Summary
In Mireles, we concluded that the expert testimony was not amenable to a strict application of the Robinson factors. 63 S.W.3d at 499.
Summary of this case from United Ser Auto v. PigottOpinion
No. 04-00-00023-CV.
Delivered and Filed: August 8, 2001.
Appeal from the 73rd Judicial District Court, Bexar County, Texas, Trial Court No. 1994-CI-14648, Honorable Peter Michael Curry, Judge Presiding.
J. Hampton Skelton, Katherine E. Bell-Moss, Skelton, Woody Arnold, Austin, for appellant.
Craig S. Smith, Law Office of Craig S. Smith, Padre Island, Stephen W. Boyd, Law Office of Stephen W. Boyd, P.C., Donald B. Edwards, Corpus Christi, for appellee.
Sitting: PHIL HARDBERGER, Chief Justice, ALMA L. LOPEZ and KAREN ANGELINI, Justices.
State Farm Lloyds appeals a jury verdict in favor of Peter and Carmen Mireles. In its sole issue, State Farm asserts there is no evidence of causation to support the judgment. We agree and reverse the trial court's judgment, rendering judgment in State Farm's favor.
Factual Procedural History
Peter and Carmen Mireles filed suit against State Farm Lloyds for breach of contract and other extra-contractual claims after State Farm denied the Mireleses' foundation damage claim. To invoke coverage under the Mireleses' insurance policy, the Mireleses were required to prove the foundation damage was caused by plumbing leaks. State Farm moved to exclude the testimony of Per Schneider, the Mireleses' causation witness, for failure to satisfy Rule of Evidence 702's reliability and relevance requirements. After holding a " Daubert" hearing, the trial court denied State Farm's motion. At trial, Schneider testified that the Mireleses' foundation damage was caused by plumbing leaks, and the jury found in the Mireleses' favor, awarding them $103,392.50 in damages for breach of contract. Although the jury found the Mireleses were not entitled to damages under the Texas Insurance Code and the Deceptive Trade Practices Act, it nevertheless awarded the Mireleses an additional $150,000 for State Farm's knowing violations of the Insurance Code and the DTPA.
State Farm filed a Motion for Judgment Notwithstanding the Verdict. The trial court entered judgment in the Mireleses favor, but reduced the recovery amount to $81,887.14 for damages caused by the leak. The trial court also disregarded the jury's findings of knowing violations of the Insurance Code and DTPA.
State Farm appeals the judgment in one issue. Specifically, it claims that the trial court erred in entering judgment in the Mireleses' favor because there was no reliable or relevant evidence of causation to support the jury's verdict. The Mireleses cross appeal, claiming that the trial court erred in reducing the jury's award and in refusing to submit their requested damages questions relating to violations of the Texas Insurance Code and DTPA.
Discussion
In State Farm's sole issue, it claims the trial court erred in entering judgment in the Mireleses' favor because the testimony of the Mireleses' expert, Per Schneider, amounted to noevidence of causation. Specifically, it claims Schneider's testimony is neither reliable nor relevant.
A. Standard of Review
In determining whether there is evidence of probative force to support a jury's finding, this court reviews all of the record evidence in a light favorable to the jury's finding, indulging every reasonable inference in the prevailing party's favor. Merrell Dow Pharms. v. Havner, 953 S.W.2d 706, 711 (Tex. 1997); Minnesota Mining and Mfg. Co. v. Atterbury, 978 S.W.2d 183, 197 (Tex.App.-Texarkana 1998, pet. denied). We will sustain a no evidence point when "(a) there is a complete absence of evidence of a vital fact, (b) the court is barred by rules of law or of evidence from giving weight to the only evidence offered to prove a vital fact, (c) the evidence offered to prove a vital fact is no more than a mere scintilla, or (d) the evidence conclusively establishes the opposite of a vital fact." Havner, 953 S.W.2d at 711.
B. Rules and Principles Governing the Admissibility of Expert Testimony
The question of the admissibility of expert testimony goes hand in hand with this court's analysis under a legal sufficiency challenge. See Havner, 953 S.W.2d at 713 (describing how the supreme court in Daubert "explained that when `wholesale exclusion' [of expert testimony] is inappropriate and the evidence is admitted, a review of its sufficiency is not foreclosed"). The factors announced in E.I. du Pont de Nemours and Co., Inc. v. Robinson, 923 S.W.2d 549 (Tex. 1995) thus guide our review of a no evidence challenge. Havner, 953 S.W.2d at 713; see also Tex.R.Evid. 702; Robinson, 923 S.W.2d at 558. Robinson teaches that, in addition to demonstrating that an expert witness is qualified to testify under Texas Rule of Evidence 702, the proponent of the evidence must demonstrate the expert's testimony is relevant to the issues and based on a reliable foundation. Robinson, 923 S.W.2d at 556.
To be relevant, the expert's testimony must bear a relationship to the issue in the case such that the testimony will aid the jury. Gammill v. Jack Williams Chevrolet, Inc., 972 S.W.2d 713, 720 (Tex. 1998). To be reliable, the scientific evidence must be grounded in scientific method and procedure such that it amounts to more than subjective belief or unsupported speculation. Id.
In Robinson, the Texas Supreme Court enumerated a list of factors to determine the reliability of expert testimony. Specifically, those factors to consider are: whether the theory has or can be tested; whether the technique relies upon the subjective interpretation of the tester; whether the theory has been subjected to peer review or publication; the technique's potential rate of error; whether the technique or underlying theory has been generally accepted by the relevant scientific community; and non-judicial uses of the technique. Robinson, 923 S.W.2d at 556.
In Gammill, however, the court made it clear that the Robinson factors do not always apply to expert testimony because they do not always fit. Gammill, 972 S.W.2d at 726. The criteria for determining reliability and relevance will vary, depending on the nature of the evidence. Id. In discharging its duty as gatekeeper, the trial court must determine how the reliability of particular testimony is to be assessed. Id. at 726. In some instances, the expert's skill and experience may support a finding that his testimony is reliable and relevant. Id. at 726. Regardless of whether the Robinson factors are applied, however, the proponent of the expert testimony must still prove the testimony is reliable and relevant. Id. at 727.
In Gammill, the supreme court declined to apply the Robinson factors to the kind of expert testimony that was being offered — mechanical engineering. Id. Nonetheless, as the court stated in Gammill, a trial court is not required to admit opinion evidence which is connected to existing data only by the "ipse dixit" or say so of the expert. Id. The court may consider whether there is too great of an "analytical gap" between the data and the expert's opinion. Id. In Gammill, the supreme court found an "analytical gap" because the expert failed to show how his observations supported his conclusions. The expert offered nothing to suggest that what he believed could have happened actually did happen and this amounted to little more than subjective belief or unsupported speculation. Id. at 728. In reaching its holding, the court stated that the trial court's duty is not to determine whether the expert's conclusions are correct, but only whether the analysis used to reach them is reliable. Id.
As when an "analytical gap" renders an opinion unreliable, so too does the failure to rule out other possible causes of a particular injury or condition. See Gammill, 972 S.W.2d at 727; Weiss v. Mechanical Associated Servs., Inc.,989 S.W.2d 120, 126 (Tex.App.-San Antonio 1999, pet. denied). In Gammill, a young girl died in a car accident because her seatbelt allegedly failed. The plaintiff's expert, in attributing the girl's death to seat belt failure, explained that he found shirt fibers in the seatbelt's webbing. Gammill, 972 S.W.2d at 727. Although the trial court found the plaintiff's expert to be qualified and permitted him to offer his opinion, the supreme court found this opinion unreliable partly on the grounds that the expert failed to explain or exclude other possibilities for the presence of any fibers in the webbing. Id. According to the court, the expert's opinions were nothing more than "subjective belief or unsupported speculation." Id. at 727-28.
This court reached a similar conclusion in Weiss v. Mechanical Associated Servs., Inc. In that case, Weiss began to suffer numerous health problems after a radiology group moved onto the floor directly beneath the doctor's office where she worked. Weiss, 989 S.W.2d at 121. Weiss sued the radiology group for negligence, claiming that the chemicals used to develop the x-rays caused her health problems. Id. at 122, 123. The radiology group moved for summary judgment, and their motion was granted. Id. at 123. On appeal, this court held that Weiss had produced no evidence of causation. Id. at 126. Specifically, this court "devalued" the experts' opinions because neither one of the experts was able to rule out other potential causes of Weiss's illness with reasonable certainty. Id.
C. Expert Testimony
State Farm challenges the testimony of Per Schneider, the Mireleses' causation witness. Specifically, it asserts that Schneider's testimony is irrelevant and unreliable because it did not meet the factors announced in Robinson and that Schneider failed to show how the facts in this case are similar to the studies he relied upon in reaching his conclusions.
Per Schneider is a licensed consulting engineer with a Bachelors of Science degree in civil engineering from the University of Texas at Austin and a Masters in Business from the University of Texas at San Antonio. Over the course of his career, Schneider has designed thousands of foundations, and at the time of trial, taught advanced foundation design at the University of Texas at San Antonio.
To prepare for his testimony, Schneider reviewed engineering reports, plumbing reports, geotechnical data, and level surveys in reaching his conclusions regarding the cause of the Mireleses' foundation damage. He also inspected the house and spoke with Mr. Mireles several years after it had been repaired. Mr. Mireles told Schneider he had seen three places where water dripped from the sewer line. However, the engineering reports Schneider reviewed indicated that only under hydrostatic conditions did water leak in the kitchen, bathroom and utility lines. The report showed that none of the plumbing lines leaked under normal operating conditions (i.e. running water, flushing toilet).
Per Schneider was not hired by the Mireleses until after the foundation damage had been repaired. Thus, he did not inspect the house at the time it was damaged and, therefore, relied upon the reports of others who had inspected the damage.
Hydrostatic testing is conducted by filling the pipes to their limit with water and then pressurizing them by blocking the pipes. These types of tests show where possible leaks might be in the plumbing lines.
Based upon his review of the data and his expertise and past experience, Schneider opined that the foundation shifted in the middle of the Mireleses' house primarily as a result of a leak in the bathroom. The bathroom, however, is approximately six to eight feet away from the heave in the foundation where the damage was located. There was no damage between the leak in the bathroom and the heave. To explain the distance between the leak and the heave, Schneider offered the following theory. According to Schneider, before a foundation is poured, plumbing pipes are placed into trenches. When the soil is replaced into the trenches, it is not compacted, but is placed loosely around the pipes. When a leak in the pipe occurs and the soil is loosely compacted, the moisture from the leak travels down the "path of least resistence" (the trenches), until the trench is filled with water. Once the soil in the trench becomes saturated, the moisture begins to move laterally into the parent soils. This can happen, according to Schneider, even if there is only vapor or moisture coming from a pipe as opposed to a dripping leak.
In Schneider's opinion, this "migration" of moisture, coupled with the soil's high clay content, caused the heave in the Mireleses' foundation. Specifically, Schneider explained that a leak in a pipe in the bathroom, which did not leak under normal flow conditions, caused the heave in the foundation six to eight feet away. According to Schneider, the clay soil drew moisture from the pipe. Once the trench was saturated, the moisture moved laterally into the parent soils causing doming in the Mireleses' foundation six to eight feet away from the moisture's origination point. Schneider was unable to explain how the leaks in the bathroom did not cause the soil to expand directly underneath the leak or how there was no trace of moisture in the soil or damage to the foundation near the leak. Further, Schneider testified he did not know where in the Mireleses' house the plumbing trenches were located.
When asked why the heave occurred toward the center of the house rather than over the plumbing leak, Schneider responded "Only God knows." Schneider stated that it is unknown how a leak chooses a spot to amplify itself in the clay soil — it is unpredictable.
Although Schneider has extensive experience in foundation design, he was unable to produce any studies or data to support his plumbing leak theory. He testified that he had seen only one other circumstance where a plumbing leak caused damage, not in the vicinity of the leak itself, but at a remote distance from the leak with no intervening damage. That occurred at North Star Mall, a large commercial structure. Schneider could produce no data related to the North Star Mall project because it occurred years ago and all the data had been destroyed. Schneider discussed several other commercial building projects he had worked on which he believed supported his opinion; however, none involved damage at a remote spot some distance from the leak, none were similar to the Mireleses' home, most did not even involve plumbing leaks, and again, he could produce no supporting data. Schneider admitted he had no studies on residential foundations.
To further support his opinion that the plumbing leak was in fact the cause of the Mireleses' foundation damage, Schneider addressed each of the four possible causes of the foundation damage: (1) foundation settlement; (2) tree influence; (3) climatic conditions; and (4) plumbing leaks. State Farm's experts agreed that these were the four possible causes for the foundation damage.
Schneider first addressed the possibility of foundation settlement. For settlement to occur, the building must be heavy, there must be moisture in the soil, and the soil must be loose. Schneider explained that the soil's capacity here was somewhere between 3,000 and 6,000 pounds per square foot. The house, however, weighed only between 100 and 150 pounds per square foot. Schneider, therefore, discarded settlement as a possible cause of the Mireleses' foundation damage because the weight of the house did not outweigh the soil's capacity to support it. Schneider admitted, however, that his theory that soils do not shrink vertically unless there is pressure on them greater than the load-bearing capacity of the soil is not a generally accepted theory in engineering.
Second, Schneider attempted to rule out tree influence as the cause of the Mireleses' foundation damage. Schneider reached his conclusion that tree influence did not affect the Mireleses' foundation because, in his opinion, trees do not affect residential foundations because the loads are not heavy enough. Schneider applied the same principles he used in concluding that settlement had not caused the damage in determining that trees do not affect residential foundations. According to Schneider, even though trees can diminish the soil's moisture content by absorbing that moisture through its roots, a foundation will not settle unless the load outweighs or is equal to the soil's capacity to support it.
Schneider also relied upon the characteristics of a neighboring home's exterior as well as the Mireleses' yard in ruling out tree influence as a possible cause of the foundation damage. There are two trees, each requiring tremendous amounts of water, halfway between the Mireleses' home and a neighboring house. On that side of the Mireleses' home, there are cracks in the exterior. The neighboring home's exterior, however, is not cracked. Thus, in Schneider's opinion, if the trees were affecting the Mireleses' home they also would affect the neighboring house. Additionally, Schneider noted that the Mireleses' yard was quite green, indicating that it was adequately watered and that the trees would have no need to suck moisture from the ground, causing the soil's load capacity to diminish. Notably, Schneider stated that his opinion is not in accordance with the majority of engineers on the subject of tree influence. Other engineers believe trees affect residential foundations; he does not.
Schneider also attempted to exclude climatic conditions as the cause of the foundation damage. He testified that climatic changes will not affect a residential foundation more than four to five feet from the foundation's perimeter. He had no data to support this conclusion however. Schneider pointed out that engineers often recommend that homeowners pour a sidewalk or some other paving around the perimeter of the house. By doing so, the amount of moisture that evaporates from the "perimeter beam" is reduced. When a home's foundation is damaged due to climatic conditions, there will be a doming effect around the exterior perimeter of the house. In this case, the Mireleses had paving around the perimeter of their home and no exterior doming. Schneider, therefore, opined that climatic conditions could not have caused the foundation damage. Schneider did, however, recognize that there was no paving along the side of the Mireleses' home with the most severe sloping. Schneider also admitted that climactic changes did cause some of the damage to the Mireleses' foundation but was unable to quantify the amount.
