734 - Addiction Incorporated

734 84 FEDERAL REPORTER, 3d SERIES

Dianne CASTANO, et al .,Pla intiffs-Appellees,

V .

The AMERICAN TOBACCO COMPANY,et al., Defendants-Appellants .

No. 95-30 7 25 .

United States Court of Appeals,Fifth Circuit .

May 23, 1996 .

Motion for class certification wasbrought in action which had been brought onbehalf of all smokers and nicotine dependentpersons and their families against tobaccocompanies. The United States DistrictCourt for the Eastern District of Louisiana,Okla Jones, II, J ., 160 F.R.D. 544, certifiedclass consisting of all nicotine dependent per-sons who have smoked and purchased ciga-rettes manufactured by tobacco companies,and appeal was taken. The Court of Ap-peals, Jerry E. Smith, Circuit Judge, heldthat multistate class would be decertified be-cause federal district court failed to considerhow variations in state law would affect pre-dominance and superiority, district court'spredominance inquiry did not include consid-eration of how trial on the merits would beconducted, and class independently failed thesuperiority requirement .

Reversed and remanded with instruc-tions .

1. Federal Civil Procedure G-171Federal district court must conduct rig-

orous analysis of the prerequisites of numer-osity, commonality, typicality, and adequacyof representation before certifying a class .Fed.Rules Civ.Proc .Rule 23,28 U.S.CA

2. Federal Civil Procedure a162Federal district court's decision to certi-

fy class is within its broad discretion, butthat discretion must be exercised withinframework of rule governing class actions.Fed.Rules Civ.Proc.Rule 23,28 U .S .C .A.

3. Federal Civil Procedure (9-172Party seeking class certification bears

burden of proof. Fed.Rules Civ.Proc.Rule23,28 U .S.CA

4. Federal Civil Procedure (3-182 .5Multistate class consisting of nicotine de-

pendent persons who have purchased andsmoked cigarettes manufactured by tobaccocompanies would be decertified because fed-eral district court failed to consider how vari-ations in state law would affect predominanceand superiority, district court's predominanceinquiry did not include consideration of howtrial on the merits would be conducted, andclass independently failed to satisfy superior-ity requirement . Fed.Rules Civ.Proc.Rule23, 28 U .S.CA

5. Federal Civil Procedure a173Although rule governing class actions

requires that class be eertified as soon aspracticable and allows court to certify a con-ditional class, it does not follow that therule's requirements are lessened when theclass is conditional . Fed.Rules Civ.Proc.Rule23(c)(1), 28 U.S .C .A .

6 . Federal Civil Procedure e-165In multistate class action, variations in

state law may swamp any common issues anddefeat predominance. Fed.Rules Civ.Proc.Rule 23,28 U.S.CA

7. Federal Civil Procedure e-1612,165Federal district court's duty to deter-

mine whether plaintiff has borne its burdenon class certification requires that court con-sider variations in state law when class actioninvolves multiple jurisdictions ; court mustconsider how variations in state law affectpredominance and superiority. Fed.RulesCiv.Proc .Rule 28,28 U.S.CA

8. Federal Civil Procedure e-173To make findings required to certify a

class action, one must initially identify thesubstantive law issues which will control out-come of the litigation . Fed.Rules Civ .Proc .Rule 23(b)(3), 28 U .S .C .A .

9. Federal Civil Procedure C=172Given plaintiffs' burden of establishing

class certification, court cannot rely on assur-

anctdomFed

10.

pleamer:beeinec(clai ;bleinglfsuecU.S

11 .

hav ;prolfactallorab;to 1U.S

689643037PRODUCED FROM B&W WEB SITEhttp://legacy.library.ucsf.edu/tid/xrw91d00/pdf

CASTANO v. AMERICAN TOBACCO CO .Cite as 84 F .3d 734 (5th Cir. 1996)

ances of counsel that any problems with pre-dominance or superiority can be overcome.Fed.Rules Civ.Proc.Rule 23, 28 U .S .C .A .

10 . Federal Civil Procedure «l72Federal district court may look past the

pleadings to determine whether require-ments of rule governing class actions havebeen met; going beyond the pleadings isnecessary as court must understand theclaims, defenses, relevant facts, and applica-ble substantive law in order to make mean-ingful determination of class certification is-sues. Fed.Rules Civ .Proc.Rule 23, 28U.S.C .A .

11 . Jury C=37Seventh Amendment entitles parties to

have fact issues decided by one jury andprohibits second jury from reexamining thosefacts and issues and thus, the Constitutionallows bifurcation of issues that are so sepa-rable that second jury will not be called uponto reconsider findings of fact by the first .U.S.C.A. Const.Amend . 7.

Bettye A. Barrios, Johnson, Johnson, Bar-rios & Yacoubian, New Orleans, LA, MelvinM. Belli, San Francisco, CA, Joseph M. Bru-no, Bruno & Bruno, New Orleans, LA, Ken-neth M. Carter, New Orleans, LA, Bruce C .Dean, New Orleans, LA, Wendell H . Gauthi-er, Daniel G. Abel, Dana Kim Cormier, JulieB. Beiser, Gauthier & Murphy, Metairie, LA,Christopher M. Guidroz, New Orleans, LA,John B. Krentel, Metairie, LA, Walter J .Leger, Jr., New Orleans, LA, Arthur R . Mil-ler, Cambridge, MA, Ronald L . Motley,Charleston, SC, Stephen B . Murray, NewOrleans, LA, Charles W . Patrick, Jr .,Charleston, SC, Robert Leland Redfearn,Jr., New Orleans, LA, Michael X. St. Martin,Houma, LA, Scott McCullen Baldwin, JohnBrowning Baldwin, Baldwin & Baldwin, Mar-shall, TX, Calvin Clifford Fayard, Jr ., Den-ham Springs, LA, Robert D . Greenbaum,Philadelphia, PA, George Febiger Riess,New Orleans, LA, Peter J . Butler, Jr.,Deutsch, Kerrigan & Stiles, New Orleans,LA, Andrew W . Hutton, Wichita, KS, WellsTalbot Watson, Lake Charles, LA, RichardM. Heimann, Lieff, Cabraser, Heimann,Bernstein, San Francisco, CA, Ralph Irving

735Knowles, Jr ., . Atlanta, GA, Arnold Levin,Levin, Fishbein, Sedran and Berman, Phila-delphia, PA, John R . Climaco, Climaco, Cli-maco, Seminatore, Lefkowitz amd Garofoli,Cleveland, OH, Jodi W. Flowers, Susan Nial,Ness, Motley, Loadholt, Richardson & Poole,Charleston, SC, Russ M. Herman, Herman,Herman, Katz & Cotlar, New Orleans, LA,Francis "Brother" Hare, Jr ., Hare, Wynn,Newell & Newton, Birmingham, AL, GayleL. Troutwine, Michael L : Williams, Williamsand Troutwine, Portlant, OR, John P. Kope-sky, Sheller, Ludwig & Badey, Philadelphia,PA, Stanley M. Chesley, Waite & Schneider,Cincinnati, OH, John P . Coale, Diane E . Coo-ley, Coale, Allen & Van Susteren, Washing-ton, DC, Margaret Moses Branch, BranchLaw Firm, Albuquerque, NM, Perry Weitz,New York City, Louie J. Roussel, III, Me-tairie, LA, Edwin Rene Murray, 'Edwin R .Murray & Associates, New Orleans, LA,Sherrill Patricia Hondorf, Waite, Schneider,Bayless & Chesley, Cincinnati, OH, ElizabethJoan Cabraser, Steven E . Fineman, Lieff,Cabraser, Heimann & Barnstein, San Fran-cisco, CA, Dianne M. Nast, Roda & Nast,Lancaster, PA, Richard Alan Daynard,Northeastern University School of Law, Bos-ton, MA, Jorge Ortiz-Brunet, Ortiz, Toro &Ortiz-Brunet,. Hato Rey, PR, Charles Zim-merman, Zimmerman & Reed, Minneapolis,MN, Jack David Maistros, Michael V . Kelly,Cleveland, OH, Martis Ann Brachtl, BrianCampf, Goodldnd, Labaton, Rudoff & Sucha-row, New York City, Kenneth S . Canfield,Atlanta, GA, Louis Gottlieb, New York City,William O . Dougherty, San Diego, CA, Ed-win David Hoskins, John C .M. Angelos, Of-fice of Peter G. Angelos, Towson, MD, DanielE . Becnel, Jr ., Becnel, Landry & Becnel,Reserve, LA, for plaintiffs-appellees .

Peter J. McKenna, Skadden, Arps, Slate,Meagher & Flom, New York City, KennethWinston Starr, Kirkland & Ellis, Washing-ton, DC, Paul R . Duke, Covington & Burling,Washington, DC, Thomas E . Silfen, LauraJean Hines, Arnold & Porter, Washington,DC, Robert C. Heim, Dechert, Price &Rhoads, Philadelphia, PA, for all defendants-appellants.

Joy Goldberg Braun, Robert E . Winn, Ses-sions & Fishman, New Orleans, LA, Bruce

I

I

ii



'jI1t

G. Shetller, Thomas E . Bezanson, Mary T .Yelenick, Chadbourne & Parke, New YorkCity, for American Tobacco Co. and Ameri-can Brands Incorporated .

Carmelite M. Bertaut, Charles L. Chas-saignac, Peter A. Feringa, Jr., Chaffe,McCall, Phillips, Toler & Sarpy, New Or-leans, LA, for Lorillard, Inc ., R.J. ReynoldsTobacco Co ., Brown & Williamson TobaccoCorp., Batus Holdings, Inc ., and Batus, Inc .

James Thomas Newsom, Shook, Hardy &Bacon, Kansas City, MO, for Lorillard, Inc .,Phillip Morris, Inc., Brown & Williamson To-bacco Corp ., Batus Holdings, Inc., Batus,Inc., Loews Corp ., Philip Morris Companies,Inc., and Lorrillard Tobacco Co .John Mason McCollam, Steven W. Copley,

Gordon, Arata, McCollam & Duplantis, NewOrleans, LA, for Lorillard, Inc., Loews Corp.,and Lorrillard Tobacco Co .Gary R. Long, James A. Wilson, Shook,

Hardy & Bacon, Kansas City, MO, for Loril-lard, Inc ., Phillip Morris, Inc., Brown & Wil-liamson Tobacco Corp., Batus Holdings, Inc .,Batus, Inc., Loews Corp., Philip Morris Com-panies, Inc ., and Lorrillard Tobacco Co .

