Supreme Court of Texas.
SOUTHWESTERN BELL TELEPHONE CO., Petitioner,
David GARZA, Respondent.
Argued Oct. 15, 2003.
Decided Dec. 31, 2004.
Rehearing Denied July 8, 2005.
Attorneys & Firms
*609 W. Carl Jordan, Houston, for Amicus Curiae.
Jorge C. Rangel, Corpus Christi, for Petitioner.
Katherine Driscoll Julia, Edinburg, for Respondent.
Justice MEDINA joined.
This is an action for a violation of the Anti–Retaliation Law, which states in part: “A person may not discharge or in any other manner discriminate against an employee because the employee has ... filed a workers’ compensation claim in good faith....”1 As the case comes to us, there are three principal issues. First: in charging the jury on liability, did the trial court err in deviating from the statutory language? Second: is there some evidence to support the jury’s finding of a statutory violation? Third: is there clear and convincing evidence to support the jury’s award of punitive damages? For this last issue, we adopt the rule stated in In re J.F.C.,2 that in reviewing the legal sufficiency of evidence to support a finding that must be proved by clear and convincing evidence, an appellate court must “look at all the evidence in the light most favorable to the finding to determine whether a reasonable trier of fact could have formed a firm belief or conviction that its finding was true.”3 By this standard of review, the punitive damages award here must be reversed. In all other respects we affirm the judgment of the court of appeals.4
David Garza and Luis Hernandez, two Southwestern Bell Telephone Co. (“SWBT”) linemen with a long history of friction working together, were hanging cable on telephone poles in Harlingen on Thursday morning, October 20, 1994. No one else was present. Hernandez was *610 working at the top of the poles from an aerial platform or “bucket” hoisted on the arm of a truck-mounted hydraulic lift. Garza was working on the ground. At one point, Hernandez dropped a rope with heavy metal hooks, narrowly missing Garza, who said nothing. A few minutes later, Hernandez lowered the bucket near where Garza had crouched down to cut a strand of wire used to hold the cable on the poles. It is not clear whether the bucket was lowered until it struck Garza in the head where he was, or whether he stood up and bumped his head on the bucket. What is clear is that Garza was very angry with Hernandez for moving the bucket so close, and hot words were exchanged. According to Hernandez, Garza also threw the strand of wire at him but hit the bucket.
Hernandez immediately left the job site and drove to the office of their “second-line” supervisor, Ruben Gonzalez, to complain of Garza’s actions. Garza tried to call Rene Robles, the two linemen’s direct, “first-line” supervisor, and then called Gonzalez, who sent another employee to bring Garza to the office. Gonzalez took written statements from both men and then interviewed them separately, asking Diana Vasquez, then vice president of the local union, to sit in and take notes. From her notes, it appears that Hernandez insisted that Garza be fired for threatening violence, but Gonzalez responded that Hernandez should see the company counselor “and get it off your chest”. Vasquez’s notes also reflect that Garza, in turn, complained that Hernandez was unsafe to work with, and that while Gonzalez agreed at least Hernandez had acted unsafely that morning, he stressed that Hernandez’s actions were no excuse for Garza’s threats or violence. At one point, according to Vasquez’s notes, Gonzalez said, “David, all my life I have been fighting bullies.” When Garza asked if that meant Gonzalez thought he was a bully, Gonzalez answered no but added, “[i]f you tell me you are going to kick my butt,” “people can perceive it as a threat.” Gonzalez finished by telling Garza that he and Hernandez could no longer work together and that he, too, should meet with the company counselor.
Garza did not complain of injury or request medical treatment until the following Monday, October 24, when he told Robles he had neck pains. Robles prepared a workers’ compensation injury report form and took Garza to a doctor, who diagnosed muscle spasms and prescribed medication. The visit and medication cost $640.77, and Garza required no further medical attention.
SWBT contends in part that a decision to discipline Garza was made before Monday and thus could not have been in response to his compensation claim. The preceding Thursday, the day of the incident, after Hernandez and Garza had left Gonzalez’s office, Gonzalez and Robles discussed what had happened. Both testified they knew that Garza, although a good worker, had earned a very poor safety record over the twenty years he had worked for SWBT, one of the worst Robles had seen among the employees who had worked for him. SWBT’s files showed that Garza had been warned following a motor vehicle accident in 1987 “that his safety record was terrible and that a future accident of any nature, be it his fault or not, could result in his being taken off the payroll”. Company files also showed that two years later, after another motor vehicle accident, Garza had been cautioned again that “if another problem arises that requires disciplinary action, you will be subject to termination.” But Garza’s record did not reflect an accident of any kind in the prior 27 months, and his 1993 performance appraisal on safety had rated him “satisfactory”. On July 22, 1994, *611 barely three months earlier, Robles had written Garza: “You made a commitment to be safe on the job and you did a good job. Keep up the good work.”
Nevertheless, at trial Robles testified that on the day of the incident he recommended to Gonzalez that Garza be disqualified from driving or working on outside jobs because “[h]e was either going to end up killing himself or killing his partner or injuring the public.” Gonzalez testified that he agreed and telephoned his superior in San Antonio, Wayne Rider, to report on the day’s events. According to Gonzalez’s and Rider’s accounts of their conversation, Gonzalez told Rider he had had a long-term problem with Garza and was “fed up” with all of Garza’s problems. Garza, he said, should no longer be allowed to drive. Although Rider supervised some 500 employees, he knew Garza from his poor safety record and told Gonzalez to put his recommendation in writing. Rider instructed Gonzalez to inquire whether Garza had threatened other employees and to request a manager outside the area to conduct an impartial investigation of the incident. But these steps, Rider testified, were “just kind of like making sure, dotting the i’s and crossing the t’s, so to speak.” The decision to disqualify Garza from driving or working outside, he said, “was made on basically the 20th” and the next day, before Garza’s worker’s compensation claim of injury the following Monday, and “everything that happened after that was just window dressing”.
Rider’s testimony is somewhat at odds with other evidence. For one thing, he also testified that he did not make a final decision for some two weeks. For another, if a decision was made Thursday, it was not immediately implemented. After meeting with Gonzalez, Garza was sent to finish the job he and Hernandez had been working on, assisted by another employee. On Friday and Saturday, Robles continued to assign Garza other outside work because, he later testified, “[m]y boss [Gonzalez] had not given me any other direction”, despite their conversation on Thursday. Meanwhile, Hernandez was given work cleaning out trucks.
Most inconsistent with Rider’s testimony are the first three written reports of the Thursday incident, none of which reflected that Garza would or should be disciplined. On Tuesday or Wednesday, October 25 or 26, Robles telephoned Rider’s clerk to provide her information for a standard form “flash report”. The report she prepared stated that no investigation committee would be appointed and that the only action proposed to be taken was “positive discipline”. Robles testified that he did not give the clerk this information, but Rider testified that the clerk would have completed the report as she was told, so that even if she added anything herself, it would only have been more than, not contrary to, what Robles told her.
About the same time, Gonzalez prepared his own report. The report stated that Gonzalez had interviewed Garza’s fellow employees as Rider had instructed, and that the nine who were contacted “stated they have never seen David threaten anyone.” Gonzalez’s report concluded:
Because there [were] no witnesses to the incident and both employees have a history of not getting along with each other and submitted conflicting stories, it is impossible to substantiate the exact events of this incident.
As a result of my investigation, I feel that both employees contributed to this confrontation and both of them should be placed on a step of positive discipline for effectiveness [with] others and be reminded that they must learn to work with each other in the future.
*612 Gonzalez’s report, it should be noted, was entirely consistent with the “flash report”, including the parts Robles disavowed, and inconsistent with the decision the two claimed to have reached the day of the accident—to disqualify Garza from driving or outdoor work.
Contrary to one statement in the “flash report”, Gonzalez did ask another manager to form a committee to investigate the incident, again as Rider had instructed. The committee of five made its report on October 27. The exact cause of the accident, they found, could not be determined because Garza and Hernandez were the only witnesses, both were reasonably credible, and their accounts differed. The committee found that both had acted unsafely, faulting Garza for continuing to work beneath the bucket after almost being hit by the rope Hernandez dropped and for not pointing out to Hernandez that he’d dropped it. The committee concluded that both employees should have communicated better and been more careful working with the aerial bucket. The committee’s only recommendations were that both employees receive further training in safety procedures and guidance in communication with others.
The first hint in SWBT’s records that Garza would be more severely disciplined came in notations Rider added to Gonzalez’s “flash report” several days after Garza’s compensation claim. In the space where the report asked whether an investigation committee would be appointed, he bracketed the typewritten “no” and wrote, “yes”. Where the report called for an explanation of action to be taken, Rider wrote: “accident will be investigated, and if employee contributed to the cause of the accident, disciplinary action will be initiated.” He then struck out all but the first four words and added instead: “Disciplinary action will be administered after investigation”.
