176 Tex

Document Sample
176 Tex Powered By Docstoc
					176   Tex.        116 SOUTH WESTERN REPORTER, 3d SERIES

exception, one of the recognized exceptions    2003). A person commits the offense of
to the warrant requirement, should be ap-      theft if he unlawfully appropriates proper-
plied to A.K.’s parents. See Carroll v.        ty with intent to deprive the owner of
U.S., 267 U.S. 132, 156, 45 S.Ct. 280, 69      property. TEX. P.CODE ANN. § 31.03 (Ver-
L.Ed. 543 (1925); Colburn v. State, 966        non Supp.2003). On the record, there is
S.W.2d 511, 519 (Tex.Crim.App.1998);           evidence showing that A.K.’s parents en-
Moulden v. State, 576 S.W.2d 817, 819          tered Jenschke’s truck without his effec-
(Tex.Crim.App.1978). However, there is         tive consent. There is, however, no evi-
no evidence A.K.’s parents were acting         dence that A.K.’s parents had the intent to
under the control of, or at the behest of,     deprive Jenschke of his property. Rather,
law enforcement rather than as plain citi-     the record shows their intent was merely
zens. Cobb v. State, 85 S.W.3d 258, 270        to obtain evidence. Their actions, then,
(Tex.Crim.App.2002).       Their    conduct,   were not committed in violation of the
therefore, does not implicate constitutional   Texas Penal Code. See TEX. P.CODE ANN.
restraints against governmental searches       §§ 30.04, 31.03 (Vernon Supp.2003).
and seizures. Id. For a search to be ille-        Because the evidence in question was
gal under federal law, the search must be      not seized in violation of Texas Penal Code
the result of state action by state agents.    § 30.04, it is not subject to exclusion under
Id.; Walter v. United States, 447 U.S. 649,    Texas Code of Criminal Procedure article
656, 100 S.Ct. 2395, 65 L.Ed.2d 410 (1980).    38.23. See TEX.CODE CRIM. PROC. ANN. art.
Neither the Fourth Amendment nor the           38.23 (Vernon 2003). We, therefore, af-
Texas Constitution requires trial courts to    firm the trial court’s denial of the motion
exclude incriminating evidence that was        to suppress.
obtained by a private citizen’s illegal
search. Cobb, 85 S.W.3d at 270–71; State
v. Johnson, 939 S.W.2d 586, 588 (Tex.
Crim.App.1996). Under the Texas Code of
Criminal Procedure, however, if a private
citizen seizes evidence in violation of any      EXXON CORPORATION, Appellant
federal or state constitutional provision or          and Cross–Appellee,
law, that evidence is subject to exclusion.                         v.
TEX.CODE CRIM. PROC. ANN. art. 38.23 (Ver-     James MAKOFSKI, Sr., as Next Friend
non Supp.2003); Johnson, 939 S.W.2d at           of James Makofski, Jr.; Barton Rus-
587–88.                                          sell, as Next Friend of John Russell;
   [11] Under Article 38.23, if A.K.’s par-      Andrea Russell; San Juanita Devora;
                                                 Felipe Devora; and Codi Stennett, Ap-
ents committed a crime when they took
                                                 pellees and Cross–Appellants.
the items from Jenschke’s truck, the trial
court abused its discretion in denying                   No. 14–00–00763–CV.
Jenschke’s motion to suppress. Cobb, 85                Court of Appeals of Texas,
S.W.3d at 271. Under Texas Penal Code                    Houston (14th Dist.).
§ 30.04, a person commits the offense of
                                                              July 24, 2003.
burglary of a vehicle if, without the effec-
tive consent of the owner, he breaks into         Rehearing Overruled Oct. 23, 2003.
or enters a vehicle or any part of a vehicle
with intent to commit any felony or theft.         Residents of subdivision sued oil com-
TEX. P.CODE ANN. § 30.04 (Vernon Supp.         pany, seeking damages for contamination
                               EXXON CORP. v. MAKOFSKI                                 Tex.   177
                      Cite as 116 S.W.3d 176 (Tex.App.—Houston [14th Dist.] 2003)

of their water supply due to oilfield leak           3. Waters and Water Courses O209
that occurred 50 years earlier. The 269th                 It was plaintiffs’ burden, in action
District Court, Harris County, John Thom-            against oil company for personal injuries
as Wooldridge, J., reduced or eliminated             allegedly resulting from contamination of
most of the jury awards for residents, and           water supply by oilfield leak that occurred
residents and oil company appealed. The              50 years previously, to present scientifical-
Court of Appeals, Scott Brister, C.J., held          ly reliable evidence of causation at trial.
that: (1) evidence was legally insufficient to
establish that child’s exposure to benzene           4. Appeal and Error O204(7), 497(1)
in subdivision’s water supply caused his                  In action against oil company for per-
acute lymphocytic leukemia (ALL); (2) evi-           sonal injuries allegedly resulting from con-
dence was legally insufficient to establish          tamination of water supply by oilfield leak
that other child’s exposure to benzene               that occurred 50 years previously, oil com-
caused his medical problems or any dis-              pany was required to object at trial so the
ease child in reasonable probability would           plaintiffs would have an opportunity to
suffer in the future; and (3) residents could        cure any defects regarding reliability of
not recover mental anguish damages.                  evidence of causation and present Court of
                                                     Appeals with a fully developed record.
    Reversed, and judgment rendered.
                                                     5. Appeal and Error O907(4)
    Charles W. Seymore, J., dissented,
and filed opinion.                                        It is generally true that unless an
                                                     appellant files a complete reporter’s rec-
                                                     ord, or a limited appeal, Court of Appeals
1. Appeal and Error O1001(3)                         presumes the omitted portions are rele-
     Court of Appeals does not take expert           vant and support the jury’s verdict.
testimony at face value; if it is unreliable,        6. Appeal and Error O516, 837(10)
it constitutes no evidence, and Court must                A complete record for purposes of
reverse any verdict based on it.                     appeal does not include matters from other
2. Appeal and Error O671(4)                          proceedings; Court of Appeals normally
                                                     tests the legal sufficiency of a jury verdict
      Oil company did not waive error of
                                                     by the evidence at the jury trial, not by
trial court when it found that evidence was
                                                     what happened before or after it.
legally sufficient to establish a link be-
tween benzene exposure and residents’                7. Appeal and Error O907(4)
specific diseases, in action by subdivision               The presumption that omitted por-
residents whose water supply had been                tions of a record support a verdict is rebut-
contaminated by oilfield leak 50 years ear-          table.
lier, by failing to obtain a record from a
                                                     8. Appeal and Error O971(2), 1001(1)
pretrial hearing at which its motions to
exclude allegedly unreliable expert testi-              Evidence O546, 570
mony were first considered; company was                 Trial O50
challenging legal sufficiency of the evi-                 Challenges to both the admissibility
dence that was presented to the jury to              and the legal sufficiency of expert testimo-
support jury’s verdict, and company was              ny involve the same substantive guidelines
not required to incorporate into trial rec-          for determining relevance and reliability;
ord all pretrial evidence that supported the         however admissibility of expert testimony
reliability of residents’ experts.                   is an initial decision made by the trial
178   Tex.         116 SOUTH WESTERN REPORTER, 3d SERIES

judge, usually outside the jury’s presence,     hood that random chance could produce a
and is subject to an abuse-of-discretion        statistically significant association where
review, while by contrast, legal sufficiency    none existed, and study made no attempt
of expert testimony tests the jury’s deci-      to distinguish among many particular
sion after the trial, looking at the whole      types of diseases with apparently different
trial record to see if any evidence supports    etiologies.
the verdict.
                                                12. Evidence O571(9)
9. Evidence O150                                     Testimony by experts was legally in-
     To benefit from exception to the nor-      sufficient to establish that child’s exposure
mal rules requiring individual proof of cau-    to benzene in subdivision’s water supply,
sation which allows scientifically reliable     contaminated as a result of oilfield leak
epidemiological studies showing more than       that occurred 50 years previously, caused
a doubling of the risk to establish a causa-    his acute lymphocytic leukemia (ALL), in
tion link, a claimant must show a similar       action against oil company; physician
exposure to the subjects studied and ex-        board-certified in toxicology testified he
clude other plausible causes with reason-       imposed only a 51 percent certainty when
able certainty.                                 testifying in court even though current
10. Waters and Water Courses O209               epidemiological standards required 95 per-
     Epidemiological studies were legally       cent, while doctor of epidemiology reinter-
insufficient to establish that child’s expo-    preted inconclusive study by the Agency
sure to benzene in subdivision’s water sup-     for Toxic Substances and Disease Registry
ply, contaminated as a result of oilfield       (ATSDR) of subdivision into a statistically
leak that occurred 50 years previously,         significant one by adding to the study.
caused his acute lymphocytic leukemia           13. Evidence O571(9)
(ALL), in action against oil company; stud-
                                                     Evidence was legally insufficient to
ies were statistically inconclusive or de-
                                                establish that child’s exposure to benzene
pended on unreliable data.
                                                in subdivision’s water supply, contaminat-
11. Waters and Water Courses O209               ed as a result of oilfield leak 50 years
     Study by the Agency for Toxic Sub-         previously, caused his anemia or other ail-
stances and Disease Registry (ATSDR) of         ments, or any disease child in reasonable
subdivision whose water supply had been         probability would suffer in the future; phy-
contaminated as a result of oilfield leak       sician admitted that medical records con-
was legally insufficient to establish that      tained no studies excluding iron-deficiency
child’s exposure to benzene in subdivision’s    anemia, which was the most common form
water supply caused his acute lymphocytic       of anemia and not associated with benzene,
leukemia (ALL); though study’s authors          and thus other plausible causes of child’s
concluded data could be used to identify        anemia were not excluded with reasonable
potential associations, their report stated     certainty, study by the Agency for Toxic
that study could not be used to identify a      Substances      and     Disease    Registry
causal relationship, plaintiffs did not pres-   (ATSDR) reported a statistically signifi-
ent any scientific evidence of whether ef-      cant decrease in asthma at subdivision, no
forts in study to mitigate possible overre-     epidemiological studies or other evidence
porting of health problems was too little or    connected benzene exposure with an ap-
too much, large number of health effects        pearance of nosebleeds, asthma and skin
were surveyed meaning increased likeli-         rashes, and physician admitted no one
                                 EXXON CORP. v. MAKOFSKI                                  Tex.   179
                        Cite as 116 S.W.3d 176 (Tex.App.—Houston [14th Dist.] 2003)

