Arnold v. State

TEAGUE, Judge,

dissenting.

I was under the impression that when this Court granted the petitions for discretionary review in the above cases, as well as others, it was the intent of this Court that it would give guidance through bright line rules to the courts of appeals of this State as to when “Rose” error would or would not call for reversible error, so that a certain amount of consistency would exist in this area of the law in all 14 courts of appeals. Sad to say, there is no such guidance in what has been written to date, and both the State and defendants are now going to be at the mercy of how the particular “jurors” of a particular court of appeals decide to vote whether or not the error was harmless under Rule 81(b)(2). Each case will therefore rest on its own merits and the winner will be decided by the equivalent of a coin flip.

In Rose v. State, 752 S.W.2d 529 (Tex.Cr.App.1987), on original submission, this Court declared Art. 37.07, § 4, V.A.C.C.P., unconstitutional as being in violation of the due course of law clause and the separation of powers doctrine of the Texas Constitution. The statute thus became void ab initio. The statute partially informed the jury in the abstract about some aspects of the law of pardons, paroles, and good conduct time. The jury was instructed, inter alia, that in assessing punishment it could consider aspects of parole law and good conduct time in assessing the defendant’s punishment.

Judge Clinton, the author of the lead opinion in Rose , on original submission, pointed out therein that “Whether jurors actually did discuss and consider parole law and good conduct time, and to what extent and effect, can never be properly discovered and adequately determined.” (at 537).

On rehearing, a majority of this Court rejected the "Almanza” charge test error, see Almanza v. State, 686 S.W.2d 157 (Tex.Cr.App.1985), in making the determination whether the unconstitutional instruction was harmless to the defendant, and opted to apply the Rule 81(b)(2) or “no contribution” test.

In the concurring and dissenting opinion that I filed on original submission in Rose , I opined that “before reversible error is shown to exist, it must be established that the instruction affected the fairness, integrity, or public reputation of the defendant’s trial that it caused the trial to be labeled ‘a miscarriage of justice.’ ” (542). In so stating, I overlooked the fact that under Rule 81(b)(2), whatever the error might be, it must not make any contribution to either the guilt or punishment phases of the trial.

Had the authors of the rule stated that the test would be “no substantial contribution”, or words to that effect, then a different analysis would be proper to make. But, they didn’t.

There is only one true way to conclusively resolve the issue, whether “Rose” error in the charge at the punishment stage of the trial is reversible error, and that is to have the jurors who heard the case, and are able to testify, testify whether the parole law instruction either did or did not make any contribution to their verdict on the punishment that was assessed. To require less amounts to little more than flipping a coin, guessing, or speculating on whether the jury actually considered the parole law instruction in assessing the defendant’s punishment.

The majority opinion by Judge Clinton in this cause does not inform the courts of appeals, or members of the bench and bar *328 of this State, when "Rose” error will or will not be reversible error.

In Rose , by relying upon the facts of the case and the defendant’s serious criminal record, I concluded that no miscarriage of justice had occurred, and with or without the instruction the jury’s sentence would have still been life imprisonment. Of course, I was merely acting as a thirteenth juror might have acted in reaching this conclusion.

On this Court’s own motion for rehearing in Rose , the unconstitutionality of the statute was reaffirmed. The first part of the opinion on rehearing was dedicated to the fact that a statute declared void is void ab initio “from its inception and cannot provide a basis for any right or relief.” (553). The Court then discussed why it was applying the Rule 81(b)(2) “harmless error” analysis, and not the Almanza analysis, and concluded that the error was harmless. In so deciding, the majority opinion did not devise or create any all-inclusive or exhaustive list of factors; instead, it invoked several of the factors that are usually used in making the determination whether the error was harmless, namely: the facts of the case; the “ignore the parole law” rebuttal instruction that had previously been given, which instruction this Court presumed the jury followed; and the five prior serious felony convictions the defendant had sustained. Thus, it would appear that what the majority opinion did on rehearing was to give the defendant a bidet shower. Of course, the opinion could have placed other factors in the record to support the jury’s verdict of 30 years’ confinement in the Department of Corrections, to show that the error was harmless, but this would still not answer the “no contribution” issue in Rule 81(b)(2).

