Summary
holding that State's burden under habitual offender statute is an issue of sufficiency of evidence and cannot be categorized as trial error
Summary of this case from Lugo v. StateOpinion
No. PD-973-06.
June 18, 2008.
Appeal from the 362nd District Court, Denton County, Robert Bruce McFarling, J.
David M. Wacker, Denton, for Appellant.
Kathleen A. Walsh, Asst. Criminal District Atty., Denton, for State.
OPINION
The court of appeals held that the application of a harm analysis is not appropriate when the State failed to prove the proper sequence of Jordan's two prior convictions for punishment enhancement purposes under the habitual felony-enhancement statute. The court reversed the trial court's judgment and remanded the case for a new punishment trial. We hold that the court of appeals properly rejected the State's contention that a harm analysis is appropriate.
Jordan v. State (Jordan II), No. 2-05-029-CR, 2006 WL 1102587, at *2, 2006 Tex.App. LEXIS 3367, at *5 (Tex.App.-Fort Worth Apr.27, 2006) (not designated for publication) (citing Jordan v. State (Jordan I), No. 2-01-530-CR, 2003 WL 21283184, at *1, 2003 Tex. App. LEXIS 4737, at *2 (Tex.App.Fort Worth June 5, 2003) (not designated for publication)).
Id. 2006 WL 1102587, at *4, 2006 Tex.App. LEXIS 3367, at *12.
Background
Jordan was charged by indictment with the felony offense of possession of 988.76 grams of cocaine with intent to deliver. The indictment also included two enhancement paragraphs alleging that Jordan had two prior sequential felony convictions for possession of a controlled dangerous substance in Oklahoma. The first enhancement paragraph alleged that Jordan had a previous conviction for possession from June 1988, and the second enhancement paragraph alleged that he had a previous conviction for possession from November 1992. Jordan pleaded guilty to the primary offense and entered pleas of not true to the two prior Oklahoma possession convictions before the jury. The jury found Jordan guilty of the primary offense and found both enhancement paragraphs to be true. The jury then sentenced Jordan to life imprisonment, the maximum sentence allowed under the habitual felony-offender statute, Texas Penal Code Section 12.42(d).
See TEX. HEALTH SAFETY CODE ANN. § 481.112(f) (Vernon 2003), last amended by Acts 2001, 77th Leg., ch. 1188, § 2, eff. Sept. 1, 2001.
TEX. PENAL CODE ANN. § 12.42(d) (Vernon 2003).
Jordan appealed, arguing, among other things, that the evidence was insufficient to prove that the November 1992 felony possession conviction was committed after the June 1988 felony possession conviction became final as required under Texas Penal Code Section 12.42(d). The Fort Worth Court of Appeals agreed with Jordan and remanded the case for a new trial on punishment. In remanding the case for a new punishment hearing, the Fort Worth Court of Appeals rejected the State's argument that the error was harmless and held that a harm analysis is inappropriate. Observing that other courts of appeals have applied a harm analysis in this context, the court concluded that this Court has never undertaken a harm analysis when holding that the evidence was insufficient to support an enhancement allegation. The court went on to note that the State incorrectly categorized the error as trial error. According to the court, the State failed "to meet its evidentiary burden of proof," and as a result, the evidence was legally insufficient to support the jury's finding that the possession offense that Jordan was convicted of in November 1992 was committed after his June 1988 possession conviction became final.
Jordan (Jordan I), 2003 WL 21283184, at *2, 2003 Tex.App. LEXIS 4737, at *5.
Id. 2003 WL 21283184, at *2-*3, *5, 2003 Tex.App. LEXIS 4737, at *6-9, 16.
Id. 2003 WL 21283184 at *2-*3, *5, 2003 Tex.App. LEXIS 4737, at *6-9.
Id. 2003 WL 21283184, at *2, 2003 Tex.App. LEXIS 4737, at *7 (citing Williams v. State, 837 S.W.2d 759, 764 (Tex.App.-El Paso 1992, no pet.); Patterson v. State, 723 S.W.2d 308, 316 (Tex.App.-Austin 1987, pet. granted)).