could state with reasonable medical cer-               the 1930s, one was found to be leaking oil
tainty that child would develop any form of            and gas into an underground aquifer in
cancer in the future.                                  1939. Humble Oil & Refining Company
14. Damages O185(3)                                    attempted to recover the lost hydrocar-
                                                       bons by ‘‘backflowing’’ the well for almost
     Recovery for future diseases is pro-
                                                       two years.
hibited unless there is a reasonable medi-
cal probability the disease will occur.                  Forty years later and a half mile to the
15. Damages O49.10                                     southeast, a water well was drilled for the
     Recovery of mental anguish damages                Three Lakes subdivision. When the well
for an increased risk of developing a dis-             was first tested for organic chemical con-
ease that is not presently manifest is pro-            taminants in 1990 (after about 10 years of
hibited.                                               use), it was discovered to contain benzene.
                                                       The subdivision’s water supply was shifted
16. Damages O192
                                                       to an uncontaminated source shortly there-
     Homeowners who lived in subdivision
whose water supply had been contaminat-
ed as a result of an oilfield leak 50 years              In 1993, several hundred residents of
previously and had been exposed to ben-                Three Lakes sued Exxon Corporation, al-
zene were not entitled to recover damages              leging the oilfield leak more than 50 years
for their mental anguish; evidence did not             before involving its predecessor contami-
establish that any of their medical condi-             nated the water well and caused them
tions were related to benzene, and evi-                various health problems. Six residents
dence did not establish there was a rea-               were selected for a bellwether trial, which
sonable medical certainty that they would              took place over the better part of six
develop a future disease as a result of ex-            weeks.
                                                         At the conclusion, the jury found Exxon
                                                       negligent and grossly negligent, and
                                                       awarded a total of almost $7 million in
                                                       actual and punitive damages. In post-
  Clinard J. Hanby, Spring, Pamela Stan-
                                                       judgment proceedings, the trial court re-
ton Baron, for appellant.
                                                       duced or eliminated most of the jury
  John O’Quinn, Robert Alan York, David                awards, rendering take-nothing judgments
Holman, W. Mark Lanier, Houston, for                   against the four adults and reduced judg-
appellees.                                             ments for the two minors.
  Panel consists of Chief Justice SCOTT                   All parties appeal. Exxon contends
BRISTER and Justices EDELMAN and                       there is no evidence its negligence caused
SEYMORE.                                               either the contamination of the water well
                                                       or the plaintiffs’ specific diseases. The
                                                       plaintiffs contend the trial court erred in
     SCOTT BRISTER, Chief Justice.                     disregarding the jury’s awards. Both are
  Among the hundreds of wells drilled in               governed by the same standard—whether
the Tomball Field northwest of Houston in              any evidence supports the jury’s verdict.1

1.    See Merrell Dow Pharms., Inc. v. Havner,           judgment in accordance with jury verdict im-
     953 S.W.2d 706, 711 (Tex.1997) (holding             proper if no scientifically reliable evidence
180         Tex.       116 SOUTH WESTERN REPORTER, 3d SERIES

Because legal standards for reviewing               miological evidence connecting a chemical
medical causation are much clearer than             exposure and a disease.6
those for underground hydrology, we ad-                [2] We also disagree with our dissent-
dress it first. Finding it dispositive, we go       ing colleague that Exxon waived error by
no further.                                         failing to obtain a record from a pretrial
                                                    hearing at which its motions to exclude
               I.   What We Review                  allegedly unreliable expert testimony were
                                                    first considered. Following this argument
  [1] Exxon challenges the opinions of
                                                    to its logical conclusion, every time a pre-
the plaintiffs’ four medical experts, argu-
                                                    trial motion to strike expert testimony is
ing they presented no scientific basis for
                                                    denied without a reporter’s record, we
concluding exposure to the water at Three
                                                    must presume any jury verdict thereafter
Lakes caused the plaintiffs’ diseases. We
                                                    is valid (no matter how scientifically un-
consider all the evidence in the light most
                                                    sound). This argument is neither raised
favorable to the verdict, and indulge every         nor briefed by the parties, apparently for
reasonable inference in that direction.2            the same reason they never requested that
But we do not take expert testimony at              the hearing be recorded—because no testi-
face value; if it is unreliable, it constitutes     mony was taken at the pretrial hearing.
no evidence, and we must reverse.3                  Not a single line in any party’s brief sug-
   We disagree with the plaintiffs that two         gests we should look to the pretrial hear-
different approaches might apply—one                ing for evidence of reliability, or that it
based on the six factors applied in Merrell         included anything other than what was
Dow Pharmaceuticals, Inc. v. Havner,4               repeated a few days later at trial.
and one that looks only for ‘‘analytical               [3, 4] While this appears to be an issue
gaps’’ in the testimony.5 Undoubtedly, the          of first impression, the Supreme Court has
tools used to test the reliability of expert        given us some guidance. First, it was the
testimony will vary depending on the field          plaintiffs’ burden to present scientifically
of expertise involved. But it is impossible         reliable evidence of causation at trial.7
to ignore the Havner factors here, as the           Second, Exxon was required to object at
field of expertise is the same—the epide-           trial (which it did repeatedly) 8 so the

     supports it); Brown v. Bank of Galveston, N.   7.     See Havner, 953 S.W.2d at 720; E.I. du Pont
     A., 963 S.W.2d 511, 513 (Tex.1998) (holding         de Nemours and Co. v. Robinson, 923 S.W.2d
     judgment notwithstanding jury verdict proper        549, 557 (Tex.1995).
     only if no evidence supports it).
                                                    8. See Maritime Overseas Corp. v. Ellis, 971
                                                      S.W.2d 402, 409 (Tex.1998). Shortly before
2.     See Havner, 953 S.W.2d at 711; see also
                                                      trial, Exxon filed a motion in limine request-
     Daubert v. Merrell Dow Pharms., Inc., 509
                                                      ing exclusion of several of the plaintiffs’ ex-
     U.S. 579, 595, 113 S.Ct. 2786, 2798, 125
                                                      perts, including 60 pages of briefing, a 28–
     L.Ed.2d 469 (1993).
                                                      page appendix applying the brief to each
                                                      plaintiff, 127 pages of deposition excerpts,
3.    See Havner, 953 S.W.2d at 714.                  and 55 pages of medical studies. Immediate-
                                                      ly before Dr. Goldstein and Dr. Shalat took
4.    Id.                                             the stand at trial, Exxon moved to exclude
                                                      their testimony, and immediately after they
5.     See Gammill v. Jack Williams Chevrolet,        finished moved to strike their testimony, each
     Inc., 972 S.W.2d 713, 727 (Tex.1998).            time requesting and the trial court allowing
                                                      the motion in limine to be incorporated by
6.     See Daubert, 509 U.S. at 591–94, 113 S.Ct.     reference in lieu of an extended oral objec-
     at 2795–97; Havner, 953 S.W.2d at 711–14.        tion.
                                   EXXON CORP. v. MAKOFSKI                                    Tex.   181
                          Cite as 116 S.W.3d 176 (Tex.App.—Houston [14th Dist.] 2003)

plaintiffs would have an opportunity to                  sufficiency of a jury verdict by the evi-
cure any defects regarding reliability and               dence at the jury trial, not by what hap-
present us with a fully developed record.9               pened before or after it.13 Moreover,
Neither of these guidelines suggests a rule              there is an important difference between
that, after its objections were overruled, it            presuming and pretending. Here, the trial
remained Exxon’s duty to incorporate into                record makes clear that no evidence was
the trial record any and all pretrial evi-               presented at the pretrial hearing.14 And if
dence that supported the reliability of the              (as discussed below) the evidence at trial
plaintiffs ’ experts.10                                  showed that scientific evidence could not
                                                         support a particular verdict, any presump-
  [5–7] It is generally true that unless
                                                         tion that science held the opposite view a
an appellant files a complete reporter’s
                                                         few days earlier has been effectively rebut-
record (or a limited appeal 11), we presume
the omitted portions are relevant and sup-
port the jury’s verdict.12 But a complete                  [8] Finally, our colleague’s reliance on
record does not include matters from other               a treatise discussing the admissibility of
proceedings; we normally test the legal                  expert testimony 16 shows he misconstrues

9.    See Maritime Overseas, 971 S.W.2d at 412.            pretrial hearing. On the critical connection
                                                           between benzene and leukemia, Dr. Bernard
10. We also disagree that this necessarily re-             Goldberg admitted he provided neither ani-
  quires repetition of pretrial evidence before            mal studies nor epidemiological articles to
  the jury. Either party at trial may incorpo-             Exxon’s counsel at any time between his de-
  rate by reference any proof from the pretrial            position and trial. Finally, the very capable
  hearing they wish made part of the trial rec-            attorneys and experienced trial judge all cer-
  ord. Here, neither did.                                  tainly knew when a record is required and
                                                           when it is not.
11.    See TEX.R.APP. P. 53(d).
                                                         15. The presumption that omitted portions of a
12. See Feldman v. Marks, 960 S.W.2d 613,
                                                           record support a verdict is rebuttable. See
  614 (Tex.1996).
                                                           Polanco v. Pan Amer. Univ., 818 S.W.2d 97,
                                                           99 (Tex.App.-Corpus Christi 1991, no writ);
13. As neither party relies on any evidence
  from the pretrial hearing, we need not decide            Candelier v. Ringstaff, 786 S.W.2d 41, 44 (Tex.
  whether a legal sufficiency review should in-            App.-Beaumont 1990, writ denied); see also
  clude evidence of scientific reliability that is         Garza v. Texas Alcoholic Beverage Comm’n, 83
  never presented to jurors. See Campbell v.               S.W.3d 161, 166 (Tex.App.-El Paso 2000),(ap-
  State, 85 S.W.3d 176, 184 (Tex.2002) (refusing           plying same rule to presumption in favor of
  to review issue of legal sufficiency without             judgments generally), aff’d, 89 S.W.3d 1 (Tex.
  adequate briefing and argument); see also                2002) cf. Vickery v. Comm’n for Lawyer Disci-
  Harvey Brown, Procedural Issues Under Dau-               pline, 5 S.W.3d 241, 252, 256 (Tex.App.-Hous-
  bert, 36 HOUS. L.REV. 1133, 1139 (1999) (sug-            ton [14th Dist.] 1999, pet. denied) (holding
  gesting parties ‘‘presumably’’ need not re-of-           denial of request for negative finding of fact
  fer evidence from pretrial hearing before jury,          contrary to court’s judgment did not rebut
  and that evidence from pretrial hearing will             presumption of validity of trial court’s ver-
  be considered ‘‘indirectly’’ in reviewing suffi-         dict). Thus, for example, Dr. Michael Gray’s
  ciency of the evidence).                                 trial testimony that there were no tests in
                                                           John Russell’s medical records excluding
14. At trial, Exxon’s counsel introduced him-              iron-deficiency anemia (a form not associated
  self to each expert by noting previous deposi-           with benzene) rebuts any presumption that
  tions or hearings at which they had crossed              the pretrial hearing included such evidence
  swords, but no mention is made of a pretrial             from the medical records.
  hearing a few days before trial. Each expert
  was examined at length about his deposition            16. Michol O’Connor, O’CONNOR’S TEXAS
  testimony, but none about testimony from any             RULES: [CIVIL TRIALS 289, § 3.5 (2002).]
182    Tex.          116 SOUTH WESTERN REPORTER, 3d SERIES

