Hood v. State

828 S.W.2d 87, 1992 Tex. App. LEXIS 51, 1992 WL 1665

• Court: Court of Appeals of Texas
• Decided: January 8, 1992
• Docket: 3-90-211-CR
• Status: Published

Opinions

ONION, Justice (Retired).

Appellant appeals his conviction for burglary of a building. After the jury found the appellant guilty, the trial court assessed his punishment at twenty years’ imprisonment and a fine of ten thousand dollars.

Appellant advances three points of error. Initially, he contends that the trial court erred in admitting into evidence, over timely objection, the written out-of-court confession of a nontestifying co-defendant which implicated the appellant. Second, appellant urges that the evidence is insufficient to sustain the conviction. Last, appellant argues that the trial court erred in permitting the jurors to separate at the guilt stage of the trial after the court’s charge had been read. We sustain the first and third points of error, reverse the judgment of conviction and remand the cause to the trial court.

The evidence reflects that a building at the Stonehill Shooting Range in Marble Palls was. burglarized on or about July 7, 1989. Cash and guns valued at $29,975.00 were taken by persons unknown. On July 12, 1989, Deputy Sheriff Conrad Nagel of Burnet County, acting with other officers, recovered some of the stolen weapons of the value of $10,000.00 at a business in Houston, known as The Bar Metals. The owner of the business, John Etheredge, had purchased the weapons for $1,000.00.

During the re-direct examination of deputy Nagel, the prosecutor offered into evidence the written statement of the nontestifying co-defendant, Fred St. Ores, who had been previously convicted of the same offense. The statement given on August 16, 1989, to deputy Nagel directly implicated the appellant. The statement was admitted into evidence over a timely objection that it was hearsay.¹ See Tex.R.Crim.Evid. 801(d) (Pamph.1991).

It is the general rule that an act or statement of a co-defendant that is made after the completion of the conspiracy is not admissible against the accused. Ward v. State, 657 S.W.2d 133, 137 (Tex.Crim.App.1983); Delgado v. State, 544 S.W.2d 929, 931 (Tex.Crim.App.1977); Chapman v. State, 470 S.W.2d 656, 662 (Tex.Crim.App.1971); Eminger v. State, 718 S.W.2d 406, 408 (Tex.App.1986, no pet.); see also United States v. Caro, 569 F.2d 411 (5th Cir.1978). This rule is not to be confused with the co-conspirator exception to the hearsay rule. Tex.R.Crim.Evid. 801(e)(2)(E) (Pamph.1991); Ward, 657 S.W.2d at 136; see also Figueroa v. State, 740 S.W.2d 537, 539 (Tex.App.1987, no pet.); Sherwood v. State, 732 S.W.2d 787, 796 (Tex.App.1987, no pet.); Bowman v. State, 704 S.W.2d 463, 465 (Tex.App.1986, pet. ref’d), disapproved in part on other grounds, 730 S.W.2d 754 (Tex.Crim.App.1987).

The statement of co-defendant St. Ores was made to a peace officer after the offense and the disposition of some of the stolen goods and in derogation of the conspiracy. We conclude that the admission of the co-defendant’s out-of-court statement to the peace officer implicating the appellant was error.

The State takes the position that the co-defendant’s statement was admissible under the rule of optional completeness. Tex.R.Crim.Evid. 107 (Pamph.1991).² Evidence which is used to fully explain a matter opened up by the other party need not be ordinarily admissible. Parr v. State, 557 S.W.2d 99, 102 (Tex.Crim.App.1977). The State claims that the appellant opened the door and it waltzed through the breach. The appellant did not, however, offer a part of St. Ores’ statement or read from it in the jury’s presence. The State notes several parts of the record where it believes the door was cracked, conveying the appropriate invitation. In each instance save one, the trial court promptly sustained the State’s objection to appellant’s questions concerning the St. Ores statement. The questions were not answered. No invitation was extended by virtue of these circumstances. The one instance where the State did not object occurred during the cross-examination of Deputy Nagel. Appellant elicited from Nagel that, based on his investigation, he had determined that Etheredge (purchaser of the stolen guns) knew St. Ores and that St. Ores had previously sold scrap metal to Etheredge. When Nagel stated that he did not know whether Etheredge knew the appellant, he was requested to look at his reports, the “Etheredge’s statement,” and the “St. Ores’ statement.” Nagel reaffirmed his earlier answer. Appellant then elicited from the witness that neither statement indicated that Etheredge knew appellant. Nagel was then asked if “anything in either statement” indicated that Etheredge knew St. Ores. Answering “Yes,” Nagel confirmed his earlier testimony. At this point the interrogation about the St. Ores statement ceased.

