tohave the most complete information. If they, their clients,and the Donziger Defendants were fully cooperative andresponsive in discovery, there might be only very limitedneed to pursue discovery from PB. But the LAPRepresentatives and the Donziger Defendants haverefused to produce documents or give fully responsivereplies to other discovery requests, claiming that theEcuadorians will not provide the information. Yet PB hasbeen interacting [*92] with the Ecuadorians for overthree years now and doing so with respect to many of thesubjects on which the LAP Representatives, theirEcuadorian lawyers and associates, and the DonzigerDefendants decline to provide discovery. As will appearbelow, it is entirely likely that PB has responsivedocuments -- especially but not only emails and othercorrespondence with the Ecuadorians about criticallyimportant events in Ecuador -- that in the circumstancesof this case are not practically available from anyone else.Moreover, PB has been a primary actor in a number ofkey events.

With these preliminary comments, the Court turns tothe Friedman factors. It begins with the question whetherenforcement of the modified Subpoena would entailattorney-client privilege and work product difficultiesthat are so substantial as to refuse enforcement altogetherand whether any such problems can be ameliorated. Itthen will proceed to the remaining factors.

A. Attorney-Client Privilege and Work Product

1. Basic Principles

(a) Attorney-Client Privilege

A party invoking the attorney-client privilege mustdemonstrate that the communication as to which privilegeis asserted was "(1) a communication between [*93]client and counsel, which (2) was intended to be and wasin fact kept confidential, and (3) made for the purpose ofobtaining or providing legal advice."219 As the Courtnoted in its opinion denying Donziger's motion to quashthe subpoena served on him in a Section 1782

proceeding:

"The 'predominant purpose' of acommunication must involve legal advice.A court should determine predominantpurpose of a communication 'dynamicallyand in light of the advice being sought orrendered, as well as the relationship,between advice that can be rendered onlyconsulting the legal authorities and advicethat can be given by a non-lawyer'."220

cuong

219 United States v. Constr. Products Research,Inc., 73 F.3d 464, 473 (2d Cir. 1996).220 In re Chevron Corp., 749 F. Supp. 2d at 165(quoting In re County of Erie, 473 F.3d 413,420-21 (2d Cir. 2007).

(b) Work Product Doctrine

The work product doctrine "provides qualifiedprotection for materials prepared by or at the behest ofcounsel in anticipation of litigation or for trial."221 Thedoctrine "is intended to preserve a zone of privacy inwhich a lawyer can prepare and develop legal theoriesand strategy 'with an eye towards litigation,' free fromunnecessary [*94] intrusion by his adversaries."222 Thedoctrine now is codified in Rule 26(b) of the FederalRules of Civil Procedure.

"The Rule states that documents'prepared in anticipation of litigation or fortrial' are discoverable only upon a showingof substantial need of the materials andinability, without undue hardship, toobtain their substantial equivalentelsewhere. Even where this showing hasbeen made, however, the Rule providesthat the court 'shall protect againstdisclosure of the mental impressions,conclusions, opinions, or legal theories ofan attorney or other representative of aparty concerning the litigation.'"223

221 In re Grand Jury Subpoenas Dated March19, 2002 & August 2, 2002, 318 F.3d 379, 383 (2d


Cir. 2003).222 United States v. Adlman, 134 F.3d 1194,1196 (2d Cir.1998) (quoting Hickman v. Taylor,329 U.S. 495, 67 S. Ct. 385, 91 L. Ed. 451(1947)).223 United States v. Adlman, 134 F.3d 1194,1197 (2d Cir. 1998) (quoting Fed. R. Civ. P.26(b)(3)).

The party invoking work product protection "bearsthe burden of establishing its applicability to the case athand."224 Thus, once a party establishes that the materialin question constitutes work product, the party seekingdisclosure of that material must show [*95] that it has a"substantial need" for the otherwise protecteddocuments225 and that the party is unable without unduehardship to obtain the substantial equivalent of thematerials by other means.'"226

224 In re Grand Jury Subpoenas Dated March19, 2002 & August 2, 2002, 318 F.3d at 384.225 In re Grand Jury Proceedings, 219 F.3d175, 190 (2d Cir. 2000).226 In re Grand Jury Subpoenas Dated March19, 2002 & August 2, 2002, 318 F.3d at 383(quoting Fed. R. Civ. P. 26(b)(3)).

"As for work-product that shows 'mentalimpressions, conclusions, opinions, or legal theories of anattorney,'" the Second Circuit has suggested that, "at aminimum such material is to be protected unless a highlypersuasive showing [of need] is made."227

227 In re Grand Jury Proceedings, 219 F.3d at190 (quoting Fed. R. Civ. P. 26(b)(3)).

(c) The Crime-Fraud Exception

"It is well-established that communications thatotherwise would be protected by the attorney-clientprivilege or the attorney work product privilege are notprotected if they relate to client communications infurtherance of contemplated or ongoing criminal orfraudulent conduct."228 Importantly, especially in thiscase, communications in furtherance of contemplated[*96] or ongoing criminal or fraudulent conduct "areproperly excluded from the scope of the privilege even ifthe attorney is unaware that his advice is sought infurtherance of such an improper purpose."229

228 In re Grand Jury Subpoena Duces Tecum

Dated Sept. 15, 1983, 731 F.2d 1032, 1038 (2dCir. 1984).229 Id.