If the reader of the opinions filed in this cause is looking for an all inclusive or exhaustive “bright line rule” in making the determination whether “Rose” error is reversible error, see Rose v. State, opinion on rehearing, I believe that he will be terribly, terribly disappointed and and will eventually declare as Justice O’Connor did in the dissenting opinion that she filed in Brooks v. State, 768 S.W.2d 481 (Tex.App.—Houston [1st Dist] 1989): “I do not understand how to apply the harm analysis the Court of Criminal Appeals tells us to use in Rose v. State. . . .” (at 483). I, too, do not find anywhere in today’s collective opinion or the individual opinions that have been filed anything that will give Justice O’Connor, other courts of appeals justices, or even judges on this Court any bright line rule. In fact, the lead opinion in this cause tells us as much: “There is no bright line rule.” There is simply no way that anyone, especially an appellate court judge, can with confidence, and conclusively, without hearing from the jurors, determine that the error in giving the parole law instruction was or was not harmless to the defendant. We are relegated to a guessing game or speculating, acting much like a rational jury might act.

There is no real consistency between the decisions of the courts of appeals which have reversed because of the parole law instruction, and those cases which were affirmed by the courts of appeals. The “luck of the draw,” as to which court of appeals the case was assigned the case, appears to be the deciding factor.

In virtually all of the cases, however, those both affirmed and reversed, the facts were heinous. However, as pointed out in Satterwhite v. Texas, 486 U.S. 249 , 108 S.Ct. 1792 , 100 L.Ed.2d 284 , the question is not whether the legally admitted evidence would support the punishment assessed but whether the State has proved beyond a reasonable doubt that the error made no contribution to the punishment that was assessed. Thus, as applied here, the issue is whether the State has satisfied its burden of proof beyond a reasonable doubt that the charge error made no contribution to the jury’s verdict on punishment. Cf. Harris v. State (Tex.Cr.App.1989).

Although it is true that there is a presumption that the jury followed the trial court’s instructions on the parole law, nevertheless, to say that a part of a jury instruction on a particular unconstitutional statute, or that a formerly approved in *329 struction added to the unconstitutional statute cures the error is ludicrous. Furthermore, the State has the burden of proof, before it can rebut the appellate presumption of harm, to establish, beyond a reasonable doubt, that the jury did not literally follow every single word of the charge, including any so-called “curative” instructions that is found in some of the parole law instructions. It is simply asking too much of any lay juror to tell him to consider the parole law instruction and the effects of good time and parole but not to do so in the case where they had just heard the instruction given to them by the trial judge. To presume that the jury was able to accomplish these mental gymnastics and require the State to rebut the presumption is or should be both shocking and totally not understandable to the average person. What difference, given the fact that the particular instruction is to be read as a whole, can one say that the “weasle words,” see Garay v. State, 755 S.W.2d 956, 957 (Tex.App.—San Antonio 1988), found in the last sentence of the “curative” instruction, have any meaning-one way or the other-to resolving the Rule 81(b)(2) harmless error issue?

Until the passage of Art. 37.07, § 4(a), it was impermissible, and usually reversible error, for the jury to consider the parole law or good conduct time in their deliberations on punishment. The Legislature, by enacting the implicated statute, however, saw fit to partially “enlighten” the jury on the parole law of this State. It failed dismally, and this Court correctly declared the statute unconstitutional. Not undaunted, the Legislature enabled the voters of this State this past November to approve a constitutional amendment, which appears at this time will cause a repeat performance of what occurred in the past in this and the intermediate courts of appeals of this State. See Art. 37.07, § 4, V.A.C.C.P., and Art. I, § 29, Texas Constitution.

In making the determination whether the “Rose” error jury instruction is reversible error, the question is actually what effect or impact the instruction had on the jury. The State has the burden to negate this effect or impact to the extent that it must establish beyond a reasonable doubt that the instruction made no contribution to the punishment that was assessed the defendant by the jury.

The lead opinion in this cause advises us that “the task of a reviewing court is to make ‘an intelligent judgment’ about whether the unconstitutional instruction ‘might have affected [or influenced] deliberations of the jury on punishment.... (Page 299 of Maj. opinion). I must ask: How can a reviewing court make an intelligent decision or judgment call if it refuses to consider the best evidence or fails to use the best source available on the subject— the testimony of the jurors.

The original lead opinion in Rose on original submission ruled: “It is now the rule that jurors may not testify to any matter or statement occurring during the course of deliberations or to the effect of anything upon their minds or emotions influencing them or concerning their mental processes. Tex.R.Cr.Evid.Rule 606(b). To gain any insight into the basis and the rationale of a verdict one must resort to evidence admitted, the charge on punishment, argument of the parties and other relevant indicia of record, if any.” (536). However, the lead opinion recognized and acknowledged the following: “Whether jurors actually did discuss and consider parole law and good conduct time, and to what extent and effect, can never be properly discovered and adequately determined_ (537). I find that the only reason for this conclusion is because a majority of this Court refuses to permit those who can give us the answer to testify.

In the “Opinion on Court’s Own Motion for Rehearing,” in Rose , this Court did not deviate from its holding that the statute was unconstitutional. The lead opinion opted to hold that the harmless error test pursuant to Rule 81(b)(2), and not the “Al-manza” test, would govern whether the error was harmless. Rule 81(b)(2) provides: “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 *330 doubt that the error made no contribution to the conviction or to the punishment.” (Emphasis supplied.) Except when the jury has assessed no more than the minimum punishment, other than through speculation or guess work, then how can it be said that the parole law instruction made no contribution to the punishment assessed?

In holding that the error was harmless, the lead opinion in Rose on rehearing concentrated on the following factors: (1) the maximum punishment that was assessed (life imprisonment); (2) the jury was given a “weasle worded” or “curative” instruction, or “ignore what I just told you” instruction, after it was given the erronous parole law instruction; (3) the facts of the primary offense were found to be heinous and aggravated; and (4) the jury was treated to the fact that the defendant had previously sustained five serious felony convictions. However, this list is not an all exclusive list, and actually bears little relationship to the issue that was before the Court at that time. The defendant’s admissible prior criminal record, as far as assessing his punishment at life imprisonment, caused me to vote that the error was harmless. I now realize that I might have misjudged the issue in light of the wording of Rule 81(b)(2).

My criticism of the lead opinion on original submission in Rose rested on the fact that it refused to hold that it was impermissible for the members of the jury to be interviewed and asked whether the parole law instruction made any contribution to their verdict on punishment and, if so, what that contribution might have been. In this regard, I agree in part with what Presiding Judge McCormick stated: “In essence, the majority finds in the same opinion that the statute violates due course of law because the jury considered the parole laws yet such was harmless because the jury did not consider the parole laws.” And Presiding Judge McCormick did not interview a single juror in this cause before he made this statement.

Thereafter, this Court remanded many, many cases to the courts of appeals to reconsider the issue of harmless error under Rule 81(b)(2) because the statute had been found to be unconstitutional by this Court. This, of course, did not set too well with most of the members of the courts of appeals, because all courts of appeals had upheld the constitutionality of the statute. Most justices, like Justice O’Connor of the Houston First Court of Appeals, exclaimed: “I do not understand how to apply the harm analysis the Court of Criminal Appeals tells us to use in Rose v. State, 752 S.W.2d 529 (Tex.Cr.App.1987) (op. on reh’gl). Nor do I, even after today.

The lead opinion in this cause informs us that “the State has the burden to show beyond a reasonable doubt that the error did not contribute to the verdict on punishment.” (Page 298 of Maj. opinion.) To this extent, the lead opinion has made a valuable contribution to the jurisprudence of this State in that at least in one area there will be consistency among the 14 courts of appeals. Of course, because the State has the burden of proof, it can never conclusively establish beyond a reasonable doubt, without the assistance of the jurors, and except where only the minimum punishment was assessed, that the error was not harmful. Because of the way that Rule 81(b)(2) is worded, and the fact that the jurors cannot testify on the issue, there is actually an irrebutable presumption of harm against the State.

When the cases under consideration were tried, the statute had not been declared unconstitutional; it was thus permissible for the trial judge to so instruct the jury; it was permissible for the attorneys to argue the instruction; and it was permissible for the jurors to consider the instruction, although they could do so only in the abstract. Cf. page 300 of Maj. opinion. “In short, the jury can take the declarations into account in considering punishment, but without regard to what authorities may later do with the defendant.” (Page 300 of slip opinion.) Except in a few rare instances, where the jury’s notes are explicit, and the notes reflect that the jury in assessing punishment is actually using the mathematical formula set out in the charge, I am unaware whether the jury considered the *331 parole law instruction, either all or part, in assessing punishment, other than in the abstract. To say, as the lead opinion does, that the “best liklihood is that a jury will consider the ‘existence,’ [of the parole law instruction], and thereby assess a term of years it believes may ensure the defendant serves more than the minimum term pres-cribd regardless of what prison and parole authorities may later decide,” is to merely guess, speculate, or act as a thirteenth juror might might have done. If the lead opinion’s statement is correct, then Rose should have been reversed, and not affirmed.

The lead opinion in this cause informs us to look to the voir dire examination as one of the factors, without informing us how much time the attorneys had or should have had to complete their voir dire examination; look to the jury arguments, without informing us how much time the attorneys had or should have had to argue their respective cases; look to see whether the jury was given the “curative” instruction, or the “ignore what I just told you” instruction. “A verdict on punishment [however] alone is not a gauge for harm; rather, it serves somewhat as a barometric measure of other pressures we have found are likely to influence the jury in assessing punishment. There is no ‘bright line’ rule.” (Page 307 of Maj. opinion.) So the majority opinion tells us.

In the opinion that I filed on rehearing in Rose , I pointed out that not all cases are going to nicely fit the facts of [Rose v. State ].” Rose, at 557 (Teague, J., concurring opinion).

One factor omitted by the lead opinion in this cause is the fact that the place of trial may be an influencing factor on the punishment assessed, i.e., what might be denominated a “misdemeanor” murder in Houston might be characterized as a “felony” murder in Montague County. Why this is so, we just don’t know. The lead opinion does consider as a factor multiple offenses on trial. Other factors that might be listed are whether the defendant had a prior criminal record, and how serious it might have been; the “good” reputation or character testimony about the defendant; whether the defendant was eligible for probation; the jury argument by the prosecution. Other factors could probably be created to exist.

The lead opinion finally concludes: “Admittedly, we cannot lay down a ‘bright line rule’ to measure the contribution a § 4 instruction will make to every result.” (Page 313 of Maj. opinion.)

The lead opinion then analyzes several opinions by the courts of appeals, namely, Arnold, Gaines, Hooper, and Payne, which are from different courts of appeals.

In Arnold, on remand, the Dallas Court of Appeals, in a double murder case (jointly tried), without deciding whether the State or the defendant had the appellate burden to establish harmless error under Rule 81(b)(2), placed heavy emphasis upon the “curative” instruction found in the defective parole law instruction, and concluded that the error was harmless.

The lead opinion, which spends an inordinately amount of time on the jury arguments of the parties, holds that the court of appeals erred in relying upon the “curative” instruction to cure the defective parole law instruction. However, it concludes: “We have reviewed every germane revelation in this record and are satisfied that they did not influence the jury adversely to appellant in assessing punishment” at thirty years confinement’ in the Department of Corrections.

In Gaines, the San Antonio Court of Appeals, only taking into consideration the facts of the case and the “curative” instruction, concluded that “the jury’s assessment at less then one-third of the possible range of punishment with no fine assessed (punishment was assessed at 25 years’ confinement in the Department of Corrections) does not reflect they were influenced by the parole law instruction. Consequently, we find beyond a reasonable doubt that the error of submission of the unconstitutional parole law provisions to the jury did not contribute to the punishment and was thus harmless error.”

The lead opinion emphasizes once again the heinous facts of the case, the fact that *332 there was no “curative” instruction given, the jury arguments of the parties, the lack of mention of parole in the § 4(a) instruction, and the time of deliberations. It held that “ ‘no part of a § 4 instruction can be reasonbly characterized and fairly regarded as ‘curative.’ Ibid.” This Court assumes that “a jury may just as likely take into account the § 4 instruction and, therefore, examines the record ‘for indicia of factors reasonably conducing to affect minds of average rational jurors in their determination of punishment.’ Ante, at 300-301.” As in Arnold, the lead opinion finds there was harmless error.

In Hooper, where the maximum penitentiary sentence and fine were assessed, the trial court gave the “curative” instruction, as occurred in Rose , although it also instructed the jury not to consider or discuss the effect of the parole law in the course of the jury’s deliberations. Although the Houston First Court of Appeals discussed the facts of the case in Hooper, it did not find them to be as egregious as those found in Rose , and other cases. However, it did find that the prosecutor discussed the parole law (this is understandable, given the fact that such was in the jury charge). The court of appeals did not find harmless error, after considering the following factors: “(1) the maximum punishment was assessed; (2) no additional curative instruction was given; and (3) the argument by the prosecutor invited consideration of parole by the jury.”

The lead opinion of Hooper in this cause affirmed the court of appeals’ reversal of the trial court’s judgment. In addition to setting out the facts of the case, the lead opinion also stressed the fact that the prosecutor “addressed the § 4 instruction” during his jury argument. But, given the fact that this was in the charge that was given, what was wrong with him so arguing the § 4 instruction at that time?

In Payne, the Dallas Court of Appeals affirmed, where the defendant received life imprisonment, enhanced, for murder. Other prior convictions, misdemeanors, were also admitted at the punishment stage of the trial. The court of appeals affirmed, holding that the facts were heinous, that there was a deadly weapon finding, that the defendant had several prior convictions, all of which caused the parole law instruction, though error, to be harmless to the defendant. The jury did send a question to the trial judge, “What is the earliest possible parole on a life sentence,” with the trial judge instructing them to only consider the evidence they had heard and the law in the court’s charge. Thus, they were not told to ignore the “weasle words.”

The lead opinion in this cause holds: “That an appellate court believes facts of an offense are heinous and a prior criminal record shows propensity for violence can not rule out that a jury may have taken into account a § 4 instruction in assessing punishment. Indeed, a jury holding similar views is just as likely to assess maximum punishment in order to compensate for possibility for parole. Thus, such matters are of little consequence in a harmless error analysis.” (Page 298 of Maj. opinion.) The lead opinion reversed the judgment of the court of appeals, holding that “the burden is on the State to demonstrate that such consideration [the jury charge] did NOT contribute to it, and we are unable to determine and to declare beyond a reasonable doubt that the error made no contribution to punishment assessed against appellant.”

In Barehill v. State, 782 S.W.2d 506 (Tex.App.—1st [Houston]), Justice O'Con-nor, in the dissenting opinion that she filed, stated reasons why she was not impressed with the “weasle words” part of the parole law instruction, which were not put in that charge. She pointed out, however, that “[W]e cannot fault appellant for acting as if it [the charge] were constitutional. In discussing the parole law, appellant’s defense cousel was acting according to the Legislature’s expressd intent. The Legislature expected attorneys for the State and the defense to discuss parole in final arguments.” I would put my amen to what Justice O’Connor has said when it comes to a jury question to the trial judge. Given the ambiguous, confusing, and misleading parole law instruction, what conscientous jury would not usually send out some note *333 or question requesting further information? It is truly amazing that more notes have not been sent by juries.

In sum, it is usually impossible to say, except where the facts of the case are bad and the defendant has a rather bad prior criminal conviction record, consisting of many violent crimes, that the statutory parole law instruction did not affect or impact the defendant’s sentence that the jury assessed, unless he received the minimum sentence, which is a rare event in most criminal cases under those circumstances. In fact, in every single case I have had occasion to read from the courts of appeals, the punishment that was assessed most certainly fit the facts of the crime.

Therefore, if the standard to be used in reviewing a “Rose” error instruction is that the entire record is to be examined in a neutral and impartial manner, given the facts of the usual case, there will be no need for this Court to review those decisions by the court of appeals that affirmed the trial court’s judgment, and should defer to those decisions that reverse. However, to do will nullify, rescind, and void Rule 81(b)(2), as far as this Court goes, which appears to be the way that this Court desires to treat that rule.

The “curative” or “weasle word” instruction that has often been invoked by some courts of appeals does not tell the jury not to consider parole and good time laws; only that they not talk about it aloud. If the “curative" or “weasel word” instruction is the “be all”, this Court could have very easily, either in Rose or some other case, ruled that that although the parole law instruction is erroneous, it cleanses itself through the “curative” instruction. Unless this Court is willing to apply some sort of “totality of the circumstances” test to the parole law instruction error, and limits its remarks to the punishment assessed, I predict that reversible error will always be present unless the absolute minimum punishment is assessed by the jury. See Rische v. State, 755 S.W.2d 477, 478 (Tex.Cr.App.1988) (Teague, J., dissenting). In fact, the only way that the State can “win” a Rule 81(b)(2) issue is through some sort of totality of the circumstances test, which the rule does not provide for. What the State and most members of this Court do not desire to accept is that they refuse to apply Rule 81(b)(2) in its literal language that was adopted by this Court. The error here does not concern any error that occurred at the guilt stage of the trial; the error lies in the court’s charge to the jury. The issue in that instance is simply whether the minds of an average, rational juror would have reached the same result had the charge error not occurred. “The issue is really what effect or impact the error had or reasonably may have made upon the jury’s decision in assessing the punisment it did [in light of the error in the charge]. The crucial thing is the impact of the thing done wrong on the minds of the other men and women, not on one’s own in the total setting.” Diaz v. State, 769 S.W.2d 307, 309 (Tex.App.—San Antonio 1989) (Carr, J., dissenting opinion).

I have concluded that in order to give our courts of appeals some much needed guidance, a bright line rule must exist. It must (1) permit the jurors to be interrogated on the issue; (2) hold that any assessment of punishment beyond the minimum is automatic reversible error; or (3) if the first two rules are inapplicable, and the court of appeals has correctly ruled, for example, that the State has the burden of proof to establish beyond a reasonable doubt that the error in the punishment charge is harmless beyond a reasonable doubt, then if the Court is not willing to apply the first two factors, this Court’s policy will always be to defer to the decision of the court of appeals, which either reversed or affirmed the verdict of the jury on punishment.

Although in Payne the appellate court erred in its statement regarding the burden of proof, I find that it would be senseless to remand this cause to that court (note: the opinion remands the cause to the trial court, not the court of appeals), because if the court of appeals should hold that the error was not harmless beyond a reasonable doubt, the State, through its local district attorney, if no one else, will surely petition this Court for review and it will *334 then be necessary for this Court to affirm the judgment of the court of appeals.