On re-direct examination the State introduced into evidence the entire extra-judicial confession of St. Ores given to the peace officers. The written statement not only implicated the appellant, it branded him the instigator of the offense. The statement did not refer to Etheredge by name and did not touch on the subject of whether Ether-edge knew the appellant. If it can be argued that appellant opened the door, can it be said that the necessity of completeness justified the introduction of the co-defendant’s statement? One of the limitations on the scope of the completeness opening is that only parts or items germane to the part or item offered “on the same subject” become admissible. See Jernigan v. State, 589 S.W.2d 681, 694-95 (Tex.Crim.App.1979); Roman v. State, 503 S.W.2d 252 (Tex.Crim.App.1974); 33 Steven Goode, Olin Guy Wellborn III and M. Michael Sharlot, Texas Rules of Evidence: Civil and Criminal § 107.1 at 30, 31 (Texas Practice 1988).³ We conclude that the appellant did not open the door by his interrogation so as to permit the admission of the co-defendant’s statement. Further, even if the door was opened, nothing in the statement was germane to “the same subject.” Cf. Pratt v. State, 748 S.W.2d 483 (Tex.App.1988, no pet.). The trial court erred in admitting the statement in question.

Having determined that the trial court erred in admitting the co-defendant’s confession, we must now decide if the introduction was harmful to the appellant. See Tex.R.App.P.Ann. 81(b)(2) (Pamph.1991); Deason v. State, 786 S.W.2d 711, 716 (Tex.Crim.App.1990). Hearsay evidence improperly admitted will constitute reversible error only if there is a reasonable possibility that the evidence is prejudicial to the defendant or may have contributed to the conviction. Goodman v. State, 701 S.W.2d 850, 863 (Tex.Crim.App.1985); Pratt, 748 S.W.2d at 487. The co-defendant’s statement is the only evidence directly linking the appellant to the burglary offense. The prosecutor offered the statement for the truth of the matter asserted and in his argument urged the jury to consider the statement as the truth.

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 doubt that the error made no contri-button to the conviction or to the punishment.

Tex.R.App.P. 81(b)(2) (Pamph.1991).

Noting that the main concern of the reviewing court should be the integrity of the criminal justice process and a defendant’s right to a fair trial, and recognizing the subjective nature of Rule 81(b)(2), the Court of Criminal Appeals has set out a basic framework for the appellate courts to consider in applying the rule. See Harris v. State, 790 S.W.2d 568, 587-88 (Tex.Crim.App.1989) (involving the admission of extraneous offenses in a murder prosecution).⁴ See also Brown v. State, 798 S.W.2d 284, 285 (Tex.Crim.App.1990). Keeping in mind the factors set out in Harris and the framework in which they are to be applied, we focus on whether the error might possibly have prejudiced the jurors’ decision-making, and not on the weight of the other evidence of guilt. Under the circumstances described, we cannot determine beyond a reasonable doubt that the error in admitting into evidence the co-defendant’s statement did not contribute to the conviction. Appellant's first point of error is sustained.

In his second point of error, appellant contends that the trial court erred in overruling his motion for an instructed verdict. Appellant offered evidence after the motion was overruled, see Kuykendall v. State, 609 S.W.2d 791, 794 (Tex.Crim.App.1980), but we shall nevertheless consider the contention as a sufficiency-of-the-evidence question. See Madden v. State, 799 S.W.2d 683, 686 n. 3 (Tex.Crim.App.1990); Miranda v. State, 813 S.W.2d 724, 728 (Tex.App.1991, pet. ref’d).