In order successfully to invoke the crime-fraudexception, a party seeking disclosure must demonstratethat there is "probable cause to believe that a fraud orcrime has been committed and that the communicationsin question were in furtherance of the fraud."230 Probablecause exists where "a prudent person ha[s] a reasonablebasis to suspect the perpetration or attempted perpetrationof a crime or fraud, and that the communications were infurtherance thereof."231 Moreover, "a finding of probablecause is not negated by 'an innocent explanation whichmay be consistent with the facts alleged."232 Thus, inorder to obtain disclosure of otherwise privileged orprotected evidence by means of the crime-fraudexception, the party seeking disclosure must show afactual basis to support a conclusion that there is probablecause to believe that (1) a crime or fraud was or is beingcommitted, [*97] and (2) the communication in questionwas or is in furtherance of the crime or fraud.233

230 United States v. Jacobs, 117 F.3d 82, 87 (2dCir. 1997).231 In re Grand Jury Subpoena Duces TecumDated Sept. 15, 1983, 731 F.2d at 1039 (2d Cir.1984).232 United States v. McDonald,01-CR-1168JSWDW, 2002 U.S. Dist. LEXIS9866, 2002 WL 31956106, at *5 (E.D.N.Y. May9, 2002) (quoting A.I.A. Holdings, S.A. v. LehmanBros. Inc., 97 Civ. 4978 (LMM) HBP, 1999 U.S.Dist. LEXIS 1183, 1999 WL 61442, at *5(S.D.N.Y. Feb. 3, 1999) (citing United States v.Farma, 758 F.2d 834, 838 (2d. Cir. 1985)).233 United States v. Jacobs, 117 F.3d at 87.Once the party has satisfied the first of theserequirements, "the decision whether to engage inan in camera review of the evidence lies in thediscretion of the district court. . . . [I]f and whenthere has been an in camera review, the districtcourt exercises its discretion again to determinewhether the facts are such that the exceptionapplies." Id.

2. Application in this Case

Although one normally would expect that a subpoenasuch as this would encounter substantial privilege andwork product obstacles, the extent to which that is so here


is very much more limited than one might expect forthree reasons. First, it is unlikely [*98] that there aremany, if any, responsive attorney-client communications.Second, Chevron has overcome the work productprotection as to any documents that contain only ordinarywork product. Third, to whatever extent there are anyattorney-client communications or documents thatconstitute opinion work product, the crime-fraudexception may well vitiate any protection.

(a) Improbability of Many Responsive Attorney-ClientCommunications, If Any

In considering whether PB's responsive documentsinclude many attorney-client communications, one mustkeep in mind the characteristics and location of theclients whose privilege PB asserts. The LAPRepresentatives have been described by their counsel as"a campesino and a canoe operator living in the remoteEcuadorian jungle."234 PB's Tyrrell has called the LAPs"indigenous people living in the jungle in Ecuador."235

As the Court noted in denying PB's motion to quash theSubpoena, "[t]he likelihood of any material number ofattorney-client communications between the LAPs andPB, which are the sole focus of the attorney-clientprivilege . . . is slim."236 PB has not suggested that it hasany responsive documents that are or memorializeconfidential [*99] communications involving any of theLAPs. Moreover, when Chevron served Donziger with asubpoena in the Section 1782 proceeding, Donziger --who has been involved in these matters since theirinception about twenty years ago -- moved to quash thesubpoena, objecting, inter alia, that the informationChevron sought from him was privileged. When he(belatedly) filed a privilege log, it did not list even asingle document said to involve a communication withany of the LAPs.

234 DI 518 (LAP Reps. Memo in Support ofMotion to Dismiss for Lack of PersonalJurisdiction), at 1.235 Smyser Decl. [DI 445] Ex. N (Transcript ofSeptember 16, 2011 oral argument in ChevronCorp. v. Naranjo, 667 F.3d 232 (2d Cir. 2011)), at88:9.236 DI 571, at 2.

A second factor further supports the view that theextent to which attorney-client privilege will play a rolehere is extremely limited. Major foci of the Subpoenainclude evidence of communications between the LAP

lawyers and non-lawyer associates, on the one hand, andCabrera or Ecuadorian judges and officials on theother.237 None of these persons stands in a privilegedrelationship either with PB or with anyone else on theLAP side of the case. Attorney-client [*100] privilege bydefinition does not extend to such materials.

237 See, e.g., Specification Nos. 14 (documentsrelated to drafting of orders by Lago Agrio court);18 (communications with Calmbacher); 19(communications with court experts); 20(communications with Cabrera); 26(communications with the Lago Agrio court); 27(communications with Ecuadorian judges); 30(documents provided by LAPs to author of LagoAgrio court).

(b) Chevron's Substantial Need Overcomes OrdinaryWork Product Protection

While it is unlikely that there are many responsivedocuments that even arguably come within theattorney-client privilege, the same is not true with respectto work product. But it is important to bear in mind thatthere are two kinds of work product -- materials that arenothing more than documents prepared in anticipation oflitigation and materials that, in addition, include mentalimpressions, conclusions, opinions, or legal theories of anattorney. As noted above, the former are discoverable ona showing of substantial need and inability to obtain theirsubstantial equivalent elsewhere without undue hardship.Chevron has made that showing.

