United States v. Felix Rengifo

BOWNES, Circuit Judge. This appeal follows from defendant-appellant Felix Rengifo’s conviction on five counts of a fifteen-count indictment which charged Rengifo and eleven other defendants with various narcotics offenses. Defendant was convicted of conspiracy to possess cocaine with intent to distribute in violation of 21 U.S.C. § 846, possession of cocaine with intent to distribute in violation of 21 U.S.C. § 841(a)(1), and three counts of unlawful use of a telephone to cause and facilitate a narcotics felony in violation of 21 U.S.C. § 843(b). There are four issues: (1) the authentication of tape recordings of telephone conversations and a composite tape admitted into evidence; (2) the admission into evidence of transcripts of recorded telephone conversations read to the jury and sending the transcripts to the jury room; (3) the admission into evidence of statements of cocon-spirators; and (4) the denial of a motion for mistrial after the jury reported it was deadlocked and giving the jury a modified Allen charge. THE FACTS Defendant’s conviction resulted from a lengthy investigation conducted by Drug Enforcement Administration (DEA) agents and state and local police in Long Island, New York, and throughout Eastern Massachusetts. The bulk of the government’s evidence consisted of a series of tape recorded telephone conversations. These were intercepted and recorded by government agents during court authorized wiretaps of a telephone located in defendant’s Long Island home and a telephone located in the Brookline, Massachusetts, home of one of the codefendants, Elmer Rodriguez. Most of the recorded conversations were in Spanish. Detective Peter Murphy (Murphy) of the Brookline Police Department compiled a composite tape, containing thirty-five selected conversations, for use during trial. Two translators, Alfredo Ayarza (Ayarza) and Guadulupe Aguilar (Aguilar) prepared English transcripts of the composite tape. At the trial, two persons read the transcripts to the jury; two readers were used so that the jury and court reporter would know when one person’s recorded conversations stopped and another’s started. The government then called Murphy, who had considerable experience in investigating narcotics offenses, to interpret the narcotics jargon used in the transcripts. 1 The government alleged that the intercepted conversations memorialized a longstanding relationship between defendant and a cocaine distribution ring based in Eastern Massachusetts, whereby defendant supplied the ring with large quantities of cocaine from his Long Island home. The taped conversations recorded members of the alleged ring periodically discussing among themselves their relationships with defendant and arranging trips from Massachusetts to New York to obtain cocaine from him. Seven Spanish conversations, between members of the group and defendant personally, evidenced defendant demanding money due him for cocaine delivered on consignment and ring members placing orders with defendant for additional quantities of cocaine. At the close of the evidence, the transcripts read in court were sent to the jury room along with the exhibits over defendant’s objection. The jury commenced its deliberations about 12:34 p.m. At about 7:27 p.m., it sent two messages to the court: *978 Judge Caffrey, we are deadlocked. Where do we go from here? /s/ Foreperson Judge Caffrey, after lengthy discussion and three ballots we are at an impasse, and the people voting for not guilty are feeling picked on and want you to know they are not changing their minds. /s/ Foreperson The judge read these messages to counsel, told them that he would give the jury a modified Allen charge, and denied defendant’s motion for a mistrial. After the charge was given, the jury resumed its deliberations and delivered a verdict of guilty about two hours later. 1. THE AUTHENTICATION OF THE ORIGINAL TAPE RECORDINGS AND COMPOSITE TAPE Defendant contends that tape recordings may only be authenticated by someone who was either a party to the recorded conversation or who overheard the recorded conversation though not personally participating in it. He argues that the recorded conversations here should not have been admitted because the witnesses relied on by the government to authenticate them could not speak or understand Spanish, the language in which most of the conversations took place. Defendant also claims that the trial court abused its discretion in admitting the composite tape in light of Murphy’s admission that he relied on a “duplicate original” to compile the composite and that he did not check that “duplicate original” against the original tape. The government relied on the, testimony of three witnesses to authenticate the tape recorded evidence: 2 Agent John Hampe (Hampe) of the Boston DEA office testified to the authenticity of the recordings obtained from the telephone of codefendant Rodriguez; Agent Thomas Cregan of the Melville, Long Island DEA office testified to the authenticity of the recordings obtained from