Allen Rennie Purvis, Kansas City, MO, forPhillip Morris Incorporated .

Scott Edward Delacroix, Charles F. Gay,Jr., Thomas J. Wyllie, Adams and Reese,New Orleans, LA, for Phillip Morris, Inc.,and Philip Morris Companies, Inc.Stephen H . Kupperman, Phillip A. Witt-

mann, Stone, Pigman, Walther, Wittmann &Hutchinson, New Orleans, LA, for RJR Na-bisco, Inc . and R .J. Reynolds Tobacco Co .

Dorothy Hudson Wimberly, Stone, Pig-man, Walther, Wittmann & Hutchinson, NewOrleans, LA, S . Ann Saucer, Powell & Asso-ciates, Dallas, TX, Paul G. Crist, Cleveland,OH, Theodore Martin Grossman, Hugh R .Whiting, Mark A. Belasic, Jones, Day, Reavis& Pogue, Cleveland, OH, for R .J. ReynoldsTobacco Co.

Madeleine M . Fischer, New Orleans, LA,Joseph J. Lowenthal, Jr ., John J. Weigel,Jones, Walker, Waechter, Poitevent, Carrere& Denegre, New Orleans, LA, James V .Kearney, Mudge, Rose, Guthrie, Alexander& Ferdon, New Orleans, LA, Aaron H .

Marks, Kasowitz, Benson, Torres & Fried-man, New York City, for Liggett Group, Inc .,Liggett & Myers, Inc ., and Brooke Group,Ltd .

Francis K . Decker, Jr ., New York City, forLiggett Group Incorporated.

Griffin B. Bell, Richard A. Schneider, King& Spalding, Atlanta, GA, Gordon A. Smith,Atlanta, GA, Steven D . McCormick, MichelleH. Browdy, Andrew R . McGaan, Chicago, IL,David M. Bernick, Chicago, IL, for Brown &Williamson Tobacco Corp.

Charles William Schmidt, III, Christovich& Kearney, New Orleans, LA, for U .S . To-bacco Co., Loews Corp ., and UST, Inc.

Alan H. Goodman, Thomas Mente Benja-min, Lenile & Kelleher, New Orleans, LA,for The Tobacco Institute, Inc.

Linda Susan Mullenix, Austin, TX, Jan S .Amundson, National Association of Manufac-turers, Washington, DC, for National Associ-ation of Manufacturers (NAM) amicus curiae.

Stephen A. Bokat, Washington, DC, RobinS. Conrad, Washington, DC, for Chamber ofCommerce of the U .S . amicus curiae .John H. Beisner, Brian David Boyle, Bar-

ton Samuel Aronson, O'Melveny & Myers,Washington, DC, Hugh F . Young, Jr., Prod-uct Liability Advisory Council, Reston, VA,for Product Liability Advisory Council(PLAC) amicus curiae .

Paul D. Kamenar, Washington LegalFoundation, Washington, DC, Daniel J . Po-peo, Washington, DC, Arvin Maskin, KonradL. Cailteux, John H. Bae, David L . Yohai,Weil, Gotshal & Manges, New York City, forWashington Legal Foundation, amicus curi-ae .

Jack R . Bierig, Bruce M . Zessar, Sidley &Austin, Chicago, IL, for American MedicalAssociation amicus curiae .

Gloria M. Janata, The Children's HealthFund, New York City, for Children's HealthFund amicus curiae.

Katherine J. Pohlman, Edina, MN, for Na-tional Association of School Nurses amicuscuriae .

Matthew L . Myers, Asbill, Junkin &Myers, Washington, DC, for American Heart

1

A:L

anB;of

Cc

Ci:

:evEcot :taiuCa~F.FWA'.Frr(anccou

2 . 1PeA .Pcrerclahu


CASTANO v. AMERICAN TOBACCO CO .Cite as 84 F .3d 734 (5th Cir. 1996)

r

II

Ass'n, American Cancer Soc., and AmericanLung Ass'n, amicus curiae .

Carol L. Galloway, Department of Healthand Hospitals for the State of Louisiana,Baton Rouge, LA, for Louisiana Departmentof Health and Hospitals amicus curiae .

Appeal from the United States DistrictCourt for the Eastern District of Louisiana .

Before SMITH, DUH8 and DeMOSS,Circuit Judges.

JERRY E. SMITH, Circuit Judge:

In what may be the largest class actionever attempted in federal court, the districtcourt in this case embarked "on a road cer-tainly less traveled, if ever taken at all,"Castano v. Ame-r-ican Tobacco Co ., 160F.R.D. 544,' 560 (E.D.La.1995) (citing En-WAR) C. LATHAM, THE POETRY Oh' ROBE!{TFROsr "THE RoAD NOT TAKEN" 105 (1969)),and entered a class certification order. Thecourt defined the class as :

(a) All nicotine-dependent persons in theUnited States . . . who have purchased andsmoked cigarettes manufactured by thedefendants ;(b) the estates, representatives, and ad-ministrators of these nicotine-dependentcigarette smokers ; and(c) the spouses, children, relatives and"significant others" of these nicotine-de-

1 . The court defined "nicotine-dependent" as :

(a) All cigarette smokers who have been diag-nosed by a medical practitioner as nicotine-dependent; and/or(b) All regular cigarette smokers who were orhave been advised by a medical practitionerthat smoking has had or will have adversehealth consequences who thereafter do not orhave not quit smoking .

Id. at 561 . The definition is based upon thecriteria for "dependence" set forth in AMERICANPsticHI, Terc Assocurtos Ducnosrnc .a .vn Sr4itsn( .u.S1 i,r v, rzt ,11~-vrat DisoaWas (4th ed .) .

2 . The original class plaintiffs were Ernest R .Pem, Sr., T. George Solomon, Jr., and DianneA. Castano. The class representatives includePerry, Gloria Scott, and Deania Jackson, all cur-rent cigarette smokers . Dianne Castano is aclass representative on behalf of her deceasedhusband, Peter Castano .

737pendent cigarette smokers as their heirs orsurvivors .

Id. at 560-61 . The plaintiffs limit the claimsto years since 1943 .1

This matter comes before us on interlocu-tory appeal, under 28 U.S .C . § 1292(b), ofthe class certification order . Concluding thatthe district court abused its discretion incertifying the class, we reverse .

I .

A. The Class Complaint

The plaintiffs 2 filed this class complaintagainst the defendant tobacco companies 3and the Tobacco Institute, Inc-, seeking com-pensation solely for the injury of nicotineaddiction. The gravamen of their complaintis the novel and wholly untested theory thatthe defendants fraudulently failed to informconsumers that nicotine is addictive and ma-nipulated the level of nicotine in cigarettes tosustain their addictive nature . The classcomplaint alleges nine causes of action:fraud and deceit, negligent misrepresenta-tion, intentional infliction of emotional dis-tress, negligence and negligent infliction ofemotional distress, violation of state consum-er protection statutes, breach of express war-ranty, breach of implied warranty, strictproduct liability, and redhibition pursuant tothe Louisiana Civil Code .

The plaintiffs seek compensatory 4 and pu-nitive damages 5 and attorneys' fees .6 In

3. The defendant tobacco companies are TheAmerican Tobacco Company, Inc ., R .J . ReynoldsTobacco Company, Brown & Williamson Tobac-co Corporation, Phillip Morris, Inc ., Liggett &Meyers, Inc ., Lorillard Tobacco Company, Inc .,and United States Tobacco Company. Prior tooral argument, Liggett & Meyers, Inc ., filed inthis court a motion conditionally to dismiss,without prejudice, its appeal because of a pend-ing settlement with the plaintiffs . We have de-clined to enter the requested dismissal .

4. The plaintiffs seek compensatory damages forfraud and deceit, negligent misrepresentation, in-tentional infliction of emotional distress, breachof express and implied warranty, strict productsliability, and redhibition .

5 . The plaintiffs seek punitive damages for fraudand deceit, intentional infliction of emotional dis-tress, negligence, and negligent infliction of emo-tional distress .

6. The plaintiffs seek attorne~s' fees fur violationsof consumer protection stanrtes nnd rrdhihiri ..,,



addition, the plaintiffs seek equitable relieffor fraud and deceit, negligent misrepresen-tation, violation of consumer protection stat-utes, and breach of express and implied war-ranty. The equitable remedies include adeclaration that defendants are financiallyresponsible for notifying all class members ofnicotine's addictive nature, a declaration thatthe defendants manipulated nicotine levelswith the intent to sustain the addiction ofplaintiffs and the class members, an orderthat the defendants disgorge any profits'made from the sale of cigarettes, restitutionfor sums paid for cigarettes, and the estab-lishment of a medical monitoring fund.

The plaintiffs initially defined the class as"all nicotine dependent persons in the UnitedStates," including current, former and de-ceased smokers since 1943 . Plaintiffs con-ceded that addiction would have to be provenby each class member; the defendants ar-gued that proving class membership will re-quire individual mini-trials to determinewhether addiction actually exists .

In response to the district court's inquiry,the plaintiffs proposed a four-phase trialplan.7 In phase 1, a jury would determinecommon issues of "core liability." Phase 1issues would include a(1) issues of law, . andfact relating to defendants' course of conduct,fraud, and negligence liability (includingduty, standard of care, misrepresentationand concealment, knowledge, intent) ; (2) is-sues of law and fact relating to defendants'alleged conspiracy and concert of action ; (3)issues of fact relating to the addictive na-tureldependency creating characteristics andproperties of nicotine; (4) issues of fact relat-ing to nicotine cigarettes as defective prod-ucts; (5) issues of fact relating to whetherdefendants' wrongful conduct was intentional,reckless or negligent; (6) identifying whichdefendants specifically targeted their adver-tising and promotional efforts to particular

7. The district court did not adopt the plaintiffs'trial plan, but its order certifying the class incor-porates many elements of it.

S. For purposes of clarity, those issues that thedistrict court did not certify as common havebeen left out of this summary of the plaintiffs'trial plan .

I

groups (e .g. youths, minorities, etc .) ; (7)availability of a presumption of reliance ; (8)whether defendants' misrepresentationS/sup-pression of fact and/or of addictive propertiesof nicotine preclude availability of a "personalchoice" defense ; (9) defendants' liability foractual damages, and the categories of suchdamages; (10) defendants' liability for emo-tional distress damages; and (11) defendants'liability for punitive damages.

Phase 1 would be followed by notice of thetrial verdict and claim forms to class mem-bers. In phase 2, the jury would determinecompensatory damages in sample plaintiffcases. The jury then would establish a ratioof punitive damages to compensatory dam=ages, which ratio thereafter would apply toeach class member.