On November 7, Gonzalez sent Rider a report far different from his report ten days earlier. It began:
This is David’s tenth recorded accident. The Investigating Committee found fault with both participants, but could not determine the exact cause because of the different versions the employees gave. Irregardless of the discrepancies, the accident did occur and David was hurt. We can no longer tolerate or should accept this kind of disregard towards safety by an employee.
The memorandum cited ten accidents in which Garza had been involved—five involving motor vehicles (1979: rear-ended a vehicle; 1980: was struck by a vehicle that failed to yield; 1986: backed into a vehicle; 1989: hit a fire hydrant; 1992: pole on trailer hit windshield of an occupied vehicle), and five requiring medical treatment (1978: stepped on a nail; 1979: strained shoulder lifting a manhole cover; 1981: slipped and fell in mud, spraining right knee; 1982: slid down a pole and got splinters in his chest; 1994: the incident with Hernandez). The report also listed eight other “miscellaneous occurrences” of unsafe conduct, from dropping a piece of cable on a parked car to allowing his foot to slip off the clutch of his truck. The report made no attempt to assess the seriousness of the accidents or the degree to which Garza had been at fault in any of them. It noted that his performance rating for safety had been “unsatisfactory” for five out of the twenty years Garza had been employed by SWBT without noting his ratings for the other fifteen years. It listed four instances of disciplinary action, all minor—one suspension for several hours, and three reminders to be more careful. The report then concluded:
*613 While this accident did not cause major harm or damage, it is another indicator of David’s unconcern towards safety and a warning of further accidents which could cause greater harm or damage. David will not admit to having a safety problem regardless of all the accidents and incidents, because to David, it is always the other persons fault. He will never change his unsafe behavior.
The company has given David many chances over the years to change his behavior. Positive discipline, training and numerous sessions between management and the local union to help him have not been successful. We have no other alternative but to take David off his present assignment. I recommend we give him 90 days to find an inside job with the company and be removed from his present job assignment. David is a hard worker and deserves the opportunity.
Rider approved Gonzalez’s November 7 report the same day. When Gonzalez was asked at trial to explain why he had changed positions so drastically from his first report some ten days earlier, he cited only the committee investigation and Garza’s safety record. But the committee investigation had concluded that fault could not be determined and therefore did not recommend serious discipline, and Gonzalez testified that he was well aware of Garza’s safety record before October 20, the day of the accident. The record simply does not explain why Gonzalez changed positions. Gonzalez’s explanations at trial for not recommending that Hernandez be disciplined were that his safety record was not as bad as Garza’s, and that shortly after the accident he was no longer under Gonzalez’s supervision, although neither was Garza.
Amidst these events, and unrelated to them, Garza’s and Hernandez’s jobs at SWBT’s Brownsville office were declared to be “surplus” and transferred to Harlingen. SWBT vice president and general manager, J.W. Galloway, stated in an October 26 memo announcing the decision that both men would be given the opportunity to work out of Harlingen. Nothing in the record reflects that Galloway was aware of Garza’s October 20 accident or October 24 medical treatment. On November 7, the same day Rider approved Gonzalez’s second report, Garza applied for a transfer. The following day, Gonzalez and Robles met with Garza to tell him that because of his safety record and the October 20 accident, he could no longer work as a lineman or drive a company vehicle. But the very next day, Robles signed off on Garza’s transfer application, giving him a job performance rating of “satisfactory” for safety. Robles testified that he simply copied the ratings from Garza’s last job performance review some months earlier.
SWBT’s written policies regarding driving and safety did not require the discipline imposed on Garza (although neither did they forbid it), and the local union lodged a grievance. A meeting was convened on November 14 to determine whether Garza’s October 20 injury was his own fault. Dennis Dobbs, the local union president, and Vasquez, its vice president, attended the hearing along with Garza, Gonzalez, and Robles. Vasquez and Dobbs took notes. At trial five years later, Vasquez (who by then had left the union for a SWBT management position) testified from her notes that Gonzalez had told Garza “the fact that he had gone to the doctor” had “caused this investigation and this action to take place”. Gonzalez testified that he did not recall making the statement, but that if he did, he meant only that absent a request for medical treatment, the accident would not have *614 been “reportable” and would not have resulted in an investigation, discipline, and the consequent union complaint. Gonzalez provided copies of the reports that had been generated after the October 20 accident, and the meeting was recessed to give Garza and the union representatives time to review them.
The meeting reconvened November 30, this time with Rider and Robles present, along with Garza, Dobbs, and Vasquez. Dobbs and Vasquez again took notes. At trial, Vasquez testified as follows:
Q: What do you recall happening that surprised you?
A: The fact that David was told he was the best employee they had ever had and that if it hadn’t been for him going to the doctor he wouldn’t be sitting there with him being in this position of getting a non-driving job.
Q: Have you ever heard management ever say anything like that before?
Q: What—how did you feel about that?
A: I was in shock. It was a real surprise to me.
Vasquez could not recall for certain who made the statement, although she believed it was Rider. Dobbs, who remembered the same statement, testified that it was made by Rider.
On December 28, Garza, Dobbs, and Vasquez met with SWBT’s human resources director, whose job it was to review grievances. Again according to Vasquez’s notes, Dobbs began by recounting the circumstances leading to Garza’s discipline and specifically stated that Rider had said earlier, “if David would not hv gone to Dr nothing would have been done.” Rider testified at trial that he did not make the statement Vasquez and Dobbs attributed to him but instead said, as reflected in Dobbs’s notes, that “[w]hat caused [10/20] to be a reportable accident is that DG went to the Dr, recvd a prescription for meds and had the Rx filled.” This statement, Rider explained, was true; OSHA regulations made such an accident “reportable”. But, he contended, his statement was not intended to, and did not, say that the reason Garza was being disciplined was because he had sought medical treatment.
Garza was willing to work out of Harlingen or any other Valley locale, but after November 8 he was disqualified from working as a lineman. SWBT at first gave him until December 30 to find an inside job with SWBT or face termination, and later it extended the deadline to January 31. In the meantime, Garza was assigned work around the Brownsville office stocking materials in the yard and storeroom, work he regarded as demeaning and “janitorial”. The work did not include things like mopping and cleaning, which SWBT contracted with others to do. Garza was also allowed to spend much of his workday studying in the library to qualify for other company positions. He tested for other jobs but did not qualify. He was offered a job in San Antonio at less pay, but he was not willing to move away from the Valley because his wife could not leave her job and his ailing father lived with them. Finally, on January 31, Garza was terminated.
After Garza left SWBT, Rider encouraged him to continue to do lineman work for SWBT as an independent contractor and even offered to help him secure a small business loan to go into business for himself. Garza did take other jobs as a lineman and contractor, often working for SWBT, sometimes doing much of the same work he had before.
In July 1996, Garza sued SWBT. Three years later, the case was tried before a jury, which found that SWBT—
*615 disqualif[ied] or discharge[d] David Garza from his Outside Plant Technician (OSPT) position because he instituted or caused to be instituted a workers compensation claim in good faith.
The jury was instructed in this regard that:
There may be more than one cause for an employment decision. An employer does not disqualify or discharge an employee for instituting or causing to be instituted a workers compensation claim in good faith, if the employer would have disqualified or discharged him when it did even if the employee had not instituted or caused to be instituted a workers compensation claim in good faith.
The jury found actual damages totaling $1,034,1085 and assessed exemplary damages of $1,000,000. The trial court rendered judgment against SWBT on the verdict, plus prejudgment interest, for $2,170,841.64. The court of appeals affirmed.6
We granted SWBT’s petition for review.7 SWBT complains of error in the jury charge and of the legal sufficiency of evidence to support liability, and alternatively, the award of exemplary damages. We examine each in turn.
We begin with the jury charge. As we noted at the outset, the Anti–Retaliation Law states:
A person may not discharge or in any other manner discriminate against an employee because the employee has ... filed a workers compensation claim in good faith ....8
The trial court inquired of the jury:
Did [SWBT] disqualify or discharge David Garza from his Outside Plant Technician (OSPT) position because he instituted or caused to be instituted a workers compensation claim in good faith?
(Emphasis added.) SWBT argues, for three reasons, that the trial court’s substitution of “disqualify” for “discriminate” was harmful error.
First, SWBT argues that it was entitled to have the question of its statutory liability submitted in the words of the statute. SWBT is correct that, as a rule, “[w]hen liability is asserted based upon a provision of a statute or regulation, a jury charge should track the language of the provision as closely as possible.”9 Certainly, “disqualify” and “discriminate” are not synonyms. But Garza responds that the charge merely specified the nature of the alleged discrimination, as encouraged by the Texas Pattern Jury Charges, which suggests the following question in this type of case:
Did Don Davis [discharge or (describe other discriminatory action) ] Paul Payne because he [filed a workers compensation claim in good faith ...]?10
*616 SWBT reads this suggestion not to dispense with the statutory language but only to focus it, such as by a question like this:
Did Don Davis discriminate against Paul Payne by reassigning him to a lower-paying job classification because he filed a workers compensation claim in good faith?