defendant’s telephone; and Murphy testified to the authenticity of the composite tape. Defendant confines his objection on appeal to the adequacy of Hampe and Murphy’s authentication. We examine defendant’s contentions concerning the authentication of the original recordings and the composite separately. A. The Original Tape Recordings This circuit does not require that the witness authenticating tape recorded conversations be someone who either participated in or personally overheard the subject matter of the recording in evidence. See United States v. Cortellesso, 663 F.2d 361 (1st Cir.1981). The government there relied exclusively on the testimony of the supervising F.B.I. agent to authenticate tape recordings of intercepted telephone conversations. The agent testified in detail to the procedures followed to obtain the recordings: his presence at the initial testing of the equipment; his personal preparation of transcripts of the tapes; and his custody of the original logs and tape recordings. We held this testimony sufficient to raise the “presumption of official regularity” and held it unnecessary to have the agents who did the actual monitoring of the calls take the stand to confirm , that they followed their supervisor’s instructions or to deny tampering with the tapes. Id. at 364. We also held that, once the government presents sufficient foundation testimony, the party challenging the tape recordings bears the burden of showing the recordings are inaccurate and that this could be done, for example, by engaging an expert to examine the tapes. Tapes are not inadmissible merely because “one can conjure up hypothetical possibilities that tampering occurred.” Id. (quoting United States v. Haldeman, 559 F.2d 31, 109 (D.C. Cir.1976), cert. denied, 431 U.S. 933, 97 S.Ct. 2641, 53 L.Ed.2d 250 (1977)). Hampe’s testimony mirrors that of the agent in Cortellesso in virtually every respect. He supervised the wiretap of the *979 Rodriguez telephone from the first day it was authorized throughout its duration. His testimony evidenced intimate knowledge of every minute detail of the procedures followed. The wiretap was controlled from a room in the DEA’s office in Boston designated the “listening post” which was fitted with the equipment necessary to intercept telephone conversations. It contained a Mitel decoder, three tape recorders and an interface box. The decoder made a record every time the headset of the target telephone was lifted and printed out any numbers dialed. It also noted the time and date of any incoming calls if those calls were answered. The three tape recorders were powered by an attached interface device which was in turn attached to the decoder. Whenever the decoder was activated, electricity would pass through the interface and cause the recorders to turn on and record. Thus, all three recorders acted simultaneously and recorded identical sounds, making in effect three original recordings. The parties stipulated that all the government’s equipment was in working order throughout the period of electronic surveillance. Hampe also described in detail the activities of the agents actually manning the listening post. He testified that he visited the “listening post” periodically and observed the agents noting the time and date of each call, whether it was incoming or outgoing, the number dialed, and whether the conversation was in English or Spanish. Individual calls on each tape were identified numerically. Hampe explained further that each tape was numbered as it was removed from the recorder and that all the tapes were stored under .lock and key at the end of each surveillance period. Ultimately, one set of tapes was sealed by court order and the other sets were stored in the vault of the Boston DEA office. Hampe testified that he had some familiarity with the voices of the speakers in the recorded conversations because he had obtained voice exemplars from defendant and Rodriguez, among others, and, in addition, had listened to some of the actual conversations. We find that Hampe’s testimony was sufficient to raise a presumption of official regularity. After Hampe’s thorough explanation of how these recordings were obtained, the burden shifted to defendant to show why the recordings should be rejected as inauthentic. The flat assertion that Hampe was unqualified to authenticate the recordings because he could not understand the language in which the recorded conversations were spoken does not discharge that burden. Hampe’s ability to authenticate these recordings did not depend on his knowledge of Spanish. It depended, rather, on his personal knowledge of the methods used to carry out the wiretaps and of the degree to which the agents conducting the surveillance complied with the methods prescribed. Once those methods had been explained, and the agents’ compliance with them established, the government had discharged its burden of authentication. Defendant has failed to make any showing of alteration, deletion or other inaccuracy in the recordings. B. The Composite Tape 3 