First, most of the events about which the Subpoenaas [*101] narrowed seeks documents -- e.g., theCalmbacher report, the appointment of Cabrera andsubmission of his report, and the authorship of theJudgment -- took place to a material degree, althoughmost certainly not entirely238 -- in Ecuador. These eventsunderlie many of the RICO and fraud allegations inChevron's complaint. "A substantial need for workproduct materials exists where the information sought is'essential' to the party's defense, is 'crucial' to thedetermination of whether the defendant could be heldliable for the acts alleged, or carries great probative valueon contested issues."239 Chevron is in substantial need ofdocuments concerning these events, and it is unlikely toobtain them from other sources, for reasons previouslydiscussed.

238 For example, the Cabrera report, or most of


it, was written by Stratus, which is located inDenver. Donziger, a central figure, is based inNew York.239 Nat'l Cong. for Puerto Rican Rights v. Cityof New York, 194 F.R.D. 105, 110 (S.D.N.Y.2000) (internal citations omitted) (plaintiffestablished substantial need where withhelddocuments contained information that was"directly relevant to plaintiffs' claims").

PB contends, however, that [*102] Chevron hasalready received "millions of document pages" from theSection 1782 proceedings, Donziger, and the third partiesit has subpoenaed in this action and does not need furtherdiscovery from PB.240 Moreover, many of the events thatChevron contends were tainted by fraud -- e.g., thewriting and filing of the Calmbacher and Cabrera reports-- occurred before PB was brought on in early 2010. PBargues that Chevron does not need from PB documentsabout events that preceded its involvement in the case.The Court disagrees.

240 DI 718, at 22.

While it is true that Chevron has received asubstantial number of documents from Donziger andothers in the Section 1782 proceedings, the documents ithas obtained do not tell the whole story of what went onin Ecuador and elsewhere leading to the Judgment.Understanding that entire story is vital to the resolution ofthis case. Moreover, as noted above, it is clear thatChevron will be unable to obtain these documents fromthe people who were on the ground in Ecuador -- e.g.,Fajardo, Prieto, Saenz, and others--and were directlyinvolved in orchestrating these events. Defendants haverefused to produce documents or other information fromtheir Ecuadorian [*103] agents.

And, although PB was not yet on the LAP teamwhen the Calmbacher and Cabrera reports were writtenand submitted, PB was brought on specifically to dealwith their fallout. Indeed, PB worked with the Ecuadorianlawyers, to whom it has referred as its "Ecuadorianco-counsel,"241 to come up with a strategy to delayproduction from Stratus and to cleanse the Cabrerareport. PB drafted the Fajardo declarations that weresubmitted to the Colorado and Ecuadorian courts. PBgave the marching orders to the new experts to "subtlyaddress" Cabrera's findings. And PB drafted the finalpost-trial briefs the LAPs submitted to the Lago Agriocourt. PB undoubtedly has documents concerning these

events and those they were intended to "cleanse." AsChevron is unable to obtain them from the LAPs'Ecuadorian counsel, it needs them from PB.

241 DI 718, at 10.

Second, sight cannot be lost of the fact that PB is analleged co-conspirator in this case. Chevron alleges thatPB "developed the RICO Defendants' strategy forpursuing the assets of Chevron and its subsidiaries aroundthe world on the basis of the fraudulent judgment inEcuador, and has also been instrumental in the cover-upand obstruction [*104] of Chevron's U.S. discoveryproceedings."242 In order for Chevron to link PB to thealleged conspiracy, which could prove essential to holdthe defendants responsible for its actions, it must shownot only that there was a conspiracy, but that PBknowingly and intentionally acted to further theconspiracy's aim. While inferences of PB's knowledgeand involvement -- particularly with respect to theFajardo Declaration -- can be drawn from certain emailsand documents produced thus far, PB is in the uniqueposition to provide documents that will shed further lighton the extent -- if any -- of PB's involvement in thealleged conspiracy.

242 ¶18(s).

Chevron has shown that it has a substantial need forthe documents responsive to the remaining specificationsof the Subpoena with which the Court has concluded thatPB should comply and that Chevron cannot obtain theirequivalent elsewhere without undue hardship. As apractical matter, therefore, the question comes down towhether there will be substantial issues as to the secondcategory of work product -- so called "opinion" workproduct -- if the Subpoena is enforced. That brings us tothe crime-fraud exception.

(c) The Crime-Fraud Exception

As the [*105] Court already has noted, Chevron hasestablished that "a prudent person ha[s] a reasonablebasis to suspect the perpetration or attempted perpetrationof a crime or fraud"243 -- in other words, probable cause-- in the following respects:

243 In re Grand Jury Subpoena Duces TecumDated Sept. 15, 1983, 731 F.2d at 1039.

First, the Guerra declaration, in and of itself,


establishes probable cause to suspect that the LAPs'representatives, including Fajardo, Donziger, and Yanza,bribed Judge Zambrano to obtain their desired result inthe Ecuadorian case and the privilege of writing theJudgment and that they took advantage of the latter. Thelatter point, moreover, is supported by abundant evidencethat portions of the Judgment are identical orsubstantially similar to internal documents prepared bythe LAPs that never were filed with the Court.

Second, as explained above, additional evidenceestablishes, at a minimum, probable cause to suspect thata crime or fraud occurred with respect to

o the Calmbacher report,

o the termination of the judicialinspection process, and

o the selection and appointment ofCabrera, the preparation and submission ofhis report to the Lago Agrio court, and itspresentation [*106] as his independentwork.