The standard for reviewing the sufficiency of the evidence is whether, viewing the evidence in the light most favorable to the prosecution, any rational trier could have found all the essential elements of the offense charged beyond a reasonable doubt. Jackson v. Virginia, 443 U.S. 307, 319, n. 12, 99 S.Ct. 2781, 2789, n. 12, 61 L.Ed.2d 560 (1979); Valdez v. State, 776 S.W.2d 162, 165 (Tex.Crim.App.1989), cert. denied, 495 U.S. 963, 110 S.Ct. 2575, 109 L.Ed.2d 757 (1990); Butler v. State, 769 S.W.2d 234, 239 (Tex.Crim.App.1989). The standard for review is the same in both direct and circumstantial evidence cases. Herndon v. State, 787 S.W.2d 408, 409 (Tex.Crim.App.1990); Chambers v. State, 711 S.W.2d 240, 244-45 (Tex.Crim.App.1986). See now Geesa v. State, 820 S.W.2d 154 (Tex.Crim.App., 1991).

Appellant argues that by removing the co-defendant’s statement from consideration, the evidence is clearly insufficient to support the conviction. In assessing the sufficiency of the evidence to support a conviction, a reviewing court must consider all the evidence the jury was permitted to hear, whether rightly or wrongly admitted. Deason, 786 S.W.2d at 716; Thomas v. State, 753 S.W.2d 688, 695 (Tex.Crim.App.1988); Beltran v. State, 728 S.W.2d 382, 389 (Tex.Crim.App.1987); Porier v. State, 662 S.W.2d 602, 606 (Tex.Crim.App.1984). In considering all of the evidence, we conclude that, viewing the evidence in the light most favorable to the verdict, any rational trier of fact could have found beyond a reasonable doubt all the essential elements of the offense charged. Appellant’s second point of error is overruled.

In his third point of error, appellant contends that the trial court erred in permitting the jurors to separate after the charge had been read to the jury at the guilt stage of the trial. See Tex.Code Crim.Proc.Ann. art. 35.23 (Supp.1991). The State argues that the record fails to show a jury separation and that the appellant has failed to sustain his burden. See McDonald v. State, 597 S.W.2d 365, 367 (Tex.Crim.App.), cert. denied, 449 U.S. 1010, 101 S.Ct. 564, 66 L.Ed.2d 467 (1980); Skillern v. State, 559 S.W.2d 828 (Tex.Crim.App.1977). The record shows that during the jury’s deliberations at the guilt stage of the trial the jury sent a note to the trial court informing the court that the jury stood ten to two and was “hung.” In open court the foreman informed the trial court that he did not believe the jury could reach a verdict. The record then reflects:

The Court: ... I’m going to ask that you deliberate more, however, I’m going to recess you until 9:00 in the morning. Be back in the jury room at 9:00.

Leave the charge here and the exhibits, Mr. Bailiff.

Now, it’s more important than ever that I instruct you that you’re not to discuss this case with anyone or allow anyone to discuss it with you.

You’re not even to discuss it among yourselves unless you’re all together back here in the jury room.

You may leave.

Just heed the instructions of the court. Leave the buttons there on the table in the jury room.

Be back in the jury room at 9:00 in the morning.

At this point appellant’s counsel stated he had a “motion to make before the jury steps out,” and asked that the motion be made out of the jury’s presence. The record then reflects:

[Whereupon the jury was retired from the courtroom, after which the following proceedings were had]:

The Court: All right.

Mr. Shell: [defense counsel] Your Honor our motion to the Court is that you not allow the jury to separate and that they continue to deliberate today.

We object to the jurors separating.

The Court: Your objection will be overruled.

We conclude that the record shows that the jury separated.

We have not found an interpretation of article 35.23 since its amendment in 1989. The 1989 amendment reflecting the statutory additions and deletions reads in pertinent part:

Art. 35.23. JURORS MAY SEPARATE. The court may adjourn veniremen to any day of the term. When jurors have been sworn in a felony case, the court may, at its discretion, permit the jurors to separate until the court has given its charge to the jury. The court on its own motion may and on the motion of either party shall, after having given its charge to the jury, order that the jury not be allowed to separate, after which the jury shall be kept together, and not permitted to separate except to the extent of housing female jurors separate and apart from male jurors, until a verdict has been rendered or the jury finally discharged [, unless by permission of the courL-with the consent of-each party]. Any person who makes known to the jury which party made the motion not to allow separation of the jury [did not consent to-separation] shall be punished for contempt of court. If such jurors are kept overnight, facilities shall be provided for female jurors separate and apart from the facilities provided for male jurors. In misdemeanor cases the court may, at its discretion, permit the jurors to separate at any time before the verdict. In any case in which the jury is permitted to separate, the court shall first give the jurors proper instructions with regard to their conduct as jurors when so separated.