Murphy testified that he compiled the composite tape by relying primarily on the logs prepared by the listening agents and a set of Spanish transcripts, which he followed roughly while listening to the tapes. He testified that he could only understand some words of the conversations but managed to identify the calls he had been told to select by checking against the data in the logs. He was also helped by hearing words of salutation and farewell at the beginning and end of each call. The clicking of the headset signalled him when each *980 call began and ended. He then checked the composite against the tape he had been given to work from to ensure that the sounds matched up. Defendant contends that the composite tape was not properly authenticated because of Murphy’s admitted failure “to compare the duplicate original reel to reel tape with the original reel to reel tape,” and because Murphy did not overhear any of the actual conversations. This argument ignores the way the tapes were made. All .three tape recordings were originals; each of them independently recorded the actual telephone conversations. This is not a situation, as defendant implies, where one tape recording was made and subsequently copies were made from it. There was no need for Murphy to check his tape against the original tape; he used an original tape. As we have already held, Agent Hampe’s authentication of the tape recordings passed muster. Murphy did not have to overhear any of the actual phone conversations to authenticate the composite tape. Defendant’s contention that Murphy was not competent to authenticate the composite because he could not understand Spanish, is equally without merit. Murphy’s testimony showed that he followed a reasonably reliable method in compiling the composite and that it was possible to make the composite without understanding what was actually being said in the recorded conversations. His testimony, like that of Hampe, placed the burden of showing some flaws in the composite tape squarely on the defendant; defendant has failed to meet that burden. II. ADMITTING THE TRANSCRIPTS AS EVIDENCE AND SENDING THEM TO THE JURY ROOM Defendant makes three contentions with regard to the admission of the transcripts: (1) that transcripts of tape recordings are not admissible as evidence under the law in this circuit; (2) that allowing the transcripts to go to the jury room enabled the government to unduly emphasize this portion of the evidence; and (3) that, even if such transcripts were admissible, the binder holding the transcripts also included summaries of the recorded conversations and identifications of the speakers and this constituted inadmissible hearsay. Contrary to what defendant contends, we have not adopted an absolute rule that transcripts of tape recordings are not admissible in evidence. In fact, the issue has not been squarely addressed in this circuit, although it did arise obliquely in United States v. Bizanowicz, 745 F.2d 120 (1st Cir.1984). The trial judge there allowed a tape recording and tape player to go to the jury room, but refused to admit the accompanying transcripts. Defendant appealed on the ground that the admission of the recording and player was error. We upheld the admission of the tape recording and player, but did not comment on the trial court’s refusal to admit the transcripts. Id. at 123. We have long upheld the practice of providing the jury with a transcript to help them follow a tape recording being played during trial. See United States v. Nashawaty, 571 F.2d 71, 75 (1st Cir.1978); United States v. DiMuro, 540 F.2d 503, 511 n. 13 (1st Cir.), cert. denied, 429 U.S. 1038, 97 S.Ct. 733, 50 L.Ed.2d 749 (1976). Our admonition in United States v. Rickman, 600 F.2d 286 (1st Cir.1979), that it is “within the discretion of the court to allow a transcript to accompany the playing of tapes so long as the court makes clear that the tapes, not the transcript constitute evidence in the case,” id. at 295, falls far short of a ruling that transcripts are inadmissible as evidence in every circumstance. We turn, therefore, to consider how other circuits have dealt with both this issue and the related one of allowing the transcripts to go to the jury room. The Second Circuit has generally allowed transcripts of tape recordings to go to the jury room provided the accuracy of the transcripts is established at trial. In United States v. Koska, 443 F.2d 1167 (2d Cir.), cert. denied, 404 U.S. 852, 92 S.Ct. 92, 30 L.Ed.2d 92 (1971), the court had to deter *981 mine whether the trial court erred when it allowed the jury to read a transcript while listening to five hours of tape recordings and later allowed twelve copies of the transcript to go to the jury room in response to the jury’s specific request. The Second Circuit upheld the trial court’s action finding “[n]o good reason ... for denying the transcript to a jury which has requested it, where ... the jury has already heard the tape and the tape is long and cumbersome.” Id. at 1169. It noted that the parties had stipulated