Third, the evidence concerning the representationsmade to the United States District Court for the Districtof Colorado in an effort to prevent the disclosure of theStratus documents that confirmed that Stratus had writtenall or most of the Cabrera report -- most importantly theFajardo declaration filed there and in many other courts --establishes probable cause to suspect that the LAPscommitted wire fraud and obstructed justice in thatrespect.

To be sure, PB contends otherwise. But itsarguments are speculative and unpersuasive when onebears in mind the relevant legal standard that governs thefirst prong of the crime-fraud analysis -- whetherChevron has provided a factual basis that would "strike aprudent person as constituting a reasonable basis tosuspect the perpetration or attempted perpetration of acrime or fraud."244 Chevron has done so, and neither theLAPs nor PB have presented any compelling evidence orarguments to the contrary.

244 Jacobs, 117 F.3d at 87 (internal citation andquotation marks omitted).

This leaves one significant question with respect to

the existence of probable cause -- whether there is asufficient basis to suspect that there [*107] was criminalor fraudulent activity with regard to the cleansing reports.

Chevron's fundamental contention is that thecleansing reports were developed at PB's instance incollaboration with the Weinberg Group as a means "tobackdoor Cabrera's findings into the Ecuadorianrecord"245 in the guise of new expert reports not subjectto the attack that had been made on Cabrera. It points tothe fact that the new experts developed no newsubstantive analysis, were instructed to rely on theCabrera report, and were not told that the Cabrera reporthad been written, at least in major part, by the LAPs andStratus. The point of the exercise, as described in anemail by a PB lawyer, was to "address Cabrera's findingsin such a subtle way that someone reading the new expertreport . . . might feel comfortable concluding that certainparts of Cabrera are a valid basis for damages."246

245 DI 541, at 12.246 See id. (quoting Hendricks Decl. [DI 36] Ex.214).

PB, for its part, says that nothing in the preparation,submission or subsequent litigation about the cleansingreports can be regarded as fraudulent. The LAPs soughtleave of the Lago Agrio court to submit additional reportsto show that the theories [*108] advanced in the Cabrerareport, whatever its authorship, were sound. While fourof the reports relied on data from the Cabrera report, thatreliance was fully disclosed.

As an initial matter, PB is wrong in asserting thatthere was nothing that can be regarded as fraudulent withrespect to the submission of the cleansing reports. Asnoted previously, the Fajardo declaration that wassubmitted to the Lago Agrio court in support of theapplication for leave to submit those reports wasdeceptive, whether or not PB knew it at the time. Asnoted earlier, "[t]here is at least probable cause to suspectthat Cabrera was handpicked by the Lago Agrio plaintiffsbecause he would "play ball" with them, that the entirereport was planned and written by the LAPs and Stratus,and that Cabrera "played ball" by simply affixing hisname to it, acting all the while under the pretense --fostered by the LAPs -- that the report was Cabrera'sindependent work." None of that was disclosed to thecourt. So there is probable cause to suspect that thereports got in through courthouse door as a result offraud.


That said, the fact remains that Chevron has notpointed to anything in the cleansing reports themselves[*109] that appears to have been fraudulent. It may bethat the experts would not have reached the sameconclusions, or would have declined their engagementsaltogether, if they had been told the full truth about therelationship among the LAPs, Stratus, and Cabrera andhis report. But their reliance on Cabrera in most caseswas disclosed in their reports.

In the last analysis, then, the Court concludes that, onthe present record, there is insufficient factual basis tosuspect that the cleansing reports themselves werefraudulent, whatever may have been done to convince theEcuadorian court to permit their filing.

* * *

In sum, the likelihood of privilege and work productissues so substantial as to defeat the attempt to enforcethe Subpoena is greatly limited. PB has not shown anymaterial likelihood of the existence of any significantnumber of responsive documents as to which theattorney-client privilege could apply. Insofar as theSubpoena seeks "ordinary" work product, Chevron hasovercome the qualified protection that enjoys, at leastwith respect to responsive documents going to thesubjects as to which it has established probable cause. Ithas satisfied the first of the two prong test [*110] thatgoverns the crime-fraud exception as to a number of keysubjects. This leaves determination with respect todocuments concerning those subjects only the questionwhether particular responsive documents -- regardless ofwhether they include so-called "opinion" work productor, less likely, attorney-client communications -- were infurtherance of a crime or fraud. And even this issue canbe further limited by a further narrowing of theSubpoena.

3. Further Limitation of the Subpoena

As noted above, Friedman requires that a Courtconfronted with an effort to obtain discovery from anattorney involved in litigating against the discoveringparty consider the need for the discovery. Similarly,Federal Rule of Civil Procedure 26(b)(2)(C) permits acourt to limit the frequency or extent of discoveryotherwise allowed if it determines that (1) the discoverysought is unreasonably cumulative or duplicative, ormore readily obtainable from another source; (2) theparty seeking discovery already has had ample

opportunity to obtain the information sought; or (3) theburden or expense of the proposed discovery outweighsits likely benefit.247

247 FED. R. CIV. P. 26(b)(2)(i)-(iii).

Here, the Subpoena [*111] seeks discovery which,although limited already to a considerable degree,touches on a variety of topics. In view of the desirabilityof focusing only on that which is most important, oflimiting the discovery required from these adversaryattorneys, and of avoiding unduly complicated or timeconsuming privilege issues to the extent that would befair and reasonable, the Court has concluded that theSubpoena should be further limited, at least at the presenttime. Chevron has established probable cause -- and thussatisfied the first of the two prong test that governs thecrime-fraud exception to attorney-client privilege andwork product protection -- as to six subjects:

o The alleged bribery of the Ecuadorianjudge and the writing of the Judgment andother judicial documents in the LagoAgrio case.

o The claim that the LAPs wrote thereports submitted over Calmbacher'ssignature and affixed signature pages,knowing that the reports did not reflect hisviews.

o The circumstances in which theLago Agrio court terminated the judicialinspection process.

o The selection and appointment ofCabrera, the preparation and submission ofhis report to the Lago Agrio court, and itspresentation as [*112] his independentwork.

o The submission of deceptiveaccounts of the LAPs' and Stratus'relationship with Cabrera in the District ofColorado and elsewhere in Section 1782proceedings.

