Summary
finding that probable cause was satisfied where recorded calls reflected named interceptee's potential involvement in drug trafficking and registers reflected frequent contacts with other members of the alleged scheme
Summary of this case from United States v. AllenOpinion
Criminal No. 09-63-ART.
September 13, 2010
MEMORANDUM OPINION ORDER
After failing to file specific objections or present alternatives, the defendants filed motions generally objecting to the government's introduction of several transcripts. Yet the government called qualified translators to testify to the transcripts' accuracy. It further corroborated voice identifications in the transcripts with witnesses subject to cross examination. And to guard against any prejudice, the Court offered repeated limiting instructions and ruled that the jury could not take the transcripts back with them during deliberations. Accordingly, the defendants' motions are denied and the government's motion to identify speakers in the transcripts is granted.
The Court previously ruled on these motions but promised this opinion.
BACKGROUND
During its investigation into a suspected drug trafficking organization, the Drug Enforcement Agency secured a series of Title III wiretaps on phones it tied to the defendants. R. 329 at 2. The agency made recordings, and many conversations were in Bosnian or Spanish. It hired translators to transcribe the recordings. In conjunction with DEA investigators, the translator/transcribers included speaker identifications in the transcripts. An indictment followed on September 23, 2009. R. 1.Anticipating transcript disputes, Magistrate Judge Robert Wier set several deadlines for exchanging transcripts (March 29, 2010); for the defendants to file objections and offer competing transcripts (May 3, 2010); and for the parties to file a status report attaching copies of "disputed or competing transcripts at issue" (May 24, 2010). R. 243. The defendants failed to file any specific objections or alternative transcripts, and the notice attaching disputed or competing transcripts was not filed. Instead, beginning on April 28, 2010, some defendants filed generalized objections. They broadly claimed that the translator/transcribers' qualifications were unknown, R. 315 at 1, and that it was inappropriate to introduce the transcripts with the speakers identified. R. 299; R. 315 at 2. One defendant additionally argued that the Court had an affirmative obligation to compare the transcripts to the recordings in camera. R. 315 at 1.
At a July 20, 2010, hearing, the Court held that the defendants waived any specific objections to the accuracy of the transcripts. R. 346 at 3. But, in recognition of the defendants' generalized objections and the usual procedure for introducing transcripts, the Court set several additional deadlines: The government was to produce the translator/transcribers' curriculum vitae by July 26, 2010; the defendants were to file any objections to the accuracy of the transcripts based on the translator/transcribers' qualifications by August 2, 2010; and by August 6, 2010, the government was to seek leave to identify the purported speakers in the transcripts. Id. at 2-3. Finally, the Court scheduled an additional pretrial hearing to resolve outstanding transcript issues. Id. The defendants filed motions elaborating upon their objections to the translator/transcribers' qualifications to accurately prepare the transcripts. R. 370, 371. And the government sought leave to identify speakers in the transcripts. R. 384.
DISCUSSION
The transcripts have been appropriately introduced as limited-use exhibits. See United States v. Segines, 17 F.3d 847, 854, 854 n. 6 (6th Cir. 1994) (holding that transcripts have "limited use" as aids in understanding the tape recording, which is the real evidence). Even though the defendants failed to properly object, the government still followed the appropriate procedure for establishing the transcripts' accuracy. It called qualified translators and transcribers to give their stamp of approval. And because the government presented witnesses to corroborate its identification of voices in the calls, it was entitled to identify those speakers in the transcripts.
I. Defendants' objections to transcript accuracy fail
The defendants protest that the government did not lay an appropriate foundation for the accuracy of the transcripts before seeking their introduction — a threshold requirement. They challenge the qualifications of the translator/transcribers, R. 370; R. 371, and argue that the Court must conduct an in camera review of the transcripts to verify their accuracy, R. 315 at 1. These objections fail for two reasons.