See 1989 Tex.Gen.Laws, ch. 825, § 1 at 3783 (S.B.351).

Under the former version of the statute the jury was not permitted to separate after the reading of the court’s charge until a verdict was rendered or the jury was discharged unless the defendant consented to the separation. See McDonald, 597 S.W.2d at 367; Rhynes v. State, 479 S.W.2d 70, 73 (Tex.Crim.App.1972). A facial interpretation of the current statute, which is not artfully drawn, shows that a jury may separate after the court’s charge is read unless a motion is made that the jury be sequestered. The court may act on its own motion and shall act to prevent separation if either party makes a motion. This interpretation is consistent with the legislative history of Senate Bill 351. After its introduction, the bill made its journey through the legislative process without an amendment. The bill analyses for the Senate Committee on Criminal Justice and the House Committee on Criminal Jurisprudence point out that Senate Bill 351 would require the court to permit a jury to separate after the charge has been given to the jury unless a motion has been made requesting that the jury not be allowed to separate.

In the instant case the appellant made a timely motion requesting the jury be sequestered and objecting to the action of the court in permitting the jury to separate. The objection was overruled. The provisions of article 35.23 and its forerunners have long been held to be mandatory. Goodall v. State, 501 S.W.2d 342, 343 (Tex.Crim.App.1973); Pearson v. State, 145 Tex.Crim. 87, 165 S.W.2d 725, 727 (Tex.Crim.App.1942); Adams v. State, 765 S.W.2d 479, 480 (Tex.App.1988, pet. ref'd). The prohibition against jury separation during deliberations is one of the most fundamental and ancient legal rules in the English-speaking world and is normally strictly enforced in order to prevent jury tampering. Adams, 765 S.W.2d at 480; Reid v. State, 749 S.W.2d 903, 909 (Tex.App.1988, pet. ref'd). When the defendant has established that an improper jury separation has occurred, it has been held that the mandatory language of article 35.23 raises a presumption of harm which the State must then seek to rebut. Reed v. State, 595 S.W.2d 856, 857 (Tex.Crim.App.1980); Trevino v. State, 565 S.W.2d 938, 940 (Tex.Crim.App.1978); Taylor v. State, 636 S.W.2d 600, 602 (Tex.App.1982, pet. ref’d). In order to afford the State an opportunity to rebut this presumption of harm, the issue of improper jury separation must be raised during trial or in a hearing on a motion for new trial. McDonald, 597 S.W.2d at 367; McIlveen v. State, 559 S.W.2d 815, 818-19 n. 1 (Tex.Crim.App.1977).

The 1989 amendment to article 35.23 clearly intended to relax the jury separation rules after the charge had been read and during the deliberations of the jury, but the legislative intent is clear that the trial court “on the motion of either party shall ,.. order that the jury not be allowed to separate.” The term “shall” is usually mandatory when used in statutes, unless legislative intent suggests otherwise. Waythe v. State, 533 S.W.2d 802, 804 (Tex.Crim.App.1976); Brinkley v. State, 167 Tex.Crim. 472, 320 S.W.2d 855, 856 (1958). Further, the legislation is presumed to have used words in a statute in the sense which they are ordinarily understood. Connors v. Connors, 796 S.W.2d 233, 237 (Tex.App.1990, writ denied). Unless the statute is ambiguous, courts must follow the clear language of the statute. RepublicBank Dallas, N.A. v. Interkal, Inc., 691 S.W.2d 605, 607 (Tex.1985); Waythe, 533 S.W.2d at 804. We hold that the provision in article 35.23 that, after the charge is read to the jury in a criminal case, the trial court shall sequester the jury upon the motion of either party, is mandatory.