Phase 3 would entail a complicated proce-dure to determine compensatory damages forindividual class members . The trial plan en-visions determination of absent class mem-bers' compensatory economic and emotionaldistress damages on the basis of claim forms,"subject to verification techniques and asser-tion of defendants' affirmative defenses un-der grouping, sampling, or representativeprocedures to be determined by the Court ."

The trial plan left open how jury trials onclass members' personal injury/wrongfuldeath claims would be handled, but the trialplan discussed the possibility of bifurcation .In phase 4, the court would apply the puni-tive damage ratio based on individual dam-age awards and would conduct a review ofthe reasonableness of the award .

B. The Class Certification Order

Following extensive briefing, the districtcourt granted, in part, plaintiffs' motion forclass certification, concluding that the pre-requisites of FEn .R.Civ.P. 23(a) had beenmet 9 The court rejected certification, under

9 . Rule 23(a) states :

One or more members of a class may sue orbe sued as representative parties on behalf ofall only if (1) the class is so numerous thatjoinder of all members is impracticable, (2)there are questions of law or fact common to

I

1


CASTANO v. AMERICAN TOBACCOCite as 84 F .3d 734 (5th Cir. 1996)

FED.R.Ctv .P . 23(b)(2), of the plaintiffs' claimfor equitable relief, including the claim formedical monitoring. 160 F.R .D . at 552 . Ap-pellees have not cross-appealed that portionof the order .

The court did grant the plaintiffs' motionto certify the class under FEn.R .Crv .P .23(b)(3),1° organizing the class action issuesinto four categories : (1) core liability ; (2)injury-in-fact, proximate cause, reliance andaffirmative defenses; (3) compensatory dam-ages; and (4) punitive damages . Id at 553-58. It then analyzed each category to deter-mine whether it met the predominance andsuperiority requirements of rule 23(b)(3) .Using its power to sever issues for certifica-tion under FED.R.Crv.P . 23(c)(4), the courtcertified the class on core liability and puni-tive damages, and certified the class condi-tionally pursuant to FED.R .CIV.P. 23(c)(1) .

1 . Core Liability Issues

The court defined core liability issues as"common factual issues [of] whether defen-dants knew cigarette smoking was addictive,failed to inform cigarette smokers of such,and took actions to addict cigarette smokers .Common legal issues include fraud, negli-gence, breach of warranty (express or im-plied), strict liability, and violation of con-sumer protection statutes ." 160 F .R.D. at553.

The court found that the predominancerequirement of rule 23(b)(3) was satisfied forthe core liability issues . Without any specificanalysis regarding the multitude of issuesthat make up "core liability," the court foundthat under Jenkins u Raymark Indus ., 782F.2d 468 (5th Cir .1986), common issues pre-dominate because resolution of core liability

the class, (3) the claims or defenses of therepresentative parties are typical of the claimsor defenses of the class, and (4) the representa-tive parties will fairly and adequately protectthe interests of the class .

10 . Rule 23(b)(3) states, in pertinent part, that aclass action may be maintained if

the court finds that the questions of law or factcommon to the members of the class predomi-nate over any questions affecting only individu-al members, and that a class action is superiorto other available methods for the fair andefficient adjudication of the controversy .

Co. 739issues would significantly advance the indi-vidual cases. The court did not discuss why"core liability" issues would be a significant,rather than just common, part of each indi-vidual trial, nor why the individual issues inthe remaining categories did not predomi-nate over the common "core liability" issues .

The only specific analysis on predominanceanalysis was on the plaintiffs' fraud claim .The court determined that it would be pre-mature to hold that individual reliance issuespredominate over common issues . Relyingon Eisen v . Carlisle & Jacquelin, 417 U .B.156, 94 S .Ct . 2140, 40 L .Ed.2d 732 (1974), thecourt stated that it could not inquire into themerits of the plaintiffs' claim to determinewhether reliance would be an issue in individ-ual trials. 160 F.R.D . at 554. Moreover, thecourt recognized the possibility that understate law, reliance can be inferred when afraud claim is based on an omission . Accord-ingly, the court was convinced that it couldcertify the class and defer the considerationof how reliance would affect predominance .

The court also deferred substantial consid-eration of how variations in state law wouldaffect predominance . Relying on two districtcourt opinions," the court concluded thatissues of fraud, breach of warranty, negli-gence, intentional tort, and strict liability donot vary so much from state to state as tocause individual issues to predominate . Thecourt noted that any determination of howstate law variations affect predominance waspremature, as the court had yet to make achoice of law determination. As for the con-sumer protection claims, the court also de-ferred analysis of state law variations, be-cause "there has been no showing that the

11 . The court cited In re Asbestos Sch . Litig . . 104F.R.D. 422, 434 (E .D .Pa .1984) (discussing thesimilarity of negligence and strict liability in U .S .jurisdictions), ajfd in part and reversed in partsub nonr . School Dist- of Lancaster v . Lake Asbes-tos of Quebec, Ltd. (In re Sch . Asbestos Litig.)("S'chool Asbestos "), 789 F .2d 996, 1010 (3dCir .), cert . denied, 479 U .S . 852, 107 S .Ct . 182, 93L .Ed.2d 117, and cert . denied, 479 U .S . 913, 107S .Ct . 318, 93 L .Ed .2d 291 (1986), and In reCordis Cardiac Pacemaker Prod . Liability Litig.,No. C-3-90-374 (S .D .Ohio Dec . 23, 1992) (un-published) (discussing similarities among negli-gence, strict liabilitv, and fraud) .


(


consumer protection statutes differ so muchas to make individual issues predominate ."Id.

The court also concluded that a class ac-tion is superior to other methods for adjudi-cation of the core liability issues . Relyingheavily on Jenkins, the court noted that hav-ing this common issue litigated in a classaction was superior to repeated trials of thesame evidence. Recognizing serious prob-lems with manageability, it determined thatsuch problems were outweighed by "thespecter of thousands, if not millions, of simi-lar trials of liability proceeding in thousandsof courtrooms around the nation." Ici, at555-t56 .

2 . Il~jury-in-fact, Proximate Cause, Reli-ance, Affirmative Defenses, and_ Com-pensatory Damages

Using the same methodology as it did forthe core liability issues, the district courtrefused to certify the issues of injury-in-fact,proximate cause, reliance, affirmative defens-es, and compensatory damages, concludingthat the -"issues are so overwhelmingly re-plete with individual circumstances that theyquickly outweigh predominance and superior-ity." Id at 556 . Specifically, the courtfound that whether a person suffered emo-tional injury from addiction, whether his ad-diction was caused by the defendants' ac-tions, whether he relied on the defendants'misrepresentations, and whether affirmativedefenses unique to each class member pre-cluded recovery were all individual issues .As to compensatory damages and the claimfor medical monitoring, the court concludedthat such claims were so intertwined withproximate cause and affirmative defensesthat class certification would not materiallyadvance the individual cases .

3. Punitive Damages

In certifying punitive damages for classtreatment, the court adopted the plaintiffs'trial plan for punitive damages : The class

12. The panel opinion in Watson has no prece-dential weight in this circuit. While the casewas awaiting rehearing en bane, it settled . Ac-cording to the Internal Operating Procedure ac-companying 5ni CiR.R . 35, "the effect of granting

jury would develop a ratio of punitive dam-ages to actual damages, and the court wouldapply that ratio in individual cases . As it didwith the core liability issues, the court deter-mined that variations in state law, includingdiffering burdens of proof, did not precludecertification . Rather than conduct an inde-pendent review of predominance or superiori-ty, the court relied on Jenkins and on Wat-son v. Shell Oil Co., 979 F2d 1014 (5thCir.1992), vacated for rehearing en banc, 990F.2d 805 (5th Cir .1993), appeal dismisseci; 53F.3d 663 (5th Cir.1994), for support of itscertification order.12

IL

[1-3] A district court must conduct a rig-orous analysis of the rule 23 prerequisitesbefore certifying a class . General Tel.' Co. v.Falcon, 457 U .S. 147, 161, 102 S .Ct : 2364,2372, 72 L.Ed2d 740 (1982); Applewhite v.Reichhold Chems., 67 F .3d 571, 573 (5th Cir.1995). The decision to certify is within thebroad discretion of the court, but that disere-tion must be exercised within the fraineworkof rule 23 . Gulf Oil Co. v. Bernard, 452 U.S .89, 100, 101 S .Ct. 2193, 2200, 68 L.Ed.2d 693(1981) . -The party seeking certification bearsthe burden of proof. Hortan v. Goose CreekInd Sch, Dist., 690 F .2d 470, 486 (5th Cir.1982), cerG denierl, 463 U .S . 1207, 103 S .Ct.3536, 77 L.Ed.2d 1387 (1983) ; In re Ameri-can Medical Sys., 75 F.3d 1069, 1086 (6thCir.1996) (concluding that district court re-versed the proper burden of proof by askingdefendants to show cause why the courtshould not certify the class) .

[4] The district court erred in its analy-sis in two distinct ways . First, it failed toconsider how variations in state law affectpredominance and superiority. Second, itspredominance inquiry did not include consid-eration of how a trial on the merits would beconducted.

Each of these defects mandates reversal .Moreover, at this time, while the tort isimmature, the class complaint must be dis-

a rehearing en banc is to vacate the previousopinion and judgment of the Court and to staythe mandate." See de Aguilar v . Boeing Co ., 47F .3d 1404, 1411 (5th Cir .), cert . denied, - U-S .-, 116 S.Ct . 180, 133 L .Ed .2d 119 (1995) .

mi:to

InCi

tioissgi,Ciancltch1111Srabe

13


CASTANO v. AMERICAN TOBACCO CO.Cite as 84 F .3d 734 (5th Cir. 1996)

missed, as class certification cannot be foundto be a superior method of adjudication .t3

A. Variations in State Law

[5] Although rule 23(c)(1) requires that aclass should be certified "as soon as practica-ble" and allows a court to certify a condition-al class, it does not follow that the rule'srequirements are lessened when the class isconditional . As a sister circuit explained :

Conditional certification is not a meanswhereby the District Court can avoid de-ciding whether, at that time, the require-ments of the Rule have been substantiallymet. The purpose of conditional certifica-tion is to preserve the Court's power torevoke certification in those cases whereinthe magnitude or complexity of the litiga-tion may eventually reveal , problems nottheretofore apparent . But in this case theDistrict Court seemed to brush aside oneof the requirements, of Rule 23(b)(3) bystating that at this time "analysis of theindividual versus common questions wouldbe for the Court to act as a seer ." Howev-er difficult it may have been for the Dis-trict Court to decide whether commonquestions predominate over individualquestions, it should not have sidesteppedthis preliminary requirement of the Ruleby merely stating that the problem of indi-vidual questions "lies far beyond the hori-zon in the realm of speculation ."