Exxon’s point. Challenges to both the ad-         checkups since 1994 have shown no indica-
missibility and the legal sufficiency of ex-      tions of recurrence.
pert testimony involve the same substan-            The plaintiffs alleged James’s ALL was
tive guidelines for determining relevance         a result of his exposure to benzene in the
and reliability.17 But admissibility of ex-       Three Lakes water supply. It was undis-
pert testimony is an initial decision made        puted that current scientific knowledge
by the trial judge, usually outside the           cannot prove what causes ALL in any
jury’s presence, and is subject to an abuse-      particular instance, and that in 95 percent
of-discretion review.18 By contrast, legal        of ALL cases the cause is never known.
sufficiency of expert testimony tests the         Because of so many uncertainties, efforts
jury’s decision after the trial, looking at       to establish liability for ALL based on
the whole trial record to see if any evi-         exposure to hazardous substances have
dence supports the verdict.19 Had Exxon           proved daunting.21 Here, the plaintiffs’
complained of the pretrial ruling on admis-       attorneys used epidemiological studies, a
sibility, we would need a record of that          government study, and expert testimony to
hearing to review it. But Exxon does not.         try to raise a fact question on causation.
It is not our place to change this legal             The jury found Exxon’s negligence
sufficiency point into something else, or         caused James’s injuries, and awarded him
dismiss it by reviewing a different decision      $1.5 million for past medical care, $1 mil-
by a different decision-maker in a different      lion for future medical care, $400,000 for
proceeding according to a different stan-         pain and anguish, and $325,000 for impair-
dard of review.                                   ment. After reducing the award for future
                                                  medical expenses to $156,603 (the only es-
           II.   Benzene and ALL                  timate of costs presented at trial), the trial
                                                  court entered judgment in accordance with
  James Makofski, Jr., was an infant when         these awards.       Exxon challenges all
his family moved to Three Lakes in 1985.          awards for lack of causation; the plaintiffs
In September 1991, he was diagnosed with          challenge the trial court’s reduction of the
acute lymphocytic leukemia (ALL).20               future medical award.
Treatment of ALL is expensive and pain-
ful; in James’s case, it included three                  The Epidemiological Studies
years of repeated rounds of chemotherapy            [9] In Havner, the Texas Supreme
and spinal injections. But the trial testi-       Court held that scientifically reliable epi-
mony also indicated current treatment is          demiological studies showing more than a
usually successful; in James’s case, annual       doubling of the risk may establish a causa-

17.   See Havner, 953 S.W.2d at 712.                from Three Mile Island incident); Whiting v.
                                                    Boston Edison Co., 891 F.Supp. 12 (D.Mass.
18.   See Robinson, 923 S.W.2d at 558.              1995) (granting motion to strike expert testi-
19.   See Havner, 953 S.W.2d at 720.                mony on Daubert grounds that attempted to
                                                    link ALL to exposure to ionizing radiation);
20. Sometimes referred to as acute lymphob-         see also JONATHAN HARR, A CIVIL ACTION (Ran-
  lastic leukemia.                                  dom House 1995) (studying trial involving six
                                                    cases of ALL allegedly caused by under-
21. See In re TMI Litigation, 193 F.3d 613 (3rd
                                                    ground leakage of trichloroethylene into com-
  Cir.1999) (affirming summary judgment strik-
                                                    munity water well in Woburn, Massachu-
  ing expert testimony on Daubert grounds that
  attempted to link ALL to radiation exposure
                                  EXXON CORP. v. MAKOFSKI                                   Tex.   183
                         Cite as 116 S.W.3d 176 (Tex.App.—Houston [14th Dist.] 2003)

tion link.22 To benefit from this exception             the record. A similar disagreement ap-
to the normal rules requiring individual                pears to exist among Texas appellate
proof of causation, a claimant must show a              courts: the Sixth Court of Appeals has
similar exposure to the subjects studied                held studies cannot be considered on ap-
and exclude other plausible causes with                 peal unless they are ‘‘admitted into evi-
reasonable certainty.23                                 dence,’’ 25 but the Fourth Court of Appeals
   Benzene has been known to be potential-              has relied on studies attached to an appel-
ly harmful to humans for more than a                    late brief.26
century. Thus, a large number of epide-                    In Havner, the Court discussed studies
miological studies have been conducted on               included in the record as well as studies
its effects. The parties’ experts agreed                that were only referred to in testimony.
that benzene has been shown to cause                    The Court found legally insufficient any
acute myelogenous leukemia (AML), the                   studies as to which there was no informa-
most common form of leukemia found in                   tion in the record (either from testimony
adults. But the same studies have not                   or the study itself) establishing confidence
reached the same conclusion regarding                   levels and intervals.27 This example sug-
ALL, the most common form found in                      gests testimony from a study may alone be
children.                                               enough, but it must be quite detailed.
   Unfortunately, almost none of the rele-                [10] Because of the uncertainty as to
vant epidemiological studies appear in the              what sources we can review, we asked
trial court record, though several were                 each party (over Exxon’s objection) to sub-
marked for identification and discussed at              mit an appendix with copies of the studies
length during the trial. As learned treat-              on which they rely. Having reviewed
ises, excerpts from these studies could be              them all, we find that even under the more
read to jurors, but none were admitted as               lenient approach (allowing these studies to
exhibits for jurors’ independent review.24              be supplemented on appeal) the evidence
But while the rules of evidence withhold                is legally insufficient under the Havner
learned treatises from jurors, that does                standards:
not mean they should be withheld entirely                 1. The Chinese Studies.28 This series
from the record. Without them, we are                   of studies included more than 100,000
hard-pressed to conduct the kind of review              workers, the majority of whom had been
Havner requires.                                        exposed to benzene in factories in China.
  Each party here says the other had the                For ALL, the study found a relative risk
responsibility to place relevant studies in             of 2.8 with a confidence interval of 0.5 to

22.   953 S.W.2d at 718.                                27.   See Havner, 953 S.W.2d at 725–26.

23.   Id. at 720.                                       28. Song–Nian Yin, et al., A Cohort Study of
                                                          Cancer Among Benzene-exposed Workers in
24.   See TEX.R. EVID. 803(18).
                                                          China: Overall Results, 29 AM. J. INDUS. MED.
25. See Minn. Mining and Mfg. Co. v. Atterbury,           227 (1996); Song–Nian Yin, et al., An Ex-
  978 S.W.2d 183, 198 (Tex.App.-Texarkana                 panded Cohort Study of Cancer Among Ben-
  1998, pet. denied).                                     zene-exposed Workers in China, 104 ENVTL.
                                                          HEALTH PERSP. 1339 (1996); Song–Nian Yin, et
26. See Texas Workers’Comp. Ins. Fund v. Lo-              al., Leukaemia in Benzene Workers: a Retro-
  pez, 21 S.W.3d 358, 364–65 (Tex.App.-San                spective Cohort Study, 44 BRIT. J. INDUS. MED.
  Antonio 2000, pet. denied).                             124 (1987).
184     Tex.         116 SOUTH WESTERN REPORTER, 3d SERIES

54.5 at a 95% confidence level. As this                tion of the cohort follow-up periodTTTT
interval includes 1.0, the finding was not             Indeed, if even one ALL case was the
statistically significant (that is, it was in-         result of misdiagnosis, the ALL [finding]
conclusive).29 The authors concluded that              would not have been statistically signifi-
risks were ‘‘markedly increased’’ for ALL,             cant. In this regard, it is noteworthy
but that ‘‘the excess risk for this condition          that one of the ALL death certificates
was not statistically significant.’’ 30                listed the cause of death as ‘‘Alymphatic
                                                       [sic] leukemia’’. As this is not a gener-
   2. The US/UK Study.31 This meta-
                                                       ally recognized diagnosis, we could not
analysis of more than 200,000 petroleum
                                                       determine the cell type of this cancer.
workers in the United States and United
                                                       Nevertheless, we coded it as acute lym-
Kingdom found a relative risk for ALL of
                                                       phocytic leukemia in our analysis and as
1.32 with a confidence interval of 0.81 to
                                                       such, the statistical significance of the
2.01 at a 95% confidence level. As this
                                                       ALL result rests on a questionable diag-
interval again includes 1.0, the finding was
                                                       nosisTTTT No epidemiologic study has
not statistically significant. The authors
                                                       demonstrated an association between
concluded that no increased risk for ALL
                                                       exposure to benzene and acute lympho-
or several other specific types of leukemia
                                                       cytic leukemia. Similarly, no other pe-
had been shown among petroleum workers
                                                       troleum refinery study has reported an
in the two countries.32
                                                       increase of ALLTTTT Thus, the finding
  3. The Port Arthur Study.33 This                     from the Port Arthur refinery was in-
study looked at the causes of death among              consistent with other studies of refinery
18,000 workers at a refinery in Port Ar-               workers with similar exposures. The
thur, Texas between 1937 and 1978. For                 likely spurious nature of the ALL find-
ALL, the study found a relative risk of 2.6            ing in the Port Arthur cohort is further
with a confidence interval of 1.1 to 5.1 at a          supported by the lack of an exposure-
95% confidence level.34 But the authors                response relationship based on length-
concluded the ALL finding was spurious                 of-employment analysis.35
because of historical difficulties in distin-           4. Other Studies Submitted. Several
guishing it from other leukemias:                    other studies were submitted that con-
  From a diagnostic perspective, it might            tained no statistical information on any
  have been difficult historically to distin-        association between benzene and ALL.
  guish between patients with CLL                    For example, one study found a significant
  [chronic lymphocytic leukemia] and                 association between ALL and other lymp-
  those with ALL. This difficulty would              hatic leukemias with exposure to benzene
  likely have prevailed during a large por-          and other solvents, but because of the

29.   See Havner, 953 S.W.2d at 723.                 33. K.P. Satin, et al., A 50–year Mortality Fol-
                                                       low-up of a Large Cohort of Oil Refinery Work-
30. Yin, Overall Results, 29 AM. J. INDUS. MED. at     ers in Texas, 38 J. OCCUP. ENVTL. MED. 492
  232, 233.                                            (1996).

                                                     34. For ease of comparison to the Havner stan-
31. Gerhard K. Raabe and Otto Wong, Leu-
                                                       dards, these figures have been converted from
  kemia Mortality by Cell Type in Petroleum
                                                       the standardized mortality ratios used in the
  Workers with Potential Exposure to Benzene,
                                                       article, which are based on a relative risk of
  104 ENVTL. HEALTH PERSP. 1381 (1996).                100.0 rather than 1.0.

32.   Id. at 1389.                                   35.   Satin, supra, at 503.
                                EXXON CORP. v. MAKOFSKI                                   Tex.   185
                       Cite as 116 S.W.3d 176 (Tex.App.—Houston [14th Dist.] 2003)

confounding chemicals and diseases the                  In 1990, ATSDR chose to study the
study tells us nothing definitive about ben-          Three Lakes subdivision shortly after the
zene and ALL.36 Another article stated it             contamination of the water supply was dis-
would be ‘‘surprising’’ if benzene caused             covered and discontinued.       Interviews
only AML, but limitations in the literature           were conducted in person and by tele-
precluded any more definitive conclusion.37           phone with about 1100 residents, with fol-
   5. Other Studies Mentioned. The vari-              low-up interviews in subsequent years.
ous experts briefly mentioned several oth-            The data was collected in a ‘‘Benzene Su-
er studies, but only in passing. For exam-            bregistry’’ report, and in 1998 two of the
ple, some testified that an author or a               primary investigators published a peer-re-
study ‘‘says benzene causes ALL’’ or ‘‘sup-           viewed study of the findings.39
ports my opinion.’’ But none of these                    In both, statistically significant increases
studies have been supplied to us (even                at a 99% confidence level were reported (at
after our specific request), and nothing              least for some age groups in some years)
gives us any indication of the scope, de-             for anemia, ulcers, gall bladder trouble,
sign, relative risk, confidence interval, con-        stomach or intestinal problems, stroke, uri-
fidence level, or even whether benzene or             nary tract disorders, skin rashes, diabetes,
ALL were actually involved. Following                 kidney diseases, and respiratory allergies.
Havner’s example, we cannot grant even a              By contrast, statistically significant de-
scintilla of weight to these passing refer-           creases were found at the same confidence
ences.38                                              level for asthma, emphysema, arthritis,
                                                      rheumatism, and hearing and speech im-
        The Three Lakes Registry
                                                      pairments. No statistically significant dif-
   [11] In addition to general epidemio-              ferences were found for liver disease, men-
logical studies, the plaintiffs rely on a             tal retardation, or cancer. There was no
study by the Agency for Toxic Substances              finding whatsoever regarding ALL.
and Disease Registry (ATSDR), a division
of the federal Department of Health and                 For several reasons, the ATSDR find-
Human Services that gathers health infor-             ings are insufficient to establish causation
mation on persons exposed to hazardous                under the Havner standards. First and
substances. In 1989, the agency selected              foremost, that is because the report itself
benzene to study because of its danger to             says they cannot:
humans, its ubiquity in the environment                  The findings in this report cannot be
(from work sites, waste dumps, gasoline,                 used to identify a causal relationship
automobile exhaust, tobacco products, and                between the health outcomes and ben-
foods), and the lack of scientific informa-              zene exposure. In addition, there are
tion on the effect of low-level, long-term               some methodological differences in data
exposure.                                                collection that may have biased the re-