First, the defendants have effectively waived these objections. Parties who object to the use of translated transcripts as exhibits have an obligation to point to specific inaccuracies — whether in the translation or transcription — or present alternative transcripts. As the Sixth Circuit has held repeatedly, failure to do so is effectively a waiver. In United States v. Garcia, for instance, the Sixth Circuit rejected a challenge to the reliability of a translated transcript in part because the "defendant [could not] complain if he [did] not offer a substitute version of the transcript." 20 F.3d 670, 673 (6th Cir. 1994) (citing United States v. Cruz, 765 F.2d 1020, 1023 (11th Cir. 1985)). Similarly, in United States v. Liddel, the Sixth Circuit upheld a trial court's decision to admit a transcript on the basis of testimony by an investigating officer that the translation was accurate. 64 F. App'x 958, 963 (6th Cir. 2003). The court held that "there can be no abuse of discretion . . . when the Defendants involved did not point to any specific inaccuracies in the translation or offer any alternative translation." Id. (citing Garcia, 20 F.3d at 673); see also United States v. Briscoe, 896 F.2d 1476, 1492 (7th Cir. 1990) ("In this case, because the defendants failed to present their own translation of the Yoruba conversations and also failed to present an expert witness to demonstrate possible inaccuracies in the translated transcripts . . . they have certainly limited their opportunity to challenge the accuracy of the government's transcripts on appeal."). Rather than offer specific objections or alternative transcripts, the defendants broadly objected to the translator/transcribers' qualifications. To this day, the defendants have pointed to no material inaccuracies in the transcripts. This is not enough.
Second, any concerns the defendants have about the proper qualification of the translator/transcribers or the procedural requirements for verifying the accuracy of the foreign language transcripts were fully addressed at a pretrial conference on August 23 and 24, 2010. In cases involving translated transcripts, the Sixth Circuit requires three measures to verify transcript reliability. First, the transcript's proponent must present a witness qualified in the relevant foreign language to verify the translation's accuracy. Garcia, 20 F.3d at 673; see also Liddel, 64 F. App'x. at 963 (affirming district court's decision to allow use of English translation transcript where qualified witness testified to accuracy and defendant neither made any objection to specific translation nor offered an alternative translation). Second, someone should testify that they have listened to the tapes and can verify the transcriptions are accurate. United States v. King, 272 F.3d 366, 374 (6th Cir. 2002) ("[I]n the absence of a stipulation, we hold that the transcriber should verify that he or she has listened to the tape and accurately transcribed its contents." (quoting United States v. Robinson, 707 F.2d 872, 878-79 (6th Cir. 1983)); United States v. Roberts, 316 F. App'x 388, 392 (6th Cir. 2008) (holding that there is no requirement that the transcriber testify; another witness can verify that the transcript accurately reflects the audio). Third, the court should screen out transcripts of tapes so inaudible that the transcriber could not have completed the transcription without the aid of a participant in the recorded conversation. See Robinson, 707 F.2d at 879 (holding that transcripts were not sufficiently reliable because "several of the tapes [were] so inaudible that it would be impossible to transcribe them without an independent recollection of the conversations"). Each requirement was met in this case.
In satisfaction of the first two requirements, the government called witnesses who both translated and transcribed the relevant transcripts at pretrial. Each was qualified to opine as to the accuracy of the translation under Rule 702. See United States v. Contreras, No. 98-4546, 1999 WL 123760, at *2 (4th Cir. 1999) ("A witness's qualifications to render an expert opinion are liberally judged by Federal Rule of Evidence 702[.]"); United States v. Abonce-Barrera, 257 F.3d 959, 965 (9th Cir. 2001) (assessing translation expert qualifications under Rule 702). That is, each enjoyed adequate personal or professional experience with the Bosnian or Spanish language to render his opinion reliable. See United States v. Bonds, 12 F.3d 540, 555 (6th Cir. 1993) (noting that Rule 702 requires that opinion testimony be reliable, on top of being helpful to the trier of fact); United States v. Vitale, No. 89-1989, 1990 WL 57235, at *1 (6th Cir. May 2, 1990) (determining translator's qualifications on basis of both personal and professional experience). And each verified the accuracy of his translations and transcriptions:
Testimony as to the accuracy of translations and transcription is appropriate at pretrial. See Garcia, 20 F.3d at 673; United States v. Wilkinson, 53 F.3d 757, 762 (6th Cir. 1995).
