Gipson v. State

Annotate this Case

844 S.W.2d 738 (1992)

Billy Eugene GIPSON, Appellant, v. The STATE of Texas, Appellee.

No. 1242-91.

Court of Criminal Appeals of Texas, En Banc.

November 18, 1992.

Rehearing Denied January 20, 1993.

*739 Allan Fishburn, Dallas, for appellant.

John Vance, Dist. Atty., Carolyn Fitz-Gerald Levin, Asst. Dist. Atty., Dallas, and Robert Huttash, State's Atty., Austin, for the State.

Before the court en banc.


MILLER, Judge.

Appellant was found guilty of murder by the trial court, then sentenced by that court to 45 years in the Texas Department of Criminal Justice, Institutional Division. On appeal, appellant challenged the sufficiency of the evidence and complained the trial court erred by admitting testimony of his custodial confession. The court of appeals found the evidence sufficient, but found trial court error in admitting the custodial confession. The Court of Appeals reversed on the basis of that error and remanded for a new trial. Gipson v. State, 819 S.W.2d 890 (Tex.App.-Dallas 1991). Although our reasons for reversing the conviction differ from those opined by the court of appeals, this Court affirms the judgment of the court of appeals.

We granted the State's petition on grounds two and three. In ground two the State complains that the court of appeals erred in reversing the conviction on the basis of the inadmissible confession in disregard of the fact that the case at bar was a trial before the court, and in its brief cites Deason v. State, 786 S.W.2d 711 (Tex. Crim.App.1990). In ground three the State complains that the court of appeals erred in implicitly holding that a harmless error analysis does not apply.

The portion of the record which the court of appeals relied on to declare the appellant's confession inadmissible is as follows:

Q. [DEFENSE ATTORNEY]: While you were taking it [confession], to be fair, of course, you told him [Gibson] it was going to be used against him or could be used against him in this trial? A. [OFFICER]: Yes. Q. Likewise, you told him it would be used in his favor? A. I told him it would be his story he could use in court. Q. For or against him? A. I just said it would be his story that he could tell in court. Q. Wouldn't that imply that it could be used in his favor? A. However you choose to use it. Q. That he could use it on his behalf or the State could use it in their behalf, that's the impression you wanted to get across to him? A. I told him it was a story he would be able to tell in court, yes, sir. Q. That would imply that it could be used in his favor, doesn't it? A. I'm saying he could imply that any way he wanted to. Q. That was your intent, wasn't it, to get across to him that what he was saying here was his story and that he could use it on his behalf in his trial? A. Yes, sir, I told him it would be presented to the Grand Jury and also presented in Court, just like he told it. Q. Either on his behalf or against him? A. Yes, sir.

The court of appeals reasoned that to warn the accused that his confession might be used for him holds out an inducement for making the confession and is an improper *740 warning; therefore, a per se rule of inadmissibility applies if this evidence is uncontroverted. The court of appeals cited Dunn v. State, 721 S.W.2d 325, 341 (Tex. Crim.App.1986). The court of appeals then purported to do a harm analysis of the error, which in its entirety is as follows:

The State contends that if it was error to admit the confession, it was harmless. See Tex.R.App.P. 81(b)(2). The State contends the erroneous admission of this type of confession is subject to the harmless error rule. See Connor v. State, 773 S.W.2d 13, 16 (Tex.Crim.App.1989). More recently, the Court of Criminal Appeals applied a harmless error analysis to the erroneous admission of an induced confession. See Sterling v. State, 800 S.W.2d 513, 520 (Tex.Crim.App.1990). Both Connor and Sterling are distinguishable. The discussion of the harmless error rule in Connor is dicta. The Sterling facts show, in addition to the inadmissible confession, Sterling made a second admissible confession to another officer and a third oral confession to a fellow inmate. These circumstances distinguish both these cases from the matter before us.

We hold that Dunn applies. We sustain Gipson's third point of error.

Gipson, 819 S.W.2d 890 at 894. Although the new Texas Rules of Appellate Procedure had been promulgated two months before the opinion in Dunn was issued, Dunn was written with no harm analysis of the error, and reversed on the basis of the inadmissible confession. Dunn, 721 S.W.2d at 327.