that the transcript was accurate and that the trial judge had taken the added precaution of instructing the jurors that they did not have to regard the transcript as controlling if they determined that any line of it differed from what they themselves heard on the tape. Id. at 1169 n. 1. The Second Circuit subsequently confirmed this ruling in United States v. Carson, 464 F.2d 424 (2d Cir.), cert, denied, 409 U.S. 949, 93 S.Ct. 268, 34 L.Ed.2d 219 (1972). In Carson, the trial court ordered counsel for both sides to listen to the tapes prior to trial to determine where they disagreed as to words on the transcripts, and then held an in camera hearing prior to the introduction of the tapes and transcripts. The parties agreed that the transcripts accurately reflected the words on the tapes with certain exceptions as to which it was agreed that the transcripts would contain the two versions, i.e., the one believed accurate by each party. After noting these precautions, the Second Circuit held it was not error to allow the jury to retain the transcripts during the trial and deliberations. Id. at 437. A similar approach has been taken by the Fifth and Eleventh Circuits, beginning with the pivotal case of United States v. Onori, 535 F.2d 938 (5th Cir.1976). Onori centered on the proper use of transcripts during trial rather than on whether they should go to the jury room. The defendant proffered the testimony of an expert on tape identification who testified, outside the presence of the jury, that there were numerous errors in the government’s transcript. The trial court offered to allow the defendant’s expert to testify to the jury but indicated that this would mean that both the government’s and the defendant’s transcript would go to the jury to help it decide which portions of each transcript were correct. The defendant rejected this offer, maintaining that the government’s version was highly prejudicial. He appealed on the ground that the trial court’s actions with respect to the transcript, and the conditions placed on the defendant’s introducing his expert’s testimony constituted reversible error. The Fifth Circuit upheld the trial court’s action and went on to lay down certain guidelines for dealing with contested transcripts. It held that the trial court should first attempt to obtain a stipulated transcript, by summoning both sides to a pretrial meeting in chambers, if necessary. Id. at 948. Citing United States v. Carson, 464 F.2d 424, it held, further, that the jury should be given both parties’ versions of the transcript if efforts to reach a stipulated transcript failed and that it should be instructed that the parties disagreed on the transcripts’ accuracy and that it could choose which version, if any, to accept. 535 F.2d at 949. The court concluded: As with other forms of potentially prejudicial evidence, the key to protecting a defendant’s rights in this situation lies in seeking limiting instructions. Upon request by a defendant, the jury should be instructed that a transcript is just another piece of evidence subject to objections, that it may have to be evaluated for accuracy, and that the jury need not accept any proffered transcript as accurate. 535 F.2d at 949. Though it did not directly address the use of transcripts by the jury during deliberations as distinct from at trial, Onori has been consistently relied upon by both the Fifth and Eleventh Circuits to allow sending transcripts to the jury room. In United States v. Costa, 691 F.2d 1358 (11th Cir.1982), for example, the defendant claimed the trial court erred in allowing *982 transcripts to go to the jury during deliberations because the jurors had not also been given the tape player and were, therefore, forced to rely on the transcripts rather than the tapes in reaching their decision. The Eleventh Circuit rejected this claim, holding that there was no error absent some showing that the transcripts were inaccurate. Id. at 1363. In United States v. Williford, 764 F.2d 1493 (11th Cir.1985), the Eleventh Circuit dismissed the contention that allowing transcripts to go to the jury during deliberations placed undue emphasis on the taped testimony: “Absent a showing that the transcripts were inaccurate or that specific prejudice occurred, there is no error in allowing transcripts to go to the jury room.” 764 F.2d at 1503 (citation omitted). And in United States v. Larson, 722 F.2d 139 (5th Cir.1983), the Fifth Circuit held that it was harmless error to allow a transcript not formally admitted as evidence, to go to the jury room, where the jurors had read the transcript during trial and been explicitly instructed to resolve any discrepancy between the transcript and the tape in favor of what they heard on the tape. 722 F.2d at 145. Citing Onori, the court stated, “transcripts are admissible as evidence subject to the court’s issuance of proper limiting instructions.” 