On the other hand, SWBT argues, if this reading is incorrect and the pattern jury question does allow omission of the statutory language, then it is improper. While it may be useful to draw the jury’s attention to the particular misconduct alleged in the case, SWBT continues, to omit the operative statutory words is to deprive the jury of the legal lens through which the misconduct must be viewed, as by asking a jury only whether the defendant turned right instead of whether turning right, if it happened, was negligent. Discrimination means treating similarly situated persons differently; to determine whether it has occurred requires a comparison between the plaintiff and others like him. To determine whether disqualification has occurred requires no such comparison. In essence, SWBT argues, the trial court incorrectly equated disqualification with discrimination. Garza answers that in this case, at least, there was no difference between asking the jury whether he was disqualified for filing a compensation claim and asking whether he was discriminated against by being disqualified for filing a compensation claim.
We agree with Garza. His disqualification was unquestionably disciplinary, and if it was imposed because Garza filed a compensation claim, as the jury found, and not because of the October 20 accident and Garza’s safety record, as SWBT contended, then it clearly violated the Anti–Retaliation Law. The jury charge would have been more accurate had it asked whether SWBT discriminated against Garza for filing a compensation claim by disqualifying him from his position, retaining the statutory term while focusing on Garza’s claim, but we do not think it reasonable to conclude that the jury was in any important way confused by the question it was asked.
Next, SWBT argues that Garza pleaded only unlawful discharge, not discrimination. Garza pleaded that SWBT “embarked upon a plan to terminate his employment” and that he suffered damages as a “consequence of the unlawful termination” or “retaliatory discharge”. SWBT contends that Garza recognized the deficiency in his pleadings by moving, after the verdict, to amend his pleadings to allege discrimination. But Garza also pleaded that “[i]n retaliation for filing” a compensation claim, he was “ordered to find a non-driving position” and was thereby “effectively disqualified from his job”. He sought damages for “retaliatory discharge under the Texas Workers’ Compensation Act and wrongful conduct” of SWBT. While Garza’s pleadings did not use the word “discrimination”, they were nevertheless sufficient to raise that charge if they gave fair notice of the facts on which Garza based his action sufficient to allow SWBT to anticipate what evidence would be relevant and to prepare its defense.11 We think Garza’s pleadings clearly met this *617 standard. Had SWBT been in doubt about Garza’s claims, it could have sought clarification through special exceptions. It did not do so.
Finally, SWBT argues that nearly half of the damages Garza—recovered for past and future lost earnings and benefits—resulted not from any discrimination against him but from his termination. SWBT contends that Garza’s discharge resulted from the “surplus” determination unrelated to his accident and compensation claim. But the “surplus” determination expressly contemplated that Garza could transfer to Harlingen and work at the same position. What kept him from doing so was his disqualification. SWBT’s argument simply rejects Garza’s assertion that because he filed a compensation claim, he was disqualified from his position and could not find another. Having agreed with Garza, the jury could find that the lost earnings and benefits resulted from his disqualification.
Accordingly, we reject SWBT’s complaint of reversible error in the jury charge.
SWBT next complains that there was no evidence to support the jury’s finding of liability. Rather, SWBT contends, the evidence establishes that it would have disciplined Garza as it did even if he had not filed a compensation claim.
SWBT has essentially two arguments. One is that the decision to discipline Garza was made on Thursday, the day of the accident, or perhaps the next day, but in any event before his compensation claim the following Monday. But as recited above, SWBT’s own records can fairly be read to the contrary. For example, Gonzalez, who was involved in the decision to disqualify Garza, recommended in a written report a day or two after the compensation claim that Garza be given only “positive discipline”. SWBT’s other argument is that it disqualified Garza because of his extraordinarily poor safety record, having specifically warned him in the past that such action might be necessary. Again, there is evidence in SWBT’s records to the contrary. For example, in the report just mentioned, made after the compensation claim, Gonzalez did not cite Garza’s safety record as a basis for discipline even though Gonzalez testified he knew of Garza’s safety record the day of the accident. Rather, he concluded that discipline should be limited because Garza’s fault in the Thursday accident could not be determined. Not until his report a week later did Gonzalez cite Garza’s poor safety record as a basis for severe discipline.
SWBT notes that there is no evidence that it discouraged compensation claims, but there is evidence that it disciplined Garza, who made a claim, more severely than it disciplined Hernandez, who did not. SWBT explains that Hernandez’s safety record was better than Garza’s, but the jury was not required to accept that explanation. SWBT acknowledges that its written safety policies did not require Garza’s disqualification, but neither, it points out, did those policies preclude disqualification. Again, SWBT’s position for going beyond its written policies is not one the jury was required to accept. Vasquez, the union vice president who had moved to management by the time of trial, testified that she was shocked and surprised by a statement made during the grievance meetings by one of Garza’s superiors that seemed to connect his discipline with his compensation claim. Dobbs, the union president, supported Vasquez’s testimony. SWBT argues that it was Garza’s injury, not his claim, that precipitated the investigations and disciplinary decision, but the jury was free to believe Vasquez and Dobbs. The *618 jury might also have discounted SWBT’s professed concerns for Garza’s carelessness when it heard Rider testify that he encouraged Garza to continue doing the same work for SWBT after his termination, as a private contractor.
From this and the other evidence we have detailed above, we have little difficulty rejecting SWBT’s argument that the jury’s liability finding lacks support in the record.
Finally, SWBT argues that there was no evidence to support the jury’s finding of actual malice as a predicate to an award of punitive damages. The jury was asked and instructed as follows:
Do you find by clear and convincing evidence that the harm to David Garza resulted from actual malice on the part of [SWBT]?
“Clear and convincing evidence” means the measure or degree of proof that produces a firm belief or conviction of the truth of the allegations sought to be established.
“Actual malice” means ill will, spite, evil motive, or purpose to injure another.
Neither SWBT nor Garza objected to this question or these instructions. Both concede that proof of actual malice was required to recover punitive damages for a violation of the Anti–Retaliation Law,12 that actual malice was correctly defined,13 that proof was required to be by clear and convincing evidence,14 and that “clear and convincing evidence was correctly defined”.15 In assessing the evidence, we *619 assume that the portions of the charge just quoted, because they were given without objection, correctly state the law.16
We must first determine the standard of review. The question is whether an elevated standard of proof at trial—here, clear and convincing evidence—requires a correspondingly elevated standard of evidentiary review on appeal. More specifically: will anything more than a scintilla of probative evidence—evidence “that would enable reasonable and fair-minded people to differ in their conclusions”—support a verdict of actual malice, as it would if proof at trial were by a preponderance of the evidence,17 or must there be more—must there be some evidence that produces a firm belief or conviction of the truth of the allegation, since that was the “clear and convincing” standard of proof the jury was required to apply? The court of appeals did not discuss the issue and applied the ordinary standard of review.18 SWBT argues that an elevated standard must be applied. Garza’s position is not clear.