Accordingly, the Subpoena in substance will be limitedfurther by confining its analysis, at least for now, to thosespecifications that seek documents relating to theforegoing subjects. The Court now proceeds to consider


the remaining aspects of the Friedman analysis248 andthe remaining arguments with respect to the Subpoena asthus effectively limited.

248 This involves consideration only of theremaining specifications, which are: 2, 14, 18, 19,21, 22, 26, 28 through 32, 35, 49, 55 though 58,as modified below.

B. PB's Role

Friedman directs the Court to focus on the role of thelawyer from whom discovery is sought both in thelitigation in question and in relation to the subjects onwhich disclosure is sought. "The first of theseconsiderations bears on the extent to which the discoverywould disrupt the litigation by injecting one of thelawyers charged with its conduct into the case as awitness or by making the advocate's conduct orknowledge an issue in the proceeding."249 The second"goes at least [*113] in part to the issue whether thelawyer is likely to have first-hand evidence that isimportant to the resolution of the lawsuit."250 Both ofthese factors cut in favor of requiring PB to produce thedocuments requested in the Subpoena as modified thusfar.

249 In re Chevron Corp., 749 F. Supp. 2d 141,at 163 (S.D.N.Y. 2010).250 Id.

(1) Whether Discovery Would Disrupt the Litigation

First, enforcement of what remains of the Subpoenawould not inject a lawyer charged with the conduct ofthis case as a witness. PB has deliberately avoidingappearing before this Court in this case, which isconducted on behalf of the LAP Representatives by othercounsel.

Second, enforcement of what remains would notplace an advocate's conduct or knowledge in issue. PB isnamed in the amended complaint as an allegedco-conspirator. Its actions and knowledge will be issuesin this case regardless of whether the Subpoena isenforced.

(2) Whether PB Likely Has Relevant Evidence

Given the nature and extent of PB's involvement, itdoubtless has firsthand knowledge that is important to theresolution of this lawsuit. PB was brought on to the LAP

team one year before Zambrano issued the Judgment thatallegedly was written [*114] by the LAPs. PB wasinvolved in hiring and overseeing the cleansing experts --whose reports Judge Zambrano ultimately claimed tohave relied upon -- and it took the lead on drafting thefinal brief filed with the court. PB worked with the LAPsto draft the Fajardo Declaration, which was intended todelay discovery and convince courts across the countrythat the LAPs' involvement in the judicial inspectionprocess and Cabrera report was entirely proper. And PBis leading the LAPs' efforts to enforce the Judgmentthroughout the world. Given PB's role as strategist sincethe time it became involved and as architect of the LAPs'closing arguments and submissions made to the LagoAgrio court, this Court has every reason to expect that PBhas documents that bear directly on the question whetherthe Judgment was obtained by fraud, includingdocuments containing statements by other personsdirectly involved in events at which PB was not present.

C. The Need for Discovery from Patton Boggs and theExtent of Discovery Already Conducted

The last two Friedman factors focus on the need toobtain the discovery from PB and the extent to which thatdiscovery has already been conducted. As the extent[*115] of discovery already conducted informs the degreeto which Chevron needs discovery from PB, the Courtanalyzes these factors together.

Chevron contends that it needs discovery from PBbecause PB "likely possesses critical relevant documentsthat Chevron cannot obtain from any other sourcesprecisely because PB played a key role in planning,executing, and perpetuating the fraud againstChevron."251 PB claims that Chevron does not needdiscovery from PB because (1) it already has amassed anenormous amount of discovery, and (2) "it is in the midstof a discovery campaign in this action that is largelycumulative and duplicative of what Chevron hasrequested from" PB.252

251 DI 713, at 19.252 DI 665, at 12.

PB is correct that, to the extent Chevron can obtainor has obtained documents from the third parties it hassubpoenaed, it does not need them from PB. But it isdoubtful that these parties will provide Chevron with thediscovery it seeks from PB. None of the parties has beennearly as involved in this overall dispute since early 2010


as PB, which has had a hand in almost every majordevelopment in this action and related ones since it joinedthe LAP team. And, as previously described, [*116]although Chevron has obtained a substantial number ofdocuments from Donziger and others in the Section 1782proceedings, it has been unable to obtain any documentsfrom PB's "co-counsel in Ecuador"-- documents whichare crucial to this case -- except to the limited extent thatthey were emails or other communications produced byDonziger. Substantial information from andcommunications with those co-counsel and non-lawyersin Ecuador -- emails, correspondence, and memoranda, ifnot more -- are likely to be in PB's files.

Especially in light of defendants' obstinate refusal toprovide Chevron with discovery from Ecuador, Chevronhas shown that it needs discovery from PB.