In re Hotel Tel. Charges, 500 F .2d 86, 90 (9thCir.1974).

[6] In a multi-state class action, varia-tions in state law may swamp any commonissues and defeat predominance . See Geor-gine v. Amch.em Prods., 83 F .3d 610, 618 (3dCir.1996) (decertifying class because legaland factual differences in the plaintiffs'claims "when exponentially magnified bychoice of law considerations, eclipse any com-mon issues in this case") ; American MedicalSys., 75 F .3d at 1085 (granting mandamus ina multi-state products liability action, in partbecause "[t]he district court . . . failed to

13. The defendants raise a number of additionalchallenges to the district court s order, includingclaims that individual issues predominate, thatthe use of a punitive damage ratio violates dueprocess, that a multi-state class action inevitablywill violate Erie R .R . v. Tompkins, 304 U .S . 64,

741consider how the law of negligence differsfrom jurisdiction to jurisdiction") .

Accordingly, a district court must considerhow variations in state law affect predomi-nance and superiority. Walsh v. Ford MotorCo., 807 F .2d 1000 (D .C .Cir.1986) (Ruth Bad-er Ginsburg, J .), cerlt denied, 482 U .S . 915,107 S.Ct. 3188, 96 L .Ed.2d 677 (1987). TheWalsh court rejected the notion that a dis-trict court may defer considering variationsin state law :

Appellees see the "which law" matter asacademic . They say no variations in statewarranty laws relevant to this case exist.A court cannot accept such an assertion"on faith." Appellees, as class action pro-ponents, must show that it is accurate .We have made no inquiry of our own onthis score and, for the current purpose,simply note the general tmstartling state-ment made in a leading treatise: "TheUniform Commercial Code is not uniform "

Id. at 1016-17 (footnotes omitted).

[7, 8] A district court's duty to determinewhether the plaintiff has borne its burden onclass certification requires that a court con-sider variations in state law when a classaction involves multiple jurisdictions. "In or-der to make the findings required to certify aclass action under Rule 23(b)(3) . . . one mustinitially identify the substantive law issueswhich will control the outcome of the litiga-tion." Alabama v. -Blue Bird Body Co ., 573F.2d 309, 316 (5th Cir .1978).

[9] A requirement that a court knowwhich law will apply before making a predo-minance determination is especially impor-tant when there may be differences in statelaw. See In re Rhone-Poulenc Rorer, Inc .("Rhorae-Poulenc "), 51 F .3d 1293, 1299-1302(7th Cir.) (mandamus) (comparing differingstate pattern instructions on negligence anddiffering formulations of the meaning of neg-ligence), cert. denied, - U.S. -, 116S.Ct. 184, 133 L .Ed.2d 122 (1995) ; In re

58 S.Ct. 817, 82 L .Ed . 1188 (1938), and thatbifurcation of core liability issues in a class ac-tion violates article III of the Constitution . Giv-en our conclusion that this matter cannot pro-ceed as a class action in any event, we find itunnecessarv to address those issues .


742 S-! FEDERAL REPORTER, 3d SERIES

"Agent Ora:npe" Prod- Liability Litig., 818 credible argument that the applicable law ofF.2d 145, 165 (2d Cir .1987) (noting possibility the different states could be categorized intoof differences in state products liability law), four patterns) ; Walsh, 807 F.2d at 1017cert. denied, 484 U .S . 1004, 108 S .Ct. 695, 98 (holding that "nationwide class action mov-L.Ed.2d 647 (1988) . Given the plaintiffs' ants must creditably demonstrate, throughburden, a court cannot rely on assurances of an 'extensive analysis' of state law variances,counsel that any problems with predomi- 'that class certification does not present insu-nance nance or superiority can be overcome . perable obstacles"') .Windham t~ Americaaa Brands, Inc., 565 A thorough review of the record demon-F.2d 59, 70 (4th Cir .1977), cert . denied, 435 strates that, in this case, the district courtU.S . 968, 98 S .Ct . 1605, 56 L.Ed.2d 58 (1978). did not properly consider how variations inThe able opinion in School Asbestos dem- state law affect predominance . The court

onstrates what is required from a district acknowledged as much in its order grantingcourt when variations in state law exist . class certification, for, in declining to make aThere, the court affirmed class certification, choice of law determination, it noted thatdespite variations in state law, because : "[t]he parties have only briefly addressed the

To meet the problem of diversity in ap- conflict of laws issue in this matter ." 160plicable state law, class plaintiffs have F .R.D. at 554 . Similarly, the court statedundertaken an extensive analysis of the that "there has been no showing that thevariances in products liability among the consumer protection statutes differ so muchjurisdictions. _ That review separates the as to make individual issues predominate ."law into four categories. Even assuming Id.additional permutations and combinations, The district court's review of state lawplaintiffs have made a creditable showing, variances can hardly be considered exten-which apparently . satisfied the district sive ; it conducted a cursory review of statecourt, that class certification does not law variations and gave short shrift to thepresent insuperable obstacles . Although defendants' arguments concerning variations .we have some doubt on this score, the In response to the defendants' extensiveeffort may nonetheless prove successful . analysis of how state law varied on fraud,

789 F.2d at 1010 ; see also Georgine, 83 F .3d products liability, affu•mative defenses, negli-at 627 & n . 13 (distinguishing School Asbes- gent infliction of emotional distress, consum-tos because it involved few individualized er protection statutes, and punitive dam-questions, and class counsel had made a ages,15 the cotut examined a sample phase 1

14. The defendants contend that this statementshows that the court erroneously placed the bur-den on them to show that the various state stat-utes differ, rather than on the plaintiffs to showthat they do not . See American Medical Systems,75 F .3d at 1085 .

15 . We find it difficult to fathom how commonissues could predominate in this case when vari-ations in state law are thoroughly considered .The Georgine court found that common issues inan asbestos class action did not predominate :

However, beyond these broad issues, theclass members' claims vary widely in charac-ter. Class members were exposed to differentasbestos-containing products, for differentamounts of time, in different ways, and overdifferent periods . Some class members sufferno physical injury or have only asymptomaticpleural changes, while others suffer from lungcancer, disabling asbestosis, or from mesothe-Iioma-a disease which, despite a latency peri-od of approximately fifteen to forty years, gen-

erally kills its victims within two years afterthey become symptomatic. Each has a differ-ent history of cigarette smoking, a factor thatcomplicates the causation inquiry.

These factual differences translate into sig-nificant legal differences . Differences inamount of exposure and nexus between expo-sure and injury lead to disparate applicationsof legal rules, including matters of causation,comparative fault, and the types of damagesavailable to each plaintiff.Furthermore, because we must apply an in-

dividualized choice of law analysis to eachplaintiffs claims, the proliferation of disparatefactual and legal issues is compounded expo-nentially . . . . In short, the number of uncom-mon issues in this humongous class action,with perhaps as many as a million class mem-bers, is colossal.

83 F .3d at 626 (citations omitted) .The Castano class suffers from many of the

difficulties that the Georgine court found disposi-

/

i

k'

i

t

juofccPth(diir.

lap :a:dtSt

st

689643045PRODUCED FROM B&W WEB SITE

http://legacy.library.ucsf.edu/tid/xrw91d00/pdf

CASTANO v. AMERICAN TOBACCO CO .Cite as 84 F.3d 734 (Sth Clr. 1996)

i

jury interrogatory and verdict form, a surveyof medical monitoring decisions, a survey ofconsumer fraud class actions, and a survey ofpunitive damages law in the defendants'home states. The court also relied on twodistrict court opinions granting certificationin multi-state class actions .

The district court's consideration of statelaw variations was inadequate . The surveysprovided by the plaintiffs failed to discuss, inany meaningful way, how the court coulddeal with variations in state law. The con-sumer fraud survey simply quoted a fewstate courts that had certified state class

tive . The class members were exposed to nico-tine through different products, for differentamounts of time, and over different time periods .Each class member's knowledge about the effectsof smoking differs, and each plaintiff begansmoking for different reasons . Each of thesefactual differences impacts the application of 'le-gal rules such as causation, reliance, compara-tive fault, and other affirmative defenses.

Variations in state law magnify the differences .In a fraud claim, some states require justifiablereliance on a misrepresentation, see Allgood v .RJ. Reynolds Tobacco Co.. 80 F .3d 168 . 171 (5thCir .1996) ; Burroughs v. Jackson Nat'l Life Ins.

Cd., 618 So.2d 1329, 1332 (Ala.1993), while oth-ers require reasonable reliance, see Parks v. Mor- "

ris Homes Corp., 245 S .C . 461, 141 S .E .2d 129,132 (1965) . States impose varying standards todetermine when there is a duty to disclose facts .See Sugarhouse Fin . Co- v- Anderson, 610 P.2d1369, 1373 (Utah 1980) (finding no duty whentransaction was made at arm's length) ; Dodd v .Nelda Stephenson Chevrolet, Inc ., 626 So.2d1288, 1293 (Ala.1993) (using a flexible standardbased on the transaction and relationship of theparties)

. Products liability law also differs among states .Some states do not recognize strict liability .

E.g., Cline v. Prowler Indus ., 418 A .2d 968, 979-80 (Del .1980) . . Some have adopted RESTATEMENT(SECOND) oF Toers § 402A. E .g ., O.S . Stapley Co.v. Miller, 103 Ariz. 556, 447 P .2d 248, 251-52(1968) . Among the states that have adopted theRestatement, there are variations . See 5 STUARTM. SPEISER ET AL., THE A.NERIGLV LAW OF TORrS§§ 18 .31, 18 :34-18 :35 (Law Co-op 1996) .

Differences in afE`irmative defenses also exist .Assumption of risk is a complete defense to aproducts claim in some states . E .g ., S .C .CODEANN . § 15-73-20 (Law Co-op 1976) . In others, itis a part of comparative fault analysis. E .g .,Coto Rev STrr. § 13-21-1 1 1 .7 (1986). Somestates utilize "pure" comparative fault, e.g ., AR¢-RFVSrATANU . § 12-2503-09 (1984) ; others followa "greater fault bar," e .g ., CvvN GEN STAT.AnN .§ 52-572h (West 1988) ; and still others use an"equal fault bar," e .g ., Aw:CooE ANS . § 16-64-122 (Michie 1991) .