The question must be answered in the context of the distinction, perhaps more deeply engrained in Texas law than in that of any other jurisdiction, between no evidence and insufficient evidence.19 The hardiness of this distinction in our jurisprudence, if not its very existence, derives from a clause included in the 1891 amendments to Article V of the Texas Constitution,20 amendments that gave the Texas judiciary its present structure.21 Section 6 created courts of civil appeals and provided that “the decision of said courts shall be conclusive on all questions of fact brought before them on appeal or error.”22 History provides no clear indication of the framers’ *620 purpose in including this “factual conclusivity clause” or the ratifiers’ purpose in adopting it.23 Since one overall purpose of the amendments was to reduce this Court’s workload, the clause may have been intended to help achieve that end, merely as a practical matter.24 More philosophically, it has been suggested that in restructuring the judiciary, the framers may have come to believe that one appeal regarding case-specific factual issues was enough, while a second appeal regarding generally applicable legal issues was necessary to assure uniform development of the law throughout the State.25 Whatever its purpose, the clause was not a conveyance of new authority to appellate courts. The year before the amendments were adopted, this Court had asserted:
“Although this court has the power to review a case upon the facts, and to set aside a verdict which has evidence to support it, that power has been reluctantly exercised. But it is the right and duty of the court to set aside a verdict, when it is against such a preponderance of the evidence, that it is clearly wrong.”26
Rather, the “factual conclusivity clause” confined this one part of the appellate function—determining whether a judgment has sufficient evidentiary support—to the courts of civil appeals, now the courts of appeals.27 The clause did not affect this Court’s power to determine whether a judgment has any evidentiary support. As we explained not long after the amendments took effect:
it is elementary that whether there be any evidence or not to support an issue is a question of law, and not of fact, and it follows that the decision of the court of civil appeals upon such a question is subject to review by this court.28
Based on this statement, no evidence is sometimes referred to as being legally insufficient, the issue being one of law, while evidence so against the preponderance of other evidence that the judgment it supports is clearly wrong is sometimes referred to as being factually insufficient. *621 The use of the word “insufficient” in both contexts has compounded confusion over the distinction.29
While the “factual conclusivity clause” requires that a distinction be made between questions of fact and questions of law, it does not prescribe where the line is to be drawn, leaving that matter for this Court. Over the years, we have concluded that no evidence means not only a complete absence of evidence but also evidence which cannot be given legal effect, either because the law does not permit it or because the evidence is too weak.30 To define evidence that is too weak to be given legal effect, we have adopted the “scintilla” rule. As Chief Justice Calvert explained in his seminal article on the subject:
The scintilla rule, although having its origin at a much earlier date, was firmed up in Texas in Joske v. Irvine [91 Tex. 574, 44 S.W. 1059 (Tex.1898) ]. The rule may be stated in these words: when the evidence offered to prove a vital fact is so weak as to do no more than create a mere surmise or suspicion of its existence, the evidence is, in legal effect, no evidence, and will not support a verdict or judgment.31
In other words, evidence which does no more than create a mere surmise or suspicion cannot prove the truth of an allegation. The rule works in tandem with the preponderance-of-the-evidence standard of proof: any evidence that does not merely create surmise or suspicion can be used to show that something is more likely than not. But when proof of an allegation must be clear and convincing, even evidence that does more than raise surmise and suspicion will not suffice unless it is capable of producing a firm belief or conviction that the allegation is true. Evidence of lesser quality is, in legal effect, no evidence. Whether evidence is of such quality is thus a question of law in a case with that elevated standard of proof, just as whether evidence is more than a scintilla is a question of law in a case proved by a preponderance of the evidence.
In reviewing a finding that must be proved by clear and convincing evidence, it makes no sense for an appellate court to determine whether the supporting evidence amounts to more than a scintilla. Even if it does, the finding is invalid unless the evidence is also clear and convincing. In such a case, the review required by the “scintilla” rule is wholly irrelevant and thus, no review at all. Thus, as a matter *622 of logic, a finding that must meet an elevated standard of proof must also meet an elevated standard of review.
We reached essentially the same conclusion in In re C.H.32 and In re J.F.C.,33 two cases involving the termination of parental rights. Because “[t]he natural right which exists between parents and their children is one of constitutional dimensions”,34 we had previously held that it could be terminated only on clear and convincing evidence,35 and shortly afterward the United States Supreme Court36 and the Texas Legislature37 reached the same conclusion. The issues in C.H. and J.F.C. were whether the higher proof standard necessitated higher standards of factual sufficiency and legal sufficiency review, respectively. In J.F.C. we explained:
We held in In re C.H. “that the appellate standard for reviewing termination findings is whether the evidence is such that a factfinder could reasonably form a firm belief or conviction about the truth of the States allegations.” We expressly “reject[ed] standards that retain the traditional factual sufficiency standard while attempting to accommodate the clear-and-convincing burden of proof.” We concluded that “the burden of proof at trial necessarily affects appellate review of the evidence.” We explained:
Under traditional factual sufficiency standards, a court determines if a finding is so against the great weight and preponderance of the evidence that it is manifestly unjust, shocks the conscience, or clearly demonstrates bias. But that standard is inadequate when evidence is more than a preponderance (more likely than not) but is not clear and convincing. As a matter of logic, a finding that must be based on clear and convincing evidence cannot be viewed on appeal the same as one that may be sustained on a mere preponderance.
The same logic dictates the conclusion that our traditional legal sufficiency standard, which upholds a finding supported by “[a]nything more than a scintilla of evidence,” is inadequate when the United States Constitution requires proof by clear and convincing evidence. Requiring only “[a]nything more than” a mere scintilla of evidence does not equate to clear and convincing evidence.38
Although the parental interest at issue was of constitutional magnitude, and the standard of proof constitutionally prescribed, we did not decide whether an elevated standard of review was likewise constitutionally mandated but applied it purely as a “matter of logic”.
*623 Similarly, the United States Supreme Court has held in other contexts that when the standard of proof at trial is constitutionally required to be more exacting, the standard of review on appeal must be more exacting as well. In Jackson v. Virginia, for example, the Court considered “what standard is to be applied in a federal habeas corpus proceeding when the claim is made that a person has been convicted in a state court upon insufficient evidence.”39 Noting that “the Due Process Clause of the Fourteenth Amendment protects a defendant in a criminal case against conviction ‘except upon proof beyond a reasonable doubt’ ”40 the Court concluded that this right could not adequately be protected by jury instructions alone;41 evidentiary review was essential42 Nor would the ordinary standard of review suffice:
“[A] mere modicum of evidence may satisfy a ‘no evidence’ standard....” Any evidence that is relevant—that has any tendency to make the existence of an element of a crime slightly more probable than it would be without the evidence—could be deemed a “mere modicum.” But it could not seriously be argued that such a “modicum” of evidence could by itself rationally support a conviction beyond a reasonable doubt.43
The elevated proof requirement necessitated a more exacting standard of review:
the critical inquiry on review of the sufficiency of the evidence to support a criminal conviction must be not simply to determine whether the jury was properly instructed, but to determine whether the record evidence could reasonably support a finding of guilt beyond a reasonable doubt.44
Texas courts use this same standard.45 We relied on Jackson in J.F.C.:
The reasoning in Jackson reinforces our conclusion that to apply our traditional no evidence standard of review in a parental termination case would not afford the protections inherent in the clear and convincing standard of proof. As the example in Jackson highlights, a parents rights could be terminated based on “but one slender bit of evidence” as long as the jury was properly instructed on the clear and convincing evidence burden of proof. Our legal sufficiency review, therefore, must take into consideration whether the evidence is such that a factfinder could reasonably form a firm belief or conviction about the truth of the matter on which the State bears the burden of proof.46
Likewise, the Supreme Court held in New York Times Co. v. Sullivan that when proof of actual malice is required in defamation cases, evidence of “convincing clarity” is required.47 Further, appellate courts cannot use ordinary standards of review but “must ‘make an independent examination of the whole record,’ so as to *624 assure [itself] that the judgment does not constitute a forbidden intrusion on the field of free expression.”48 Texas law adheres, of course, to both requirements.49
In Jackson and in Sullivan, as in J.F.C. and C.H., the elevated standard of evidentiary review on appeal was not itself an independent constitutional guarantee but was instead a necessary concomitant to a constitutional right to proof by more than a preponderance of the evidence. Whatever the source of a right to an elevated standard of proof, whether by constitutional provision, by federal50 or state51 *625 statute, or by court decision,52 the problem of evidentiary review on appeal is the same: a finding that cannot be made on evidence of lesser quality should not be affirmed on such evidence.
Three objections, primarily, are raised to the use of an elevated standard of evidentiary review on appeal. None, we think, has merit. The first is that such a standard invites the appellate court to weigh evidence and resolve issues of credibility, thus invading the province of the jury. But the required weighing of evidence is neither more nor different than that involved in an ordinary factual sufficiency review; the only difference is that a higher quality of evidence is necessary to tip the scales. Issues of credibility that depend on appearance and demeanor cannot be weighed by the appellate court; the witnesses are not present. And even when credibility issues are reflected in the written transcript, the appellate court must defer to the jury’s determinations, at least so long as those determinations are not themselves unreasonable. For example, the Supreme Court explained in Jackson that an elevated standard of review—
does not require a court to “ask itself whether it believes that the evidence at the trial established guilt beyond a reasonable doubt.” Instead, the relevant question is whether, after viewing the evidence in the light most favorable to the prosecution, any rational trier of fact could have found the essential elements of the crime beyond a reasonable doubt. This familiar standard gives full play to the responsibility of the trier of fact fairly to resolve conflicts in the testimony, to weigh the evidence, and to draw reasonable inferences from basic facts to ultimate facts. Once a defendant has been found guilty of the crime charged, the factfinder’s role as weigher of the evidence is preserved through a legal conclusion that upon judicial review all of the evidence is to be considered in the light most favorable to the prosecution. The criterion thus impinges upon “jury” discretion only to the extent necessary to guarantee the fundamental protection of due process of law.53
Similarly, we held in J.F.C. and C.H. that an appellate court must assume that the jury resolved evidentiary conflicts in favor of its verdict if it reasonably could have done so.54
The second objection is that an elevated no-evidence review standard is unnecessary because the appellate court can assure adherence to the proof standard through a review of the sufficiency of the evidence. Indeed, this Court has occasionally said that “[t]he requirement of clear and convincing evidence is merely another method of stating that a cause of action must be supported by factually sufficient evidence”,55 although we have not *626 said this in a case in which clear and convincing was a requirement for proof at trial.56 The requirement that a jury use a heightened standard of proof is a relatively recent phenomenon.57 An insufficiency review cannot substitute for a no-evidence review because the consequences are different. As a rule, if a party with the burden of proof offers no evidence on an issue, the opposing party is entitled to judgment.58 If, however, there is some evidence to prove an issue but that evidence is factually insufficient, the only remedy is a new trial.59 To police adherence to an elevated proof standard solely through insufficiency review is to deny the remedy of rendition to those whose interests deserve increased protection. This, of course, makes no sense.