* * *

In sum, the Friedman factors cut strongly in favor ofrequiring this limited production from PB, particularly inlight of the fact that PB has never appeared before thisCourt.

II. Alleged Undue Burden and Cost Shifting

In a recent submission to the Court, PB estimatedthat "it will take between 30 and 40 weeks to completethe review, production, and privilege logging of the emailand non-email [electronically stored information]"required to comply with the Subpoena, even as modifiedby the September 2012 hearing and [*117] theNovember Order.253 Moreover, PB contends thatChevron should bear the costs associated with PB'sefforts to comply with the Subpoena because PB "is not aparty to this action and does not have a direct interest inthe outcome of the RICO action."254

253 DI 847, at 2-3. Chevron submitted aresponse on March 7, 2013. Given that the Courtdid not ask for or permit a response fromChevron, and that Chevron waited nearly twoweeks to submit it, the Court does not considerChevron's response, nor any submissions made inreply to it.254 DI 527, at 30-31.

PB has not sustained its burden of persuasion.

As an initial matter, PB is no ordinary, unrelated

non-party witness. It is an alleged co-conspirator andsome of its actions are at issue in this case regardless ofwhether the Subpoena as narrowed is enforced.Moreover, it stands to reap a fee that has been estimatedat hundreds of millions of dollars if the Judgment isenforced and collected. Like any lawyer's contingent feematter, whether the contingency is all or just part of thecompensation arrangements, certain investments of timeand money are necessary in order to obtain the potentialbenefit of a successful outcome. PB's attempt to [*118]portray itself as a nonparty with no interest in the matteris unsupportable.

Second, PB has overstated the burden of compliance,in terms both of the cost and the required time and hasavoided engaging with options that give strong promiseof reducing that cost and burden. For example:

o At the September 2012 hearing, theCourt urged the parties to analyze, in theirsubsequent submissions with respect toburden, whether and to what extentpredictive coding255 could "reduce theburden and effort" required to complywith the Subpoena.256 Apart from onefootnote,257 PB's submission ignored thesubject entirely. The logical inference isthat PB failed to address the subjectbecause it would not have aided itsargument.

o PB's cost and time estimatespresuppose that every document would bereviewed at least twice by "Patton Boggsattorneys"258 and that it would take 15 to20 lawyers working an average of 40hours per week approximately 40 weeks tocomplete the review. But the Court sees nolegitimate reason why (1) far less costlycontract attorneys could not do all or mostof the review, as is common in the legalcommunity today, (2) two or certainlythree levels of review are necessary,[*119] or (3) more reviewers could not beused.

255 Predictive coding is an automated methodthat credible sources say has been demonstrated toresult in more accurate searches at a fraction of


the cost of human reviewers. See e.g., Moore v.Publicis Groupe, 287 F.R.D. 182, 191 (S.D.N.Y.2012) adopted sub nom. Moore v. PublicisGroupe SA, 11 Civ. 1279 ALC AJP, 2012 U.S.Dist. LEXIS 58742, 2012 WL 1446534 (S.D.N.Y.Apr. 26, 2012) ("Computer-assisted reviewappears to be better than the available alternatives,and thus should be used in appropriate cases.");DI 713, at 9; Joe Palazzolo, How a Coputer Didthe Work of Many Lawyers, WSJ LAW BLOG,Jan. 17, 2013.256 Tr., Sept. 27, 2012, at 138:24-139:7.257 DI 665, at 9 n.10 ("the use of predictivecoding is still under investigation").258 DI 847, at 2-3. It estimates that 15 percent ofthe documents would be reviewed a third time by"senior attorneys." Id. at 3.

Third, all of the cost and time estimates upon whichPB relies antedate the further and substantial narrowingof the Subpoena effected by this opinion. They areobsolete.

But putting all that aside, PB's last estimate of thecost of reviewing and logging the documents in order tocomply with the Subpoena was in the range of [*120]$1,060,000 and $1,290,000.259 Even assuming that wereaccurate today, and the Court does not so assume, theburden on PB would not be undue given its role in thiscase, its size (reportedly over $300 million in grossrevenues in 2012), and its economic interest in thiscontroversy. Indeed, there is no persuasive evidence thatthe compliance costs are out of line with what would betypical for nonparty witness in complex commerciallitigation.

259 DI 665, at 2-3.

Nor is the Court persuaded that any part of the costshould be shifted to Chevron. Many of the events inwhich it was involved underlie Chevron's mainallegations. "Where a nonparty was substantiallyinvolved in the underlying transaction and could haveanticipated that [it] would reasonably spawn somelitigation, expenses should not be awarded."260 Here, PBwas well aware of Chevron's fraud allegations when itjoined the LAP team -- indeed it was brought on tocombat them -- and understood Chevron's intention tofight this matter vigorously. Any failure to haveanticipated that its involvement could lead to discoveryobligations and expenditures on its own behalf, if there

was such a failure, would have reflected an uncommonlack [*121] of foresight.261

260 In re First Am. Corp., 184 F.R.D. 234, 242(S.D.N.Y. 1998) (internal citation and quotationmarks omitted).261 Furthermore, PB has initiated three separatesuits against Chevron. Where "nonparties wereinvolved in litigation arising out of the same facts,courts have viewed such parties as 'not neutral' forpurposes of awarding costs." Id. (internal citationand quotation marks omitted).