Id. 2003 WL 21283184, at *2-*3, *5, 2003 Tex.App. LEXIS 4737, at *7-8 (citing McCrary v. State, 604 S.W.2d 113, 116 (Tex.Crim.App. 1980); Williams v. State, 596 S.W.2d 903, 904 (Tex.Crim.App. 1980); Hickman v. State, 548 S.W.2d 736, 737 (Tex.Crim.App. 1977); Johnson v. State, 784 S.W.2d 413, 414-15 (Tex.Crim.App. 1990)).
Id. 2003 WL 21283184, at *2, 2003 Tex. App. LEXIS 4737, at *8.
Id.
At his second punishment trial, Jordan again entered pleas of not true to the enhancement paragraphs. The State offered, and the trial judge admitted, Jordan's Oklahoma penitentiary packet that contained information about Jordan's prior Oklahoma convictions into evidence. The information in the packet, however, failed to indicate when Jordan committed the possession offense for which he was convicted in November 1992. Nevertheless, the jury found both enhancement allegations to be true and sentenced Jordan to ninety-nine years' imprisonment under Section 12.42(d).
Id.
Jordan appealed again, claiming, among other things, that the evidence was legally insufficient to prove that Jordan committed the second possession offense after his June 1988 possession conviction was final. The State argued that the jury could infer from the information contained in the Oklahoma penitentiary packet that Jordan committed the second possession offense after his first possession conviction became final. The Fort Worth Court of Appeals disagreed, stating that there was "no evidence in the record reflecting the date on which [Jordan] committed the second enhancement felony[.]" The court then rejected, for a second time, the State's claim that the finding was not harmful. Relying on its opinion in Jordan I, the court held that a harm analysis is not appropriate in this context. The court then reversed the trial court's judgment and remanded the case for a new trial on punishment.
Jordan (Jordan II), 2006 WL 1102587, at *1, 2006 Tex.App. LEXIS 3367, at *2-3.
Id. at *1, 2006 Tex.App. LEXIS 3367, at *3-4.
Id. at *1-*2, 2006 Tex.App. LEXIS 3367, at *4-5.
Id. at *2, 2006 Tex.App. LEXIS 3367, at *5.
Id.
Id. at *4, 2006 Tex.App. LEXIS 3367, at *12.
State's Petition for Discretionary Review
The State petitioned us for review challenging the court of appeals's refusal to apply a harm analysis. We granted review to address the following question:
Did the court of appeals err in refusing to conduct a harm analysis in this case where the evidence supporting an enhancement allegation was found to be insufficient when this court has previously held that only "structural" constitutional errors are categorically immune to harmless error analysis?
The State argues that our decision in Cain v. State outlines the current law regarding the applicability of a harm analysis. In Cain, we held that "[e]xcept for certain federal constitutional errors labeled by the United States Supreme Court as `structural,' no error, whether it relates to jurisdiction, voluntariness of a plea, or any other mandatory requirement, is categorically immune to a harmless error analysis." A "structural" error "affect[s] the framework within which the trial proceeds, rather than simply an error in the trial process itself and "render[s] a trial fundamentally unfair." Accordingly, such errors are not amenable to a harm analysis. Conversely, constitutional trial error, which "occur[s] during the presentation of the case," is subject to a harm analysis because the error can "be quantitatively assessed in the context of other evidence presented in order to determine" whether the error "was harmless beyond a reasonable doubt."
947 S.W.2d 262 (1997).
Id. at 264.
Arizona v. Fulminante, 499 U.S. 279, 310, 111 S.Ct. 1246, 113 L.Ed.2d 302 (1991).
Rose v. Clark, 478 U.S. 570, 577, 106 S.Ct. 3101, 92 L.Ed.2d 460 (1986).