743actions . The survey of punitive damages waslimited to the defendants' home states .Moreover, the two district court opinions onwhich the court relied did not support theproposition that variations in state law couldbe ignored.16 Nothing in the record demon-strates that the court critically analyzed howvariations in state law would affect predomi-nance .

The court also failed to perform its duty todetermine whether the class action would bemanageable in light of state law variations .The court's only discussion of manageability

Negligent infliction of emotional distress alsoinvolves wide variations. See Douglas B . Mar-low, Negligent Infliction of Mental Distress: AJurisdictional Survey of Existing Limitation De-vices and Proposal Based on an Analysis of Objec-tive Versus Subjective Indices of Distress, 33 VitL .L.Rav . 781 (1988) . Some states do not recognizethe cause of action at all. See Allen v. Walker,569 So .2d 350, 352 (Ala.1990) . Some require aphysical impact. See OB-GYN Assocs. v. Little-

ton, 259 Ga. 663, 386 S.E.2d 146, 148 (1989) .Despite these overwhelming individual issues,

common issues might predominate- We are,however, left to speculate . The point of detailingthe alleged differences is to demonstrate the in-quiry the district court failed to make .

16 . Both the plaintiffs and the district court citeCordis and School Asbestos for the definitiveproposition that state law does not vary enoughin negligence, strict liability, or fraud to preventcertification, See Castano, 160 F .R.D . at 554 .Putting aside the obvious objection that a courtmust independently analyze the case before it todetermine predominance, such reliance is mis-placed .

In Cordis, the court specifically recognized thatthere are differences in the law of strict liabilityand fraud in different jurisdictions . The courtcertified the class despite those differences be-cause the differences did not eliminate predomi-nance in that particular case . Such a findingcannot be reflexively applied to the case subjudice.

The same is true of School Asbestos . Like thecourt in Cordis, the district court there foundlittle variation in state negligence law . TheThird Circuit agreed that, the variations in strictliability would not make the class unmanageable .789 F .2d at 1009 . See also Georgine, n . 13, 83F.3d at 627 & n . 13 (acknowledging that thecourt in School Asbestos certified the class de-spite variations in state law, but limiting thereach of the decision to cases where variationscan be broken down into a small number ofpatterns) . It is a stretch to characterize thesetwo cases as standing for the proposition thatstate law does not vary on negligence, strictliability, or Fraud .


(

1


is a citation to Jenkins and the claim that"[wJhi]e manageability of the liability issuesin this case may well prove to be difficult, theCourt finds that any such difficulties pale incomparison to the specter of thousands, if notmillions, of similar trials of liability proceed-ing in thousands of courtrooms around thenation." Id- at 555-56.

The problem with this approach is that itsubstitutes case-specific analysis with a gen-eralized reference to Jenkins . The Jenkinscourt, however, was not faced with managinga novel claim involving eight causes of action,multiple jurisdictions, millions of plaintiffs,eight defendants, and over fifty years of al-leged wrongful conduct . Instead, Jenkinsinvolved only 893 personal injury asbestoscases, the law of only one state, and theprospect of trial occurring in only one dis-trict: Accordingly, for purposes of the in-stant case, Jenkins is largely inapposite .

In summary, whether the specter of mil-lions of cases outweighs any manageabilityproblems in this class is uncertain when thescope of any manageability problems is un-known. Absent considered judgment on themanageability of the class, a comparison tomillions of individual trials is meaningless .

B . Predominance

The district court's second error was thatit failed to consider how the plaintiffs' addic-tion claims would be tried, individually or ona class basis . See 160 F.R.D. at 554. Thedistrict court, based on Eisen v. Carlisle &Jacquelin, 417 U .S. 156, 177-78, 94 S .Ct .2140, 2152-53, 40 L.Ed.2d 732 (1974), and

17. See Falcon" 457 U .S . at 160, 102 S .Ct. at 2372("Sometimes the issues are plain enough frontthe pleadings . . . and sometimes it may be neces-sary for the court to probe behind the pleadingsbefore coming to rest on the certification ques-tion .") ; Coopers & Lybrand v. Livesay. 437 U .S .463, 469, 98 S .Ct . 2454, 2458, 57 L .Ed .2d 351(1978) (reasoning that "the class determinationgenerally involves considerations that are 'en-meshed in the factual and legal issues compris-ing the plaintiff's cause of action .' ") ; id . at 469n . 12, 98 S .Ct. at 2458 n . 12 ('"'Evaluation ofmany of the questions entering into determina-tion of class action questions is intimately in-volved with the merits of the claims . The typi-cality of the representative's claim or defenses

. and the presence of common questions oflaw or fact are obvious examples . The more

Miller v. Mackey Int'l, 452 F .2d 424 (5thCir.1971), believed that it could not go pastthe pleadings for the certification decision .The result was an incomplete and inadequatepredominance inquiry.

The crux of the court's error was that itmisinterpreted Eisen and Miller. Neithercase suggests that a court is limited to thepleadings when deciding on certification .Both, instead, stand for the unremarkableproposition that the strength of a plaintiffsclaim should not affect the certification deci-sion . In Eisen, the Court held that it wasimproper to make a preliminary inquiry intothe merits of a case, determine that theplaintiff was likely to succeed, and conse-quently shift the cost of providing notice tothe defendant. 417 U.S . at 177, 94 S .Ct. at2152. In Miller, this court held that a dis-trict court could not deny certification basedon its belief that the plaintiff could not pre-vail on the merits. 452 F.2d at 427 .

[10] A district court certainly may lookpast the pleadings to determine whether therequirements of rule 23 have been met ."Going beyond the pleadings is necessary, asa court must understand the claims, defens-es, relevant facts, and applicable substantivelaw in order to make a meaningful determi-nation of the certification issues . See MANU-AL FOR COMPLEX LITIGATION § 30 .11 (3d ed .1995).

The district court's predominance inquirydemonstrates why such an understanding isnecessary. The premise,of the court's opin-ion is a citation to Jenkins and a conclusionthat class treatment of common issues wouldsignificantly advance the individual trials .

complex determinations required in Rule23(b)(3) class actions entail even greater entan-glement with the merits."') ; Love v. Turlington,733 F .2d 1562, 1564 (11th Cir .1984) ("While it istrue that a trial court may not properly reach themerits of a claim when determining whether theclass certification is warranted, this principleshould not be talismanically invoked to artificial-Iv limit a trial court's examination of the factorsnecessary to a reasoned determination of wheth-er a plaintiff has met her burden of establishingeach of the Rule 23 class action requirements ." ) ;Hc¢'f v. N.D. Cass Co ., 485 F .2d 710, 713 (5thCir .1973) (en bane) (,'"It is inescapable that insome cases there will be overlap between thedemands of [rule] 23(a) and (h) and the questionof whether plaintiff can succeed on the merits .") .

I

1


CASTANO v. AMERICAN TOBACCO CO.Cite w 84 F .3d 734 (5th Clr.

Absent knowledge of how addiction-as-injurycases would actually be tried, however, it wasimpossible for the court to know whether thecommon issues would be a "significant" por-tion of the individual trials . The court justassumed that because the common issueswould play a part in every trial, they must besignificant.1a The court's synthesis of Jen-kins and Eisen would write the predorni-nance requirement out of the rule, and anycommon issue would predominate if it werecommon to all the individual trials .19

The court's treatment of the fraud claimalso demonstrates the error inherent in itsapproach .20 According to both the advisorycommittee's notes to Rule 23(b)(3) and thiscourt's decision in Simon v. Merrill Lyneh,

18. The district court's approach to predomi-nance stands in stark contrast to the methodolo-gy the district court used in Jenkins . There, thedistrict judge had a vast amount of experiencewith asbestos cases . He certified the state of theart defense because it was the most significantcontested issue in each case. Jenkins v. RayrnarkIndustries, Inc., 109 F.RD. 269, 279 (E .D .Tex.1985). To the contrary, however, the districtcourt in the instant case did not, and could not,have determined that the common issues wouldbe a significant part of each case. Unlike thejudge in Jenkins, the district judge a quo had noexperience with this type of case and did noteven inquire into how a case would be tried todetermine whether the defendants' conductwould be a significant portion of each case .

19 . An incorrect predominance finding also impli-cates the court's superiority analysis : The great-er the number of individual issues, the less likelysuperiority can be established. American Medi-cal Sys ., 75 F .3d at 1084-85 (distinguishing asingle disaster mass tort from a more complexmass tort) . The relationship between predomi-nance and superiority in mass torts was recog-nized in the Advisory Committee's note to rule23(b)(3), which states :

A'"mass accident " resulting in injuries to nu-merous persons is ordinarily not appropriatefor a class action because of the likelihood thatsignificant questions, not only of damages butof liability and defenses to liability, would bepresent, affecting the individuals in differentways. In these circumstances an action con-ducted nominally as a class action would de-generate in practice into multiple lawsuits sep-arately tried.

FEO-RCrv.P . 23(b)(3) advisory committee's note(citation omitted), reprinted in 39 F .RD. 69, 103(1966) . See also Georgine, 1996 WL 242442, at' 12-' 13, - F.3d at --- (relying on theAdvisory Committee's note) ; American MedicalSys ., 75 F .3d at 1084-85 .

7451996)

Pierce, Fenner & Smith, Ine., 482 F2d 880(5th Cir.1973), a fraud class action cannot becertified when individual reliance will be anissue . The district court avoided the reachof this court's decision in Siman by an erro-neous reading of Eisen ; the court refused toconsider whether reliance would be an issuein individual trials .

The problem with the district court's ap-proach is that after the class trial, it mighthave decided that reliance must be proven inindividual trials. The court then would havebeen faced with the difficult choice of decerti-fying the class after phase 1 and wastingjudicial resources, or continuing with a classaction that would have failed the predomi-nance requirement of rule 23(b)(3)?t

The plaintiffs assert that Professor Charles Al-len Wright, a member of the Advisory Committeehas now repudiated this passage in the notes .See H. NEWBERG . 3 NEWBERG ON Ct.t55 ACIIONS§ 17 .06 (3d ed . 1992) . Professor Wright's recentstatements, made as an advocate in School Asbes-tos, must be viewed with some caution. As Pro-fessor Wright has stated:

I certainly did not intend by that statementto say that a class should be certified in allmass tort cases. I merely wanted to take thesting out of the statement in the Advisory Com-mittee Note, and even that said only that aclass action is "ordinarily not appropriate" inmass-tort cases . The class action is a complexdevice that must be used with discernment. Ithink for example that Judge Jones in Louisi-ana would be creating a Frankenstein's mon-ster if he should allow certification of whatpurports to be a class action on behalf ofeveryone who has ever been addicted to nico-tine .