At no time during the trial did the State attempt to rebut the presumption of harm arising from the jury separation.⁵ No motion for new trial was filed. Before the 1989 amendment to article 35.23 and the adoption of the Texas Rules of Appellate Procedure, the law in Texas was clear that if a jury was allowed to separate in violation of the statute, the separation was presumed harmful and reversal was required unless the State had presented evidence which affirmatively negated the presumption. Harris v. State, 738 S.W.2d 207, 222 (Tex.Crim.App.1986); Skillern v. State, 559 S.W.2d 828, 830 (Tex.Crim.App.1977); Reid, 749 S.W.2d at 910.

There can be no doubt that when a party makes a motion to keep the jury together after the charge has been read to the jury, the 1989 version of article 35.23 is still mandatory. The trial court has no discretion in the matter. Where the jury is permitted to separate despite a motion by the defendant to sequester the jurors, and the State does not offer any evidence to affirmatively rebut and negate any harm resulting from the jury separation, is such error subject to a harm analysis? The cases cited above were decided before the adoption of Rule 81(b)(2), which requires the reviewing court to conduct a harm analysis.

We have not found any appellate court opinion that has specifically ruled on the question of whether the type of error at issue is now subject to a harm analysis under Rule 81(b)(2). At first blush, a reading of the rule gives the impression that the rule is applicable to any “error in the proceedings below.” This is not so. As explained by the Court of Criminal Appeals in Arnold, v. State, 786 S.W.2d 295, 297 n. 1 (Tex.Crim.App.1990):

Independent of Rule 81(b)(2) there are statutes specially providing a standard of review of some errors, e.g., Almanza v. State, 686 S.W.2d 157, at 171 (Tex.Cr.App.1985); see Rose v. State, supra [752 S.W.2d 529], at 537, 553 [Tex.Cr.App.1987]. Moreover, because, as well as violations of mandatory statutes, some constitutional violations “by their very nature cast so much doubt on the fairness of the trial process that, as a matter of law, they can never be considered harmless, [e.g.,] Sixth Amendment violations that pervade the entire proceeding.” Satterwhite v. Texas, 486 U.S. 249, at 253, 108 S.Ct. 1792, at 1794, 100 L.Ed.2d 284, at 293 (1988). Rule 81(b)(2) presupposes that the error in question is subject to a harmless error analysis not provided elsewhere.

(emphasis in original).

A number of recent cases have found that a particular error was not subject to a harm analysis under Rule 81(b)(2). See, e.g., Nunfio v. State, 808 S.W.2d 482, 485 (Tex.Crim.App.1991) (error by the court in the denial of a proper voir dire question which prevents the intelligent exercise of one’s peremptory challenges is an abuse of discretion and is not subject to a harm analysis under Rule 81(b)(2)); Russell v. State, 790 S.W.2d 655 (Tex.Crim.App.1990) (harmless error analysis inapplicable to the State’s failure to meet its burden of proving finality of a conviction alleged for the purpose of enhancement of punishment); Hernandez v. State, 785 S.W.2d 825 (Tex.Crim.App.1990) (denial of a complete statement of facts in capital murder case on appeal not subject to harmless error analysis); Scarbrough v. State, 777 S.W.2d 83 (Tex.Crim.App.1989) (denial of right to self-representation not subject to harmless error analysis); Wedlow v. State, 807 S.W.2d 847 (Tex.App.1991, pet. ref’d) (denial of a timely presented motion for severance of causes pending against the defendant was pretrial error which permeated the entire trial and was not the type of error subject to harm analysis).

In Roberts v. State, 784 S.W.2d 430 (Tex.Crim.App.1990), the Court was confronted with a severance of parties question. See Tex.Code Crim.Proc.Ann. art. 36.10 (1981). The trial court granted the severance motion, but violated the order of trial upon which the defendants had agreed in accordance with the statute. In Roberts, the Court discussed the harm analysis required by Rule 81(b)(2) in conjunction with mandatory statutes and observed the following:

Prior to promulgation of the Rules of Appellate Procedure not every “error in the proceedings below” was deemed subject to a harm analysis. Violation of some procedural provisions, “mandatory” in nature, was held by the Court to justify reversal of the conviction without an inquiry into harmfulness, vel non, of the error. A ready example is violation of Article 35.11, V.A.C.C.P., the jury shuffle statute. E.g., Smith v. State, 648 S.W.2d 695 (Tex.Cr.App.1983). In cases involving breach of many procedural statutes the record will contain no concrete data from which an appellate court can meaningfully gauge the likelihood that the error did or did not contribute to the conviction or punishment of the accused. This Court has yet to address the broad question whether violation of any statute formerly held “mandatory” will now be subject to Rule 81(b)(2), and we do not today. We do hold that, though its precursors have been designated “mandatory,” violation of Article 36.10, supra, may in some cases prove harmless.