36. Anthony J. McMichael, et al., Solvent Expo-       38.   See Havner, 953 S.W.2d at 725–26.
  sure and Leukemia among Rubber Workers: An
  Epidemiologic Study, 17 J. OCCUP. MED. 234          39. Jeanne R. Burg and Ginger L. Gist, The
  (1975).                                               National Exposure Registry: Analyses of
                                                        Health Outcomes From the Benzene Subregis-
37. David A. Savitz and Kurtis W. Andrews,
  Review of Epidemiologic Evidence on Benzene           try, 14 TOXICOLOGY & INDUS. HEALTH 367 (1998).
  and Lymphatic and Hematopoietic Cancers, 31
  AM. J. INDUS. MED. 287 (1997).
186    Tex.          116 SOUTH WESTERN REPORTER, 3d SERIES

  porting rates, resulting in false positive        large number of health effects are sur-
  findings. The findings of this report do,         veyed, there is an increased likelihood that
  however, reinforce the need to continue           random chance will produce a statistically
  regular followup of this population.              significant association when in fact none
Given this conclusion, the plaintiffs cannot        exists.42
use the subregistry data to establish cau-
                                                       Fourth, although the authors concluded
sation. Havner does not allow us to give
                                                    the Three Lakes data could identify poten-
any weight to in-court conclusions about a
                                                    tial associations,43 that does not justify
study that disagree so diametrically with
                                                    making causation conclusions from it. The
the published conclusions of the research-
                                                    ATSDR’s express purpose for studying
ers who conducted it.40
                                                    Three Lakes was the lack of epidemiologi-
  Second, the ATSDR’s reference to po-
                                                    cal information on the health effects of
tential bias recognizes the difficulty in
                                                    long-term exposure to low levels of ben-
gathering accurate data when many of the
                                                    zene. As Havner notes, even statistically
respondents are parties in this litigation.
                                                    significant associations in an isolated study
Experts for all parties agreed that those
                                                    are no evidence of causation, as scientific
involved in litigation are likely to report
                                                    methodology requires some confirmation.44
higher levels of medical problems. The
ATSDR researchers attempted to mitigate                Finally, the ATSDR data is too general
this problem by asking residents not                to support the kind of specific uses the
whether they had a particular health con-           plaintiffs try to make of it. In their brief,
dition, but whether a physician had told            the plaintiffs make the rather shocking
them they had it. What is missing from              assertion that Exxon did not dispute gen-
the plaintiffs’ efforts to use the ATSDR            eral causation, despite days of cross-ex-
data is any scientific evidence of whether          amination and counter witnesses called
that mitigation was too little or too much.         entirely for that purpose. Apparently,
   Third, the ATSDR also noted another              this is a reference to Exxon’s concession
potential source for false positives—the            that benzene is associated with one type
large sample size and number of compari-            of leukemia (AML), but Exxon never con-
sons in the study. This comment appar-              ceded the same as to ALL. While lumping
ently refers to the unfortunately-named             distinct diseases together as ‘‘leukemia’’
‘‘Texas Sharpshooter Fallacy,’’ in which            may yield a statistical increase as to the
natives of this state are alleged to shoot at       whole category, it does so only by ignor-
the side of a barn and then draw a target           ing proof that some types of the disease
where the most holes are located, thereby           have a much greater association with ben-
establishing the accuracy of their mark-            zene than others. The ATSDR made no
smanship.41 Epidemiologists use the term            attempt to distinguish among many par-
to identify the phenomenon that when a              ticular types of diseases with apparently

40. 953 S.W.2d at 729–30 (holding expert’s            of the Law, 81 CORNELL L.REV. 953, 960 n. 41
  conclusion that substance caused birth de-          (1996) (book review).
  fects, based on report in which researchers
  said it was only potentially capable of doing     42.   Id.
  so, was not evidence of causation).

41. See Boughton v. Cotter Corp., 65 F.3d 823,      43.   Burg and Gist, supra, at 383.
  835 n. 20 (10th Cir.1995); Robert F. Blom-
  quist, Bottomless Pit: Toxic Trials, The Ameri-   44.   See Havner, 953 S.W.2d at 727.
  can Legal Profession, and Popular Perceptions
                                 EXXON CORP. v. MAKOFSKI                                 Tex.   187
                        Cite as 116 S.W.3d 176 (Tex.App.—Houston [14th Dist.] 2003)

different etiologies. The categories of                view’’ requires 95 percent certainty, he
diseases the ATSDR uses—such as ‘‘ane-                 imposed only a 51 percent certainty when
mia or other blood disorders,’’ ‘‘skin rash-           testifying in court. Havner expressly re-
es, eczema, and other skin allergies,’’ or             jects that argument:
‘‘cancers’’—are simply too general to meet                The generally accepted significance level
Havner’s standards.                                       or confidence level in epidemiological
                                                          studies is 95% TTT Virtually all the pub-
                 The Experts
                                                          lished, peer-reviewed studies on Bendec-
   Finally, the plaintiffs also presented two             tin have a confidence level of at least
experts to establish a causal connection                  95%TTTT We think it unwise to depart
between benzene and ALL: Dr. Bernard                      from the methodology that is at present
Goldstein, a physician board-certified in                 generally accepted among epidemiolo-
internal medicine, hematology, and toxicol-               gists.48
ogy, and Dr. Stuart Shalat, a doctor of
epidemiology. Together, their testimony                  Dr. Goldstein justified his departure
illustrates most of the practices Havner               from current epidemiological standards by
declares to be inconsistent with sound                 pointing to lower standards used by
methodology among epidemiologists.                     OSHA, the EPA, or physicians who treat
   [12] At trial, both experts testified that
                                                          [Plaintiffs’ counsel]: If you wait till you
benzene causes ALL, despite the many
                                                          have this 10–to–1 exactitude that [Exxon
published, peer-reviewed epidemiological
                                                          is] urging, what’s going to really happen
studies finding no statistically significant
                                                          in this world?
relationship. Neither, however, appears to
have ever offered the same opinion in an                  [Dr. Goldstein]: There will be a lot of
established scientific journal. In his own                sick people TTT I just wouldn’t be prac-
textbook, Dr. Goldstein listed the associa-               ticing good medicine or public health if I
tion between benzene and ALL as ‘‘sug-                    waited till I was absolutely certainTTTT I
gested but unproven’’ in 1983, as ‘‘suggest-              can’t wait as a physician until I’m 95
ed’’ in 1986, and in 1993 as a ‘‘biomedical               percent sure of something before I can
plausibility.’’ 45  Plausibility    is    not             deal with the patient’s potential prob-
enough.46 Havner instructs us to be espe-                 lems.
cially skeptical of scientific evidence that              Again, the Havner Court expressly dis-
has not been published or subjected to                 tinguished between Type I (false positive)
peer review.47                                         and Type II (false negative) statistical er-
  Dr. Goldstein explained his divergence               rors when deciding what confidence level
from the published literature on his belief            is appropriate.49 As noted in the examples
that a lower confidence level should be                given in Havner, differences in costs and
used in court. Although he admitted cau-               benefits make false positives acceptable in
sation ‘‘from a scientifically proven point of         some situations but unacceptable in oth-

45. Dr. Goldstein testified he was upgrading           47.   Id. at 726–27.
  this designation to ‘‘probable’’ in an edition
  not yet published, but did not indicate what         48.   Id. at 723–24.
  (if any) peer-review was involved in the publi-
  cation.                                              49.   Id. at 722–23.
46.   See Havner, 953 S.W.2d at 729.
188     Tex.           116 SOUTH WESTERN REPORTER, 3d SERIES

ers.50 For example, it may be appropriate                                Conclusion
for the EPA to protect people from chemi-
                                                       In sum, the plaintiffs presented evidence
cal exposure on weak evidence that it will
                                                    that several experts believe a causal con-
cause any harm, but that does not make it
                                                    nection exists between benzene and ALL,
equally appropriate to impose a judgment
                                                    while others do not. But an expert’s belief
of several million dollars on weak evidence
                                                    that a substance might cause a disease is
that a defendant caused any harm. Hav-
                                                    no evidence that in reasonable probability
ner specifically bars this effort to lower
                                                    it did.54 No epidemiological study estab-
epidemiological confidence levels for court-        lishes a statistically significant doubling of
room use.51                                         the risk of ALL from exposure to benzene;
  Additionally, Dr. Goldstein suggested             accordingly there was no evidence that—
causation conclusions regarding AML                 more probably than not—James’s ALL
should lead to the same conclusions re-             was caused by exposure to benzene in the
garding ALL, as the diseases have ‘‘similar         Three Lakes water supply. We hold the
characteristics.’’ But he admitted all epi-         evidence is legally insufficient to support
demiological studies in the last twenty             James Makofski’s damage awards.
years invariably treat these malignancies
separately, and every study filed with us                 III.   Benzene and Other Diseases
shows significantly different findings as to
                                                       The only other plaintiff who obtained
each. We cannot treat an epidemiological
                                                    judgment against Exxon was John Russell.
study regarding one disease as if it applied
                                                    Born in November 1990, he was found to
to another.52
                                                    be anemic in a blood test taken five
   Finally, Dr. Shalat reinterpreted the            months after birth. Subsequent blood
ATSDR’s inconclusive finding regarding              tests were occasionally ‘‘borderline,’’ but
cancer into a statistically significant one by      there has been no diagnosis of anemia
adding to the study—which specifically did          since his early infancy. His parents testi-
not study any deaths in the Three Lakes             fied he has suffered from recurrent nose-
subdivision—statistics for the general pub-         bleeds, asthma, and skin rashes.
lic regarding deaths from cancer. Havner               The jurors found John had suffered ‘‘$0’’
expressly prohibits reanalysis of a study to        for all damages alleged in the past, but
derive a significant finding if sound scienti-      awarded $200,000 for future pain and men-
fic methodology would not draw the same             tal anguish, $100,000 for future impair-
inference.53 Even if the incidence of all           ment, and $176,000 for future medical care.
types of cancer was relevant (which it is           Exxon asserts there is no evidence John’s
not for the reasons stated above), there is         exposure to benzene caused any of his
no evidence that Dr. Shalat’s reconfigura-          ailments, or that this exposure more than a
tion to reach a conclusion that the authors         decade ago will in reasonable probability
did not is scientifically sound.                    result in any future damages.