The fourth possible cause of foundation damage was ground swelling due to plumbing leaks. According to Schneider, clay soils contain a negative charge; water contains a positive charge. When the two come into contact, they are "attracted magnetically . . . and . . . then the whole soil mass expands . . ." If the house load is not enough to compress the soil and moisture together, doming will result. In Schneider's opinion, all the elements were in place here to have ground swelling: a leak, dense clay, and a light house load. Thus, according to Schneider, the Mireleses' "foundation . . . moved upward in the middle as a result of [a] leak in the bathroom."
State Farm presented expert testimony that varied considerably from Schneider's testimony. Ralph Kean Jenner, an engineer who works for Haag Engineering Company, has both a bachelor's and master's degree in civil engineering and has been a practicing engineer since 1985. State Farm hired him to do an analysis of the Mireleses' residential foundation.
Jenner expressly disagreed with Schneider's opinion that even if trees remove a large amount of moisture from the soil, the soil will not shrink because the house's load does not exceed the soil's capacity to support it. Jenner stated that "[Schneider] is the only engineer I know of who believes that." To the contrary, Jenner testified that trees growing near a foundation cause settlement. Specifically, Jenner stated that a tree's roots seek out moisture underneath a slab, potentially removing hundreds of gallons of moisture a day. As the trees dry out or "desiccate" the soil, a foundation will settle irrespective of the soil's load capacity and the foundation's weight.
Jenner relied on several studies to support the proposition that trees can, and often do, cause foundations to settle. One, titled "A Study of Behavior of Slabs Founded on Active Clay Soils," was admitted into evidence. The study was commissioned by the Federal Department of Housing and Urban Development and reviewed sixty-nine houses. In that study, the authors concluded that "the biggest thing causing these houses to move around was tree growth. Every time a tree was growing half the height from the foundation, it caused a low spot in the house. The lowest spot in the entire house in every single case was near those trees." Although there are thousands of articles describing the effects of trees on foundations, there are no studies agreeing with Schneider's opinion on trees.
The documents Jenner relied on are as follows: A Study of Behavior of Slabs Founded on Active Clay Soils, published by the Department of Housing and Urban Development; Design and Construction of Light Foundations on Expansive Clay Soils, a paper from a continuing education seminar sponsored by the Texas Section of the American Society of Civil Engineers; Maintenance of Existing Foundations on Expansive Clay Soils, distributed to the public by the Texas Agricultural Extension Service and The Texas AM University System; and the HOW Expansive Soils Manual, authorized by the Home Owners Warranty Corporation.
Jenner also testified that Schneider's plumbing leak theory was incorrect. According to Jenner, the foundation will always heave where the leak is located. If there is damage at a distant location from the leak, there will be intermediate damage. He agreed that water can migrate, but stated the focus of the damage will be around the leak. Water traveling through the soil will swell along the way. In this case, the plumbing lines weren't anywhere near the high point in the house.
Jenner was of the opinion the damage was caused by foundation settlement, climatic conditions and tree influence, but not plumbing leaks.
State Farm also offered the testimony of John Mach. He also disagreed with Schneider's theories and opined trees caused the damage. He stated also that Schneider's theory regarding trees is not a generally accepted foundation theory and that there is no literature to support Schneider's views. He further testified that Schneider's views on whether soils shrink vertically, whether a residential foundation will settle, and the migratory vapor theory are not generally accepted.
D. Analysis
We now consider whether Schneider's expert opinion is reliable and relevant and, therefore, amounts to some evidence that the Mireleses' foundation damages were caused by a plumbing leak.