(1) Gordana Smith testified that she is a native Bosnian and attended school there. She owns her own translation agency and is one of four people in this country certified by the American consular court to test the Bosnian language skills of those seeking employment with the government. Ms. Smith has offered translation or interpretation services to the United States Army and other government agencies. She also attended workshops to interpret courtroom proceedings and has conducted in-court interpretations. Finally, she boasts a degree in teaching English as a second language and taught English to students in Bosnia. She certified that, after she translated and transcribed calls, she proofread them and stands by "every word" of her translations and transcriptions.
(2) Isabel Church is a native Spanish speaker who grew up in a bilingual household. She took Spanish courses at the University of Texas, testing at a very proficient level. Certified by the National Security Agency in transcription, translation, and deciphering Spanish-to-English conversations, Ms. Church serves as an intelligence analyst for the DEA. She described the method she used to translate and transcribe the calls. Ms. Church did not begin transcribing until she was sure what was said on the audio. She took note of inaudible portions of the recording. Then, after transcribing, she checked her work "numerous times." Ms. Church added that she has since reviewed her transcripts for accuracy. Like Ms. Smith, she certified the accuracy of her translations and transcriptions.
(3) David Jake Kavosevic grew up in a bilingual Serbo-Croatian/English household. Mr. Kavosevic has performed translation and monitoring services for several federal government agencies. He was stationed in Bosnia for roughly a year for military intelligence. At pretrial, he explained his process of translation and transcription. First, he listened to the call all the way through, writing notes summarizing what he heard. He said that he listened to the calls two or three times in the process of translation and transcription. He marked inaudible portions of the transcript. Later, he went back and verified the accuracy of his work. While listening to the calls, he read his transcript. He certified the accuracy of "everything" he did.
(4) Haxhi Gjonbalaj grew up in the former Yugoslavia. Both of his parents spoke Bosnian. He spoke Albanian and Bosnian in school. Mr. Gjonbalaj moved to the United States after six years of schooling in Yugoslavia, where he spoke English in school and Bosnian at home. He works for an interpreting company that provides interpretation, translation, and foreign language transcription services. His clients have included the FBI, ICE, and DEA. In translating and transcribing calls, Mr. Gjonbalaj initially wrote down everything verbatim. He confirmed that after translating and transcribing calls, he checked the quality of his own work by listening to the audio "over and over" while following along with the transcript. He certified the accuracy of any transcript bearing his initials. Asked directly by the Court, defense counsel were unable to point to any inaccuracies in Mr. Gjonbalaj's work.
(5) Radmilo Bozinovic is from the former Yugoslavia; Serbo-Croatian is his native language. He is currently a part-time interpreter for the judicial council of California, providing English/Serbo-Croatian interpreting services in the California court system. He is certified by the California court system to do that work. Here, Mr. Bozinovic began by taking notes on a given call. Next, he re-ran the call and filled in any blanks that remained. He then went back over his work "as many times as necessary to get the complete text." Once he was satisfied that he had captured the call, he would later re-listen to the call in its entirety while reviewing the transcript. When available, other call monitors reviewed his work. He certified that both the translations and transcriptions were accurate.
Several witnesses testified that Bosnian and Serbo-Croatian are essentially the same language, separated only by slight differences. Kavosevic, for instance, testified that there is "barely any difference" between the two. "There might be one or two words that may differentiate," he added, "but it's not really much of a difference at all."