Roberts, 784 S.W.2d at 436-36.

The Roberts court held that, under the particular facts of that case including the fact that one co-defendant testified and the other co-defendant was not called because “in all likelihood” he was impeachable with prior convictions, Rule 81(b)(2) was applicable to the error committed in violation of article 36.10. The court warned, however, that under different circumstances when the co-defendant did not testify it would be rare that an appellate court could say with the required level of confidence that a violation of an agreed order of trial did not contribute to a defendant’s conviction. The Court further explained that:

Ordinarily it will be impossible to know from the record whether a co-defendant’s testimony would have become available had he been tried first, or whether, if available, it would materially affect a jury’s deliberations. Because under Rule 81(b)(2), a conviction must be reversed “unless” the appellate court can determine error is harmless beyond a reasonable doubt, such uncertainty in the record would counsel reversal.

Roberts, 784 S.W.2d at 438.

In Sodipo v. State, 815 S.W.2d 551 (Tex.Crim.App.1991) (opinion on rehearing), the Court held that the trial court erred in permitting the State to amend the indictment on the date of the trial prior to the commencement of the trial on the merits, and that such error was not subject to a harm analysis. The Court found that article 28.10 of the Texas Code of Criminal Procedure “is written with clarity and is not ambiguous” but that a “vacancy” exists as to the amendment of the State’s pleadings on the date of trial prior to the commencement of the trial on the merits between the ’time slots for amendments provided in sections (a) and (b) of the statute. See Tex.Code Crim.Proc.Ann. art. 28.10 (1989). In absence of any “specific criteria” in article 28.10 for the time slot in question, the court concluded that the State was not permitted to amend the indictment on the date of trial before the commencement of the trial on the merits, and that the trial court erred in authorizing such amendment. There was no in-depth discussion for the conclusion that the error was not subject to a harm analysis.

In the Sodipo opinion on rehearing the Court held that its discussion of Section 28.10(a) in its original opinion was improvident. The opinion on original submission had held that the denial of a ten-day delay after the State’s amendment of the indictment as authorized by article 28.10(a) was error which was not subject to a harm analysis. While the Court has disavowed some of the discussion, the original opinion is still instructive as to harm analysis. That opinion found article- 28.10(a) to be mandatory in nature and that the ten-day requirement could not be subjected to a harm analysis in any meaningful manner because the record will not reveal any concrete data from which an appellate court can meaningfully gauge or quantify the effect of the error. The opinion relied in part on Roberts, 784 S.W.2d at 435, and stressed that the court was not deciding that all “mandatory” statutes are immune to a harmless error analysis. The Sodipo opinion on original submission did not terminate its discussion at that point but continued:

By comparison, the United States Supreme Court has made the distinction between its federal harmless error rule and the use of the automatic reversal rule by focusing on the reliability of the outcome of the trial. Those rights that require automatic reversal, according to the Court, are so basic or fundamental that in their absence “a criminal trial cannot reliably serve its function as a vehicle for determination of guilt or innocence ... and no criminal punishment may be regarded as fundamentally fair.” Rose v. Clark, 478 U.S. 570, 106 S.Ct. 3101, 92 L.Ed.2d 460 (1986). The distinctive characteristic of these violations is that it is extremely difficult to know whether they might have affected the outcome, or the likelihood that they have influenced the outcome is so strong that it is not worth expending the judicial resources necessary to evaluate the effect of the error in particular cases.⁶ United States v. Cronic, 466 U.S. 648, 659-60, 104 S.Ct. 2039 [2047-48], 80 L.Ed.2d 657 (1984). We find this reasoning to be persuasive in our determination of whether Rule 81(b)(2) applies in the instant case. Therefore, we hold that the Court of Appeals erred when it held that a harmless error analysis under Rule 81(b)(2) should be applied to the ten day provision of Art. 28.10(a).

815 S.W.2d at 554-55.

We conclude that the type of error presented in violation of the mandatory provision of article 35.23 in the instant case is not an error subject to harm analysis under Rule 81(b)(2). The error cannot be subjected to a harm analysis in any meaningful manner because this record does not reveal any concrete data from which an appellate court can meaningfully gauge or quantify the effect of the error. See Wedlow, 807 S.W.2d at 852. This does not mean that every violation of article 35.23 is to be excluded from a harm analysis under Rule 81(b)(2). Our holding is limited to the violation of article 35.23 as shown by the particular facts and circumstances of the instant case.

Alternatively, we attempt to apply harm analysis under Rule 81(b)(2). It must be remembered that there is a presumption of harm in the rule itself, and the State has the burden of demonstrating that the error is harmless beyond a reasonable doubt. See Arnold, 786 S.W.2d at 298, 311. The burden is certainly not on the appellant to demonstrate harm. Roberts, 784 S.W.2d at 437, n. 2. We have examined the factors set out in Harris, 790 S.W.2d at 587-88 and the framework in which the factors are to be applied. We do not find the Harris guidelines to be of much assistance in light of the nature of the error and the record in the instant case. The uncertainty would counsel reversal. We cannot determine beyond a reasonable doubt that the jury separation error made no contribution to the conviction. The third point of error is sustained.

The judgment is reversed and the cause remanded.

1 . The "hearsay" objection was apparent from the context and constituted a sufficient objection. Long v. State, 800 S.W.2d 545, 548 (Tex.Crim.App.1990); Carter v. State, 717 S.W.2d 60, 76 (Tex.Crim.App.1986); Zillender v. State, 557 S.W.2d 515, 517 (Tex.Crim.App.1977); Tex.R.Crim.Evid.Ann. 103(a)(1) (Pamph.1991); Tex.R.App.P.Ann. 52(a) (Pamph.1991). We do not understand the State to contend otherwise.

2 . Rule 107 of the Texas Rules of Criminal Evidence provides:

When part of an act, declaration, conversation, writing or recorded statement is given in evidence by one party, the whole on the same subject may be inquired into by the other, as when a letter is read, all letters on the same subject between the same parties may be given. When a detailed act, declaration, conversation, writing or recorded statement is given in evidence, any other act, declaration, writing or recorded statement which is necessary to make it fully understood or to explain the same may also be given in evidence. "Writing or recorded statement’ includes depositions.

3 . Another limitation is if the prejudicial effect of the matter offered on the justification of completeness substantially outweighs its probative value, it may be excluded under Rule 403. Tex.R.Crim.Evid. 403 (Pamph.1991).

4 . The factors set out in Harris to assess the effect of the error include: 1) the source and nature of the error; 2) the extent to which the state used the error throughout the course of the trial; 3) the probable collateral implications of the error; 4) the probable weight a juror would put on such an error; and 5) the likelihood that a finding of -harmlessness would encourage the State to repeat the error with impunity. These factors should be applied within a two-step framework: 1) isolate the error and its effects; and 2) ask whether a rational trier of fact might have reached a different result if the error and its effects had not resulted. See Charles P. Bubany, Criminal Procedure: Trial and Appeal, 45 Sw.L.J. 293, 316 (1991).

5 . On the morning following the jury separation, the trial court gave an “Allen” or dynamite charge to the jury. See Allen v. United States, 164 U.S. 492, 17 S.Ct. 154, 41 L.Ed. 528 (1896); Duc Vu v. State, 750 S.W.2d 8 (Tex.App.1988, pet. ref'd). We express grave doubts as to this particular charge being an appropriate "Allen" charge.

6 See e.g. Vasquez v. Hillery, 474 U.S. 254, 106 S.Ct. 617, 88 L.Ed.2d 598 (1986) (discrimination by a grand jury); Hollaway v. Arkansas, 435 U.S. 475, 98 S.Ct. 1173, 55 L.Ed.2d 426 (1978) (Sixth Amendment assistance of counsel); Davis v. Alaska, 415 U.S. 308, 94 S.Ct. 1105, 39 L.Ed.2d 347 (1974) (Sixth Amendment right to a fair and impartial jury); Payne v. Arkansas, 356 U.S. 560, 78 S.Ct. 844, 2 L.Ed.2d 975 (1958) (coerced confession).