722 F.2d at 145. The particular circumstances of the present case, where the jury did not speak the language of the recorded conversations and, therefore, had to rely exclusively on the transcripts, both at trial and during deliberations, was recently addressed by the Eleventh Circuit in United States v. Cruz, 765 F.2d 1020 (11th Cir.1985). In Cruz, as in the present case, the government introduced taped Spanish conversations and English transcripts. The defendant claimed that the trial court erred in admitting the transcripts as evidence on two grounds: (a) because of the unreliability inherent in translating foreign languages; and (b) because the jury’s lack of conversance with Spanish meant they necessarily considered the transcripts as substantive evidence rather than as a mere aid for use in listening to the tapes. The court rejected both arguments. It held that the defendant’s failure to submit his own version of the transcripts precluded him from complaining about the unreliability of the transcripts on appeal, and cited Onori as authority for upholding the admissibility of the transcripts as substantive evidence. Id. at 1023. 4 We conclude our survey by noting that these issues have not yet been squarely addressed in the other circuits. We now turn to defendant’s contention that allowing the transcripts to go to the jury room prejudiced him because it unduly emphasized the wiretap evidence. We recognize that double exposure to transcripts of crucial evidence, i.e., during the trial and in the jury room, can be prejudicial to a defendant. Relevant evidence, however, always prejudices the party against whom it is offered. It can hardly be argued that accurate transcripts of properly authenticated and relevant wiretap recordings are not relevant evidence. Under Federal Rule of Evidence 402, all relevant evidence is admissible, unless excluded by the Constitution, an Act of Congress, Rule of the Supreme Court, or the Rules of Evidence. The only ground for excluding transcripts would be that their “probative value is substantially outweighed by the danger of unfair prejudice.” Fed.R.Evid. 403. The Notes of the Advisory Committee on Proposed Rules relative to Rule 403 state: “In reaching a decision whether to exclude on grounds of unfair prejudice, consideration should be given to the probable effectiveness or lack of effectiveness of a limiting instruction.” The question of unfair prejudice is an evidentiary one that falls within the *983 discretion of the district court. We think, however, that some guidelines should be formulated for the use of wiretap transcripts. Drawing upon the precedent established in the Second, Fifth and Eleventh Circuits, we, therefore, hold as follows. The district court, in the exercise of its discretion, should decide whether properly authenticated transcripts should be admitted as evidence and go to the jury room initially along with the rest of the exhibits. If the jury requests the transcripts after it has started its deliberations, it is within the district court’s discretion to decide whether the jury’s request should be granted. We believe that it is advisable for the district court to try to obtain a stipulated transcript from the parties before trial or, at least, before a transcript is used. Failing such stipulation, each party should be allowed to introduce its own transcript of the recording provided that it is properly authenticated. When the jury receives two transcripts of the same recording, it should, of course, be instructed that there is a difference of opinion as to the accuracy of the transcripts and that it is up to them to decide which, if any, version to accept. The jurors should also be instructed that they can disregard any portion of the transcript (or transcripts) which they think differs from what they hear on the tape recording. Further limiting instructions will depend on the circumstances of each case. We now turn to the specific issue before us, whether the defendant was unfairly prejudiced and the district court abused its discretion by sending the English translation transcripts to the jury room. We find that there was no abuse of discretion. As already indicated, we do not accept defendant’s contention that the transcripts were suspect merely because they were based on the composite tape. The qualifications of the government’s translators and the word-for-word accuracy of the transcripts were explored in depth at trial. Both translators, Ayarza and Aguilar, were called as witnesses and defense counsel cross-examined them extensively on their familiarity with different Spanish dialects and on the accuracy of the tapes generally. They both testified that the transcripts were fair and accurate and explained that they first prepared Spanish transcripts from the tapes and then translated those transcripts into English. Defendant, on the other hand, produced no transcripts to rebut the government’s transcripts. Nor did he request any instruction on how the jury should treat the transcripts once his objection to their being sent to the jury room was overruled. Indeed, in the absence of any alternative transcript, there was no meaningful instruction the trial judge could give the jury