Conclusion

PB shall produce documents responsive to thefollowing specifications, as modified by the November2012 Order: 2, 14, 18, 19, 21, 22, 26, 28 through 32, 35,49, and 55 though 58 and, in addition, specification 20insofar as it seeks documents described in the margin.262

262 Specification 20 is modified to seek "AllDOCUMENTS or draft DOCUMENTS or otherwritings filed with the LAGO AGRIO COURTunder the signature of or in the nameCABRERA." Thus, PB is being compelled tocomply with only 19 of the Subpoena's 58specifications.

To the extent PB claims that any documentsresponsive to these requests are protected by theattorney-client privilege or the protection afforded to"opinion" work product, such claims shall be asserted inconformity with S.D.N.Y. Local Civil Rule 26.2. [*122]To the extent PB claims that any documents responsive tothese requests constitute fact work product, Chevron hasovercome its burden, and PB must produce them.

Production both of documents and the privilege logshall take place on a rolling basis commencing no laterthan March 28, 2013, with continuing production of eachto occur no less than weekly. The Court recognizes thatthere is some uncertainty in present circumstances as tohow quickly compliance reasonably can be achieved. Forthe present, the complete privilege log shall be due andproduction of all responsive documents not scheduled onthe privilege log shall be completed on or before May 1,2013. While the Court will consider a well supportedrequest for additional time, any motion for an extensionof the May 1 date shall made no later than April 14, 2013.


SO ORDERED.

Dated: March 15, 2013

/s/ Lewis A. Kaplan

Lewis A. Kaplan

United States District Judge


UNITED STATES DISTRICT COURTNORTHERN DISTRICT OF INDIANA

SOUTH BEND DIVISION

IN RE: BIOMET M2a MAGNUM HIP )IMPLANT PRODUCTS LIABILITY )LITIGATION (MDL 2391) ) CAUSE NO. 3:12-MD-2391

))

)This Document Relates to All Cases ) )

ORDER REGARDING DISCOVERY OF ESI

Biomet has produced 2.5 million documents to plaintiffs in this docket’s

constituent cases, and the Plaintiffs’ Steering Committee believes production

should run to something closer to 10 million documents. The parties have set

forth their positions on the procedures or protocols that should be used to

facilitate identification, retrieval, and production of electronically stored

information in submissions filed on April 1 and 5. The parties seek my guidance

as to the direction discovery of ESI should take, and I believe the parties need a

prompt ruling more than they need extensive discussion of each point they raise.

Biomet began producing documents in cases eventually centralized here in

the summer of 2012. Some plaintiffs’ counsel, anticipating this docket’s formation,

told Biomet (occasionally in forceful terms) not to begin document production until

the Judicial Panel on Multidistrict Litigation decided whether to centralize.

Biomet, neither sold on centralization nor free of judicial exhortations in other

cases against it, started the process of identifying and producing documents.

Page 1 of 7

Biomet used a combination of electronic search functions to identify

relevant documents. Keyword culling was used first, reducing the universe of

documents and attachments from 19.5 million documents to 3.9 million

documents, comprising 1.5 terabytes of data. Removal of duplicates left 2.5

million documents and attachments. Statistical sampling tests of a random

sample projected, with a 99 percent confidence rate, that between .55 and 1.33

percent of the unselected documents would be responsive and (with the same

confidence level) that between 1.37 and 2.47 percent of the original 19.5 million

documents were responsive. In comparison, Biomet’s keyword/deduplication

approach had identified 16 percent of the original 19.5 million.

Biomet then employed technology-assisted review, or predictive coding, to

identify the relevant documents to be produced from the 2.5 million that emerged

from the keyword and deduplication processes. Predictive coding has found many

uses on the Internet. Under predictive coding, the software “learns” a user’s

preferences or goals; as it learns, the software identifies with greater accuracy just

which items the user wants, whether it be a song, a product, or a search topic.

Biomet used a predictive coding service called Axelerate and eight contract

attorneys to review a sampling of the 2.5 million documents. After one round of

“find more like this” interaction between the attorneys and the software, the

contract attorneys (together with other software recommended by Biomet’s e-

discovery vendor) reviewed documents for relevancy, confidentiality, and privilege.

Page 2 of 7

To date, Biomet’s e-discovery costs are about $1.07 million and will total

between $2 million and $3.25 million.

Biomet invited the Plaintiffs’ Steering Committee to suggest additional

search terms and offered to produce the rest of the non-privileged documents from

the post-keyword 2.5 million so the Steering Committee can verify that Biomet is

producing the relevant documents. The Steering Committee has declined those

offers, believing they are too little to assure proper document production.

The Steering Committee contends Biomet’s initial use of the keyword

approach has tainted the process. They point to a recent article that mentioned

unidentified “literature stating that linear review would generate a responsive rate

of 60 percent and key word searches only 20 percent, and [the defendants in the

case being discussed] proposed that predictive coding at a 75 percent responsive

rate would be sufficient.” Barry Kazan and David Wilson, TECHNOLOGY-ASSISTED

REVIEW IS A PROMISING TOOL FOR DOCUMENT PRODUCTION, New York Law Journal

(Mar. 18, 2013). The Steering Committee sees Biomet’s approach as insufficient

because, although it employed predictive coding, Biomet began with the less

accurate keyword search. The Steering Committee sees Biomet’s offer to let the

Steering Committee propose search terms as unhelpful because the Steering

Committee’s unfamiliarity with Biomet terminology prevents them from making

suggestions advisedly.