In Texas, in assessing punishment, the jury is free to take into consideration the facts of the case. Furthermore, the jury is free to assess punishment anywhere within the range of punishment provided for that offense. Thus, based solely on the facts of the case, and the State did not present any evidence at the punishment stage of the trial, the jury is free to assess a 17 year old’s first degree conviction, who had no prior convictions, and the State did not present any evidence at the punishment stage of the trial, at life imprisonment, just as the jury could assess the minimum punishment for some old “con” who had spent most of his lifetime committing criminal wrongs, who committed the same offense as the 17 year old did.

In Gabriel v. State, 756 S.W.2d 68 (Tex.App.—Houston [1st Dist.] 1988), Justice Cohen perhaps hit the nail on the head when he made the following observations:

Determining whether the parole charge is harmless is one of the most difficult and least objective decisions we are called upon to make. Different courts (and different panels of the same court) are unlikely to achieve consistent results under this highly subjective standard. Moreover, current law makes it impossible for the parties to prove harm or lack of harm. Tex.R.Crim.Evid. 606(b) prohibits a juror from testifying about any statement occurring during deliberation or its effect on his vote or mental processes. The sole source of objective guidance on the harm issue is thus unavailable. We are left with two unpleasant choices: we must either reverse every case where punishment exceeded the minimum or make poorly educated guesses about whether the defendant was harmed_ (at 69-70).

Justice Cohen argued that the escape route from this dilemma was to apply the contemporaneous objection rule to the issue. However, this Court in Rose held that it was unnecessary to object to the parole law charge.

Much like a jury deciding the answers to the special issues in a death penalty case, oftentimes the punishment assessed in a non-capital case depends upon the “luck of the draw” as to what jury hears the case. One jury may be, for whatever reason, sympathetic towards the defendant, and assess a punishment that no rational jury would even think about assessing, whereas another jury may assess a punishment that “shocks the conscience” of any rational individual.

Because of the error in the charge, there is an appellate presumption that the error was harmful. The State has the burden to rebut this presumption of error by establishing beyond a reasonable doubt that the error made no contribution to the punishment verdict. Justice Carr pointed out in Diaz v. State, supra, “it is not an appellate court’s function to determine punishment. Nor is it permissible to speculate upon probable reassessment of punishment ... The question is not was the jury right in their assessment of punishment, regardless of the error or its effect upon the jury. It is rather what effect the error had or reasonably may be taken to have had upon the jury’s decision. The crucial thing is the impact of the thing done wrong on the minds of other men and women, not on one’s own, in the total setting ... The question is whether or not we can say that the error made no contribution beyond a reasonable doubt given the number of years assessed by the jury....”

What causes the problem to become difficult to resolve is two fold: (1) the wording of Rule 81(b)(2), i.e., the case will be reversed “unless the appellate court determines beyond a reasonable doubt that the error made no contribution to the conviction or to the punishment” (emphasis added), and (2) the fact that the jurors are not permitted to inform us of what effect, if any, the parole law instruction might have had on the punishment they assessed. Compare Rule 52(a) and (b) of the Federal Rules of Criminal Procedure. As previously pointed out, it is the State in this instance which has the burden of proof to establish beyond a reasonable doubt that the charge error was harmless beyond a reasonable doubt.

*335 The problem with the “Rose” error charge rests in the fact that to resolve the issue, an appellate court must engage in either speculation or guess work as to what contribution, if any, the error had on the jury that assessed the punishment. Without being able to communicate with the jury, it appears that the error as to punishment is reversible per se. Therefore, given the wording of Rule 81(b)(2) and the fact that the parties may not have the jurors testify as to whether the “Rose” charge contributed to the punishment they assessed, the error is reversible per se.

In conclusion, although this Court did not draft Rule 81(b)(2), it adopted same. Thus, until the rule is changed, this Court must bite the bullet on all of the “Rose” charge cases, and reverse and remand those cases to the trial court.

It is therefore necessary to reverse all of the above cases listed in this consolidated cause and remand them to the trial court.