Finally, even if Texas courts of appeals must use an elevated standard of no-evidence review, objection is made to this Court’s authority to do so, based on the “factual conclusivity clause” of the Texas Constitution with which we began this discussion. But as we have explained, the “factual conclusivity clause” does not define *627 the “questions of fact” on which the courts of appeals’ decisions are conclusive. We held, shortly after the clause was adopted, that “whether there be any evidence or not to support an issue is a question of law, and not of fact”.60 Evidence that does not produce a firm belief or conviction does not support an issue that must be proved by clear and convincing evidence. Whether evidence supports an issue is no less a question of law simply because the standard of proof is heightened. We essentially reached this conclusion in defamation cases requiring proof of actual malice. The Supreme Court has stated that “[t]he question whether the evidence in the record in a defamation case is sufficient to support a finding of actual malice is a question of law.”61 We have agreed and treated the question as such,62 although we might have concluded for purposes of the “factual conclusivity clause” that the question was one of fact exclusively for the courts of appeals. Such a conclusion would not have contradicted the Supreme Court but would simply have assigned its required evidentiary review solely to our courts of appeals. By undertaking the review ourselves, we in effect determined that the required evidentiary review was within our jurisdiction.
In sum, we think that whenever the standard of proof at trial is elevated, the standard of appellate review must likewise be elevated. Because the “clear and convincing” standard of proof in this case is the same as in J.F.C., we follow the procedure outlined there for conducting a no-evidence review:
In a legal sufficiency review, a court should look at all the evidence in the light most favorable to the finding to determine whether a reasonable trier of fact could have formed a firm belief or conviction that its finding was true. To give appropriate deference to the factfinder’s conclusions and the role of a court conducting a legal sufficiency review, looking at the evidence in the light most favorable to the judgment means that a reviewing court must assume that the factfinder resolved disputed facts in favor of its finding if a reasonable factfinder could do so. A corollary to this requirement is that a court should disregard all evidence that a reasonable factfinder could have disbelieved or found to have been incredible. This does not mean that a court must disregard all evidence that does not support the finding. Disregarding undisputed facts that do not support the finding could skew the analysis of whether there is clear and convincing evidence.
If, after conducting its legal sufficiency review of the record evidence, a court determines that no reasonable factfinder could form a firm belief or conviction that the matter that must be proven is true, then that court must conclude that the evidence is legally insufficient.63
We need not consider here whether a different procedure for review (as opposed to the standard of review) of an award of punitive damages is required by the United States Supreme Court’s recent decision *628 in Cooper Industries, Inc. v. Leatherman Tool Group.64
We now turn to the evidence of actual malice—that is, ill will, spite, evil motive, or purposeful injury. Garza complains mostly of Gonzalez and Rider. SWBT argues that neither was proved to be a vice-principal whose conduct can be attributed to SWBT for the purpose of awarding exemplary damages. We need not reach this argument and assume for purposes of our analysis that Gonzalez and Rider were vice principals.
Garza points to the following evidence:
• Gonzalez—implied Garza was a bully; told Rider he had long-term problems with Garza and was “fed up” while allowing Garza to return to outside work; went out of his way to see that Garza was punished, even after Garza was no longer under his supervision; distorted Garza’s safety record, blowing small incidents out of proportion, even though Garza continued to be rated as satisfactory on safety two days after discipline was imposed; gave Rider a second report that was a diatribe against Garza;
• Rider—admitted he was personally familiar with Garza, although he had some 500 employees under his supervision, and pressed for having Garza disciplined without a determination that Garza was at fault in the accident with Hernandez; severely disciplined Garza while calling him one of SWBT’s best linemen; faulted Garza’s safety record as a lineman but offered Garza the same work as an independent contractor;
• the discipline imposed on Garza was not required by SWBT policies;
• the union meetings following Garza’s discipline were a sham, or as Rider put it, “window dressing”;
• after discipline was imposed, Garza was assigned demeaning work, humiliating him in front of his co-workers;
• Hernandez was not disciplined.
With this evidence, however, we must also consider other evidence that was undisputed. Garza unquestionably had a lengthy record of safety violations and had been strictly warned twice before the October 20 accident that he could be terminated for any further mishap, irrespective of fault. SWBT’s investigation of the accident, which included a thorough and impartial inquiry by a committee headed by an outside manager, resulted in differences of opinion. Although not required by SWBT policies, the discipline imposed on Garza was not prohibited and was consistent with the warnings he had previously received. The meetings with union officials were at their behest, and SWBT employees merely attended as requested. SWBT gave Garza two months at first, and then an additional month, to relocate within the company, offering him the opportunity to use work time to study in order to qualify for other positions. While Garza considered the shopwork to which he was assigned during that period demeaning, it was not janitorial in the sense that it did not involve mopping and cleaning. And Hernandez’s safety record was not as poor as Garza’s.
Viewing all of this evidence—as well as the evidence we have detailed earlier along with the entire record—in the light most favorable to the verdict, we cannot conclude that a reasonable trier of fact could form a firm belief or conviction that SWBT acted toward Garza with ill will, spite, evil motive, or purposeful injury. While there *629 are some indications that it might have done so, there are a great many others that it did not. At most, the record reflects that SWBT mishandled the situation; it does not produce a reasonable conviction that SWBT intended to punish Garza without cause.
Therefore, we conclude that there is no clear and convincing evidence of actual malice to support an award of punitive damages.
The concurring opinion agrees that an elevated standard of review must be applied, agrees that no evidence of actual malice meets that standard, and then castigates the Court for “impermissibly weigh[ing] conflicting evidence”.65 The opinion gives only two examples.
First, the concurring opinion states that “the jury could reasonably have inferred that Garza’s record was not as poor as SWBT suggested”, implying that we disagree.66 We do not. We absolutely agree. But while our review must acknowledge reasonable inferences that may have been drawn by the jury, it cannot, under the procedures set out in J.F.C., disregard undisputed evidence. It is undisputed not—merely unchallenged, but undisputed—that Garza’s safety record was accurately reflected in his personnel file. Garza conceded at trial that every incident for which he was cited actually occurred, and that on most occasions he was at fault. He vigorously disputes that his safety record justified, or was the actual reason for, the discipline he was given, and the jury was certainly entitled to believe him. The standard of review requires us to credit the jury’s reasonable inference, but it does not allow us to disregard the undisputed fact that Garza’s record of safety violations was lengthy.
Next, the concurring opinion states that we “suggest[ ] that SWBT actually aided Garza by conducting a ‘thorough and impartial’ investigation.”67 We suggest no such thing. We say only, as the concurring opinion does, that “[s]ome parts of the investigation may indeed have been impartial, and may even have favored Garza.”68 We agree with the concurring opinion that the jury may have believed that the investigation, as a whole, was not thorough and impartial. Again, however, the standard of review requires that we credit evidence the jury could have believed without disregarding undisputed evidence.
We agree with the concurring opinion that the jury could have believed: that SWBT overstated the significance of Garza’s safety record, that the investigation was in large part unfair, that Garza’s discipline was inconsistent with SWBT’s written policies, and that SWBT was insincere in helping Garza qualify for other jobs with the company. But in determining whether there was any clear and convincing evidence that SWBT acted with actual malice, we must also consider the undisputed evidence that Garza had a lengthy record of safety violations, overstated or not, that part of the investigation was impartial and tended to exonerate Garza, that no written company policy forbade the discipline Garza was given, and that Garza was given the time and opportunity to qualify for another job, of which he has never complained. This analysis, as we have explained, is required by the elevated standard of proof and does not *630 transgress constitutional limits on our authority.
The concurring opinion cannot point to any evidence or inference favorable to Garza that we have failed to consider. Without so much as a single example, the concurring opinion accuses the Court of decision-making in excess of its authority, then reaches exactly the same result we do. Its charges must be viewed in that light.
* * * * * *
Accordingly, the judgment of the court of appeals is reversed insofar as it affirms the award of exemplary damages, and in all other respects is affirmed.
Justice O’NEILL filed an opinion concurring only in the judgment.
Justice O’NEILL, concurring.
“[The Pirate’s] Code is more what you’d call ‘guidelines’ than actual rules.”