50.   Id.                                             could not support a finding regarding limb
                                                      reduction defects).
51.   Id. at 722–24.

52. Id. at 725 (holding published studies show-     53.    Id. at 720, 727.
  ing statistically significant results regarding
  birth defects other than limb reduction defects   54.    Id. at 729.
                               EXXON CORP. v. MAKOFSKI                                 Tex.   189
                      Cite as 116 S.W.3d 176 (Tex.App.—Houston [14th Dist.] 2003)

   [13] As with leukemia, experts for both           John had rather than diseases, and argued
parties agreed that some types of anemia             these symptoms were similar to those that
have been shown to be associated with                preceded James Makofski’s ALL diagno-
benzene exposure and other types have                sis. His entire closing argument for
not. Dr. Michael Gray, an internist board            John’s damages was an attempt to connect
certified in occupational medicine who tes-          the two boys’ ailments:
tified for the plaintiffs, said he believed             You look at John RussellTTTT What sum
John’s anemia was attributable to benzene,              of money for John Russell? He’s eight
but admitted the medical records con-                   years old. He’s at the age that James
tained no studies excluding iron-deficiency
                                                        was. He’s got the same nosebleeds.
anemia—the most common form, and one
                                                        He gets told about the same flus. He’s
not associated with benzene. As John has
                                                        also got an asthma problem, not related
not been diagnosed with anemia since
                                                        in the sense that the asthma is caused
then, there has been nothing to support
                                                        by the benzene, but it makes every cold
this conclusion thereafter. Without estab-
                                                        he has worse for the asthma. He’s got
lishing the type of John’s anemia, Dr. Gray
                                                        the anemia. as Dr. Gray told you, you
did not exclude other plausible causes of it
                                                        know, this is a boy—you might check his
with reasonable certainty.55
                                                        anemia one day and it may be okay.
   Nor was there any other reliable scienti-            Other days it’s notTTTT You won’t find
fic evidence connecting benzene to any of               many eight year olds that sit that still.
John’s other complaints. Dr. Gray admit-                That’s not just well-mannered. That’s a
ted asthma was not caused by benzene                    boy who is a little different because his
(and as noted above, the ATSDR reported                 body does not have the energy and be-
a statistically significant decrease in asth-           cause he’s got some problems and Dr.
ma at Three Lakes). He also testified
                                                        Gray’s testified to them. At this point is
benzene had suppressed John’s immune
                                                        it leukemia? No.
system, but offered no epidemiological
studies or other scientific evidence con-            Jurors obviously followed the unstated
necting benzene exposure at levels John              conclusion (‘‘but he might get it in the
experienced with an appearance of nose-              future’’) when they awarded no past dam-
bleeds, asthma, and skin rashes more than            ages but substantial future ones.
a decade later (as would be necessary to               But not a single expert at trial sup-
support the jurors’ awards of future dam-            ported this analogy. All agreed John did
ages only). Although nosebleeds, respira-            not have ALL, and none could say he
tory infections, and skin rashes have many           probably ever would. His testifying ex-
causes, there is nothing in the trial record         pert admitted that no one could state with
to exclude those alternate explanations but          reasonable medical certainty that John
Dr. Gray’s naked assurance.                          would develop any form of cancer in the
  Finally, none of this evidence explains            future, much less ALL. Moreover, as Exx-
why the jury awarded nothing to John for             on points out, the analogy with Makofski
past damages but a substantial amount for            was misleading, as there was no evidence
damages in the future. Clearly, the award            the latter had anemia or nosebleeds for
was based on no disease he has exhibited             several years before his ALL diagnosis.
to date. Instead, throughout the trial the           To the contrary, the only testimony at trial
plaintiffs’ counsel pointed to symptoms              was that Makofski suffered nosebleeds and

55.   Id. at 720.
190     Tex.       116 SOUTH WESTERN REPORTER, 3d SERIES

anemia immediately before his ALL diag-         claim for damages except mental anguish
nosis, due to the massive destruction of        (which they awarded in amounts ranging
bone marrow cells that the disease engen-       from $50,000 to $100,000). The plaintiffs
ders.                                           do not challenge the jury’s verdict that
   [14, 15] Texas law prohibits recovery        they have suffered nothing but mental an-
for future diseases unless there is a rea-      guish. But they do challenge the trial
sonable medical probability the disease will    court’s order setting aside the mental an-
occur.56 It also prohibits recovery of men-     guish awards.
tal anguish damages for an increased risk         Dr. John Wilson, an expert in post-trau-
of developing a disease that is not present-    matic stress syndrome (‘‘PTSD’’), admitted
ly manifest.57 We find there was no legal-      that while none of the four adults would
ly sufficient evidence to support a causa-      meet current medical standards for that
tion connection between benzene exposure        diagnosis, they nevertheless showed some
and any disease John Russell in reasonable      or all of its symptoms. The plaintiffs ar-
probability will suffer in the future.          gue the verdict could not have been based
                                                merely on their fear of contracting future
  IV.    Benzene and Mental Anguish             diseases, as the trial court excluded such
   The remaining four adult plaintiffs as-      evidence. While it is true the trial court
serted a wide range of medical problems         made that order (over the plaintiffs’ re-
continuing up to the time of trial, though      peated objections), the record shows the
by that time none of them had lived in          evidence from the plaintiffs’ testifying psy-
Three Lakes for several years, or been          chologist was nothing but:
exposed to benzene in the subdivision’s           [As to Janie Devora]: She began to have
water supply for a decade. Andrea Rus-               worries about her family. She had
sell (John’s mother) presented evidence              worries about was she going to get a
that she suffers from anemia and keratos-            blood disorder. She began to have
es (benign precancerous skin lesions). Phil-         sleep disturbance. She began to ex-
lip Devora complained of skin rashes, acne,          press more stress and anxiety reac-
and fungal infections on his fingernails.            tions. She began to have concerns
His wife Janie sought damages for anemia,            learning that other people in the com-
benign fibroid growths in her uterus, and            munity had cancer.
excessive menstrual bleeding (which led to        [As to Phillip Devora]: He was a little
her decision to undergo a hysterectomy).             bit what I would term guarded in his
Codi Stennett presented evidence she suf-            presentation; but underneath that, he
fers from skin rashes, leukoplakia (depig-           also expressed, again, specifically,
mentation of the skin), fungal infections,           fears and anxieties and apprehensions
bronchial infections, and anemia. Several            about what was happening to him,
of these plaintiffs attributed some or all of        what was happening to his wife and
their complaints to ‘‘immune suppression’’           his familyTTTT But his anxiety was
caused by exposure to benzene.                       apparent.
  Jurors apparently attributed almost             [As to Codi Stennett]: I think that the
none of these complaints to the water at             most succinct way I can put it is that
Three Lakes, awarding ‘‘$0’’ as to every             she carries around this anxiety that

56. See Pustejovsky v. Rapid–American Corp.,    57. See Temple–Inland Forest Prods. Corp. v.
  35 S.W.3d 643, 652 (Tex.2000).                  Carter, 993 S.W.2d 88, 93 (Tex.1999).
                                 EXXON CORP. v. MAKOFSKI                                       Tex.   191
                        Cite as 116 S.W.3d 176 (Tex.App.—Houston [14th Dist.] 2003)

      she knows she’s been exposed to the                Because Texas law does not permit re-
      benzene, that it’s had an effect on her,         covery of mental anguish damages for
      that it affects her cells and that she’s         fears related to developing a disease that
      going to carry that with her.                    has not occurred and has not been shown
  [As to Andrea Russell]: She has anxiety.             to be likely,58 we hold the trial court did
    She has depression. She’s trauma-                  not err in disregarding the plaintiffs’ men-
    tized by this experience, and she has              tal anguish damage awards.59
    current concerns about her physical
    health and where that’s going to go
    starting now and in—from there on.                    Someday, medical science may find that
                                                       benzene causes ALL; but on the record
He agreed with the plaintiffs’ counsel that
                                                       before us it is just as likely to find that it
each of the plaintiffs suffered mental an-
                                                       does not. Some of the other ailments the
guish because ‘‘they’ve been poisoned,’’
                                                       plaintiffs have suffered may also be con-
and while he admitted their experiences
                                                       nected to their exposure to benzene; but
fell short of accepted PTSD causes such as
                                                       on the record before us we simply cannot
combat or concentration camps, there was
                                                       tell. We cannot rush to impose liability
a ‘‘possible parallel’’ with victims of the
                                                       when scientifically reliable evidence is un-
latter because the plaintiffs here had been
‘‘gassed and survived.’’
                                                       Accordingly, we reverse the judgment of
   [16] Clearly, there was substantial evi-            the trial court, and render judgment that
dence the plaintiffs were anguished about              the plaintiffs take nothing.
their exposure to benzene in the Three                      SEYMORE, J., dissenting.
Lakes water supply. But all of that evi-
dence related to their fears about what it               CHARLES W. SEYMORE, Justice,
might mean for their health. Adopting the              dissenting.
vivid metaphor used by plaintiffs’ counsel,               The majority erroneously parses the ap-
people suffer mental anguish from drink-               pellate record and fashions a result by
ing poison not because it tastes bad, but              reducing the gatekeeper’s Robinson hear-
because they fear what it will do to them.             ing to a nullity. My colleagues have decid-
The jury found these plaintiffs suffered no            ed to treat reliability as an evidentiary
illness from their exposure to benzene at              jury issue; moreover, they misapply appel-
Three Lakes in the ten years since it                  late rules and case precedent. I dissent.
stopped, or in reasonable probability will
do so in the future. At this point so long                     THE ISSUE IS RELIABILITY AND THE
after their exposure, it is clear that their                      GATEKEEPER IS ESSENTIAL
anguish is less related to the water than to             The issue on appeal is reliability: Exxon
what they have been told about it.                     contends that plaintiffs’ experts’ opinions 1

58.   See Carter, 993 S.W.2d at 93.                    60.    See Havner, 953 S.W.2d at 728.

59. Jurors also awarded each of the six plain-         1.     Exxon challenges the contamination causa-
  tiffs $500,000 in punitive damages. As we                 tion testimony presented by Professor Patrick
  have found there is no evidence supporting                Agostino (a geologist), Dr. Joseph Hughes (a
  any of the plaintiffs’ actual damage awards,              professor of environmental science and engi-
  we also set aside the awards of punitive dam-             neering), Dr. Phillip Gschwend (a chemical
  ages. See Twin City Fire Ins. Co. v. Davis, 904           oceanographer), Dr. Dennis McLaughlin (a
  S.W.2d 663, 665 (Tex.1995).                               hydrologist), and Dr. Jack Matson (an envi-
192    Tex.         116 SOUTH WESTERN REPORTER, 3d SERIES