State Farm contends that Schneider's testimony is unreliable because it fails to meet any of the Robinson factors. The Mireleses, however, make no attempt to meet the Robinson factors through Schneider's testimony. Instead, they contend that Schneider's experience and his analysis of causation render his opinion reliable and relevant. They essentially argue that, as in Gammill, Schneider's opinions don't "fit" within the Robinson analysis and, therefore, the court must determine reliability based on whether there is an "analytical gap" between the data and the opinions offered. Gammill, 972 S.W.2d. at 727; Ford Motor Co. v. Aguiniga, 9 S.W.3d 252, 263 (Tex.App.-San Antonio 1999, pet. denied).We find that Schneider's expert testimony is not amenable to a strict application of the Robinson factors. See Gammill, 972 S.W.2d at 726 (holding Robinson factors inapplicable to mechanical engineering expert testimony). The Mireleses were entitled, as in Gammill, to attempt to show the reliability and relevance of Schneider's testimony through his experience, skill and consideration of whether his analysis is grounded in scientific methods and procedure. Id.
We find Schneider's expert testimony unreliable and irrelevant. Although he is an experienced engineer, Schneider's testimony is unreliable because it is nothing more than subjective belief and unsupported speculation. Further, his reliance on foundation experience with large commercial buildings and projects which had nothing in common with the facts of this case renders his opinions irrelevant and unhelpful to the jury.
The evidence shows that Schneider has not actually experienced the phenomenon he claims occurred in the Mireleses' home on a regular basis. In fact, he can only point to one undocumented occasion — North Star Mall — where, he says, a plumbing leak caused remote damage with no intervening evidence of damage. One would expect that if his theory of how a plumbing leak caused the damage to the Mireleses' house is valid, an experienced foundation expert would have seen it more than once. Certainly, if he is primarily depending on his experience to support his opinion, he would have to have seen it more than once. See Gammill, 972 S.W.2d at 726 (citing the testimony of an experienced beekeeper to be admissible to show bees always take off into the wind). The remainder of Schneider's opinion in this regard is based on his knowledge of how clay soils generally react to moisture and how foundations and plumbing lines are traditionally placed. However, he was not able to adequately explain how a plumbing leak can cause a heave at a remote distance without showing signs of intervening damage other than to say "Only God knows." And, he did not even know where the plumbing trenches were located. As in Gammill, Schneider left an "analytical gap" between his observations and his conclusions. He offered nothing to show that what he believes caused the damage actually did cause it. His analysis is simply unreliable.
Further, it is undisputed that Schneider stands alone among foundation experts in his opinion and theories. Both he and the State Farm experts agree that Schneider's theories are not generally accepted among experts in the field. And, it is undisputed that, although many foundation studies exist, none support Schneider's theories. Simply put, Schneider failed to show any basis for his opinions.
Schneider failed in his attempt to rule out other causes of the damage as well. With regard to his attempts to rule out settlement and tree influence, Schneider made assumptions that he could not support. Although he believes soil does not shrink vertically unless the load of the house is heavier than the soil's capacity, he had nothing to base this opinion on and admits that his theory is not generally accepted in the engineering community. With regard to climatic conditions, Schneider's opinion likewise lacked support and, in fact, he admitted that climatic changes did cause some of the damage to the Mireleses' home.
And, Schneider's testimony is irrelevant because it bears little relationship to the issue in this case. The issue here is whether a plumbing leak in the Mireleses' bathroom caused damage some six to eight feet away from the leak with no intervening damage. Throughout his testimony, Schneider failed to show that any of his experiences with plumbing leaks causing damage were similar in nature to that of the Mireleses' leak and damage. He discussed many foundation projects he had worked on, but only one — North Star Mall — involved a leak causing remote damage with no intervening damage. Schneider could not show similarities between a large mall and the Mireleses' residence. He produced no data nor could he testify that the two cases involved similar conditions. Thus, his testimony was irrelevant and not helpful to the jury.
Conclusion
Schneider's causation theory is unreliable and irrelevant. Because we are barred by rules of law and of evidence from giving weight to Schneider's testimony, there is no evidence of causation. Havner, 953 S.W.2d at 711. We, therefore, need not reach the Mireleses' cross issues on appeal. We reverse the trial court's judgment and render judgment in favor of State Farm Lloyds.