PIRATES OF THE CARIBBEAN
The Court purports to recognize the constitutional limitation that the factual conclusivity clause imposes upon our jurisdiction, yet proceeds to weigh the evidence unfettered by any such constraint. Because the Court usurps the factual sufficiency review power that our Constitution reserves to the courts of appeals, I cannot join its opinion. I agree, however, that the judgment awarding punitive damages should be reversed in this case. Under the standard announced in Continental Coffee Products Co. v. Cazarez, 937 S.W.2d 444, 453 (Tex.1996), I would hold that Garza presented no evidence of actual malice apart from the statutory violation itself. Accordingly, I concur in the Court’s judgment but not its evidentiary analysis.
The Court correctly notes that a higher threshold needs to be applied in determining whether there is some clear and convincing evidence to support a verdict. When applying this standard, however, the Court should not weigh conflicting evidence; instead, it should disregard any evidence “that a factfinder reasonably could have disbelieved.” TEX. CONST. art. V, § 6.
The Court’s opinion describes how legal and factual sufficiency review developed in Texas, characterizing the distinction between the two as “perhaps more deeply engrained in Texas law than in that of any other jurisdiction.” To appreciate “[t]he hardiness of this distinction in our jurisprudence,” I believe that it is important to demonstrate the unique role that each type of review plays in practice, and how both types of review perform an important role in avoiding an unjust result. See Clewis v. State, 922 S.W.2d 126 (Tex.Crim.App.1996). A footnote in the Court of Criminal Appeals’ opinion in Clewis colorfully illustrates the difference between legal and factual sufficiency review in the context of a criminal trial. It compares the testimony of a paid informant—who testifies that the defendant committed the crime—against the testimony of forty nuns who testify that the defendant could not have committed the crime and that the informant himself is guilty:
*631 The prosecution’s sole witness, a paid informant, testifies that he saw the defendant commit a crime. Twenty nuns testify that the defendant was with them at the time, far from the scene of the crime. Twenty more nuns testify that they saw the informant [and not the defendant] commit the crime.
Id. Members of the jury may choose whether or not to believe the paid informant, just as they may choose whether or not to believe the nuns; consequently, it is a case where reasonable minds may differ.
If the jury chooses to believe the paid informant and thus convicts the defendant, then the court of appeals may conduct both a legal and a factual sufficiency analysis of that verdict. In a legal sufficiency analysis, the testimony of the forty nuns must be disregarded, however convincing it may be; the informant’s testimony satisfies the prosecution’s burden of production, and a legal sufficiency review will not weigh the countervailing evidence against that testimony. In a factual sufficiency analysis, however, the entire record is considered and the weight of the evidence can make a difference—the nuns’ testimony may so outweigh the informant’s testimony as to make a guilty verdict manifestly unjust.
When the burden of proof is heightened beyond a mere preponderance of the evidence—for example, when a fact must be proved either by clear and convincing evidence or beyond a reasonable doubt—then the standard for legally sufficient evidence will be correspondingly heightened. See Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 257, 106 S.Ct. 2505, 91 L.Ed.2d 202 (1986).
This heightened standard of review, however, does not abrogate the traditional distinction between factual and legal sufficiency review: as part of their jurisdiction over factual matters, the courts of appeals may weigh countervailing evidence in order to reverse a verdict that is manifestly unjust, but this Court may not. Article V, section 6 of the Texas Constitution makes the court of appeals’ determination of the factual insufficiency of the evidence in a case ‘conclusive,’ ” and stating that this Court “has repeatedly acknowledged, as it must, that it has no authority to review that determination except to ascertain that it was lawfully made—that is, that the appeals court considered all the evidence and stated its reasons for judging the evidence insufficient to support a finding of fact.”).
*632 We preserved this distinction in J.F.C. when we held that a legal sufficiency review—even in light of a heightened evidentiary burden—“should disregard all evidence that a reasonable factfinder could have disbelieved or found to have been incredible.” 96 S.W.3d at 266. The Court’s opinion stretches the definition of “undisputed evidence” to include any testimony that is not directly contradicted. The Court then weighs this “undisputed evidence” contrary to the verdict with the evidence supporting the verdict, and finds that the evidence supporting the jury’s finding of malice was outweighed by “a great many other” indications contrary to the malice finding. There are two significant problems with this analysis.
First, the Court’s opinion inaccurately characterizes certain evidence as “undisputed.” It is true that certain facts were uncontested, but the inferences arising from those facts were hotly disputed by the parties at trial. For example, the Court suggests that Garza’s “undisputed ... lengthy record of safety violations” contributed to make the jury’s verdict unreasonable. I agree that it is undisputed that Garza had two prior safety warnings; however, the latest of these warnings occurred more than two years before Garza was disqualified from driving. In the intervening two years, the record reflects that Garza’s safety record improved to the point that his supervisor wrote a letter stating “[y]ou made a commitment to be safe on the job and you did a good job. Keep up the good work.” Garza also introduced evidence that his most recent performance appraisal rated him “satisfactory” for safety. Thus, the question of what inference should be drawn from Garza’s safety record was directly disputed at trial, and the jury could reasonably have inferred that Garza’s record was not as poor as SWBT suggested.
The Court also suggests that SWBT actually aided Garza by conducting a “thorough and impartial” investigation. Again, this characterization was hotly disputed. Certainly, SWBT representatives testified that the investigation was both thorough and impartial. Garza, however, contended that SWBT manipulated the results of the investigation by (1) looking at incidents that occurred over a twenty-year period instead of the typical two-year time period in order to make Garza’s safety record appear worse; (2) including incidents in which Garza was not at fault; and (3) ignoring Garza’s exemplary safety record over the two years prior to the accident. Garza may not have provided evidence that every single aspect of the investigation lacked impartiality, but evidence need not be controverted by introducing directly contradictory evidence; instead, evidence may also be disputed by “circumstances in evidence tending to discredit or impeach” it. Lofton v. Tex. Brine Corp., 777 S.W.2d 384, 388 (Tex.1989) (Hecht, J., dissenting) (“The appeals court calls this evidence undisputed, and it was, in the sense that the Court can point to no directly conflicting evidence. This is not to say that the evidence was not vigorously challenged.... [D]iscrepancies in the witnesses’ testimony allowed the plaintiff to dispute the defendants’ position.”). Some parts of the investigation may indeed have been impartial, and may even *633 have favored Garza. Nevertheless, Garza vigorously challenged the evidence that SWBT’s investigation as a whole was “thorough and impartial,” and the jury could reasonably have inferred that it was not.
Even if selected facts were not directly controverted at trial, the inferences arising from those facts were hotly disputed. As a result, this evidence cannot be fairly characterized as “undisputed,” and it should not have been included in the Court’s legal sufficiency review. See J.F.C., 96 S.W.3d at 266.
In addition to mischaracterizing the evidence that it terms “undisputed,” the Court misstates the role that truly undisputed evidence should play in a legal sufficiency analysis and impermissibly expands the Court’s scope of review. It is not clear from the Court’s opinion that there is any difference between a factual sufficiency review and a legal sufficiency review in cases governed by the clear and convincing standard. But there is a difference, and that difference is whether countervailing evidence should be weighed as part of the court’s review. See J.F.C., 96 S.W.3d at 266. In a factual sufficiency review, on the other hand, the reviewing court may weigh the disputed evidence to see if it is “so significant” that a factfinder could not reasonably have formed a firm belief or conviction. Id. Thus, while both types of review attempt to answer the same question, they consider the evidence differently; the reviewing court takes a less deferential view of the evidence when performing a factual sufficiency review.
This is not to say, however, that undisputed evidence should not be included, or that it has no effect on the scope of review; indeed, undisputed evidence contrary to the verdict must be reviewed in order to determine whether the evidence supporting the verdict has been conclusively negated. See Powers & Ratliff, No Evidence” and “Insufficient Evidence” Points of Error, 38 TEX. L.REV. 361, 364 (1960).
In recent years, commentators have noted that the Court’s scope of review appears to be unclear; in some cases, the Court states that it reviews “all of the evidence” in a legal sufficiency review, while in others it reviews “only the evidence and inferences that tend to support the finding.” Hall, Reeves v. Sanderson Plumbing Prods., Inc., 530 U.S. 133, 150, 120 S.Ct. 2097, 147 L.Ed.2d 105 (2000). I believe that this distinction is an important one. Of course this Court can review all the evidence; indeed, as noted above, it must do so in order to establish whether a fact has been conclusively proved. The Court’s opinion, however, does not merely review the entire record to see whether malice was conclusively negated; instead, the Court gives credence to evidence contrary to the verdict (i.e., Garza’s purportedly poor safety record, SWBT’s purportedly impartial investigation, and various other “undisputed” facts), weighs this evidence against the evidence supporting the verdict, and decides that the evidence contrary to the verdict outweighs the evidence supporting it.
Although the Court’s opinion suggests that it is merely adopting the same scope of review used by federal courts, the United States Supreme Court has made it clear that in a legal sufficiency review, a court should not weigh the evidence contrary to the verdict against the evidence supporting the verdict. Id. (holding that, when reviewing legal sufficiency as part of a motion for judgment as a matter of law, “the court should review all of the evidence in the record” but that it also “must draw all reasonable inferences in favor of the nonmoving party, and it may not make credibility determinations or weigh the evidence ”) (emphasis added); Liberty Lobby, Inc., 477 U.S. at 255, 106 S.Ct. 2505 (“Credibility determinations, the weighing of the evidence, and the drawing of legitimate inferences from the facts are jury functions, not those of a judge.”).
Even in the criminal context, where the burden of proof is at the highest level, the Court of Criminal Appeals will not weigh evidence in a legal sufficiency review. Instead, that Court has held that “[d]irect evidence of ‘X’ fact is always legally sufficient to support a finding of ‘X’ fact.” Clewis, 922 S.W.2d at 132.
Even if the hypothetical were modified so that the undisputed evidence demonstrated that the weather was foggy at the time the informant claimed to witness the crime, that fact should not affect the legal sufficiency analysis. This fact, although undisputed, can make a difference in the appellate court’s review only to the extent it makes the informant’s purported eyewitness testimony less believable. If a reviewing court considers the evidence of foggy weather and relies on it to determine that no reasonable juror could have reached a “firm belief or conviction” that the informant indeed witnessed the crime, then that court would be weighing the evidence in a manner that we said in J.F.C. was impermissible in a legal sufficiency review. J.F.C., 96 S.W.3d at 268 (holding that a legal sufficiency review should “not consider evidence that a factfinder reasonably could have disbelieved.”). Just as a jury may disbelieve disputed evidence, it may also disbelieve certain inferences arising from even undisputed evidence. For example, while the jury could not have disbelieved the undisputed evidence of foggy weather, it could nevertheless have believed the informant’s testimony that he could view the crime even through the fog; thus, the jury could have disregarded the potential inference that the fog blocked the informant’s view. Consequently, the court conducting a legal sufficiency review must also disregard the potential inference that the foggy weather blocked the informant’s view. Of course, the court of appeals *635 would be free to consider the evidence of foggy weather—and any other evidence contrary to the verdict—in conducting a factual sufficiency review.
Just as the hypothetical jury could have disbelieved that the foggy weather blocked the informant’s view, the jury in this case could have disbelieved that SWBT’s actions were truly motivated by Garza’s safety record and that SWBT’s investigation was truly impartial. The jury could similarly have disregarded inferences arising from the other evidence cited by the Court; Garza challenged, at least indirectly, the conclusion that the disciplinary action was consistent with company policy and the conclusion that SWBT was sincere in offering him the opportunity to qualify for other jobs within the company. But the Court does not disregard these inferences, even though the jury could have disbelieved them. Instead, the Court’s opinion in this case holds that, although the evidence included “some indications” of malice, that evidence was outweighed by “a great many other” indications contrary to the malice finding. 164 S.W.3d 607. Because the Court compares conflicting evidence and weighs the competing inferences drawn from that evidence, it exceeds the parameters of a legal sufficiency review and engages instead in a factual sufficiency review. This type of analysis is appropriate only in a factual sufficiency review.
Although I disagree with the manner in which the Court conducts its legal sufficiency review, I agree that the Court’s result in this case is correct. Under the standard announced in Id. at 454.
Garza argues that he presented such evidence in this case. I disagree. Much of the evidence that he characterizes as showing “malice” actually goes to the threshold question of liability for retaliation; for example, Garza argues that the company distorted his safety record in order to fire him, that the company disciplined him but did not discipline Hernandez, that the discipline was not required under SWBT’s policies, and that the company used his safety record as a pretext for firing him. Garza argues that this evidence demonstrates the company’s ill-will toward him and its “specific intent to cause injury” to him. If we allowed evidence of the statutory violation to also serve as evidence of malice, however, then punitive damages would be available in every employment retaliation case. See, e.g., *636 Cazarez, 937 S.W.2d at 454.
Under the logic of Cazarez, evidence of malice must go beyond the level of ill-will or intent that is inherent in the statutory violation itself. Such a requirement is not unusual; the West Virginia Supreme Court has applied a similar requirement for punitive damages that evidence of malice be proven in addition to the statutory violation itself. See Harless v. First Nat’l Bank, 169 W.Va. 673, 289 S.E.2d 692, 703, n. 19 (1982) (allowing punitive damages only upon proof of the employer’s wanton, willful or malicious conduct, and providing examples of such conduct “where the employer circulates false or malicious rumors about the employee before or after the discharge or engages in a concerted action of harassment to induce the employee to quit or actively interferes with the employee’s ability to find other employment”). In this case, Garza presented evidence that SWBT retaliated against him for filing a workers’ compensation claim, but none of that evidence demonstrates a level of ill-will beyond that which is inherent in the act of retaliation itself.
If we look only at the evidence going beyond the violation itself, then we are left with evidence that (1) Gonzalez implied that Garza was a bully, and stated that he was “fed up” with Garza, and (2) Garza was assigned janitorial work, which he felt was demeaning. These facts do not amount to legally sufficient evidence of “ill-will, spite, or a specific intent to cause injury to the employee.” Gonzalez’s statements certainly indicate a level of dissatisfaction with Garza, but those statements alone do not demonstrate ill-will or spite, and Garza does not allege that Gonzalez said anything more that would indicate such ill-will.
Garza cites this Court’s opinion in Id. at 614. In this case, by contrast, there was no evidence that the janitorial assignment was intended to humiliate Garza, that he was ridiculed, or that the purpose of the assignment was otherwise improper. Consequently, I agree with the Court’s conclusion that Garza failed to present legally sufficient proof of malice in this case.
Because the Court impermissibly weighs conflicting evidence and thus performs a factual sufficiency review under the guise of legal sufficiency, I cannot join its opinion. However, because I conclude that Garza presented no clear and convincing evidence of “ill-will, spite, or a specific intent to cause injury,” apart from evidence of the statutory violation itself, I concur in the Court’s judgment reversing the award of punitive damages.
TEX. LAB.CODE § 451.001(1).
96 S.W.3d 256, 264–268 (Tex.2002).
Id. at 266.
58 S.W.3d 214 (Tex.App.Corpus Christi 2001).
Past lost earnings of $129,538; past lost benefits of $7,100; future lost earnings of $174,732; future lost benefits of $122,738; past mental anguish of $300,000; future mental anguish of $300,000.
58 S.W.3d 214 (Tex.App.Corpus Christi 2001).
46 Tex. Sup.Ct. J. 776 (June 12, 2003).
TEX. LAB.CODE § 451.001(1) (emphasis added).
Spencer v. Eagle Star Ins. Co. of Am., 876 S.W.2d 154, 157 (Tex.1994).
Comm. on Pattern Jury Charges, State Bar of Tex., Texas Pattern Jury Charges—Business, Consumer, Insurance, Employment PJC 107.5 (2003) (italics, brackets, and parentheses in original).
See Roark v. Allen, 633 S.W.2d 804, 810 (Tex.1982).”).
TEX. CIV. PRAC. & REM.CODE § 41.002(b)(3)).
TEX. CIV. PRAC. & REM.CODE § 41.001(7) (now defining “malice”—as distinct from actual malice—as a predicate for recovery of exemplary damages to mean “a specific intent by the defendant to cause substantial injury or harm to the claimant”).
Cf. Act of April 11, 1995, 74th Leg., R.S., ch. 19, §§ 1–2, 1995 Tex. Gen. Laws 108, 109–110, 113 (adopting and defining “clear and convincing evidence” standard, now codified as 41.003(a); removing inapplicability of chapter 41 to actions under the workers’ compensation laws, formerly codified as id. § 41.002(b)(3); and providing: “This Act takes effect September 1, 1995, and applies only to a cause of action that accrues on or after that date. A suit filed before the effective date of this Act is governed by the law applicable to the claim that existed immediately before the effective date of this Act, and that law is continued in effect for that purpose.”); Act of May 27, 1995, 74th Leg., R.S., ch. 260, § 9, 1995 Tex. Gen. Laws 2207, 2473 (amending but retaining provision making chapter 41 inapplicable to actions under the workers’ compensation laws); Act of May 8, 1997, 75th Leg., R.S., ch. 165, § 4.01, 1997 Tex. Gen. Laws 325, 328 (adopting amendment contained in Act of April 11, 1995, 74th Leg., R.S., ch. 19, § 1, 1995 Tex. Gen. Laws 108, 109, and not the later amendment contained in Act of May 27, 1995, 74th Leg., R.S., ch. 260, § 9, 1995 Tex. Gen. Laws 2207, 2473).
See Addington v. Texas, 441 U.S. 418, 431–32, 99 S.Ct. 1804, 60 L.Ed.2d 323, (1979)).
Osterberg v. Peca, 12 S.W.3d 31, 55 (Tex.2000) (stating that “it is the court’s charge, not some other unidentified law, that measures the sufficiency of the evidence when the opposing party fails to object to the charge”).
Burroughs Wellcome Co. v. Crye, 907 S.W.2d 497, 499 (Tex.1995).
58 S.W.3d at 227 (“In reviewing the evidence for legal sufficiency, we consider only the evidence and inferences, viewed in their most favorable light, which support the jury’s findings and disregard all evidence and inferences to the contrary. Anything more than a scintilla of evidence that is more than a basis for mere surmise or suspicion is legally sufficient to support the finding. If the evidence provides a rational basis for reasonable minds to differ as to the existence of a vital fact, there is some evidence or more than a scintilla.” (Citations omitted.)).
See generally W. St. John Garwood, Another Look at “No Evidence” and “Insufficient Evidence”, 69 TEX. L.REV. 515 (1991).
Tex. S.J. Res. 16, 22nd Leg, R.S., 1891 Tex. Gen. Laws 197.
See Garwood, supra note 19, 29, at 805 (“Prior to the [1891 constitutional amendments] the distinction seems to have been rarely mentioned.”).
TEX. CONST. art. V, § 6.
See Powers & Ratliff, supra note 19, at 540 (“There is a paucity of evidence about the framers’ precise rationale for giving courts of appeals final authority over questions of fact”).
See 1 GEORGE D. BRADEN, ET AL., THE CONSTITUTION OF THE STATE OF TEXAS: AN ANNOTATED AND COMPARATIVE ANALYSIS 399 (1977) (“The courts of civil appeals were the major innovation of the 1891 reform package. The supreme court was falling so far behind that either the right to appeal had to be severely curtailed or the system had to be radically revised.... The theory in creating the courts of civil appeals apparently was that their decisions would be final in most civil cases. The supreme court’s primary function was to be the resolution of conflicts.”).
See Powers & Ratliff, supra note 19, at 540 (“Why should we give the supreme court jurisdiction over questions of law—thereby giving litigants two levels of appeal but—not give it jurisdiction over questions of fact—thereby limiting litigants to only one level of appeal? The reason, we suggest, is that the supreme court needs jurisdiction over questions of law to insure uniformity throughout the state, uniformity that is not as important on issues of fact that are specific to a particular case.”).
Missouri Pac. Ry. v. Somers, 78 Tex. 439, 14 S.W. 779, 779 (1890).
Cropper v. Caterpillar Tractor Co., 754 S.W.2d 646, 648 (Tex.1988) (stating that the Conclusivity Clause “functions not as a grant of authority to the courts of appeals but as a limitation upon the judicial authority of this court.”).
Muhle v. New York, T. & M. Ry., 86 Tex. 459, 25 S.W. 607, 608 (1894) (stating that “whether or not there was evidence from which the jury might have [made a finding] is a question of law which we are called upon to determine.”).
See, e.g. Garwood, supra note 19, at 805 (“[T]here still exists a modicum of confusion in the profession about the difference between the so-called fact question of ‘insufficient’ evidence and the law question of ‘no’ evidence. Indeed, considering the matter on principle, it is perhaps surprising that we do not have more confusion than we do; and I sometimes wonder if there is much logical basis for the distinction. But ... it is almost ‘hornbook law’ in Texas that the distinction is recognized and has important consequences.”); Calvert, supra note 19, at 359 (“Expressions in points of error such as ‘no evidence,’ ‘insufficient evidence,’ ‘no sufficient evidence,’ ‘no legally sufficient evidence,’ ‘against the great weight of the evidence,’ ‘contrary to the preponderance of the evidence,’ ad infinitum, have definite connotations in the mind of an appellate judge, but, except in a very limited way, they are not, or at least should not be, controlling.”).
See Calvert, supra note 19, at 362–363 (“ ‘No evidence’ points must, and may only, be sustained when the record discloses one of the following situations: (a) 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; (d) the evidence establishes conclusively the opposite of the vital fact.”).
Calvert, supra note 19, at 363.
89 S.W.3d 17 (Tex.2002).
96 S.W.3d 256 (Tex.2002).
Stanley v. Illinois, 405 U.S. 645, 651, 92 S.Ct. 1208, 31 L.Ed.2d 551 (1972) (“The integrity of the family unit has found protection in the Due Process Clause of the Fourteenth Amendment, the Equal Protection Clause of the Fourteenth Amendment, and the Ninth Amendment.” (Citations omitted.)).
In re G.M., 596 S.W.2d 846, 847 (Tex.1980).
Santosky v. Kramer, 455 U.S. 745, 769, 102 S.Ct. 1388, 71 L.Ed.2d 599 (1982).
Act of Act of May 24, 1995, 74th Leg., R.S., ch. 709, § 1, 1995 Tex. Gen. Laws 3745, and Act of May 26, 1995, 74th Leg., R.S., ch. 751, § 65, 1995 Tex. Gen. Laws 3910–3911, both amending TEX. FAM.CODE § 11.15).
96 S.W.3d at 264–265 (footnotes and citation omitted).
443 U.S. 307, 309, 99 S.Ct. 2781, 61 L.Ed.2d 560 (1979).
Id. at 315, 99 S.Ct. 2781 (citation omitted).
Id. at 316–317, 99 S.Ct. 2781.
Id. at 317–318, 99 S.Ct. 2781.
Id. at 320, 99 S.Ct. 2781 (citations omitted).
Id. at 318–319, 99 S.Ct. 2781.
Wilson v. State, 7 S.W.3d 136, 141 (Tex.Crim.App.1999) (“In evaluating legal sufficiency, we must view the evidence in the light most favorable to the verdict and determine whether any rational trier of fact could have found the elements of the offense beyond a reasonable doubt.”).
96 S.W.3d at 265–266.
376 U.S. 254, 285–286, 84 S.Ct. 710, 11 L.Ed.2d 686 (1964).
Bose Corp. v. Consumers Union, 466 U.S. 485, 511, 104 S.Ct. 1949, 80 L.Ed.2d 502 (1984) (stating that appellate review must determine whether the evidence is “sufficient to cross the constitutional threshold that bars the entry of any judgment that is not supported by clear and convincing proof of ‘actual malice.’ ”)
Turner v. KTRK Television, Inc., 38 S.W.3d 103, 119–120 (Tex.2000)
See, e.g., 49 U.S.C. § 28103(a)(1) (harm caused by rail carrier).
See, e.g., TEX.REV.CIV. STAT. art. 179(e), § 6.04(a) (prerequisite to issuing race track license); id. art. 6243e.2(1), § 6(d)(rebuttal of presumption of on-duty firefighter disability).
E.g., Bogart v. Somer, 762 S.W.2d 577, 577 (Tex.1988) (per curiam) (requiring clear and convincing proof to refute the presumption of gift).
443 U.S. at 318–319, 99 S.Ct. 2781 (emphasis in original).
In re C.H., 89 S.W.3d 17, 25 (Tex.2002).
The Clear and Convincing Evidence Standard in Texas: A Critique, 48 BAYLOR L.REV. 391, 393–403 (1996).
Omohundro, 341 S.W.2d at 411 (trespass to try title case).
See Keever, 888 S.W.2d at 793.
Holt Atherton Industries, Inc. v. Heine, 835 S.W.2d 80, 86 (Tex.1992) (“As a general matter, when we sustain a no evidence point of error after a trial on the merits, we render judgment on that point.”); see Calvert, supra note 19, at 362 (“ ‘No evidence’ points of error are inherently and fundamentally points which call for reversal of a trial court’s judgment and rendition of judgment for the appellant.”).
Garza v. Alviar, 395 S.W.2d 821, 823 (Tex.1965) (“If the contention [that the evidence is factually insufficient] is sustained, the finding under attack may be set aside and a new trial ordered.”); see Calvert, supra note 19, at 365 (“ ‘Insufficient evidence’ points of error are points which call for a reversal of the trial court’s judgment and remand of the case for retrial.”).
Muhle v. New York, T. & M. Ry., 86 Tex. 459, 25 S.W. 607 (1894) (stating that “whether or not there was evidence from which the jury might have [made a finding] is a question of law which we are called upon to determine.”).
Harte–Hanks Communications, Inc. v. Connaughton, 491 U.S. 657, 685, 109 S.Ct. 2678, 105 L.Ed.2d 562 (1989).
96 S.W.3d at 266.
532 U.S. 424, 431, 121 S.Ct. 1678, 149 L.Ed.2d 674 (2001).
Post at 636.
Post at 632.
Post at 632.
Post at 632.