‘‘are not evidence because they are unrelia-      the underlying facts or data do not provide
ble and founded on unproven assump-               a sufficient basis for the expert’s opinion
tions.’’ The majority reviews the report-         under Rule 702 or 703, the opinion is inad-
er’s record from trial and scientific studies     missible.’’).
that are not included in this appellate rec-
ord and concludes the plaintiffs presented           The majority, without citation to any
no evidence of reliability before the jury        authority, brushes aside the evidentiary
and thus no evidence of medical causation.        Robinson hearing in this case as ‘‘a differ-
There are two glaring errors in the majori-       ent decision-maker in a different proceed-
ty’s approach. First, they simply ignore          ing.’’ They could not be more wrong.
the evidentiary Robinson hearing held by          First, the roles of the trial court and jury
the trial court, disregarding it as though it     as decision-makers are not coterminous;
is a nullity. Second, the majority confuses       the trial court is not a fact finder: ‘‘The
the jury’s role, mistaking reliability as an      trial court’s role is not to determine the
element to be proved before the jury.             truth or falsity of the expert’s opinion.’’
                                                  Robinson, 923 S.W.2d at 558; Weingarten
1.    Trial Court Determines Reliability,         Realty Investors v. Harris County Ap-
            Preferably Pretrial                   praisal Dist., 93 S.W.3d 280, 285 (Tex.
   Because the issue is reliability, it is im-    App.-Houston [14th Dist.] 2002, no pet.).
perative that we review both the trial rec-       To the contrary, ‘‘the trial court must be
ord and the pretrial Robinson hearing.            careful not to step into the role of the fact
TEX.R.APP. P. 34.1. The trial court has the       finder in weighing credibilityTTTT’’ Green
heightened responsibility to act as a gatek-      v. Tex. Workers’ Comp. Ins. Facility, 993
eeper in a Robinson hearing to screen out         S.W.2d 839, 843 (Tex.App.-Austin 1999,
unreliable expert evidence. Gen. Motors           pet. denied). Instead, determination of
Corp. v. Sanchez, 997 S.W.2d 584, 590             admissibility is a question of law, not a
(Tex.1999); Maritime Overseas Corp. v.            jury decision. See N. Dallas Diagnostic
Ellis, 971 S.W.2d 402, 409 (Tex.1998); E.I.       Ctr. v. Dewberry, 900 S.W.2d 90, 93–94
du Pont de Nemours & Co., Inc. v. Robin-          (Tex.App.-Dallas 1995, writ denied); Wein-
son, 923 S.W.2d 549, 556 (Tex.1995). This         garten, 93 S.W.3d at 285 (‘‘Reliability is an
is true because reliability of an expert’s        admissibility issue for the trial court, not a
opinions is a question of admissibility.          weight-of-the-evidence issue for the fact
Helena Chem. Co. v. Wilkins, 47 S.W.3d            finder.’’).
486, 499 (Tex.2001); Gammill v. Jack
Williams Chevrolet, 972 S.W.2d 713, 726             Second, a Robinson hearing, whether
(Tex.1998) (‘‘All expert testimony should         held pretrial or at trial, is the same pro-
be shown to be reliable before it is admit-       ceeding as trial. An opponent of expert
ted.’’). ‘‘Preliminary questions TTT con-         testimony may object to reliability when
cerning admissibility of evidence shall be        the evidence is offered at trial. Maritime,
determined by the courtTTTT’’ TEX.R. EVID.        971 S.W.2d at 409. At trial, a challenge to
104(a); see also Gammill, 972 S.W.2d at           an expert is typically made through a voir
718; Robinson, 923 S.W.2d at 556; TEX.R.          dire examination outside the presence of
EVID. 705(c) (‘‘If the court determines that      the jury:

  ronmental engineer). Further, Exxon chal-         matologist), Dr. Stuart Lloyd Shalat (an epi-
  lenges the medical causation testimony pro-       demiologist), and Dr. Michael Gray (a medi-
  vided by Dr. Marvin Legator (a toxicologist),     cal doctor).
  Dr. Bernard Goldstein (a toxicologist and he-
                               EXXON CORP. v. MAKOFSKI                                 Tex.   193
                      Cite as 116 S.W.3d 176 (Tex.App.—Houston [14th Dist.] 2003)

  Prior to the expert giving the expert’s               2.   If Held Reliable, No Repetition
  opinion or disclosing the underlying                                Necessary
  facts or data, a party against whom the               The majority also states that when Exx-
  opinion is offered TTT in a civil case may         on renewed its objections at trial, the
  [ ] be permitted to conduct a voir dire            plaintiffs had the opportunity to prove sci-
  examination directed to the underlying             entific reliability and develop a full record.
  facts or data upon which the opinion is            The impact of this holding is, frankly, a
  based. This examination shall be con-              waste of judicial resources and an eviden-
  ducted out of the hearing of the jury.             tiary absurdity. First, having passed mus-
                                                     ter in the gatekeeper’s hearing, a propo-
TEX.R. EVID. 705(b); see, e.g., Guadalupe–
                                                     nent of expert testimony does not need to
Blanco River Auth. v. Kraft, 39 S.W.3d
                                                     parrot the evidence of reliability at trial.
264, 266 (Tex.App.-Austin 2001), rev’d, 77
                                                     See Judge Harvey Brown, Procedural Is-
S.W.3d 805 (Tex.2002) (both opinions iden-
                                                     sues Under Daubert, 36 HOUS. L.REV. 1133,
tifying voir dire challenge to reliability);
                                                     1148 (1999). In fact, the rules of evidence
N. Dallas Diagnostic Ctr., 900 S.W.2d at
                                                     permit an expert to testify at trial without
93, 96. However, through case law, Texas
                                                     disclosure of underlying facts or data:
courts have been urged to determine Rob-
                                                     ‘‘The expert may testify in terms of opin-
inson issues pretrial: ‘‘it is imperative to         ion or inference and give the expert’s rea-
ventilate any Robinson issues as early as            sons therefor without prior disclosure of
possible, preferably as a pretrial matter.’’         the underlying facts or data, unless the
Maritime, 971 S.W.2d at 412 (Gonzalez, J.,           court requires otherwise.’’ TEX.R. EVID.
concurring); see Regan v. Schlumberger               705(a) (emphasis added).
Tech. Corp., No. 01–00–00026–CV, 2001
                                                        Second, addressing reliability pretrial,
WL 1344077, at *3 (Tex.App.-Houston [1st
                                                     rather than before the jury, allows a full
Dist.] Nov. 1, 2001, pet. denied) (not desig-
                                                     presentation of the issue. As the Texas
nated for publication) (‘‘We, too, urge early
                                                     Supreme Court has noted, judges may be
Robinson/Havner hearingsTTTT’’). A pre-
                                                     better-equipped to determine reliability
trial Robinson hearing is no different than
                                                     than juries. See Robinson, 923 S.W.2d at
one held during trial. We appropriately
                                                     558. This is because a judge has the
consider an evidentiary Robinson hearing             benefit of documents and briefs, whereas
to be the same as a voir dire held during            similarly detailed information could easily
trial. Naturally, we would consider a voir           overwhelm the issues before the jury. Id.
dire directed at an expert’s reliability upon        The judge ‘‘can freely ask questions in a
appellate review of the issue. See Guada-            preliminary hearing and thus glean more
lupe–Blanco River Auth., 77 S.W.3d at                information’’ without the risks associated
807. Accordingly, it is imperative that we           with explaining reliability to a jury. Id. In
consider the evidentiary hearing held pre-           so doing, the trial court independently
trial. We should not punish those litigants          evaluates the underlying data. See Hav-
who ventilate reliability issues early by            ner, 953 S.W.2d at 713. Additionally, a
refusing to consider the gatekeeper’s hear-          hearing before the trial court, as opposed
ing. Under the majority’s approach, the              to trial before the jury, solves rule-based
trial court is converted from a gatekeeper           problems with developing the record. For
to ‘‘idle spectator.’’ See Maritime, 971             instance, in reviewing the trial record, the
S.W.2d at 412 (quoting Robinson, 923                 majority states it is hard-pressed to con-
S.W.2d at 554).                                      duct a Havner review without the epidemi-
194      Tex.          116 SOUTH WESTERN REPORTER, 3d SERIES

ological studies on which plaintiffs’ experts        preservation of error, see United Parcel
relied. However, such studies are not ad-            Serv., Inc. v. Tasdemiroglu, 25 S.W.3d 914,
missible at trial even under the learned             916 (Tex.App.-Houston [14th Dist.] 2000,
treatise hearsay exception. See TEX.R.               pet. denied), and gave the trial court the
EVID. 803(18) (statements from learned               opportunity to reconsider its previous rul-
treatises may be read into evidence, but             ings.3 Cf. Clark v. Trailways, Inc., 774
not received as exhibits). In contrast, in           S.W.2d 644, 647 (Tex.1989) (renewed ob-
determining reliability in hearing, a trial          jections promote efficient administration of
court ‘‘is not bound by the rules of evi-            justice by providing a trial court with a
dence except those with respect to privi-            final opportunity to prevent the erroneous
leges.’’ TEX.R. EVID. 104(a). In the case            admission of evidence).
before us, this explains why the trial court
examined epidemiological studies, but such
                                                               3.   Jury Determines Weight
studies were not admitted at trial before
                                                                       and Credibility
the jury.
   Third, a proponent of expert testimony               The majority concludes that a no-evi-
has the burden to prove reliability the first        dence challenge to reliability ‘‘tests the
time the opponent objects. See Guadalu-              jury’s decision after trial, looking at the
pe–Blanco River Auth., 77 S.W.3d at 807              whole trial record to see if any evidence
(‘‘Once the [opponent] made its objection,           supports the verdict.’’ Although the ma-
the burden was on [the proponent] to es-             jority examines the trial record for more
tablish that [the expert’s] opinion was reli-        than a scintilla of reliability,4 reliability
able.’’); Robinson, 923 S.W.2d at 557 (at            need not be proved before the jury like an
point of motion to exclude, proponents               element of plaintiffs’ cause of action. A
bore burden to establish admissibility); see         negligence cause of action has just four
also Frias v. ARCO, 104 S.W.3d 925, 927              elements: (1) a legal duty; (2) breach of
(Tex.App.-Houston [14th Dist.] 2003, no              that duty; (3) proximate cause; and (4)
pet.) (summary judgment opinion). In this            damages. Mission Petroleum Carriers,
case, Exxon objected pretrial and, because           Inc. v. Solomon, 106 S.W.3d 705, 710 (Tex.
the court denied Exxon’s motions pretrial,           2003). Causation evidence only becomes
the plaintiffs met their burden to the trial         ‘‘no evidence’’ if it is unreliable. See Wein-
court’s satisfaction. No authority then re-          garten, 93 S.W.3d at 285. Further, as
quires repeated proof of reliability just            explained above, reliability is a question of
because objections are renewed. Nor is               law for the trial court. If an opponent
there any rule requiring plaintiffs to incor-        never objects, experts can testify at trial to
porate by reference at trial their proof             causation without ever proving their opin-
from the pretrial hearing.2 Here, Exxon’s            ions are based on reliable foundations.
repeated objections to reliability accom-            See Maritime, 971 S.W.2d at 411 (failure
plished two things only. They assured                to timely object waives issue).