Letter of Dec. 22, 1994, to N . Reid Neureiter,Williams & Connolly, Washington, D .C .

20. The court specifically discussed reliance inthe context of a fraud claim . Reliance is also anelement of breach of warranty claims in somestates, see, e .g., Modern Farm Serv., Inc. v. BenPearson, Inc., 308 F .2d 18, 23 (5th Cir.1962)(Arkansas) ; Caruso v. Celsius Insulation Re-

sources, Inc., 101 F.R.D . 530, 536 (M .D.Pa.1984),and an element of consumer protection statutesin others, see, e .g., Louisiana ex rel. Guste v.General Motors Corp., 370 So .2d 477, 489 (La.

1978) .

21 . Severing the defendants' conduct from reli-ance under rule 23(c)(4) does not save the classaction . A district court cannot manufacture pre-dominance through the nimble use of subdivision(c)(4) . The proper interpretation of the interac-tion between subdivisions (b)(3) and (c)(4) is that

D-89643048?RODUCED FROM B&W WEB SITEhttp://legacy.library.ucsf.edu/tid/xrw91d00/pdf

746

I

84 FEDERAL REPORTER, 3d SERIES

In addition to the reasons given above,regarding the district court's procedural er-rors, this class must be decertified because itindependently fails the superiority require-ment of rule 23(b)(3). In the context of masstort class actions, certification dramaticallyaffects the stakes for defendants . Class cer-tification magnifies and strengthens the num-ber of unmeritorious claims . Agent Orange,818 F.2d at 165-66 . Aggregation of claimsalso makes it more likely that a defendantwill be found liable and results in significant-ly higher damage awards. MANUAL FOR Conl-PLEx LITIGATIOIv § 33 .26 n. 1056; Kenneth S .Bordens and Irwin A . Horowitz, Mass TortCivil Litigation: The Impact of Procedural

a cause of action, as a whole, must satisfy thepredominance requirement of (b)(3) and that(c)(4) is a housekeeping rule that allows courts tosever the common issues for a class trial . See Inre N.D .Cal. Dalkon Shield IUD Prods . LiabilityLitig ., 693 F .2d 847, 856 (9th Cir .1982) (balanc-ing severed issues against the remaining individ-ual issues), cert . denied, 459 U .S . 1171, 103 S .Ct .817, 74 L .Ed .2d 1015 (1983) ; see also Jenkins,109 F.R.D. at 278 (comparing state of the artdefense to individual questions of exposure anddegree of injury in a class action certified only onthe common issue of the state of the art defense) .Reading rule 23(c)(4) as allowing a court to severissues until the remaining common issue pre-dominates over the remaining individual issueswould eviscerate the predominance requirementof rule 23(b)(3) ; the result would be automaticcertification in every case where there is a com-mon issue, a result that could not have beenintended .

22. In re Gen. Motors Corp. Pick-Up Truck FuelTank Prods . Liability Litig ., 55 F .3d 768, 784-85(3d Cir .), cert. denied, - U .S . -, 116 S.Ct . 88,133 L .Ed .2d 45 (1995) ; Rhone-Poulenc, 51 F .3dat 1299-1300. See also Georgine, 83 F .3d at 625n . 10 (rejecting the argument that the possibilitvof settlement should be factored positively inapplying rule 23(b)(3)) . But see In re A .H. Rob-ins Co., 880 F .2d 709, 740 (4th Cir.1989) (treat-ing the fact that certification may foster settle-rnent as a positive factor when applying rule23(b)(3)) (dicta ), cert . denied, 493 U .S . 959, 110S .Ct. 377, 107 L .Ed .2d 362 (1989) .

23. At the time rule 23 was drafted, mass tortlitigation as we now know it did not exist .Schuck, supra, at 945 . The term had been ap-plied to single-event accidents. Id. Even inthose cases, the advisory committee cautionedagainst certification . See supra note 19 . Asmodern mass tort litigation has evolved, courtshave been willing to certify simple single disaster

Changes ota Jury Decisions, 73 JUDICATURF22 (1989) .

In addition to skewing trial outcomes, classcertification creates insurmountable pressureon defendants to settle, whereas individualtrials would not . See Peter H . Schuck, MassTorts : An Institutional Evolutionist Per-spective, 80 Co1tvEL[, L .Rev. 941, 958 (1995) .The risk of facing an all-or-nothing verdictpresents too high a risk, even when theprobability of an adverse judgment is low .Rlaone-Poulenc, 51 F .3d at 1298 . These set-tlements have been referred to as judicialblackmail .'°

It is no surprise then, . that historically,certification of mass tort litigation classes hasbeen disfavored.'a The traditional concern

mass torts, see Sterling v . Velsicol Chem . Corp.,855 F.2d 1188, 1197 (6th Cir .1988), but havebeen hesitant to certify more complex mass torts,see Georgine, 83 F.3d at 627-628, 632 (discussingthe trend in certification and decertifying anasbestos class action); American Medical Sys ., 75F.3d at 1084-85 . See also Rhone-Poulenc, 51F.3d 1293 (decertifying class) ; In re Joint E. & S .Dist . Asbestos Litig., 14 F .3d 726 (2d Cir.1993)(vacating limited fund class action) ; In re Ben-dectin Prod. Liability Litig., 749 F .2d 300 (6thCir.1984) (granting mandamus reversing classcertification) ; Dalkon Shield IUD Prods . LiabilityLitig., 693 F.2d at 856 (decertifying class for lack

overclas.Ournottraccou:malysi:cer,higma .tiot

\oriexrtclaofclagufm

of commonality and superiority) ; Harding v .Tambrands Inc., 165 F .R.D. 623, 629 (D .Kan .1996) (denying certification of nationwide classof persons alleging toxic shock syndrome) ; Kurc-zi v. Eli Lilly & Co ., 160 F.R.D. 667 (N .D .Ohio1995) (denying nationwide class certification) ;Hurd v. Monsanto Co., 164 F .R .D . 234 (S .D .Ind .1995) (refusing to certify class of persons alleg-ing PCB exposure at one plant); Bethards v. BardAccess Sys ., Inc ., 1995 WL 75356 (N .D .Ill .1995)(recommending deniall of class certification in 2products liability action regarding catheters) ;Ikonen v . Hartz Mountain Corp ., 122 F .R .D . 258(S .D .Cal .1988) (denying class certification in fleaand tick spray products liability action) ; In reTetracycline Cases, 107 F,R .D. 719 (W .D.Mo .1983) (denying certification because class actionis not superior method of adjudication) ; Mertensv. Abbott Laboratories, 99 F .R.D . 38 (D .N .H .1983)(denying certification of class in DES litigation) ;Ryan v . Eli Lilly & Co ., 84 F .R .D. 230 (D.S .C .1979) (denying certification of class of womenwho took synthetic estrogen during pregnancy) ;Yandle v. PPG Indts ., 65 F .R .D . 566 (E .D .Tex .1974) (denying asbestos claims class certifica-tion) . But see Central Weslevan College v . W.R .Grace & Co ., 6 F .3d 177 (4th Cir .1993) (affirmingcertification of class of colleges in suit againstasbestos rnanufacturer) ; Agent Orange, 818 F .2d



CASTANO v. AMERICAN TOBACCO CO. 747Cite as 84 F.3d 734 (5th Cir. 1996)

over the rights of defendants in mass tortclass actions is magnified in the instant case .Our specific concern is that a mass tort can-not be properly certified without a priortrack record of trials from which the districtcourt can draw the information necessary tomake the predominance and superiority anal-ysis required by rule 23 . This is becausecertification of an immature tort results in ahigher than normal risk that the class actionmay not be superior to individual adjudica-tion.

We first address the district court's superi-ority analysis. The court acknowledged theextensive manageability problems with thisclass . Such problems include difficult choiceof law determinations, subclassing of eightclaims with variations in state law, Erieguesses, notice to millions of class members,further subclassing to take account of tran-

at 166-67 (certifying class despite manageabilitydifficulties because of centrality of military con-tractor defense) ; School Asbestos, 789 F.2d 996 ;In re Teletronics Pacing System, Inc ., AcufixAtrail "7" Leads Prod. Liability Litig., No . C-1-95-094 (S .D.Ohio, Nov. 17, 1995) (certifyingclass against manufacturer of alleged defectivepacemaker leads) (unpublished) ; In re CopleyPharmaceutical, Inc., 161 F.R.D. 456 (D.Wyo.1995) (certifying nationwide class for limitedthreshold liability issues regarding prescriptiondrug albuterol, but refusing to certify class forindividual issues of liability and causation orpunitive damages) ; Craft v. Vanderbilt Univ., No .3:94-0090 (M .D .Tenn . July 14, 1994) (certifyingclass for exposure to a radioactive isotope inmedical experiments) (unpublished) ; In re CordisCardiac Pacemaker Prod. Liability Litig ., No. C-3-90-374 (S .D .Ohio Dec . 23, 1992) (unpublished) .

24. There is reason to believe that even a masstort like asbestos could be managed, withoutclass certification, in a way that avoids judicialmeltdown . See Georgine, 83 F .3d at 634 (sug-gesting methods, short of a nationwide class ac-tion, that would be more efficient than individualtrials) ; John A. Siliciano, Mass Torts and theRhetoric of Crisis, 80 CoaNSt.c L.RF.v . 980, 1010-l2 (1995) (suggesting that stringent "gate keep-ing" by courts at the outset would have prevent-ed asbestos from becoming a monstrous masstort). In a case such as this one, where causa-tion is a key element, disaggregation of claimsallows courts to dismiss weak and frivolousclaims on summary judgment.

Where novel theories of recovery are advanced(such as addiction as injury), courts can aggres-sively weed out untenable theories. See, e .g.,

sient plaintiffs, and the difficult procedurefor determining who is nicotine-dependent .Cases with far fewer manageability problemshave given courts pause . See, e.g., Georgine,83 F.3d at 632 ; In re Hotel Tel., 500 F2d at90.

The district court's rationale for certifica-tion in spite of such problems--i.e., that aclass trial would preserve judicial resourcesin the millions of inevitable individual trials-is based on pure speculation. Not everymass tort is asbestos, and not every masstort will result in the same judicial crises?'The judicial crisis to which the district courtreferred is only theoretical .

What the district court failed to consider,and what no court can determine at this time,is the very real possibility that the judicialcrisis may fail to materialize.g The plain-

Allgood v. RJ. Reynolds Tobacco Co., 80 F .3d168, 172 (5th Cir.1996) (rejecting failure-to-warnclaim against tobacco companies based on inade-quate proof of reliance and, alternatively, on"common knowledge" theory). Courts can usecase management techniques to avoid discoveryabuses . The parties can also turn to mediationand arbitration to settle individual or aggregatedcases .

25. The plaintiffs, in seemingly inconsistent posi-tions, argue that the lack of a judicial crisisjustifies certification ; they assert that the reasonwhy individual plaintiffs have not filed claims isthat the tobacco industry makes individual trialsfar too expensive and plaintiffs are rarely suc-cessful. The fact that a party continuously losesat trial does not justify class certification, howev-er. See American Medical Systems, 75 F .3d at1087 and n. 20 (granting mandamus in partbecause judge's comments that class treatmentwas warranted because the defendant had great-er litigation resources than the plaintiff demon-strated a bias in favor of certification by thejudge) . The plaintiffs' argument, if accepted,would justify class treatment whenever a defen-dant has better attorneys and resources at itsdisposal.

The plaintiffs' claim also overstates the defen-dants' ability to outspend plaintiffs . Assumingarguendo that the defendants pool resources andoutspend plaintiffs in individual trials, there is noreason why plaintiffs still cannot prevail . Theclass is represented by a consortium of well-financed plaintiffs' lawyers who, over time, candevelop the expertise and specialized knowledgesufficient to beat the tobacco companies at theirown game . See Francis E . McGovern, An Analy-sis of Mass Torts for Judges, 73 TEc .L .2av. 1821,1834-35 (1995) (suggesting that plaintiffs can

89643050RODUCED FROM B&W WEB SITEhttp://legacy.library.ucsf.edu/tid/xrw91d00/pdf

748 8-1 FEDERAL REPORTER, 3d SERIES

tiffs' claims are based on a new theory ofliability and the existence of new evidence .Until plaintiffs decide to file individualclaims, a court cannot, from the existence ofinjury, presume that all or even any plaintiffswill pursue legal remedies.26 Nor can acourt make a superiority determinationbased on such speculation . Arnericmt Medi-cal Sys., 75 F .3d at 1085 (opining that superi-ority is lacking where judicial managementcrisis does not exist and intlividual trials arepossible) .

Severe manageability problems and thelack of a judicial crisis are not the onlyreasons why superiority is lacking. Themost compelling rationale for finding superi-ority in a class action-the existence . of anegative value suit-is missing in this case .Accord Phillips Petroleuna Co. v. Shutts, 472U.S. 797, 809, 105 S .Ct. 2965, 2973, 86L.Ed.2d 628 (1985) ; Rhoue-Poulenc, 51 F .3dat 1299 .

As he stated in the record, plaintiffs' coun-sel in this case has promised to inundate thecourts with individual claims if class certifi-cation is denied . Independently of the relia-bility of this self-serving promise, there isreason to believe that individual suits arefeasible . First, individual damage claims arehigh, and punitive damages are available inmost states. The expense of litigation doesnot necessarily turn this case into a negativevalue suit, in part because the prevailingparty may recover attorneys' fees undermany consumer protection statutes. SeeBoggs v. Alto Trailer Sales, 511 F.2d 114,118 (5th Cir.1975) (acknowledging that theavailability of attorneys' fees is a commonbasis for finding non-superiority) .

In a case such as this one, where eachplaintiff may receive a large award, and feeshifting often is available, we find Chief

overcome tobacco defendants' perceived advan-tage when a sufficient number of plaintiffs havefiled claims ar d shared discovetyl . Courts canalso overcome the defendant's alleged advan-tages through coordination or consolidation ofcases for discoven' and other pretrial matters .See MANUAL FOR Coa eLEC LmcnnuN at § 33,21-25 .

26. There are numerous reasons whv plaintiffswith positive-value suits opt out of the tort svs-tem, including risk aversion to engaging in litiga-tion, privacy concerns, and alternative avenues

Judge Posner's analysis of superiority to bepersuasive :

For this consensus or maturing of judg-ment the district judge proposes to substi-tute a single trial before a single jury . . .,One jury . . . will hold the fate of an indus-try in the palm of its hand ., . . That kindof thing can happen in our system of civiljustice . . . . But it need not be toleratedwhen the alternative exists of submittingan issue to multiple juries constituting inthe aggregate a much larger and morediverse sample of decision-makers . Thatwould not be a feasible option if the stakesto each class member were too slight torepay the cost of suit . . . . But this is notthe case . . . . ' Each plaintiff if successful isapt to receive a judgment in the millions .With the aggregate stakes in the tens orhundreds of millions of dollars, or even inthe billions, it is not a waste of judicialresources to conduct more than one trial,before more than six jurors, to determinewhether a major segment of the interna-tional pharmaceutical industry is to followthe asbestos manufacturers into Chapter11 .

Rhone-Poulenc, 51 F.3d at 1300. So toohere. we cannot say that it would be a wasteto allow individual trials to proceed, before adistrict court engages in the complicated pre-dominance and superiority analysis necessaryto certify a class .

Fairness may demand that mass torts withfew prior verdicts or judgments be litigat-ed first in smaller units-even single-plain-tiff, single-defendant trials-until generalcausation, typical injuries, and levels ofdamages become established . Thus, "ma-ture" mass torts like asbestos or DalkonShield may call for procedures that are notappropriate for incipient mass tort cases,such as those involving injuries arising

for medical treatment, such as Medicaid. SeeMcGo.ern, supra, at 1827-28 . In a case whcrccomparative negligence is raised, plaintitls havethe best insight into their own relative fault .Ultimately, a court cannot extrapolate . from thenumber of potential plaintiffs, the actual numberof cases that will be filed . See id. at 1823 & n . 8(contending that onlr 10 to 20% of persons whosuCfer harm actually invoke the tort litigationprocess) .

M

juccjuin

a

1 .

2



CASTANO v. :',MERIC:VN TOBACCO CO .Ciieastl4 F .3d 734 (5thCir . 1996)

ft•om new products, chemical substance.s,or pharmaceuticals .

MAtiUAl, FoE Co,,lPl .t:x LITICATtuv § 33 .26 .The remaining rationale for superiority-

judicial efficiency ''-is also lacking . In thecontext of an immature tort, any savings injudicial resources is speculative, and anyimagined savings would be overwhelmed bythe procedural problems that certification ofa srci ge.reeri ., cause of action brings with it.

Even assuming argueiido that the tort sys-tem will see many more addiction-as-injuryclaims, a conclusion that certification willsave judicial resources is premature at thisstage of the litigation . Take for example thedistrict court's plan to divide core liabilityfrom other issues such as comparative negli-gence and reliance . The assumption is thatafter a class verdict, the common issues willnot be a part of follow-up trials . The courthas no basis for that assumptiori .

It may be that comparative negligence willbe raised in the individual trials, and theevidence presented at the class trial will haveto be repeated . The same may be true forreliance .2s The net result may be a waste,not a savings, in judicial resources . Onlyafter the courts have more experience withthis type of case can a court certify issues ina way that preserves judicial resources . SeeJenkins, 782 F .2d 468 (certifying state of theart defense because experience had demon-strated that judicial resources could be savedby certification) .

Even assuming that certification at thistime would result in judicial efficiencies inindividual trials, certification of an immaturetort brings with it unique problems that mayconsume more judicial resources than certifi-cation will save. These problems are notspeculative ; the district court faced, and ig-nored, many of the problems that immaturetorts can cause .

The primary procedural difficulty createdby immature torts is the inherent difficulty adistrict court will have in determining wheth-

27 . See Sterlirtg, 855 F .2d at 1196 ("The procedur-al device of Rule 23(b)(3) class action was de-signed not solelti as a means for assuring legalassistance in the %indication of small claints but,rather, to achieve the economics of time . effort .and expense .") .

749er the requirements of rule 23 have beenmet. We have already identified a numberof defects with the district court's pretlomi-nance and manageability inquires, defectsthat will continue to exist on remand becauseof the unique nature of the plaintiffs' claim .

The district court's predominance inquiry,or lack of it, squarely presents the problemsassociated with certification of immaturetorts. Determining whether the common is-sues are a "significant" part of each individu-al case has an abstract quality to it when nocourt in this country has ever tried an injury-as-addiction claim . As the plaintiffs admit-ted to the district court, "we don't have thelearning curb [sic] that is necessary to say toYour Honor 'this is precisely how this casecan be tried and that will not run afoul of theteachings of the 5th Circuit ."'

Yet, an accurate finding on predominanceis necessary before the court can certify aclass . It may turn out that the defendant'sconduct, while common, is a minor part ofeach trial . Premature certification deprivesthe defendant of the opportunity to presentthat argument to any court and risks decerti-fication after considerable resources havebeen expended .

The court's analysis of reliance also dem-onstrates the potential judicial inefficienciesin immature tort class actions . Individualtrials will determine whether individual reli-ance will be an issue. Rather than guessthat reliance may be inferred, a district courtshould base its determination that individualreliance does not predominate on the wisdomof such individual trials . The risk that adistrict court will make the wrong guess, thatthe parties will engage in years of litigation,and that the class ultimately will be decerti-fied (because reliance predominates overcommon issues) prevents this class actionfrom being a superior method of adjudica-tion .

The complexity of the choice of law inquiryalso makes individual adjudication superior

28. See, e .g ., Allgood, 80 F .3d at 171 (holding thatunder Texas law, reliance is an essential elementof both affirmative fraud and fraudulent conceal-menl) .


(


to class treatment. The plaintiffs have as-serted eight theories of liability from everystate . Prior to certification, the districtcourt must determine whether variations instate law defeat predominance . While thetask may not be impossible, its complexitycertainly makes individual trials a more at-tractive alternative and, ipso facto, rendersclass treatment not superior . See Georgine,83 F.3d at 634 (recommending that Congresssolve the problems inherent in multi-stateclass actions by federalizing choice of lawrules, but rejecting such legislation when itmasquerades as judicial innovation) .

Through individual adjudication, the plain-tiffs can winnow their claims to the strongestcauses of action.' The result will be aneasier choice of law inquiry and a less compli-cated predominance inquiry . State courtscan address the more novel of the plaintiffs'claims, making the federal court's Erieguesses less complicated . It is far moredesirable to allow state courts to apply anddevelop their own law than to have a federalcourt apply "a kind of Esperanto [jury] in-struction." Rh,arre-PaulereG 51 F .3d at 1300;MANUAL FOR COMPLE% LITIGATION § 3326(discussing the full cycle of litigation neces-sary for a tort to mature).

The full development of trials in everystate will make subclassing an easier process .The result of allowing individual trials toproceed is a more accurate determination ofpredominance. We have already seen theresult of certifying this class without individ-ual adjudications, and we are not alone inexpressing discomfort with a district court'scertification of a novel theory . See Rhone-Poulenc, 51 F .3d at 1300 .

Another factor weighing heavily in favor ofindividual trials is the risk that in order tomake this class action manageable, the courtwill be forced to bifurcate issues in violation

29. State courts are more than capable of provid-ing definitive statements regarding the validity ofaddiction-as-injury claims . See, e .g., Joseph E.Seagram & Sons v. McGuire, 814 S .W.2d 385(Tex.1991) (accepting "common knowledge" the-ory and holding no cause of action for alcoholaddiction claim based on products liability, mis-representations, negligence, breach of impliedwarranties of merchantability and fitness, viola-tions of consumer protection statutes, and con-

of the Seventh Amendment. This class ac-tion is permeated with individual issues, suchas proximate causation, comparative negli-gence, reliance, and compensatory damages .In order to manage so many individual is-sues, the district court proposed to empanela class jury to adjudicate common issues. Asecond jury, or a number of "second" juries,will pass on the individual issues, either on acase-by-case basis or through group trials ofindividual plaintiffs .

[11] The Seventh Amendment entitlesparties to have fact issues decided by onejury, and prohibits a second jury from reex-amining those facts and issues30 Thus, Con-stitution allows bifurcation of issues that areso separable that the second jury will not becalled upon to reconsider findings of fact bythe first:

[Tlhis Court has cautioned that separationof issues is not the usual course thatshould be followed, and that the issue to betried must be so distinct and separablefrom the others that a trial of it alone maybe had without injustice. This limitationon the use of bifurcation is a recognition ofthe fact that inherent in the SeventhAmendment guarantee of a trial by jury isthe general right of a litigant to have onlyone jury pass on a common issue of fact .The Supreme Court recognized this princi-ple in Gasoline PToducts [Co., Inc v.Champlin Refining Co., 283 U.S . 494, 51S.Ct. 513, 75 L .Ed. 1188 (1931)] . . . . TheCourt explained . . . that a partial new trialmay not be "properly resorted to unless itclearly appears that the issue to be retriedis so distinct and separable from the othersthat a trial of it alone may be had withoutinjustice ." Such a rule is dictated for thevery practical reason that if separate juriesare allowed to pass on issues involvingoverlapping legal and factual questions the

spiracy) ; see also Allgood, 80 F .3d at 171-72(rejecting failure-to-warn claim against tobaccocompanies based on inadequate proof of relianceand, alternatively, on "common knowledge" the-ory) (citing Joseph E. Seagram ) .

30. "[N]o fact tried by jury, shall be otherwise re-examined in any Court of the United States . . ."U .S . Const. amend. VII.

32 .C


verdicts renderedinconsistent .

CASTANO v. AMERICAN TOBACCO CO. 751Cite as 84 F .3d 734 (5th Cir. 1996)

by each jury could be findings of a first jury . The risk of such

Alabama v. Blue Bird Body Co ., 573 F .2d309, 318 (5th Cir.1978) (citations and foot-notes omitted) .The Seventh Circuit recently addressed

Seventh Amendment limitations to bifurca-tion . In Rhone-Poulenc, 51 F.3d at 1302-03,Chief Judge Posner described the constitu-tional limitation as one requiring a court to"carve at the joint" in such a way so that thesame issue is not reexamined by differentjuries . "The right to a jury trial . . . is aright to have jm-iable issues determined bythe first jury impaneled to hear them (pro-vided there are no errors warranting a newtrial), and not reexamined by another finderof fact." Id. at 1303.

Severing a defendant's conduct from com-parative negligence results in the type of riskthat our court forbade in Blace Bird. Com-parative negligence, by definition, requires acomparison between the defendant's and theplaintiff's conduct. Rhone-Pouleaac, 51 F .3dat 1303 ("Comparative negligence entails, asthe name implies, a comparison of the degreeof negligence of plaintiff and defendant.") .At a bare minimum, a second jury will rehearevidence of the defendant's conduct . Thereis a risk that in apportioning fault, the secondjury could reevaluate the defendant's fault,determine that the defendant was not atfault, and apportion 100% of the fault to theplaintiff. In such a situation, the second jurywould be impermissibly reconsidering the

31 . The plaintiffs argue that any risk that a bifur-cation order would violate the Seventh Amend-ment is speculative, as the plaintiffs may prevailon causes of action that either do not requirebifurcation or do not contain issues that are sointertwined that the Seventh Amendment will beimplicated. In essence, plaintiffs' argumentboils down to a repudiation of the class com-plaint's negligence and strict products liabilityclaims .

32. This contention is disingenuous at best . Atoral argument, the plaintiffs asserted that time isof the essence, because plaintiffs who die cannotpartake in a medical monitoring fund. What theplaintiffs failed to mention was that the districtcourt refused to certifsa medical monitoringfund, and the plaintiffs have not cross-appealedthat decision . Moreover-, for the remainder ofthe claims a plaintiff's familv or estate can suebased on survi~orship statutes . The plaintiffs'

reevaluation is so great that class treatmentcan hardly be said to be superior to individu-al adjudication .3t

The plaintiffs' final retort is that individualtrials are inadequate because time is runningout for many of the plaintift's 3' They pointout that prior litigation against the tobaccocompanies has taken up to ten years to windthrough the legal system . ~Vhile a compel-ling rhetorical argument, it is ultimately in-consistent with the plaintiffs' own argumentsand ignores the realities of the legal system .First, the plaintiffs' reliance on prior person-al injury cases is unpersuasive, as they admitthat they have new evidence and are pursu-ing a claim entirely different from that ofpast plaintiffs.

Second, the plaintiffs' claim that time isrunning out ignores the reality of the classaction device. In a complicated case involv-ing multiple jurisdictions, the conflict of lawquestion itself could take decades to work itsway through the courts.' Once that issuehas been resolved, discovery, subclassing,and ultimately the class trial would takeplace . Next would come the appellate pro-cess . After the class trial, the individualtrials and appeals on comparative negligenceand damages would have to take place. Thenet result could be that the class action de-vice would lengthen, not shorten, the time ittakes for the plaintiffs to reach final judg-ment .

class complaint envisions survivor lawsuits . Infact, the named plaintiff in this case, DianneCastano, is a non-smoker who is suing both forthe ~,~rongful death of her husband and as arepresentative in a survival action .

33. The plaintiffs rely on School Asbestos for theproposition that variations in state law do notpreclude predominance. Putting that issuceaside, the case is instructive fur .Ohat happenedafter the Third Circuit remanded to the districtcourt. Almost nine years after the first com-plaint was filed, and eight years after the court ofappeals had affirmed certification, the conflict oflaw issues had yet to be resolved . See In rc Sch .Asbestos Litig ., 977 F .2d 764, 771 (3d Cir .1992)(granting mandamus to disqualify judge but re-fusing to address whether district court's trialplan properhy resolved any problems with varia-tions in state law because new judge may adopt adifferent trial plan) .



IV.

The district court abused its discretion byignoring variations in state law and how atrial on the alleged causes of action would betried . Those errors cannot be corrected onremand because of the novelty of the plain-tiffs' claims . Accordingly, class treatment isnot superior to individual adjudication .

We have once before stated that "tradition-al ways of proceeding reflect far more thanhabit . They reflect the very culture of thejury trial . . . ." In re Fibreboard Corp., 893F2d 706, 711 (5th Cir.1990). The collectivewisdom of individual juries is necessary be-fore this court commits the fate of an entireindustry or, indeed, the fate of a class ofmillions, to a single jury . For the forgoingreasons, we REVERSE and REMAND withinstructions that the district court dismissthe class complaint .

UNITED STATES of America,Plaintiff-Appellant,

V.

Bernard SCHUCHII'L4NN,Defendant-Appellee .

No. 95-10212.

United States Court of Appeals,Fifth Circuit.

May 24, 1996 .

Defendant was charged with conspiringto defraud the United States and the FederalHome Loan Bank Board, making false en-tries in bank records, and willful misapplica-tion of funds based on his alleged partic-ipation in scheme to circumvent bank's loans-to-one-borrower limit. The United StatesDistrict Court for the Northern District ofTexas, Joe Kendall, J., granted defendant'smotion for judgment of acquittal, on theorythat government had failed to satisfy its bur-

den of proving that defendant acted withrequisite guilty knowledge, and governmentappealed. The Court of Appeals, Duhe, Cir-cuit Judge, held that government failed toshow that at time defendant approved loan,which was ostensibly to administrative assis-tant of one of his business colleagues, defen-dant knew that administrative assistant wasmere nominee and that loan proceeds wouldactually be disbursed to business colleague inviolation of bank's loans-to-one-borrower lim-it.

Affirmed .

1. Banks and Banking e-509 .25Conspiracy e-47(6)

Government failed to show that at timebank officer approved loan, which was osten-sibly to administrative assistant of one of hisbusiness colleagues, officer knew that admin-istrative assistant was mere nominee of hisbusiness colleague, and that colleague wouldin fact receive loan proceeds in violation ofbank's loans-to-one-borrower . limitation ;guilty knowledge could not be inferred, asrequired to support charges against bankofficer for conspiring to defraud the UnitedStates and the Federal Home Loan BankBoard, malting false entries in bank records,and willfully misapplying funds, merely fromfact that administrative assistant's income,standing alone, would not have qualified herfor loan, given evidence that administrativeassistant came from wealthy family, and thatlenders would consider factors such as familywealth in deciding whether to approve loanapplication . 18 U .S .C .A . §§ 371, 657, 1006 .

2. Criminal Law e-753.2(3 .1)

Trial judge has duty to grant motion forjudgment of acquittal when evidence, viewedin light most favorable to government, is soscant that jury can only speculate as to de-fendant's guilt . Fed.Rules Cr.Proc.Rule 29,18 U .S .C.A.

3. Criminal Law 0-1144.13(2.1), 1159.2(7)In reviewing challenges to sufficiency of

evidence, Court of Appeals views evidence inlight most favorable to jury verdict and willaffirm if rational trier of fact could have


734 - Addiction Incorporated

Documents

Transcript of 734 - Addiction Incorporated