In the petition before us, the State complains that the court of appeals erred in reversing the conviction in disregard of the fact that the case at bar was a trial before the court, and in its brief cites Deason, supra. To better understand the forces at work in the State's ground for review, we need to go back to Arnold v. State, 161 Tex.Cr. 344, 277 S.W.2d 106 (1955), which shaped Tolbert v. State, 743 S.W.2d 631 (Tex.Crim.App.1988), which was the authority relied on in Deason. Arnold says, as does Tolbert and Deason, that:

When a cause is tried before the court and there is nothing to show that the judgment was based upon the inadmissible evidence (such as by findings or conclusions of fact or law) it will be presumed that the trial judge disregarded incompetent evidence admitted at the trial and the judgment will not be reversed on appeal on the ground of the admission of incompetent evidence if sufficient proper evidence was admitted to sustain the judgment.

Arnold, 277 S.W.2d 106 at 107. This was a type of harmless error test, created on the "civil side of the street", instituted before the promulgation of Rule 81(b)(2) of the Rules of Appellate Procedure.[1] While Arnold was authored pre-81(b)(2), Tolbert and Deason were not, and the standard for evaluating harm in a trial before the court should have changed from the Arnold line of cases to an 81(b)(2) analysis with the advent of that rule. Rule 81(b)(2) speaks to us in very plain terms and does not distinguish between trials before juries and trials before the court. It states:

If the appellate record in a criminal case reveals error in the proceedings below, the appellate court shall reverse the judgment under review, unless the appellate court determines beyond a reasonable doubt that the error made no contribution to the conviction or to the punishment.

While Deason employs an 81(b)(2) analysis after applying the Arnold "presumption test", there's no authority to apply such multipronged, conditional testing to an error.

"The confusing majority opinion is, however, understandable when one considers that its authority is Tolbert v. State, which opinion, given the issue that this Court had to resolve, is truly one of this Court's most incomprehensible, illogical, and nonsensical opinions ever handed down. [Cites omitted]. Perhaps the idea of Tolbert is that if this Court follows Tolbert a sufficient number of times, *741 then by sheer number that opinion will become authoritative, rather than simply incomprehensible, illogical, and nonsensical."

Deason, 786 S.W.2d at 720, Teague, J., concurring. The promulgation of Rule 81(b)(2) implicitly voided the presumption test, and we now expressly do so. The conditional presumption test language in the Arnold/Deason line of cases is expressly disavowed. The State's ground for review that the court of appeals erred in reversing the conviction on the basis of the inadmissible confession in disregard of the fact that the case at bar was a trial before the court is overruled.

The State also complains that the court of appeals erred in implicitly holding that a harmless error analysis does not apply. The State is correct; a harmless error analysis does apply and in that respect we sustain the State's ground for review. However, we must reverse based on Rule 81(b)(2).

Rule 81(b)(2) is clear. It says that we must reverse when error is found in the proceedings below unless we are able to analyze the error and its effects and determine beyond a reasonable doubt that error made no contribution to the conviction or punishment. In other words, Rule 81(b)(2) mandates reversal unless strong argument is brought to bear which enables the appellate court to make an intelligent determination that the error and all its tangential effects made no contribution to the conviction or punishment.

By its very nature, Rule 81(b)(2) requires a subjective analysis by which the appellate court must calculate as much as possible the probable impact of the error on the jury, or as in this case, the judge. Each case must be examined on its own merits by the appellate court, and if the record does not expressly show, as it does in this case, that the error complained of clearly contributed to the conviction, then the error must undergo an in-depth analysis for harm to the defendant.

In doing a harm analysis, the reviewing court should examine the source of the error, the nature of the error, whether or to what extent it was emphasized by the State, and its probable collateral implications. Further, the court should consider how much weight a judge or juror would probably place upon the error. In addition, the court must also determine whether declaring the error harmless would encourage the State to repeat it with impunity. Harris v. State, 790 S.W.2d 568 (Tex.Crim. App.1989).

However, in the cause before us, we need not apply the subjective methodology set out in Harris, supra. We do not need to determine whether the error contributed to the conviction because at the pronouncement of guilt, the trial judge specifically states that her judgment was based at least in part on the inadmissible confession. The record reflects the following on page 242 of the trial transcript:

THE COURT: Thank you. Based on the medical evidence, the Defendant's confession, the Defendant's actions, based on testimony from the hotel management workers, based on the accomplice testimony and also Mr. White's testimony, the Court is going to find the Defendant guilty of the offense of murder as now charged in the indictment.

(Emphasis added.) Under Rule 81(b)(2), this Court is constrained to finding the error did contribute to the conviction. Had the trial court not specifically said "Based on ... the Defendant's confession", an argument could have been made by the State that the error was harmless, in part because the confession did not admit to the murder, and the information garnered by the confession was repeated in other admissible testimony. However, in this case, the trial judge specifically said her judgment was based on the confession to some degree. Rule 81(b)(2) is unambiguous and says the error can make no contribution to the conviction. The trial judge says plainly that it did. The defendant has been harmed, the conviction tainted. The State's ground for review is overruled. The judgment of the court of appeals is affirmed.

CLINTON, J., agrees with the majority on the merits, but would remand for an 81(b)(2) harm analysis.

*742 BENAVIDES, Judge, concurring.

I concur in the Court's judgment for reasons different than those expressed in the majority opinion. Particularly, I take exception to the Court's analysis in two main respects. First, I do not believe that the present case calls for a reexamination of the rule recently confirmed in Tolbert v. State, 743 S.W.2d 631 (Tex.Crim.App.1988), and I would thus dispose of the case in conformity with that rule rather than overrule it. Second, I am generally opposed to this Court's assessing the harmfulness of an error for the first time on discretionary review, and accede to the task in this case only because it seems to me that harm is conclusively established on the record.


The rule elaborated by this Court in Tolbert and followed thereafter in cases such as Deason v. State, 786 S.W.2d 711 (Tex. Crim.App.1990), is not really a special version of the harmless error rule and should not be abandoned upon the ground that rule 81(b)(2) of our Rules of Appellate Procedure supplanted it. All harmless error rules have in common the underlying belief that criminal convictions ought not to be reversed for errors which affect only the fairness of procedures, events, or rulings in trials, but do not contribute to their outcomes. In each case, it is the task of a reviewing court to determine whether an acknowledged error actually influenced deliberations of the factfinder. The rule of Tolbert simply assumes that there are no such influences in bench trials. Thus, its underlying rationale is not that verdicts should be protected from trial errors that make no difference, but rather that judges should be permitted without consequence to overrule even well-founded objections to evidence in trials without a jury.

I suspect that Tolbert is wrongheaded, not so much because it superficially resembles a harmless error rule different than the one now prescribed by our Rules, but because it indulges what may turn out to be an unwarranted presumption about the actual intentions of trial judges. Tolbert assumes that judges cure their own errors in the admission of evidence by privately and spontaneously instructing themselves to disregard such evidence before reaching a decision in the case. But a ruling that deliberately receives evidence over objection may actually indicate an opposite intention on the part of the judge.

Nevertheless, the presumption that trial judges disregard inadmissible evidence received over objection has been the law for a good many years, and its recent reaffirmation in Tolbert and Deason makes it a poor candidate for reexamination under any sound application of stare decisis. The credibility of this Court depends in no small measure upon its willingness to approach the business of ultimate review with sensitivity to the actual problems of criminal litigation. Trial judges and lawyers alike have a right to know the rules before they begin trial. Thus, to the fullest extent consistent with judicial restraint and discretion, it is the obligation of this Court to make the law consistent, coherent, and enduring. One of the most powerful devices for accomplishing this end in the common law tradition has always been the doctrine now known as stare decisis. Unlike civil law countries, English-speaking jurisdictions officially accept that decisions of the judiciary are binding in future cases. Among other things, this principle of precedent helps assure that the law will not be implemented or applied capriciously, that it will be largely even-handed and comprehensible.

Consequently, I take exception to overruling Tolbert in the present context because disposition of this case does not depend upon the continued vitality of that holding, and because it does not seem to me that the holding has been supplanted by rule 81(b)(2). Nevertheless, I agree that the trial judge should not be presumed to have disregarded the objectionable evidence in this case because the record clearly reflects that she considered it in fact. In short, the presumption is unequivocally rebutted. This, I think, is altogether consistent with the rule of Tolbert. I would, therefore, wait to reexamine Tolbert, if at *743 all, until the rule is necessarily implicated, and the issue is properly joined and briefed.


When we have found error in the decision of a lower appellate court, it is my belief that the case should then be remanded to that court for further proceedings not inconsistent with our opinion, at least whenever any further factual or legal analysis must be performed before final disposition of the case and such analysis has not already been acceptably performed by the court of appeals or by this Court on discretionary review. Clearly, that includes evaluating the harmfulness under Rule 81(b)(2) of trial errors. Here, on the other hand, this Court addresses the question of harm sua sponte, and finds the error not to be harmless.


Sometimes the discharge of our responsibility on discretionary review will necessarily mean the reversal or affirmance of a conviction in the trial court. But that is rarely so in cases where we review a court of appeals only for legal mistakes in its assessment of trial error. Whether we ultimately reverse or affirm its decision in this respect, it is usually our duty to remand for a harm analysis rather than to perform that analysis ourselves.

The issue is far more serious than one of judicial economy. The law sometimes requires awkward or undesirable procedures which may not, for that reason alone, be ignored by those agencies charged with enforcing or applying them. In some quarters, for example, the existence of two high courts in Texas is thought uneconomical, a circumstance which it has been suggested should result in the abolition of this Court. Yet surely such objection would not excuse exercise by the Texas Supreme Court of criminal jurisdiction while the Court of Criminal Appeals still exists.

Similarly, the Constitution of Texas provides that "the Court of Criminal Appeals may, on its own motion, review a decision of a Court of Appeals in a criminal case as provided by law." Tex. Const, art. 5, § 5. The applicable law provides that we may review such a decision "with or without a petition for discretionary review being filed by one of the parties[.]" Art. 4.04, § 2, V.A.C.C.P. This is the whole of our discretionary review jurisdiction, and it limits our authority to the review of "decision[s]" made in the lower courts.

Little has been done to elaborate the meaning of this puzzling word, even though it is plainly at the heart of our constitutional jurisdiction. Certainly, it is rationally susceptible of at least two readings. For example, one might take "decision" to mean "disposition." In that sense, our authority would extend to the review of all affirmances and reversals of criminal convictions, and would reach any rationale for doing so, even if not discussed by the lower appellate court. On the other hand, "decision" might instead be taken to mean merely the conclusion or group of conclusions reached by the lower court in resolving one or more points of error. In such case, our power of review would be limited to those questions of law actually passed upon by the court of appeals.

I am not prepared to express an opinion in the present context whether this Court is utterly without jurisdiction in noncapital cases to pass upon the harmlessness of errors before a court of appeals has done so. Suffice it to say for the present that I would not regard such a claim as frivolous. But I must insist that a due regard for the authority of our intermediate appellate courts at least militates strongly in favor of restraint in such matters. Deciding the question of harm in most cases where the Court of Appeals has not yet done so produces a result which is at least potentially different than the result which would have been produced had the Court of Appeals instead resolved the question itself.

After an entire decade of multi-tiered appellate review in criminal cases, it is nothing short of astonishing that the court of last resort in such matters can still not come to grips with its own place in the "new" system. During the past ten years, we have remanded to the Court of Appeals for a harm analysis in nearly ten per cent *744 of all cases on discretionary review decided by published opinion. Of course, there have also been a number of cases, far fewer to be sure, in which the harm analysis, as here, was done for the first time in this Court. What is so embarrassing is not that we have sometimes handled cases differently, but that we have done so capriciously. The long and the short of it is that similarly situated litigants are being treated inconsistently by this Court without any principled basis or rational explanation. Surely it is not too much to expect that a tribunal with the stature of this one develop some protocol for the conduct of its business, if for no other reason to avoid looking foolish or arbitrary.

There is no cogent reason under the existing system to expect that an ultimate decision about the viability of any given criminal conviction will be made by this Court on discretionary review, since there is nothing in the law to suggest that we have any such authority. Rather, it is the various courts of appeals that are charged with providing the appellate review to which the laws of this State entitle persons who have been convicted of crimes. Our job is to scrutinize those decisions sparingly for deficiencies of analysis in order to see whether the criminal jurisprudence of Texas is working as it should. And the criteria for success of the system is not whether we would have done the same as an appellate court any more than it is whether we would have done the same as a jury. When the person or tribunal who is charged under the law with performing a task has performed it as the law prescribes, it is our duty to leave it undisturbed, even if we would not have performed it the same way. And for like reason, it is our duty to forbear from preemptively executing the task assigned to another by law. It is just not possible to review a matter which has not yet been viewed in the first place.


In the instant cause, a majority of this Court fails to notice that the analysis for harm it performs should ordinarily be done in the first instance by the Court of Appeals. Accordingly, I object to the implication of its opinion that a decision about the harmfulness of error may be made routinely by this Court for the first time on discretionary review. Nevertheless, because it seems to me that harm is conclusively established on the record by the trial judge's admission that objectionable evidence was used to convict Appellant, I am willing to concede that a finding of harm necessarily follows in this case from a finding of error. Under the circumstances presented, there is no potential that the result reached in the Court of Appeals, were it to review the matter, could possibly be different than the result reached by the majority here. For this reason I concur in the judgment of the Court.

McCORMICK, P.J., and CAMPBELL and WHITE, JJ., join.


[1] See Deason, supra, Teague, J., concurring.