2.     Nor did Exxon need to incorporate by refer-   3.     And the trial court did reconsider its rulings
     ence evidence from the pretrial hearing.             as to four plaintiffs, finding no reliability and
     However, by suggesting that either party             thus no evidence of causation.
     could incorporate such proof by reference at
                                                     4.     For instance, the majority states ‘‘it cannot
     trial, the majority again implicitly acknowl-
                                                          grant even a scintilla of weight’’ to general
     edges our need to review the evidence which
                                                          references to studies that are not in the trial
     the trial court reviewed as gatekeeper.
                                   EXXON CORP. v. MAKOFSKI                                  Tex.   195
                          Cite as 116 S.W.3d 176 (Tex.App.—Houston [14th Dist.] 2003)

   The trial court’s gatekeeping function,               fied, ‘‘Most of what I’m relying on already
however, does not supplant the opportuni-                exists in the peer-reviewed literature.’’
ty to expose weaknesses in expert testi-                 Dr. Marvin Legator testified that ‘‘science’’
mony before the jury. Weingarten, 93                     helped him form the opinion that ALL
S.W.3d at 285. Under Robinson stan-                      could result from benzene exposure be-
dards, the jury continues ‘‘to assess the                cause ‘‘[a]s we did more and more studies,
weight and credibility of the proffered tes-             we added more and more leukemias and
timony.’’ Robinson, 923 S.W.2d at 558.                   lymphomas in terms of what benzene can
This is consistent with the rules of evi-                do.’’
dence: ‘‘This rule [that the trial court de-                The majority concludes, via Exxon’s
termines admissibility] does not limit the               cross-examination, that no epidemiological
right of a party to introduce before the                 study meeting Havner’s standards sup-
jury evidence relevant to weight and cred-               ports the experts’ opinions. Aside from
ibility.’’ TEX.R. EVID. 104(e). Cross-ex-                epidemiological studies, perhaps the ex-
amination remains ‘‘the traditional and ap-              perts relied upon in vitro human cellular
propriate means of attacking shaky but                   studies. Or in vivo studies. Or animal
admissible evidence.’’      Gammill, 972                 studies. The cross-examination does not
S.W.2d at 728; cf. Keo v. Vu, 76 S.W.3d                  foreclose these possibilities. While cross-
725, 734 (Tex.App.-Houston [1st Dist.]                   examination of Dr. Goldstein regarding ep-
2002, pet. denied) (‘‘Further, factual weak-             idemiological studies was very effective,
nesses underlying an expert’s causation                  perhaps Dr. Shalat or Dr. Legator relied
opinion generally go to the testimony’s                  upon an epidemiological study meeting
weight, rather than its admissibilityTTTT’’).            Havner standards. Again, this was not
                                                         foreclosed by cross-examination. In my
       4.   Exxon’s Cross–Examination                    opinion, it’s extremely difficult to deter-
   Given the generalities of direct testimo-             mine reliability solely based on cross-ex-
ny proffered by plaintiffs, the majority                 amination at trial if a plaintiff meets its
should hesitate to conclude that the ex-                 burden of proof in a pretrial hearing. And
perts’ testimony was unreliable based on                 in this case, while the cross-examination
cross-examination and studies absent from                was quite damaging, without reviewing the
the appellate record. As I pointed out                   Robinson hearing record, cross-examina-
above, an expert may testify on direct                   tion goes only to weight and credibility.
without revealing underlying facts and                   Cf. Maritime, 971 S.W.2d at 414 (Gonzalez,
data and without repeating the evidence of               J., concurring) (stating cases in which ex-
reliability from the pretrial hearing. See               pert testimony is untenable on its face, and
TEX.R. EVID. 705(a). Thus, plaintiffs’ ex-               thus excusing a Robinson hearing, ‘‘will be
perts testified on direct at trial in generali-          comparatively rare’’);      Gammill, 972
ties, without offering studies, literature, or           S.W.2d at 728 (availability of cross-exami-
data on which they relied. For instance,                 nation does not relieve the trial court of its
Dr. Goldstein testified that since his depo-             threshold responsibility as gatekeeper).
sition in this case, he found literature sup-              Further, the Robinson and Havner fac-
porting a relationship between benzene                   tors are flexible and non-exhaustive. See
and ALL.5 Dr. Stuart Lloyd Shalat testi-                 Havner, 953 S.W.2d at 718–19; Robinson,

5.     The majority emphasizes cross-examination           Exxon since his deposition. However, this
     of Dr. Goldstein in which he admits he pro-           goes to supplementation of discovery, not the
     vided no animal or epidemiological studies to         existence or non-existence of studies.
196    Tex.        116 SOUTH WESTERN REPORTER, 3d SERIES

923 S.W.2d at 557; see also Daubert v.          that admissibility and sufficiency reviews
Merrell Dow Pharm., Inc., 509 U.S. 579,         are distinct, although both involve the reli-
593–94, 113 S.Ct. 2786, 125 L.Ed.2d 469         ability of expert testimony. See Austin v.
(1993) (promoting flexible inquiry rather       Kerr–McGee Ref. Corp., 25 S.W.3d 280,
than a definitive checklist); Coastal Tank-     284 (Tex.App.-Texarkana 2000, no pet.).
ships, U.S.A., Inc. v. Anderson, 87 S.W.3d         However, others have recognized that
591, 599, 611 (Tex.App.-Houston [1st Dist.]     review of reliability, even in the context of
2002, pet. denied) (en banc) (‘‘pertinent,      a no-evidence challenge, requires review of
suggested inquiries in assessing the relia-     the Robinson hearing. The pretrial Rob-
bility of expert testimony are applied flexi-   inson hearing ‘‘should be recorded by the
bly and are not exclusive or required.’’).      court reporter. Without a record from the
Trial courts are also granted deference in      hearing, the appellant will not be able to
their selection of the factors to use in        show harm.’’ MICHOL O’CONNOR, O’CON-
determining reliability. Kumho Tire Co.         NOR’S TEXAS RULES: CIVIL TRIALS 289, § 3.5

v. Carmichael, 526 U.S. 137, 152–53, 119        (2002). Further, an appeal ‘‘from a final
S.Ct. 1167, 143 L.Ed.2d 238 (1999). With-       judgment rendered in a trial in which the
out the pretrial hearing record, we do not      trial court ruled on a motion to exclude an
know if the trial court used additional fac-    expert TTT must include the record from
tors or which Robinson and Havner fac-          the hearing in the appellate record.’’ Id.
tors it assessed.                               290, § 5.1. This is because a party may
                                                offer evidence at the hearing and, having
   Because there was an evidentiary Rob-
                                                prevailed, not present it before the jury.
inson hearing, the full bases of the ex-
                                                See Judge Harvey Brown, Procedural Is-
perts’ opinions should have been revealed       sues Under Daubert, 36 HOUS. L.REV. 1133,
there. It is unsound and poor precedent         1139 (1999). ‘‘[T]he evidence from the
to assume that cross-examination was con-       hearing will be considered in determining
clusive without consideration of the very       both admissibility and, indirectly, the suffi-
hearing in which reliability was tested.        ciency of the evidence, even if the evidence
Using a partial record contradicts the Tex-     is not fully presented to the jury.’’ Id.
as Supreme Court’s admonition ‘‘that            Indeed, the Texas Supreme Court has
courts should make a determination of reli-     stated that reliability issues go to ‘‘the
ability from all the evidence.’’ Havner,        admissibility of expert evidence rather
953 S.W.2d at 720 (emphasis added).             than the legal sufficiency of the evidence.’’
                                                Gen. Motors Corp., 997 S.W.2d at 590.
      5.   What We Review and How
                                                   Finally, as to standard of review, the
   Throughout this dissent, I have empha-       Texas Supreme Court has recognized that
sized review of the Robinson hearing, even      a no-evidence complaint may be sustained
when it is held pretrial. The majority          when the record shows one of the follow-
contends I have misconstrued admissibility      ing: (1) a complete absence of a vital fact;
of expert testimony with its legal sufficien-   (2) the court is barred by rules of law or
cy. I am not the first to be accused of         evidence from giving weight to the only
eroding the distinction. See Lucinda M.         evidence offered to prove a vital fact; (3)
Finley, Guarding the Gate to the Court-         the evidence offered to prove a vital fact
house: How Trial Judges are Using Their         is no more that a mere scintilla; or (4)
Evidentiary Screening Role to Remake            the evidence establishes conclusively the
Tort Causation Rules, 49 DEPAUL L.REV.          opposite of the vital fact. Maritime, 971
335, 376 (1999). Litigants have also urged      S.W.2d at 409. By challenging reliability
                                    EXXON CORP. v. MAKOFSKI                                 Tex.   197
                           Cite as 116 S.W.3d 176 (Tex.App.—Houston [14th Dist.] 2003)

of the experts’ testimony, and thus admis-                that the transcript and evidence from it
sibility, Exxon contends the testimony                    are not in the appellate record.6 Under
amounts to no evidence because a ‘‘rule of                established appellate law, we cannot con-
law or evidence,’’ specifically Rule of Evi-              sider whether error exists unless the rec-
dence 702, precludes giving the evidence                  ord relating to the alleged error is before
any weight. See id.; Exxon Pipeline Co.                   us. Foust v. Estate of Walters, 21 S.W.3d
v. Zwahr, 35 S.W.3d 705, 718 (Tex.App.-                   495, 504 (Tex.App.-San Antonio 2000, pet.
Houston [1st Dist.] 2000) (Taft, J., dis-
                                                          denied). Further, as appellant, Exxon had
senting), rev’d, 88 S.W.3d 623 (2002). Ac-
                                                          the burden to bring forth a sufficient rec-
cordingly, I would begin our analysis by
                                                          ord showing error. See Melendez v. Exx-
examining the reliability, and therefore
                                                          on Corp., 998 S.W.2d 266, 274 (Tex.App.-
admissibility of the testimony, Zwahr, 35
S.W.3d at 718 (Taft, J., dissenting), recog-              Houston [14th Dist.] 1999, no pet.). ‘‘Ab-
nizing the trial court’s discretion in deter-             sent such a record, the reviewing court
mining the matter. See id. at 719.                        must presume that the evidence before the
                                                          trial judge was adequate to support the
   I recognize that some courts conduct de
novo or something similar to a de novo                    decision.’’ Simon v. York Crane & Rigging
review in this context. See Mo. Pac. R.R.                 Co., 739 S.W.2d 793, 795 (Tex.1987); see
Co. v. Navarro, 90 S.W.3d 747, 750 (Tex.                  also Foust, 21 S.W.3d at 504.
App.-San Antonio 2002, no pet.); Austin,                    The effect of these precepts on this case
25 S.W.3d at 285. However, I would re-
                                                          ends in an affirmance: two plaintiffs keep
view for an abuse of discretion because the
                                                          their verdict and four plaintiffs collect
Texas Supreme Court has done so in the
                                                          nothing. For Exxon’s appeal, because we
three post-Havner cases in which it ad-
                                                          are missing the pretrial Robinson record,
dressed reliability in the context of a no-
evidence challenge. See Exxon Pipeline                    we must assume that evidence of reliabili-
Co. v. Zwahr, 88 S.W.3d at 628–29 (revers-                ty presented at the Robinson hearing sup-
ing Zwahr, 35 S.W.3d at 705); Guadalupe–                  ports the trial court’s decision to admit
Blanco River Auth., 77 S.W.3d at 807 (re-                 the experts’ testimony. Because we must
versing Guadalupe–Blanco River Auth.,                     assume the testimony was properly admit-
39 S.W.3d at 264); Helena Chem. Co., 47                   ted, the experts’ causation evidence is le-
S.W.3d 486, 499 (Tex.2001) (affirming Hel-                gally sufficient for plaintiffs James Makof-
ena Chem. Co. v. Wilkins, 18 S.W.3d 744                   ski, Jr. and John Russell. The lack of a
(Tex.App.-San Antonio 2000)). In dissent-                 Robinson record cuts both ways in this
ing, however, I find that the degree of                   case, however, and similarly hinders the
deference to give on appellate review is a                cross-appeal filed by plaintiffs Andrea
less pressing question than whether we                    Russell, Janie Devora, Phillip Devora, and
can ignore evidence from the gatekeeper’s                 Codi Stennett. They contend that one
hearing.                                                  study, the ATSDR, supports causation
                                                          testimony regarding benzene exposure
                                                          and their health problems. However, the
              1.   The Record Filed                       ATSDR is not in the trial record. By the
  The primary reason the majority dis-                    same standards that stymie Exxon’s ap-
counts the pretrial Robinson hearing is                   peal, we must assume the Robinson hear-