beyond instructing them, as he did, that they could consider the transcripts “like any other evidence in the case.” Finally, we hold that the government’s use of readers as an alternative to playing the tapes and giving the jury the transcripts to follow did not change the essential nature of the transcript evidence. The readers were not witnesses and their reading of the transcripts did not constitute “testimony” in the sense of “evidence given by a competent witness, under oath or affirmation.” See Black’s Law Dictionary 1646 (rev. 4th ed. 1978). The readers performed an essentially mechanical task. Defendant’s additional objections to the admission of the transcripts are without merit. His claim that they constituted inadmissible hearsay because they contained summaries and identifications does not withstand scrutiny. We have examined the binder of transcripts submitted to the jury and have been unable to find any summaries of recorded conversations therein. As to identification, the speakers were identified in the transcripts by Ayarza, who testified that he was present when a voice exemplar of the defendant was taken, that he had examined voice exemplars of the other key speakers, and that he had heard defendant, and the other key coconspirators, speak in person several times. Ayarza testified that he was, therefore, able to recognize the speakers’ voices on the tapes as he prepared the transcripts and explained that he relied on factors such as tone, diction, accent and speed of delivery *984 to identify the voices as they came across on the tapes. Identifications based on personal knowledge of this sort do not constitute hearsay. Further, the speakers often identified themselves in the course of a call and so the jurors were in a position, for the most part, to confirm Ayarza’s identifications themselves. The judge instructed them to that effect, telling them that they could accord such weight to this “internal evidence” as they deemed appropriate. III. THE ADMISSION OF COCONSPIR-ATORS’ STATEMENTS Defendant claims that the admission of twenty-five of the recorded conversations was error on the ground that they did not meet the requirements of Federal Rule of Evidence 801(d)(2)(E) and were, therefore, inadmissible hearsay. He also contends that his sixth amendment confrontation clause rights were violated by the admission of these particular recordings. Rule 801(d)(2)(E) provides that certain statements made by coconspirators, ordinarily considered hearsay, are not to be deemed hearsay and may be admitted into evidence. Specifically, it provides: (d) Statements which are not hearsay. A statement is not hearsay if— (2) Admission by party-opponent. The statement is offered against a party and is ... (E) a statement by a coconspir-ator of a party during the course and in furtherance of the conspiracy. In United States v. Martorano, 557 F.2d 1, reh’g denied, 561 F.2d 406 (1st Cir.1977) (explaining holding), cert. denied, 435 U.S. 922, 98 S.Ct. 1484, 55 L.Ed.2d 515 (1978), this circuit outlined the conditions under which the rule is met: The district judge will admit the hearsay declarations if he determines, by a preponderance of the evidence, that a conspiracy existed, that the declarant and the defendant were members of it at the time the statements were made, and that the declarant’s statements were made in furtherance of the conspiracy. 557 F.2d at 11. Defendant urges us to reexamine the law surrounding Rule 801(d)(2)(E), including our own decision in Martorano, and its relation to Glosser v. United States, 315 U.S. 60, 74-75, 62 S.Ct. 457, 467, 86 L.Ed. 680 (1942). Defendant first challenges the district court’s determination that Rule 801(d)(2)(E) was satisfied. He maintains that the requirement that the existence of a conspiracy be proved by a preponderance of the evidence cannot be met if the only evidence of conspiracy is the hearsay statements themselves. He argues that the seven recorded conversations in which he personally participated are not sufficient in themselves to establish the existence of a conspiracy and that the twenty-five taped conversations containing hearsay may not be considered as evidence of a conspiracy under the rule. We reject the first premise of defendant's argument. Our review of the seven conversations involving the defendant personally has convinced us that there was ample independent proof to establish by a preponderance of the evidence the existence of a conspiracy involving defendant and various of his associates. In those conversations, defendant discussed shipments and deliveries of cocaine, the payment of proceeds from sales, and various personal problems arising among the conspirators. With such evidence of a conspiracy before it, the district court did not err in finding that the other telephone conversations among the coconspirators could be admitted under Rule 801(d)(2)(E). Defendant also challenges the admission of the coconspirator hearsay statements on the ground that they violated his rights under the confrontation clause. Defendant