Fulminante, 499 U.S. at 309-10, 111 S.Ct. 1246.
Id. at 307-08, 111 S.Ct. 1246; see Neder v. United States, 527 U.S. 1, 8-15, 119 S.Ct. 1827, 144 L.Ed.2d 35(1999).
The State asserts that the court of appeals erred in refusing to apply a harm analysis. The State argues that the court of appeals erroneously relied on precedent that predates Cain as support for its decision that a harm analysis is not appropriate. The State maintains that when a jury makes an erroneous finding of true to an enhancement paragraph, the error is not "structural." The only right implicated, according to the State, is the defendant's statutory right to have the jury instructed on the correct range of punishment. The error at issue, therefore, is the "jury's consideration of the wrong punishment range." Finally, claiming that the evidentiary insufficiency is harmless, the State argues that the punishment evidence may still support the punishment assessed by the jury.
Jordan argues that the State failed to meet its evidentiary burden of proof as required under Section 12.42(d) of the Texas Penal Code. And because the State failed to meet its burden, the evidence is insufficient and categorically immune from a harm analysis.
Law
Texas Penal Code Section 12.42(d) governs punishment enhancement for habitual felony offenders:
[I]f it is shown on the trial of a felony offense other than a state jail felony punishable under Section 12.35(a) that the defendant has previously been finally convicted of two felony offenses, and the second previous felony conviction is for an offense that occurred subsequent to the first previous conviction having become final, on conviction he shall be punished by imprisonment in the institutional division of the Texas Department of Criminal Justice for life, or for any term of not more than 99 years or less than 25 years.
TEX. PENAL CODE ANN. § 12.42(d). See generally Brooks v. State, 957 S.W.2d 30, 32 (1997).
Explaining how Section 12.42(d) operates, we have said that "[t]he [chronological] sequence of events must be proved as follows: (1) the first conviction becomes final; (2) the offense leading to a later conviction is committed; (3) the later conviction becomes final; (4) the offense for which defendant presently stands accused is committed." The State carries the burden of proving beyond a reasonable doubt that a defendant's second previous felony conviction was committed after the defendant's first previous felony conviction became final. And when "there is no evidence to show that the offenses were committed and became final in the proper sequence, the defendant's sentence may not be enhanced under the State's habitual offender statutes."
Tomlin v. State, 722 S.W.2d 702, 705 (Tex.Crim.App. 1987); see also Valdez v. State, 218 S.W.3d 82, 84 (Tex.Crim.App. 2007).
Id.; Porter v. State, 566 S.W.2d 621, 622 (Tex.Crim.App. 1978) (citing Wiggins v. State, 539 S.W.2d 142, 143 (Tex.Crim.App. 1976); Hickman, 548 S.W.2d at 737).
Tomlin, 722 S.W.2d at 705 (emphasis in original).
As noted by the court of appeals, our cases predating our 1997 decision in Cain demonstrate that we have never considered the application of, or conducted, a harm analysis after concluding that the State failed to present sufficient evidence to prove the proper sequence of the defendant's prior felony convictions for enhancement purposes under Section 12.42(d). Similarly, we have never applied a harm analysis where the State failed to establish that an enhancement conviction was final before the defendant committed the primary offense. In fact, we have explicitly disavowed the application of a harm analysis where the State failed to meet its burden of showing finality. In doing so, we said: "A harmless error analysis should not be undertaken when the State fails to meet its burden of proof."
Hickman, 548 S.W.2d at 736; Porter, 566 S.W.2d at 622; Williams, 596 S.W.2d at 903-04; McCrary v. State, 604 S.W.2d 113, 113-16 (Tex.Crim.App. 1980); Jefferson v. State, 611 S.W.2d 102, 103 (Tex.Crim.App. 1981); Ex parte Augusta, 639 S.W.2d 481, 483-86 (Tex.Crim.App. 1982), rev'd on other grounds.
Jones v. State, 711 S.W.2d 634, 636 (Tex.Crim.App. 1986); Russell v. State, 790 S.W.2d 655, 656 (Tex.Crim.App. 1990); Scott v. State, 553 S.W.2d 361, 364 (Tex.Crim.App. 1977).
Russell, 790 S.W.2d at 656.
W.(citing Jones, 711 S.W.2d at 636).
Analysis
At the sentencing phase, neither party carries the burden of proving what punishment should be assessed within the statutorily prescribed range applicable to a given offense. Generally, the factfinder's decision of what particular sentence to assess is a "normative, discretionary function" that does not depend on the resolution of specific facts. However, when the State seeks to enhance a defendant's sentence for the primary offense by alleging that a defendant has a prior conviction, and the defendant enters a plea of not true, the factfinder must decide whether the State has sustained its burden by entering a finding that the enhancement allegation is either true or not true. In essence, the assessment of punishment involves two types of deliberations when the State has alleged, and the defendant has entered a plea of not true to, a prior conviction used for enhancement purposes. First, the factfinder engages in a deductive, discrete fact-finding process to determine whether the State has proved that the enhancement allegation is true. And second, considering all of the evidence admitted during the guilt and punishment phases, the factfinder engages in a normative process that is uninhibited by any required, specific fact determination to decide what particular punishment to assess within the range prescribed by law. When a reviewing court determines that the State's evidence fails to show that an enhancement allegation is true, the Double Jeopardy Clause does not bar the use of the enhancement conviction during a retrial on punishment.
Murphy v. State, 777 S.W.2d 44, 63 n. 10 (Tex.Crim.App. 1989) (plurality on rehearing).
Barrow v. State, 207 S.W.3d 377, 379-80 (Tex.Crim.App. 2006).
See TEX. PENAL CODE ANN. §§ 12.42-12.43.
See Wilson v. State, 671 S.W.2d 524, 526 (Tex.Crim.App. 1984).
See Murphy, 777 S.W.2d at 63 n. 10; Preston v. State, 457 S.W.2d 279, 281 (Tex.Crim.App. 1970).
Murphy, 777 S.W.2d at 62-63, 63 n. 10.
Monge v. California, 524 U.S. 721, 734, 118 S.Ct. 2246, 141 L.Ed.2d 615 (1998).
Turning to the case before us, we cannot accept the State's and the dissent's argument that the error at issue involves the jury's consideration of the incorrect range of punishment as opposed to the State's failure to present sufficient evidence. Here, in accord with our precedent, the issue framed by Jordan in the court of appeals presented a direct challenge to the legal sufficiency of the evidence to support the jury's finding of true to the second enhancement allegation. In this case, both the State and the dissent are attempting to transform Jordan's claim into something that it is not — trial error. The evidentiary insufficiency, rather than ensuing from action or inaction on the part of the trial judge, resulted from the jury's unsupported determination that the State's evidence supported a finding of true to the second enhancement allegation.
See Burks v. United States, 437 U.S. 1, 15, 98 S.Ct. 2141, 57 L.Ed.2d 1 (1978) (discussing the difference between trial error and evidentiary sufficiency).
Having determined that Jordan's point of error questioned the legal sufficiency of the evidence, we now consider whether the application of a harm analysis is appropriate. Advocating for the application of a harm analysis, the State asserts that the court of appeals, in refusing to conduct such an analysis, equated the State's failure to meet its burden of proof for punishment enhancement under Section 12.42(d) with the State's failure to meet its burden of proof to support a conviction. The State claims that, although a harm analysis can never be used where there is insufficient evidence to support a conviction because the evidence will never justify a conviction, a harm analysis can be applied where the State fails to sustain its burden of proving an enhancement allegation. A harm analysis is applicable in this context because the evidence may still support the jury's punishment assessment.
As recognized by the State, sufficiency of the evidence with regard to a conviction and sufficiency with regard to a prior conviction alleged for enhancement purposes involve very different considerations. However, despite these differences, we conclude that, even if the State's failure to prove the chronological sequence of punishment enhancement allegations as required under Section 12.42(d) is subject to a harm analysis, such a deficiency will never be considered harmless.
TEX.R.APP. P. 44.2.
Here, had the jury properly found that the State failed to prove that the November 1992 possession conviction (the second enhancement allegation) was committed after the June 1988 possession conviction (the first enhancement allegation) became final, the applicable range of imprisonment would have been life or anywhere between fifteen years and ninety-nine years. Notably, the first enhancement paragraph, by itself, had no effect on the term of imprisonment available for the primary offense. But with the jury's unsupported finding of true to the second enhancement paragraph, the lower end of the, punishment range was increased from fifteen years to twenty-five years.
TEX. HEALTH SAFETY CODE ANN. § 481.112(f).
Compare TEX. HEALTH SAFETY CODE ANN. § 481.112(f) with TEX. PENAL CODE ANN. 12.42.
With the dual deliberation process involved in the assessment of punishment in mind, the jury's historical fact determination that the sequential enhancement allegations were true set the range of punishment applicable to the primary offense. It was then within those parameters that the jury was free to exercise its discretionary, normative function. And in exercising its normative judgment about what particular sentence to assess, the jury's unsupported finding of true to the second enhancement allegation could very well have generated an inherent, institutional pressure to impose a greater sentence. Indeed, the legislative purpose behind the habitual felony-enhancement statute is to punish more harshly persons who repeatedly commit crimes. Recognizing that society sanctions increased punishment for habitual offenders, a jury may be inclined to assess a more lengthy sentence in an effort to conform to society's expectations.
State v. Allen, 865 S.W.2d 472, 474 (Tex.Crim.App. 1993).
Cf. Uribe v. State, 688 S.W.2d 534, 538 (Tex.Crim.App. 1985), overruled on other grounds by Cartwright v. State, 833 S.W.2d 134, 135 (Tex.Crim.App. 1992); see also Dix DAWSON. 43 TEXAS PRACTICE AND PROCEDURE §§ 36.53, 36.55 (2nd ed. 2001).
Mullins v. State, 409 S.W.2d 869, 872 (Tex.Crim.App. 1966) (opinion on rehearing).
Id.; see also DIX DAWSON, 43 TEXAS PRACTICE AND PROCEDURE §§ 36.53, 36.55 (2nd ed. 2001).
The problem we confront is that in the absence of discrete, objective facts decided by a jury in assessing punishment, there is no way to quantify what impact the unsupported finding of true had on the jury's normative sentencing function. Accordingly, any attempt to calculate how the jury exercised its normative function in assessing Jordan's sentence would necessarily entail pure speculation. Under these circumstances, the State's failure to meet its burden of proof, even if subjected to a harm analysis, can never be deemed harmless.
Cf. Colella v. State, 915 S.W.2d 834, 845 (Tex.Crim.App. 1995).
Conclusion
Based on the foregoing, we therefore conclude that the court of appeals was correct in refusing to conduct a harm analysis in this case. The judgment of the court of appeals is affirmed, and this case is remanded to the trial court for a new punishment hearing.
KELLER, P.J., filed a dissenting opinion.
The Court holds that automatic reversible error results from the legal insufficiency of the evidence of an enhancement allegation. This is the first time since our decision in Cain that we have held any error to be automatically reversible except those errors so designated by the Supreme Court. For this reason, and others, I respectfully dissent.
Cain v. State, 947 S.W.2d 262 (Tex.Crim.App. 1997).
A conclusion on appeal that the evidence is legally insufficient to support a particular verdict or finding serves to remove that verdict or finding from the case. Period. Any adverse consequences that depend entirely upon the verdict or finding disappear because there is no a longer a verdict or finding to support them. When the evidence is legally insufficient to support a verdict of guilt, for example, the judgment of conviction, which depends entirely upon the verdict of guilt, is overturned. A similar result occurs with respect to deadly weapon findings. The assessment of a deadly weapon finding adversely affects a defendant's eligibility for parole and mandatory supervision. If the evidence is legally insufficient to support a deadly weapon finding, then the finding is removed and those adverse consequences disappear.
TEX. GOV'T. CODE §§ 508.145(d), 508.149(a)(1).
The same is true when the evidence supporting an enhancement allegation is legally insufficient. In that case, the finding of "true" is removed so that the judgment reflects a simple unenhanced finding of guilt. If, without that enhancement, the term of years assessed is no longer within the range of punishment, then a remand for a new punishment hearing would be appropriate.
But if an adverse consequence does not depend entirely upon the verdict or finding for which the evidence is insufficient, then the matter is a question of trial error, not sufficiency of the evidence. If the evidence in this case was legally insufficient to support the enhancement, then the enhancement should not have been submitted to the jury, and as a result, the jury should not have been instructed on the 25-year minimum for repeat offenders. Appellant may wish to phrase the issue as one of evidentiary insufficiency, but the real error upon which he is predicating his request for relief is a jury charge error that flows from the insufficiency.
See Malik v. State, 953 S.W.2d 234, 237 (Tex.Crim.App. 1997) (when the evidence is legally insufficient, "the case should never have been submitted to the jury").
We should not assume that every meritorious legal insufficiency claim will result in relief to a defendant. For example, the relief for insufficiency of the evidence to support a deadly weapon finding is deletion of the finding. But if the offense is a 3(g) offense, the defendant may obtain no practical benefit from the appellate holding of insufficiency of the evidence.
TEX. GOV'T CODE §§ 508.145(d), 508.149(a)(2)-(16).
Drug-free-zone findings produce both trial and post-trial effects, and provide a good example of the difference between a claim of legally insufficient evidence and a claim of jury charge error that arises from the fact of legally insufficient evidence. A drug-free-zone finding affects the range of punishment and also affects eligibility for parole and mandatory supervision. A legal insufficiency claim, if found to be meritorious, would result in removing the drug-free-zone finding, and the adverse parole and mandatory supervision consequences would disappear. If the term of years assessed by the jury exceeded the range of punishment for the unenhanced offense, then a remand for resentencing would also be required. But if the term of years assessed was within the range of punishment for the unenhanced offense, then a remand for resentencing could be based only on a claim of jury charge error — that the jury was given an incorrect instruction on the range of punishment, which may in turn have affected the jury's assessment of punishment.
TEX. GOV'T CODE §§ 508.145(e), 508.149(a)(14).
The Court notes numerous cases in which no harm analysis was conducted after a finding of insufficient evidence to support enhancement allegations. But those cases all far pre-date Cain, in which we held that the only errors categorically immune from a harm analysis are those that have been designated as such by the United States Supreme Court.
This leads me to the harm analysis. Depending upon whether appellant objected, the error in the jury charge should be analyzed for "some harm" or "egregious harm." Regardless of which standard is applied in this case, I would hold that the error is harmless.
Almanza v. State, 686 S.W.2d 157, 171 (Tex.Crim.App. 1984).
The Court concludes that a harm analysis should not be done because it would involve speculation. Though there is always speculation in a harm analysis, it is perhaps true that in most cases like this one such speculation would lead to a finding of harm. But the fact that a particular error may almost always be harmful does not mean that it is immune from a harm analysis.
Cain v. State, 947 S.W.2d at 264.
In this case the harmlessness seems apparent. Even if the jury had been correctly instructed, it would have heard exactly the same evidence. The prior convictions would still have been presented as "convictions" rather than simple bad acts. The only differences are that the jury would have been permitted to assess a lower minimum punishment, and the range of punishment would have been expanded by ten years. But the jury gave appellant 99 years — the highest term of years possible. I could readily conclude that the ten-year difference in the lower end of the punishment range might have mattered if the jury had assessed 25 years or 50 years, or even 75 years. But not 99 years.
I respectfully dissent.