The Steering Committee wants Biomet to go back to its 19.5 million

documents and employ predictive coding, with plaintiffs and defendants jointly

Page 3 of 7

entering the “find more like this” commands. Biomet objects on a variety of

grounds, including its estimate that virtually starting over would cost it millions

more than the millions it already has spent in document production. The Steering

Committee responds that Biomet gambled when it spent millions on document

production that several of plaintiffs’ counsel warned Biomet not to undertake until

the Panel had centralized the cases.

The issue before me today isn’t whether predictive coding is a better way of

doing things than keyword searching prior to predictive coding. I must decide

whether Biomet’s procedure satisfies its discovery obligations and, if so, whether

it must also do what the Steering Committee seeks. What Biomet has done

complies fully with the requirements of Federal Rules of Civil Procedure 26(b) and

34(b)(2). I don’t see anything inconsistent with the Seventh Circuit Principles

Relating to the Discovery of Electronically Stored Information. Principle 1.02

requires cooperation, but I don’t read it as requiring counsel from both sides to

sit in adjoining seats while rummaging through millions of files that haven’t been

reviewed for confidentiality or privilege. Both sides cite reports from the Sedona

Conference project, e.g., The Sedona Conference, The Sedona Conference

Commentary on Proportionality in Electronic Discovery (Jan. 2013); The Sedona

Conference, The Sedona Conference Best Practices Commentary on the Use of

Search and Information Retrieval Methods in E-Discovery, 8 Sedona Conf. J. 189

(2007); and The Sedona Conference, Conducting E-Discovery After Amendments:

Page 4 of 7

The Second Wave, 10 Sedona Conf. J. 215 (2009), and I don’t see Biomet’s

approach as running afoul of any of the principles set forth in those publications.

In contrast, the Steering Committee’s request that Biomet go back to Square

One (more accurately Square Two, since Biomet first collected the 19.5 million

documents) and institute predictive coding at that earlier stage sits uneasily with

the proportionality standard in Rule 26(b)(2)(C). Doing so would entail a cost in

the low seven-figures. The confidence tests Biomet ran as part of its process

suggest a comparatively modest number of documents would be found. The

Steering Committee challenges that conclusion by pointing to studies (one in

1985) indicating that, on average, Boolean searches identify less than a quarter

of the relevant documents in a set of documents. Boolean language provides the

basis for keyword searches, though I can’t find anything in this record that

equates today’s keyword searches to Boolean searches. In contrast, the Steering

Committee says predictive coding identified 75 to 95 percent of the relevant

documents — about four times more efficient than keyword searches. The 75

percent figure appears to come from the previously-cited recent New York Law

Journal article about technology-assisted review as part of document production.

The article itself doesn’t vouch for the accuracy of the 75 percent figure; the article

simply notes that in a Virginia state court case, Global Aerospace v. Landow

Aviation, No. CL 61040 (Va. Cir. Ct., Loudon County, Apr. 23, 2012), the

defendants “proposed that predictive coding at a 75 percent responsive rate would

be sufficient.” I can find no source for the 95 percent figure.

Page 5 of 7

It might well be that predictive coding, instead of a keyword search, at Stage

Two of the process would unearth additional relevant documents. But it would

cost Biomet a million, or millions, of dollars to test the Steering Committee’s

theory that predictive coding would produce a significantly greater number of

relevant documents. Even in light of the needs of the hundreds of plaintiffs in this

case, the very large amount in controversy, the parties’ resources, the importance

of the issues at stake, and the importance of this discovery in resolving the issues,

I can’t find that the likely benefits of the discovery proposed by the Steering

Committee equals or outweighs its additional burden on, and additional expense

to, Biomet. FED. R. CIV. P. 26(b)(2)(C).

The Steering Committee appears to argue that Biomet is estopped from

relying on proportionality arguments based on the incremental cost of what the

Steering Committee seeks because Biomet embarked on its document

identification in disregard of pre-centralization warnings and advice from some

counsel for plaintiffs in individual cases. It might be that the Steering Committee’s

argument could carry the day in some cases, but this one doesn’t seem to be such

a case. The Steering Committee hasn’t argued (and I assume it can’t argue) that

Biomet had no disclosure or document identification obligation in any of the cases

that were awaiting a ruling on (or even the filing of) the centralization petition.

Until the MDL Panel enters a centralization order under 28 U.S.C. § 1407 (or

transfers a tag along pursuant to an earlier centralization order), a transferee

court is free to act on pending matters. Indeed, through its conditional transfer

Page 6 of 7

orders, the Panel regularly encourages transferee courts to do so. To hold that a

party that behaves as the transferee court directs, or that follows the transferee

court’s standing procedures, does so only by forfeiture of the proportionality

provision of Rule 26(b)(2)(C), seems an uncongenial exercise of whatever discretion

I have. It also would seem inconsistent with the purposes of centralization under

§ 1407.

In making this ruling, I assume that Biomet will remain open to meeting

and conferring on additional reasonably-targeted search terms and to producing

the non-privileged documents included in the statistical sample. Beyond that, if

the Steering Committee wishes production of documents that can be identified

only through re-commenced processing, predictive coding, review, and production,

the Steering Committee will have to bear the expense.

ENTERED: April 18, 2013

/s/ Robert L. Miller, Jr. Judge, United States District CourtNorthern District of Indiana

Page 7 of 7