relies on Ohio v. Roberts, 448 U.S. 56, 100 S.Ct. 2531, 65 L.Ed.2d 597 (1980), a case involving the contested admission of statements made by a witness in prior judicial proceedings. The Court held: In sum, when a hearsay declarant is not present for cross-examination at trial, the Confrontation Clause normally requires a showing that he is unavailable. Even then, his statement is admissible *985 only if it bears adequate “indicia of reliability.” Reliability can be inferred without more in a case where the evidence falls within a firmly rooted hearsay exception. In other cases, the evidence must be excluded, at least absent a showing of particularized guarantees of trustworthiness. Id. at 66, 100 S.Ct. at 2539 (footnote omitted). Defendant contends that the cocon-spirator hearsay statements admitted here violated his confrontation clause rights under Roberts because the prosecution made no showing that the declarants were unavailable to testify at trial. This reading of Roberts was, however, recently rejected by the Court in United States v. Inadi, — U.S. -, 106 S.Ct. 1121, 89 L.Ed.2d 390 (1986): “Roberts cannot fairly be read to stand for the radical proposition that no out-of-court statement can be introduced by the government without a showing that the declarant is unavailable.” Id. at-, 106 S.Ct. at 1126. It went on to hold: “[W]e continue to affirm the validity of the use of co-conspirator statements, and we decline to require a showing of the declarant’s unavailability as a prerequisite to their admission.” Id. at-, 106 S.Ct. at 1129. The Court expressly noted that the reliability of the out-of-court statements was not an issue. Id. at- n. 3, 106 S.Ct. at 1124-25. So also, defendant here has not challenged the reliability of the coconspirator statements. IV. THE DENIAL OF DEFENDANT’S MOTION FOR A MISTRIAL Defendant claims that because the jury’s notes to the court showed that it was deadlocked and at an impasse it was error for the trial court to refuse to grant a mistrial, and that giving the modified Allen charge worked an impermissible influence on the jurors. Defendant does not take issue with the substance of the charge. While it has long been the rule that a court commits reversible error if it inquires into the nature and extent of the jury’s division during deliberations, Brasfield v. United States, 272 U.S. 448, 47 S.Ct. 135, 71 L.Ed. 345 (1926), this court has held that it is not reversible error for the jury to reveal its division voluntarily. United States v. Hotz, 620 F.2d 5 (1st Cir.1980). The court there dismissed the jury after four and a quarter hours of deliberation when it learned from information volunteered by the jury that they were at an impasse and divided 11-1. The purpose of shielding the jury from inquiry by the judge into their numerical division is to prevent actual or inferred pressure from the judge towards one particular group to change its position. Brasfield v. United States, 272 U.S. at 450, 47 S.Ct. at 135-36. As this court pointed out in United States v. Hotz, 620 F.2d 5, it is always best for the trial judge not to know the extent and nature of a division among the jurors and to instruct the jury not to reveal that information. Id. at 7. Nevertheless, “if the jury does volunteer its division, the court may rely and act upon it.” Id. Here, the numerical division among the jurors was not revealed to the court. The notes from the jury foreperson only stated that the jury was “deadlocked” and “at an impasse.” At no time did the court poll the jury or inquire into its numerical division. Indeed, the second charge admonished the jurors to respect those among them holding different views. In these circumstances, we hold that giving the modified Allen charge was the correct response to the information that the jury was at an impasse. There was no error in refusing to grant a mistrial. Affirmed. Footnotes [1] The transcripts used such terms as "girl" which Murphy testified were commonly used by drug dealers to refer to cocaine. [2] It is undisputed that the government bears the burden of authenticating this evidence. See Fed.R.Evid. 901. [3] This court has long upheld the use of composite tapes as evidence. See United States v. DiMuro, 540 F.2d 503, 512 (1st Cir.), cert. denied, 429 U.S. 1038, 97 S.Ct. 733, 50 L.Ed.2d 749 (1976) (government’s presentation of only twenty-five out of a total of over six hundred intercepted gambling conversations held not unfair where plaintiffs had access to the original tapes and ample opportunity in cross-examination to bring out the total number of intercepted conversations in which each defendant was involved). [4] Cruz differed from the present case in two respects. First, the government there did not have readers read the transcripts, but distributed them to the jury to read while the Spanish recording was played and had an interpreter signal the jury when it was appropriate to turn the pages. Additionally, in Cruz, the transcripts were only sent to the jury room after the jury had made a specific request for them.