6.     Nor can it appear in the record. On file             es that there was no reporter’s record made
     with this court at our request is a letter from        by her on September 30, 1999, the day of the
     the official court reporter in which she advis-        Robinson hearing.
198       Tex.         116 SOUTH WESTERN REPORTER, 3d SERIES

ing record supports the trial court’s ulti-           originally sought a hearing by submission,
mate decision that expert testimony in                Exxon protested and sought an evidentiary
their cases was unreliable.                           hearing, which was held 11 days before the
                                                      start of trial. The majority also fails to
             2.   Legal Contortions                   acknowledge that plaintiffs filed written
   To avoid the effect of the missing rec-            objections to some of the exhibits attached
ord, the majority engages in legal contor-            to Exxon’s motions. We do not know
tions. First, the majority excuses Exxon              whether these objections were ruled upon.
from bringing forward a complete appel-               And we also do not know whether the
late record because neither party raised or           exhibits attached to Exxon’s motions en-
briefed the argument. Certainly, inade-               compass those offered and admitted in
quate briefing can result in waiver of an             hearing. I do not imply that a trial court
appellant’s issue. See Campbell v. State,             must hold an evidentiary hearing instead
85 S.W.3d 176, 184 (Tex.2002) (cited by               of ruling by submission.8 However, if an
majority). However, waiver of error due               evidentiary hearing is held, one party’s
to inadequate briefing is distinct from the           filings in the clerk’s record are not a sub-
burden to bring forth a record on appeal.             stitute on appeal for the reporter’s record
Additionally, because it is an appellant’s            of the hearing.
burden to bring forth a record showing                   Third, the majority surprisingly and
error, I do not assign blame to the appel-            wrongly concludes that there was no evi-
lees, as the majority does, for failing to            dence presented at the pretrial hearing.9
brief the omissions in the appellate record.          To this end, the majority assumes that a
I also do not excuse the missing record, as           trial court would hold an evidentiary hear-
the majority does, because ‘‘the very capa-           ing only to receive live testimony from the
ble attorneys and experienced trial judge             experts. I disagree. I do not suggest that
all certainly knew when a record is re-               the pretrial hearing in this case included
quired and when it is not.’’ There is no              live testimony. Nor do I imply that all
deference on appellate review based on                evidentiary hearings on reliability must in-
attorneys’ or judges’ reputations.                    clude live testimony. Instead, evidence in
  Second, the majority implies that 270               a pretrial evidentiary hearing could include
pages of Exxon’s motions to exclude and               depositions, studies, and documents.10 In
supporting exhibits suffice in lieu of the            this case, the trial record and clerk’s rec-
reporter’s record from the Robinson hear-             ord reflect that the trial court held an
ing.7 In this approach, the majority pre-             evidentiary hearing, but do not delineate
tends that the Robinson hearing was held              what evidence was offered or admitted in
by submission. In fact, when the plaintiffs           that hearing.

7.     If the plaintiffs filed responses to Exxon’s   9.     The majority reaches this conclusion be-
     motions to exclude, Exxon did not bring them          cause Exxon’s counsel introduced himself to
     forward on appeal. The clerk’s record is en-          the experts at trial and cross-examined them
     tirely one-sided.                                     without referring to the pretrial hearing.

                                                      10. For instance, with regard to contamination
8.     Cf. Tanner v. Westbrook, 174 F.3d 542, 546       causation, the plaintiffs contend they used
     (5th Cir.1999) (where trial court refused to       ‘‘Exxon’s numbers’’ to calculate the amount
     hold pretrial hearing, appellate court re-         of benzene lost in the blowout. Exxon’s doc-
     viewed documents that were submitted pre-          uments evidencing such numbers could be
     trial).                                            presented without live testimony.
                                 EXXON CORP. v. MAKOFSKI                                    Tex.   199
                        Cite as 116 S.W.3d 176 (Tex.App.—Houston [14th Dist.] 2003)

   Fourth, the majority wrongly concludes              2002, no pet.) (sufficiency challenge waived
that under my analysis, ‘‘every time a pre-            where party did not request partial record
trial motion TTT is denied without a report-           and failed to file exhibits in appellate rec-
er’s record, we must presume any jury                  ord); see also Sandoval v. Comm’n for
verdict thereafter is valid (no matter how             Lawyer Discipline, 25 S.W.3d 720, 722
scientifically unsound).’’ We will only need           (Tex.App.-Houston [14th Dist.] pet. de-
a reporter’s record of evidentiary Robin-              nied).
son hearings, not hearings by submission.
                                                          Sixth, in discounting the missing Robin-
Further, I leave open the possibility that
                                                       son record, the majority states it was
in rare cases, expert testimony will be
                                                       plaintiffs’ ‘‘burden to present scientifically
untenable on its face. See Maritime, 971
S.W.2d at 414 (Gonzalez, J., concurring).              reliable evidence of causation at trial.’’ In
                                                       citing Havner and Robinson for this prop-
   Fifth, the majority presumes that the
                                                       osition, the majority ignores Maritime, de-
parties ‘‘never requested that the hearing
                                                       cided after those cases. In Maritime, the
be recorded.’’ The record is silent as to
                                                       court encouraged pretrial resolution of re-
whether the parties requested or excused
                                                       liability issues. 971 S.W.2d at 409. Thus,
the court reporter from the pretrial hear-
                                                       the majority’s citations to Havner and
ing. However, failure to request that a
                                                       Robinson are taken out of context. Nei-
hearing be recorded does not excuse an
                                                       ther case supports the contention that we
appellant’s burden to bring forth a record
                                                       review only the evidence presented at trial
showing error. Further, Exxon specifical-
                                                       before the jury.
ly designated the pretrial Robinson record
as part of the appellate record:                          Seventh, the majority circularly con-
   The pre-trial ‘‘Robinson ’’ hearing held            cludes that the trial cross-examination
   on September 30, 1999TTTT This request              both proves unreliability and rebuts the
   includes all exhibits offered or admitted           presumption we give to the missing record.
   into evidence either for the Court or for           However, the majority fails to cite a single
   the jury at the hearing of September 30,            case in which the presumption given the
   1999 or at trial on the merits beginning            missing record is rebutted by that which is
   October 11, 1999 and ending November                available.11 I have already discussed the
   19, 1999.                                           limitations of cross-examination in this
Thus, we cannot presume the record we                  case given the generalities of direct testi-
have is the entire record pertinent to relia-          mony. The cross-examination simply did
bility. See Andrews v. Sullivan, 76                    not foreclose the bases of the experts’
S.W.3d 702, 704 (Tex.App.-Corpus Christi               opinions. If the bases of their opinions are

11. The majority miscites Candelier v. Ring-             dence from trial in determining the post-trial
  staff, 786 S.W.2d 41, 44 (Tex.App.-Beaumont            issue. Thus, the Beaumont appellate court
  1990, writ denied). Unlike this case, in               realized it did not have the entire record
  Candelier, the appellant requested a partial           pertinent to the issue.
  reporter’s record from two post-trial hearings           Here, we know the record pertinent to reli-
  only. Under the rules of appellate procedure,          ability is incomplete because (1) during trial,
  the ‘‘appellate court must presume that the            the trial court and attorneys referred to the
  partial reporter’s record designated by the            pretrial objections, rulings, and evidence; (2)
  parties constitutes the entire record for pur-         the clerk’s record reflects that Exxon de-
  poses of reviewing the stated points or is-            manded and received an evidentiary Robinson
  sues.’’ TEX.R.APP. P. 34.6(c)(4). In one of the        hearing; and (3) in oral argument, the attor-
  post-trial hearings, the Candelier trial court         neys acknowledged the trial court examined
  clearly stated that it was considering the evi-        studies not a part of the trial record.
200   Tex.          116 SOUTH WESTERN REPORTER, 3d SERIES

not foreclosed, we presume that the miss-        the minority (though perhaps leading)
ing record provides the evidence of reliable     edge of science and, at worst, absolute
bases. I think a stronger case for rebuttal      quacks. As judges, we are also aware of
would be presented, for example, where           cases discussing the limitations of science,
the existing record proves that no evidence      benzene and toxic exposure, and cancers.
was offered, admitted, or submitted in the       See Frias, 104 S.W.3d at 925; Daniels v.
missing pretrial hearing. Or where the           Lyondell–Citgo Ref. Co., Ltd., 99 S.W.3d
existing record clearly delineates what the      722, 726 (Tex.App.-Houston [1st Dist.]
trial court reviewed in pretrial hearing.        2003, no pet.); Navarro, 90 S.W.3d at 747;
In truth, it is the majority who pretends in     Austin, 25 S.W.3d at 280. However, jus-
this case, allowing its reactions to the trial   tice should not give into visceral reactions
testimony to blur the requisites for appel-      to a case.
late review.                                        We must follow procedural rules. We
   Finally, if all the above approaches fail     must enforce requisites for appellate re-
the majority’s reasoning, the majority cre-      view. If procedural and appellate law pre-
ates a safety net by simply making its own       clude a review on the merits, we must
appellate record. Before oral argument,          write accordingly. And in this case, the
we became aware of some of the deficien-         incomplete record precludes a true review
cies in the record. During argument, the         on the merits. Whether an appellate court
parties acknowledged that the trial court        can disregard the gatekeeper’s hearing in
examined scientific studies not included in      a no-evidence context is the material issue
the appellate record. At Chief Justice           in this case, upon which a take nothing
Brister’s request, the parties filed fourteen    judgment or an affirmance rests. The an-
studies as exhibits to their briefing. His       swer to the issue will also clarify for all
request was an implicit acknowledgment           courts and litigants the correct procedure
that we must review the evidence which           for challenging an expert and proving reli-
the trial court reviewed. Exxon properly         ability. I thus urge review of this case by
objected, asserting that such studies were       the Texas Supreme Court. Although we
outside the trial record. We may not cre-        may have decided the case differently were
ate a new record on appeal. Richards v.          we jurors, appellate court judges must ap-
Comm’n for Lawyer Discipline, 35 S.W.3d          ply rules and law consistently and even-
243, 251 (Tex.App.-Houston [14th Dist.]          handedly. Otherwise, we diminish rule of
2000, no pet.); Intermarque Auto. Prod.,         law, predictability, and regularity neces-
Inc. v. Feldman, 21 S.W.3d 544, 547 n. 3         sary to guide future litigants.
(Tex.App.-Texarkana 2000, no pet.). Ma-            Under the standards I have discussed in
terials attached to briefing, but not in the     this dissent, I would overrule Exxon’s is-
appellate record, cannot be considered by        sues, overrule plaintiffs’ cross-issues, and
an appellate court. Till v. Thomas, 10           affirm the trial court’s judgment. Because
S.W.3d 730, 733–34 (Tex.App.-Houston [1st        the majority has not followed established
Dist.] 1999, no pet.); TEX.R.APP. P. 34.1.       rules and standards for appellate review, I
   Judges are human and are affected by
the facts of the cases before them. In this
case, the trial record imparts a gut reac-
tion that plaintiffs’ experts are, at best, on

Shared By: