IN THE COURT OF CRIMINAL APPEALS
OF TEXAS
NO. PD-0551-10
W ILLIAM TH O M AS LEO NARD, Appellant
v.
TH E STATE O F TEXAS
O N DISCRETIONARY REVIEW
FROM TH E ELEVENTH CO URT O F APPEALS
TARRANT CO UNTY
Womack, J., delivered the opinion of the Court, in which Price, Johnson, Cochran, and
Alcala, JJ., joined. Johnson, J., filed a concurring opinion, in which Cochran, J., joined.
Keasler, J., filed a dissenting opinion, in which Keller, P.J., and Hervey, J., joined. Meyers, J.,
did not participate.
Opinion on Rehearing
We withdraw our opinion of March 7, 2012, and issue this opinion in its place.
The appellant’s deferred-adjudication community supervision was revoked based on the
results of polygraph examinations. The Court of Appeals held that this evidence was
inadmissible, and therefore the trial court abused its discretion. We shall affirm the Court of
2
Appeals.
I. Background
A. Trial Court Proceedings
The appellant was indicted for aggravated sexual assault. In November 2004, he pled
guilty to injury to a child and was placed on deferred-adjudication community supervision for
five years. After several amendments, the terms of the appellant’s community supervision
included two provisions relevant to this appeal:
Submit to sex offender treatment and evaluation as directed by the supervision
officer. Attend and participate fully in and successfully complete psychological
counseling, treatment, and aftercare sessions for sex offenders with an individual
or organization as specified by or approved by the Court or the supervision
officer. Pay all costs of evaluation, counseling, treatment and aftercare. Treatment
must be completed within three years of its initiation, with at least one-third of
treatment completed each year.
And:
Must submit to, pay all costs for, and show no deception on any polygraph examination
and other diagnostic test or evaluation as directed by the court or supervision officer.
In October 2008, the State petitioned the trial court to proceed to adjudication, alleging
that the appellant had violated three terms of his community supervision.1 In the first count, the
State alleged that the appellant had violated the “Sex Offender Treatment” condition in two
ways:
1
W hen a defendant is placed on deferred adjudication community supervision, he pleads guilty, but the trial
court defers any adjudication of guilt. If the defendant is found to be in violation of the terms of his community
supervision, the trial court may proceed to adjudicate him guilty in accord with his plea. Thus in such cases the
hearing on whether the defendant violated the terms of community supervision may be called an “adjudication
hearing,” but it is governed by the same rules as a hearing to revoke community supervision and is, in practical
terms, a hearing on whether to revoke the defendant’s deferred adjudication community supervision. See T EX . C O D E
C RIM . P RO C . art. 42.12 § 5(a)-(c).
Throughout this opinion, we will use the terms “adjudication” and “revocation” in such a manner as to
reflect the fact that adjudication hearings are a subset of revocation hearings.
3
a) In violation of this condition, [the appellant] was unsuccessfully discharged
from sex offender treatment, on or about October 21, 2008.
b) In violation of this condition, [the appellant] failed to complete one-third of sex
offender treatment at Michael Strain and Associates within the first year.
In the second count, the State alleged that the appellant had violated the “Polygraph” condition:
In violation of this condition, [the appellant] submitted to polygraph testing and
revealed significant criteria indicative of deception, on or about [April 11, 2008,
and December 12, 2007].
In the third count, the State alleged that the appellant had not promptly paid his assessed fine and
supervision fees.
At the adjudication hearing, the appellant pled not true to the first and third counts, and
the State waived the second count. The State presented two witnesses: Dr. Michael Strain and the
appellant’s probation officer.
Strain was a psychotherapist in whose program the appellant had been enrolled until
Strain discharged him from the program. When the State asked Strain why he had discharged the
appellant, the appellant objected, citing two cases from this Court that had held that, because
polygraphs were unreliable as a matter of law, the results of polygraph examinations were
inadmissible.2 The trial court then asked Strain to answer the question. Strain said that he had
discharged the appellant “because he had failed five polygraphs.” The appellant renewed his
objection; the trial court overruled it, granting the appellant a running objection to all polygraph
testimony. When Strain resumed testifying about polygraph results, the appellant requested a
2
See Romero v. State, 493 S.W .2d 206, 213 (Tex. Cr. App. 1973) (“[T]he results of polygraph tests should
not be received into evidence, over objection, even if there had been a prior agreement or stipulation” by the parties
to admit the results); Tennard v. State, 802 S.W .2d 678, 683 (Tex. Cr. App. 1990) (“The existence and results of a
polygraph examination are inadmissible for all purposes.”).
4
Daubert3 hearing to test the validity of the polygraph results. After a lengthy discussion with
defense counsel, the trial court held in abeyance the motion for a Daubert hearing and never
revisited the matter.
Strain then proceeded to testify that he had discharged the appellant from treatment
because he believed that the appellant was not being truthful with him. The appellant had failed
three polygraphs prior to April 2007, at which point the appellant “made several admissions” and
“cleared” a polygraph.4 After that “cleared” polygraph, the appellant then failed five polygraphs,
three of which “questioned him if he had had sexual contact with children, if he had done a sex
crime or if he had isolated children,” and two of which “asked only if he had had sexual contact
with children or done a sex crime.” Strain said that he believed these polygraph results showed
that the appellant was engaged in “secret keeping.”
Strain said that he had no basis, other than the failed polygraphs, to discharge the
appellant from the program or to believe that the appellant was being dishonest. The appellant
was making satisfactory progress,5 the appellant attended all required sessions, and there was no
reason other than the polygraphs to believe the appellant had committed any offenses. Strain
testified that he did not conduct the polygraphs, but he did not say who did or give any
information about the polygrapher. Strain did not testify as to what specific questions the
3
See generally Daubert v. Merril Dow Pharm., Inc., 509 U.S. 579 (1993).
4
W hen the State asked what the April 2007 admissions were, the appellant objected on the basis that there
was no connection between the 2007 polygraphs and the five polygraphs for which the appellant had been discharged
from the program. The trial court sustained that objection.
5
Strain said that while the length of his program might vary from patient to patient, his normal schedule was
a three-year program, with patients expected to complete one-third of the program each year. Strain said that the
appellant had completed one-third of the program in the first year and was on schedule for the second year of the
program at the time Strain discharged him.
5
appellant was asked or what answers the appellant gave, nor did he provide any details on how
the polygraphs were conducted. Strain said that the appellant was the only patient in the history
of his clinic to have been discharged solely for failing polygraph examinations. According to
Strain, each time prior patients had failed this many polygraphs, there had been other violations
of the conditions of therapy.
The appellant’s community-supervision officer then testified that the appellant had failed
to pay some of his supervision fees on time, but had eventually paid the fees. During cross-
examination, defense counsel seemed to have elicited that the periods in which the appellant was
behind on his fees corresponded to periods of unemployment.
After this testimony, the parties rested. The State waived the third count of the motion to
adjudicate, leaving only the first count.
In his argument, defense counsel observed that Strain had testified that the appellant had
completed one-third of his treatment in the first year, showing that count 1b was not true, and
that the only possible basis for revocation was count 1a. Defense counsel then argued that
Strain’s testimony regarding polygraph results was “per se inadmissible.” While polygraphs
might be acceptable tools for therapists to use during therapy, they could not be “the only reason
for putting a man in the penitentiary ….”
In issuing its ruling, the trial court distinguished “between admitting polygraphs outright
to show that someone did or did not tell the truth and … considering them as the basis of an
expert opinion as to whether someone poses a danger and a risk to this community ….” The trial
court then explained that it was
very concerned that if polygraphs cannot be considered as a tool by the expert who
6
receives them to be able to monitor someone on sex offender probation in this
community, then we’ll be losing a valuable tool to monitor someone’s honesty,
which is the crux of allowing people to continue … in the community on sex
offender probation when they obviously have shown that they are a danger to
children in the past.
The trial court found count 1a to be true and count 1b to be false. It then adjudicated the
appellant guilty of injury to a child and sentenced him to seven years’ confinement.
B. Court of Appeals
1. Arguments of the Parties
The appellant brought five points of error to the Eleventh Court of Appeals. In three
points he argued that the trial court erred by admitting and considering the polygraph testimony,
and without this testimony there was insufficient evidence to show that the appellant had violated
the terms of his community supervision. In a fourth point, the appellant argued that the trial court
erred in denying his request for a Daubert hearing. In his final issue, the appellant argued that the
polygraph results were inadmissible hearsay because Strain did not conduct the polygraph
examinations.
The State made two arguments for admitting the polygraph results. First, under Rule of
Evidence 703, “the facts or data” upon which an expert bases his opinion “need not be
admissible in evidence.” The State argued that the import of Strain’s testimony was not that the
appellant failed polygraphs, but that an expert had formed the opinion that the appellant was not
being truthful, and, on the basis of this opinion, had discharged the appellant from his treatment
program.
In its second argument for admissibility, the State argued that because requiring the
appellant to take polygraphs and “show no deception” was a valid condition of his community
7
supervision, the results of the polygraphs must be admissible or else there would be no way for
the trial court to enforce the condition. “Requiring that Appellant ‘show no deception’ on the
polygraph is just a logical extension of the already permissible condition of submitting to a
polygraph as part of community supervision.” While acknowledging this Court’s longstanding
rule that polygraph results are inadmissible, the State said that this created a “conundrum” for
trial courts that imposed polygraph examinations as a condition of community supervision.
In a third point, the State argued that the appellant’s request for a Daubert hearing was
not preserved for review because the trial court never made a final ruling on the matter.
2. Opinion of the Court
Characterizing the adjudication proceeding as “trial by polygraph,” the Court of Appeals
sustained the appellant’s first three points of error and reversed the trial court.6 The Court of
Appeals rejected the State’s Rule 703 argument by noting that the reason polygraph results are
inadmissible is because “the Court of Criminal Appeals has determined that, as a matter of law,
they are unreliable.”7 Thus, even if the polygraph results were the basis for an expert’s opinion,
the trial court did not have the discretion to consider them. Because the failed polygraph
examinations were the sole basis for Strain’s opinion that the appellant was being dishonest, that
opinion was itself inadmissible evidence from the polygraph examinations.
The Court of Appeals did note, however, that there was a “paradox” regarding
polygraphs, because a trial court had the authority to require polygraph examinations as a
condition of community supervision, but the results of those court-ordered polygraph
6
Leonard v. State, 315 S.W .3d 578, 581 (Tex. App.–Eastland 2010).
7
Id., at 580.
8
examinations may not be shown to the trial court.8 The Court of Appeals seemed to resolve this
paradox at the end of the opinion by suggesting that while the results of polygraphs are
inadmissible, the fact that a defendant failed to take a court-ordered polygraph could be
admissible at a revocation hearing.
II. Discretionary Review
We granted the State’s petition for discretionary review on five grounds. The first four
grounds for review reassert the State’s Rule 703 argument and ask whether the Court of Appeals
erred in holding that our cases stating that polygraph results are inadmissible “for all purposes”
should take precedence over Rules 703 and 705, which, respectively, allow an expert to base his
opinion on inadmissible information and to explain to the trial judge the basis for his opinion
even if the information is otherwise inadmissible. In its fifth ground for review, the State asks
whether the Court of Appeals “err[ed] in holding that polygraph results are always inadmissible
notwithstanding a valid condition of community supervision [requiring that] the Appellant
submit to and show no deception on a polygraph examination ….”
III. Abuse-of-Discretion Review
Before we address the arguments of the parties, we first address a point passed over by
the parties and the Court of Appeals. The arguments both to us and the Court of Appeals deal
with the admissibility of the fact that the appellant failed five polygraphs examinations and
whether an expert may base an opinion solely on the results of polygraph examinations. But the
trial court did not adjudicate the appellant’s guilt based on the failed polygraph examinations.
Although the State initially pled the failed polygraph exams as a basis for adjudication, it
8
Ibid.
9
abandoned that allegation before the hearing began.
The appellant was adjudicated guilty because he failed to complete a sex-offender
treatment program. The only issue for an appellate court in this case is whether the trial court
abused its discretion in proceeding to adjudication.
The basic facts relevant to that matter do not seem to be in dispute. A condition of the
appellant’s community supervision was that he “successfully complete psychological counseling,
treatment, and aftercare sessions for sex offenders with an individual or organization as specified
by or approved by the Court or the supervision officer.” The individual in whose treatment
program the appellant was enrolled testified that the appellant was discharged from the program
without successfully completing it.
This appeal has been merely about the admissibility of evidence regarding why the
appellant was discharged. Is that even relevant to determining whether the trial court abused its
discretion?
During the adjudication hearing, the trial court seemed interested in this question. At one
point, the trial court asked defense counsel:
If we were to proceed hypothetically [and] Mr. Strain got up here and said, “I
discharged him,” and that’s all he can say because it’s based on polygraphs which
are inadmissible, would you feel like that would be the way to proceed in this
case? … [D]ischarge is a violation of his probation, and if it’s proved to me by a
preponderance [of the evidence] that your client was discharged, the Court has the
ability, as you well know, to revoke his probation and send him to prison. Now,
we can either explore the reasons for that or we cannot.
In a revocation proceeding, the trial court has discretion to revoke community
supervision when a preponderance of the evidence supports one of the State’s allegations that the
10
defendant violated a condition of his community supervision.9 Though defendants are not
entitled to community supervision as a matter of right, once a defendant is assessed community
supervision in lieu of other punishment, this conditional liberty “should not be arbitrarily
withdrawn by the court ….”10 On appeal from a trial court’s decision to revoke, therefore,
appellate courts review the record only to ensure that the trial court did not abuse its discretion.
Upon close examination, however, it is not obvious how an abuse-of-discretion standard
applies in this case. The trial court ordered the appellant to “[a]ttend and participate fully in and
successfully complete” a program. The evidence at the adjudication hearing showed that the
appellant did “[a]ttend” and “participate fully” in the program, both of which were within his
power to do. The appellant did not have full control over his ability to “successfully complete”
the program, however; he was discharged because the therapist came to believe that he was being
dishonest. Thus it was the therapist’s discretion that caused the appellant to be in violation of a
term of his community supervision.
As the trial court suggested, under an ordinary abuse-of-discretion review, our analysis
would already be over: Discharge caused the appellant to be in violation, and there was evidence
that the appellant was discharged. Such an analysis on our part, however, would be inadequate.
Revocation involves the loss of liberty and therefore implicates due process.11 “The
central issue to be determined in reviewing a trial court’s exercise of discretion in a [community
9
See Scamardo v. State, 517 S.W .2d 293, 298 (Tex. Cr. App. 1974); Caddell v. State, 605 S.W .2d 275,
277 (Tex. Cr. App. 1980).
10
DeGay v. State, 741 S.W .2d 445, 449 (Tex. Cr. App. 1987).
11
Caddell, 605 S.W .2d, at 277.
11
supervision] revocation case is whether the [defendant] was afforded due process of law.”12 It
would surely offend due process if a defendant were discharged from his therapy program for a
wholly inappropriate reason – such as illegal discrimination or mere caprice – and the bare fact of
that discharge were used as a basis to revoke the defendant’s community supervision. Yet, by an
ordinary abuse-of-discretion review, such a revocation would be sustained.
What has happened here is that the trial court, through a condition of the appellant’s
community supervision, made the appellant’s compliance with the terms of his community
supervision subject to the discretion of a third party. In such a case, to determine whether the trial
court abused its discretion we must also examine the third party’s use of its discretion to ensure
that it was used on a basis that was rational and connected to the purposes of community
supervision.13
With the standard of review explained, the issues in this case become clearer. If the
polygraph results were inadmissible, then the record would not contain a basis for Strain’s
decision to discharge the appellant, and the trial court abused its discretion by adjudicating the
appellant guilty.
IV. Polygraph Evidence
The history of this Court’s dealings with polygraph evidence is long but not very
complicated. For more than sixty years, we have not once wavered from the proposition that the
results of polygraph examinations are inadmissible over proper objection because the tests are
12
DeGay, 741 S.W .2d, at 450.
13
See T EX . C O D E C RIM . P RO C . art. 42.12 §11(a) (“The judge may impose any reasonable condition that is
designed to protect or restore the community, protect or restore the victim, or punish, rehabilitate, or reform the
defendant.”).
12
unreliable.
Our first discussion of the reliability of “‘lie detector’ tests” was a passing note in McKay
v. State.14 In that case, a defendant charged with driving while intoxicated challenged the
admissibility of a “Harger Drunkometer” test that showed his blood alcohol level to be .27. After
we affirmed the conviction, McKay asked for rehearing. As examples of appellate courts
disallowing scientific evidence when scientists were “not generally agreed” regarding the
reliability of the evidence, he cited several cases from other states holding that “lie detector tests”
were inadmissible. Though the issue of lie detectors bore little relation to the case then before us,
we stated that we were “in accord with such holdings, and until such tests can be shown to fairly
show the guilt or innocence of an accused, we see no reason for their introduction in evidence.”15
In Peterson v. State, we weighed in on the admissibility of lie detector results in a more
relevant context.16 Peterson had sought to call as a witness “one D.E. Wheeler, the operator of the
Keeler polygraph, or lie detector machine, for the Texas Department of Public Safety,” in order
to have Wheeler “testify as to the facts observed by him in the giving of such test to the
14
McKay v. State, 235 S.W .2d 173, 176 (Tex. Cr. App. 1951) (op. on motion for reh’g). Prior to McKay,
we had mentioned the use of “lie detector” machines in criminal investigations, but the admissibility of the results
does not seem to have been an issue. See, e.g., Lott v. State, 148 S.W .2d 1102, 1103 (Tex. Cr. App. 1941) (Kendall
County theft suspect taken to San Antonio “for the purpose of having the lie detector used on her.”); McCumber v.
State, 160 S.W .2d 73, 73 (Tex. Cr. App. 1942) (W illacy County man suspected of receiving and concealing stolen
property “was taken to San Antonio, and there subjected to a trial by a ‘lie-detector’ machine, but such machine
failed to work when applied to appellant.”). Cavazos v. State, 172 S.W .2d 348, 349 (Tex. Cr. App. 1943) (Texas
Ranger took W illacy County murder suspect “to San Antonio to a lie detector machine.”); Loya v. State, 172 S.W .2d
508, 509 (Tex. Cr. App. 1943) (W illacy County murder suspect “taken to San Antonio and placed before what was
called a ‘lie detector.’”); Prince v. State, 231 S.W .2d 419, 420 (Tex. Cr. App. 1950) (Panola County “uxoricide”
suspect “was carried to Austin, a distance of 280 miles, where he was subjected to the ‘lie detector’ test at the
Department of Public Safety.”).
15
McKay, 235 S.W .2d, at 176.
16
Peterson v. State, 247 S.W .2d 110 (Tex. Cr. App. 1952).
13
defendant.”17 The trial court had excluded the testimony because “the facts concerning the test
would invade the province of the jury and constituted a mere expression of opinion as to the guilt
or innocence of the defendant.”18 Though we held that the appellant had not properly preserved
the matter for our review, we went on to note “that this Court has not yet authorized the
admission of such evidence on behalf of the State, and, until this is done, it cannot be admitted
on behalf of the defendant.”19
Several cases over the next decade cited back to this apparent dictum in Peterson as a
holding of the Court. In Stockwell v. State, after determining that there had been enough evidence
to require the trial court to hold a hearing on juror misconduct, we noted that the misconduct in
the case – reading a newspaper article that implied the defendant had failed four lie detector tests
– was particularly harmful because lie detector results were inadmissible.20 In Davis v. State, we
held that the trial court erred in not granting the defendant a hearing outside the presence of the
jury regarding the voluntariness of his confession.21 After the trial court had denied Davis’s
motion for a hearing, he had presented evidence of the voluntariness of his confession, including
the fact that he failed two polygraph examinations, to the jury. At the end of the opinion, we
noted that we were not “condemning the use of the polygraph as a means of interrogation,” but
“upon another trial, the testimony which in effect revealed the results of the polygraph tests
17
Id., at 111.
18
Ibid.
19
Ibid.
20
301 S.W .2d 669, 671 (Tex. Cr. App. 1957) (citing Peterson for the proposition that “[e]vidence as to
such tests would have been inadmissible upon the trial on behalf of either the State or the appellant.”)
21
308 S.W .2d 880, 883 (Tex. Cr. App. 1957).
14
should be excluded from the jury.”22 In Placker v. State, the defendant had tried to admit
evidence that he had taken and passed a polygraph examination.23 Citing to Peterson, Stockwell,
and Davis, we reaffirmed that “evidence of the results of a polygraph test is not admissible on
behalf of either the state or the defendant.”24
It was not until 1964, 23 years after the mention of lie detector tests began appearing in
our cases and 12 years after our statement in Peterson, that we had occasion to rule on a case
where polygraph evidence had been admitted over objection.25 Nichols v. State was an appeal
from a conviction for statutory rape.26 After defense counsel “rather extensively” cross-examined
the complaining witness, on redirect the prosecutor asked her, “Without telling me any results,
did you take a lie detector test about this?”27 As defense counsel objected, the witness answered
in the affirmative. The trial court instructed the jury to disregard, but it did not rule on the
defense motion for mistrial.
On appeal, we found “without a single exception, that every court of last resort that has
22
Id., at 883 (citing Peterson).
23
350 S.W .2d 546, 546-47 (Tex. Cr. App. 1961).
24
Id., at 547.
25
There are at least two reported cases from this time period in which the results of polygraph examinations
(or, at least a plain implication of the results of a polygraph examination) were admitted into evidence, but proper
objections were not made. See Gasway v. State, 248 S.W .2d 942, 946 (Tex. Cr. App. 1952) (defense counsel
introduced fact that defendant had failed polygraph examination, and did not object to later testimony regarding
reliability of polygraph); Nixon v. State, 266 S.W .2d 150, 151 (Tex. Cr. App. 1954) (defense witness strongly
implied that defendant had passed a polygraph examination; no objection is mentioned).
The dearth of cases in which polygraph results were admitted, even while numerous cases from this era
mention polygraph examinations being used during police investigations, implies a general understanding that
polygraph results were not admissible.
26
378 S.W .2d 335, 336 (Tex. Cr. App. 1964).
27
Ibid.
15
been called upon to decide the question has ruled that results obtained from the so-called lie
detector test are not admissible as evidence.”28 Even though the trial court did not deny the
motion for a mistrial, we held that it was error not to have granted it because the prejudicial
effect of admitting the results of a polygraph examination29 was such that “no amount of
instructions on the part of the trial court could remove from the jury the harmful effect created in
their minds.”30
Nine years after Nichols, we issued our most important and thorough opinion on
polygraph evidence. The defendant in Romero v. State had entered into a signed agreement with
the State that he would take a polygraph examination and the results, whatever they showed,
would be admitted at trial.31 After he failed the examination, the defendant objected to the
admission of the results, claiming that the State had violated the terms of the agreement by giving
him methadone shortly before the examination and by using an unqualified polygraph examiner.
28
Ibid.
29
“An impression must have been implanted in the minds of the jurors that the result of the lie detector test
had been unfavorable to appellant or else appellant’s counsel would not have objected to the question propounded by
the state.” Id., at 337.
Our conclusion in Nichols that the jury inferred the results of a polygraph examination from the context in
which it was mentioned is at odds with several contemporary cases in which we held that the mention of a polygraph
examination, when the results were not explicitly mentioned, was insufficient to require a reversal or mistrial so long
as the reference was not to the defendant taking or not taking a polygraph examination. See Renesto v. State, 452
S.W .2d 498, 500 (Tex. Cr. App. 1970) (so holding, and citing several other cases where it was so held). In Roper v.
State, 375 S.W .2d 454 (Tex. Cr. App. 1964), a witness gave an unresponsive statement that the defendant took a
polygraph examination during the police investigation, which, combined with his being on trial, would seem to imply
the results. W e held that because the question was unresponsive and we detected no bad faith on the part of the
prosecutor, the trial court’s instruction to the jury to disregard was sufficient. Id., at 456-57.
Perhaps the holding in Nichols can be explained by our note that “the state’s whole case depended entirely
upon the testimony of this one 14 year old witness,” and the result of mentioning the polygraph examination was to
bolster her credibility. Nichols, 378 S.W .2d, at 338.
30
Id., at 338.
31
493 S.W .2d 206, 207 (Tex. Cr. App. 1973).
16
The trial court admitted the polygraph results over this objection. Aside from the polygraph
examiner’s testimony that the defendant “gave deceptive answers” regarding his role in the
offense, the only evidence against the defendant came from accomplice testimony.32
We began our discussion of polygraphs by noting, “No machine or device has perhaps
caused greater controversy among ‘experts,’ judges, lawyers, physicians, psychologists,
government officials, and the public in general than the polygraph or lie-detector machine.”33 In
comparing our cases barring the admission of polygraph results to those of other jurisdictions, we
noted our rule “followed the almost unanimous view of American courts.”34
We then listed several potential problems with polygraph examinations that might distort
the results, such as the test subject’s emotional tension, physiological abnormalities, mental
abnormalities, unresponsiveness, and unobserved muscular movements that produce ambiguities
or misleading indications in the blood-pressure tracing.35 Additionally, we noted several potential
problems that were unrelated to the test subject, such as potential errors by the examiner, the lack
of standardization in procedures, the difficulty in jury evaluation of examiners’ opinions, and the
tendency of the trier of facts to treat polygraph evidence as conclusive.36
Although we noted that our bar on polygraph evidence had received some criticism and
32
Id., at 209.
33
Ibid.
34
Id., at 210.
35
Id., at 210 (citing Henderson v. State, 230 P.2d 495, 501-2 (Okla. Cr. App. 1951); Commonwealth v.
Fatalo, 191 N.E.2d 479 (Mass. 1963)
36
Id., at 210-11 (citing State v. Valdez, 371 P.2d 894, 895 (Ariz. 1962); Samuel L. Highleyman, The
Deceptive Certainty of the “Lie Detector,” 10 H ASTIN GS L.J. 47, 51-61 (1958); Richard A. Sternbach, Lawrence A.
Gustafson, and Ronald L. Collier, Don’t Trust the Lie Detector, 40 H ARVARD B US . R EV . 127 (1962)).
17
that at least two federal district courts (one of which was later overruled) had admitted polygraph
evidence,37 we concluded that these authorities were insufficient to overcome the unreliability of
polygraphs: “While we are aware of efforts to promote the reliability of polygraph tests and of
the claims that they are now more reliable than before, we are convinced at this time that we
should adhere to the general rule of exclusion.”38
But would a stipulation between the parties be sufficient to allow the admission of
otherwise unreliable evidence? After citing to opinions from other jurisdictions, we agreed with
an opinion of the Alaska Supreme Court that regardless of what the parties agreed to, it was our
duty to keep unreliable evidence out of trials: “A stipulation for admission does not increase the
reliability of polygraph results and therefore should not lead to any deviation from the
exclusionary policy.”39 On that basis, we reversed the judgment of the trial court.
Romero was the last time we examined in detail the science and jurisprudence
surrounding polygraph examinations, and, in the 39 years since, we have never departed from its
holding.40
37
Id., at 211 (citing Ferguson, Polygraph v. Outdated Precedent, 35 T EXAS B.J. 531 (1972); United States
v. Ridling, 350 F.Supp. 90 (E.D. Mich. 1972); United States v. Zeiger, 350 F.Supp. 685 (D.D.C. 1972) (rev’d
without op., 472 F.2d 1280 (D.C. Cir. 1972)).
38
Ibid.
39
Id., at 213 (quoting Pulakis v. State, 476 P.2d 474, 479 (Alaska 1970).
40
See, e.g., Robinson v. State, 550 S.W .2d 54, 60-61 (Tex. Cr. App. 1977) (reversing capital murder
conviction because jury was told that the State’s primary witness had passed a polygraph examination); Shiflet v.
State, 732 S.W .2d 622, 630 (Tex. Cr. App. 1985) (where defendant, who was a sheriff’s deputy, claimed admission
to murder was coerced in part by failed polygraph examination, we held polygraph results were not coercive because
“as an experienced police officer, [the defendant] either knew or should have known that the results of the tests were
not admissible evidence against him in a court of law. It has long been the consistent holding of this Court, which we
believe is taught to law enforcement officers throughout this state, that evidence of the results of a polygraph
examination is not admissible evidence on behalf of either the State or the accused.”); Nesbit v. State, 227 S.W .3d
64, 66 n.4 (Tex. Cr. App. 2007) (where the State filed a motion to revoke community supervision based on failing a
polygraph examination, we noted in dicta that “[n]either the results of a polygraph test nor the ‘fact’ of failing a
18
V. Rule 703 and Reliability
In its first four issues, the State argues that, while polygraph results are inadmissible on
their own, an expert who testifies under Rule 702 may, according to Rule 703, base his opinion
on inadmissible evidence and, under Rule 705,41 inform the trial court of the basis for his
opinion. So, in this case, it would have been permissible for Strain to (1) testify regarding his
opinion of the appellant’s honesty while the appellant was in treatment, (2) base that opinion
entirely on polygraph results, and (3) because his opinion was based on the polygraph results, tell
the trial court of the polygraph results. The State does not ask us to revisit our holding in Romero
and, in its brief on rehearing, concedes that the results of polygraph examinations should
continue to be inadmissible “for a number of reasons .…”42
Rule of Evidence 702 allows “a witness qualified as an expert by knowledge, skill,
experience, training, or education” to give opinion testimony based on “scientific, technical, or
other specialized knowledge.”43 Rule 703 specifies the bases on which an expert may base his
testimony, and specifically allows that “[i]f of a type reasonably relied upon by experts in the
particular field in forming opinions or inferences upon the subject, the facts or data [relied upon]
need not be admissible in evidence.”44
polygraph test are admissible in a Texas criminal proceeding.”).
41
Texas Rule of Evidence 705(a) allows a testifying expert to “disclose on direct examination, or be
required to disclose on cross-examination, the underlying facts or data.”
42
The State specifically mentions Rule of Evidence 403, which allows a trial court to exclude relevant
evidence “if its probative value is substantially outweighed by the danger of unfair prejudice, confusion of the issues,
or misleading the jury, or by considerations of undue delay, or needless presentation of cumulative evidence.”
43
T EX . R. E VID . 702.
44
T EX . R. E VID . 703.
19
We believe the State’s interpretation stretches Rule 703 beyond its limits. Rule 703
allows opinions based only upon inadmissible evidence if the inadmissible evidence is of a sort
“reasonably relied upon.” “The use of [the word] ‘reasonably’ rather than ‘customarily’ or
‘regularly’ implies that judicial oversight was intended.”45 While Strain did make the conclusory
statement that those in his field reasonably rely on polygraph results, the sole basis of his opinion
was the results of a test that we have held inadmissible because it is not reliable. “Total reliance
on inadmissible and untrustworthy facts cannot be reasonable. Nor would such an opinion
achieve the minimum level of reliability necessary for admission under Rule 702.”46 Rule 703 is
not a conduit for admitting opinions based on “scientific, technical, or other specialized
knowledge” that would not meet Rule 702’s reliability requirement. If the methodology or data
underlying an expert’s opinion would not survive the scrutiny of a Rule 702 reliability analysis,47
Rule 703 does not render the opinion admissible. Thus Rule 703 did not provide a basis for the
trial court to admit Strain’s testimony, and we overrule the State’s first four grounds for review.
VI. The “Show No Deception” Provision
In its final ground for review, the State argues that the Court of Appeals erred by stating
that polygraph results could not be admitted, even to litigate the provision in the appellant’s
probation condition requiring him to take and “show no deception” on polygraph examinations.
The requirement that the appellant “show no deception” on the polygraph, however, strikes us as
questionable. At the very least, we do not believe it presents a sufficient basis to compel the
45
Steven Goode, Olin Guy W ellborn III, & M. Michael Sharlott, Guide to the Texas Rules of Evidence:
Civil and Criminal, 2 Tex. Prac. §703.3 (1993).
46
Ibid.
47
See generally Kelly v. State, 824 S.W .2d 568 (Tex. Cr. App. 1992); Daubert, 509 U.S. 579.
20
admission of polygraph results.
The Court of Appeals cited to a string of lower court cases where “taking a polygraph
exam was a term or condition of community supervision,”48 and the State directs us to a lower
court opinion that held that a trial court was within its discretion to require polygraph
examinations as a condition of community supervision.49 However, the court in that case, after
stating our rule that polygraph results were inadmissible, noted that the polygraph condition at
issue “was imposed … not to gather possible evidence for use during a trial but solely to serve as
a catalyst for further investigation.”50
One of the suggested terms of community supervision contained in Article 42.12, Section
11 of the Code of Criminal Procedure also points at how testing, as a requirement of community
supervision, should operate. Section 11(a)(14) would require that the defendant “[s]ubmit to
testing for alcohol or controlled substances;” it would not require the defendant to “pass” such
testing. The testing for alcohol and controlled substances would allow the trial court to monitor
compliance with other conditions of community supervision, such as to not consume alcohol or
controlled substances. That is, even though many tests for alcohol and controlled substances are
quite accurate and their results admissible in court, the testing is a means to an end, not an end to
itself.
If polygraph examinations were completely accurate, requiring that the appellant “show
no deception” might make sense. As we discussed in Romero, though, there are several potential
48
Leonard, 315 S.W .3d, at 580 n.1.
49
See Ex parte Renfro, 999 S.W .2d 557, 560 (Tex. App.–Houston [14th Dist.] 1999, pet. ref’d).
50
Id., at 561.
21
sources of error in polygraph examination, few (if any) of which are within the test subject’s
control.51 The “show no deception” requirement, by its terms, would cause the appellant to be in
violation of his community supervision whenever the polygraph results indicated deception,
whether those results were accurate or not. Moreover, it would allow the appellant to comply
with the terms of his community supervision if he were to succeed in defeating the polygraph.
The trial court “may impose any reasonable condition [of community supervision] that is
designed to protect or restore the community, protect or restore the victim, or punish, rehabilitate,
or reform the defendant.”52 Whether requiring the appellant to submit to a polygraph examination
is a reasonable condition is an issue we are not called on to decide today. What we decide is that
any such requirement does not justify admitting legally unreliable evidence.53
VII. Conclusion
Because the trial court adjudicated the appellant guilty based on a third party’s exercise of
discretion, we must look at the basis for that underlying exercise of discretion.
Here, the only basis for the third party’s decision was the results of polygraph
examinations.
We hold that Rule 703 does not allow an expert to present opinion testimony based on
scientifically unreliable facts or data. Additionally, the “show no deception” requirement in the
terms of the appellant’s community supervision does not provide a basis for admitting unreliable
evidence. Therefore, the only evidence supporting revocation was inadmissible. The trial court
51
See Romero, 493 S.W .2d, at 210-11.
52
T EX . C OD E C RIM . P RO C . art. 42.12 §11(a).
53
See Ex parte Doan, 369 S.W .3d 205, at 210 (Tex. Cr. App. 2012) (“The Rules of Evidence … apply fully
in a Texas probation revocation hearing.”).
22
abused its discretion by adjudicating the appellant guilty.
We affirm the decision of the Court of Appeals
Delivered November 21, 2012.
Publish.