All of this was more than enough to meet the requirement that a witness qualified in the relevant language testify to the accuracy of the government's translations. In United States v. Gomez, 67 F.3d 1515, 1525-26 (10th Cir. 1995), the court held it was not an abuse of discretion for the district court to qualify a witness who minored in Spanish and spent twenty-three months on a religious mission in South America. The witnesses in this case far exceeded that. Likewise, the testimony of these witnesses also satisfied the requirement that the transcriber verify the accuracy of the transcriptions. Each witness, all of whom transcribed as well as translated, vouched for the correctness of his or her work.
One cooperating defendant, Ernestina Hajiric, also corroborated the accuracy of some translations. She testified that transcripts of her calls played at trial were accurate, including the profanity.
The fact that these witnesses assisted with the investigation leading to trial is no obstacle. In United States v. Liddel, for instance, the Sixth Circuit held that the district court did not abuse its discretion by allowing a bilingual officer who otherwise served as a fact witness to also provide expert testimony as to a transcript's accuracy. 64 F. App'x at 963. The court acknowledged that this approach involved some risk but noted that "there can be no abuse of discretion in this type of situation when the Defendants involved did not point to any specific inaccuracies in the translation or offer any alternative translation." Id. (citing Garcia, 20 F.3d at 673); see also Abonce-Barrera, 257 F.3d at 965 (holding it was not plain error for court to allow active participant in investigation to testify as translation expert).
In the same vein, the defendants argue it would be inappropriate to permit some of these witnesses to offer expert translation testimony because they might be Serbian. People from Serbia, they contend, may harbor ethnic animosity towards Bosnians. This generalized, unsubstantiated claim of bias surely cannot disqualify them if witnesses apt to suffer a more particularized bias — translators who helped investigate the case — can certify transcript accuracy. What's more, most testified they were Bosnian, not Serbian, and the defendants failed to expose any bias.
None of the defendants' arguments is convincing. First, they insist that, to qualify as translators under Rule 702, the witnesses must also qualify as interpreters under 28 U.S.C. § 1827. R. 371, Attach. 1 at 2. Not so. Section 1827 has nothing to do with the qualification of translating witnesses under Rule 702. Rather, it lays the ground rules for "a program to facilitate the use of certified and otherwise qualified interpreters in judicial proceedings instituted by the United States." 28 U.S.C. § 1827. These witnesses are not real-time courtroom interpreters. Furthermore, the Sixth Circuit has indicated that would-be translation experts need only qualify under Rule 702. In United States v. Vitale, No. 89-1989, 1990 WL 57235, at *1 (6th Cir. May 2, 1990), the court upheld a district court's decision to qualify an FBI agent as a translation witness because the decision was not clearly erroneous. It then cited to United States v. August, 745 F.2d 400, 407 (6th Cir. 1984), which held that "[t]he decision to allow a witness to testify as an expert pursuant to Fed.R.Evid. 702 is left to the sound discretion of the trial judge and will not be disturbed . . . [unless] that ruling was clearly erroneous or an abuse of discretion." (emphasis added). Other courts have more directly held that Rule 702 governs whether a witness qualifies as a translator. See United States v. Contreras, No. 98-4546, 1999 WL 123760, at *2 (4th Cir. March 9, 1999) ("A witness's qualifications to render an expert opinion are liberally judged by Federal Rules of Evidence 702[.]"); Abonce-Barrera, 257 F.3d at 965 (assessing translator's qualifications under Rule 702).
For the same reason, Fed.R.Evid. 604 does not apply: "An interpreter is subject to the provisions of these rules relating to qualification as an expert and the administration of an oath or affirmation to make a true translation." (emphasis added).
Relatedly, another defendant argues — without citation to legal authority — that these witnesses must have been previously qualified as competent to testify in a proceeding in federal court. R. 370 at 5. But, as the Ninth Circuit recognizes, there is no such requirement. Abonce-Barnera, 257 F.3d at 965 ("However, there is nothing in Rule 702 that requires an expert [in Spanish-language transcription and translation] to have been previously qualified as an expert; such an approach would lead to absurd results.").
Next, the defendants contend that these witnesses' testimony is insufficiently reliable under Rule 702 because 1) the translators have not been "tested" — i.e., the defendants have not been given a "literal transcript of the phone calls in their original language so as to allow a comparison to the translation offered by the government," and 2) before trial, the government provided no indication as to the "methods or principles by which these transcribed translations were created, let alone any means by which to test the application of those methods and principles to the given facts." R. 371, Attach. 1 at 1-2. Yet neither Garcia nor Liddel hint that either of these steps is required. Furthermore, there is no requirement that these witnesses' opinions be subject to testing to be deemed reliable under Rule 702. As the Supreme Court explained, the criteria used to determine reliability will vary from case to case. Where the witness's expertise does not "rest[] upon scientific foundations," the "relevant reliability concerns may focus upon personal knowledge or experience." Kumho Tire Co. v. Carmichael, 526 U.S. 137, 149-50 (1999). In addition, the defendants did have the opportunity to test the accuracy of the translations themselves, but declined to make any specific objections or offer an alternative transcript. And the translators testified to the method of compiling these transcripts at the pretrial hearing.
A final objection: One defendant argues that the Court has an obligation to review the transcripts itself in camera to verify their accuracy. R. 315 at 1. It is true that, for recordings and transcripts entirely in English, courts have an affirmative obligation to compare the transcript to the audio when the parties do not stipulate to its accuracy. United States v. Rodriguez, 98 F. App'x 469, 470-71 (6th Cir. 2004); see also United States v. Robinson, 707 F.2d 872, 879 (6th Cir. 1983) ("The court should also make an independent determination of accuracy by reading the transcript against the tape."). But in foreign language cases, the court only needs to do this for disputed portions of the transcript. In Rodriguez, for example, the district court only compared the transcript with an in-court translation of the audio for the disputed portion of the transcript. 98 F. App'x at 470-71. And in United States v. Garcia, the Sixth Circuit affirmed a trial court's decision to admit a translated transcript merely because a professional translator testified to its accuracy at pretrial. 20 F.3d 670, 673 (6th Cir. 1993). The court made no mention of any effort by the trial court to read the transcript while an interpreter translated the audio in real time. Id. Here, the defendants made no specific objections to the foreign language transcripts. And for the English language transcripts, the Court asked defense counsel which specific transcripts they contest as inaccurate. In response, counsel confirmed the English language transcripts were accurate in all material respects. The Court still monitored the English recordings and transcripts during trial, striking one transcript sua sponte for inaccuracy.
Turning to the third requirement — that the Court screen out transcripts based on inaudible tapes — the Court instructed the defendants to file any objections to the transcripts on the basis of the underlying recording's audibility. None did so. What's more, the procedure used to prepare the transcripts mitigated any problems. The translator/transcribers testified that they flagged inaudible calls on the transcripts.
Even though the defendants failed to properly object to introduction of the transcripts, and even though the government followed the appropriate procedure for introduction, the Court took two prophylactic measures to guard against any lingering prejudice. It repeatedly offered limiting instructions — separate instructions for foreign language and English language transcripts — advising the jury that the transcripts were mere aids and that the tapes were the evidence. See United States v. Jacob, 377 F.3d 573, 582 (6th Cir. 2004) ("Even if the judge erred in failing to make explicit findings concerning the accuracy of the transcript, any error was harmless. Not only did the district court repeatedly instruct the jurors that the transcript was not evidence, but it also emphasized to the jurors that if they did not hear what Agent Riolo said he heard, then they should disregard what he wrote."); United States v. Wilkinson, 53 F.3d 757, 762 (6th Cir. 1995) ("Finally, any potential prejudice stemming from the use of the transcript was also remedied by a cautionary jury instruction."); United States v. Hughes, 895 F.2d 1135, 1147 n. 22 (6th Cir. 1990) ("In its instruction to the jury, the district court properly cautioned that the transcripts were presented for the purposes of aiding their understanding of the tape, clarifying portions that were difficult to hear, and identifying speakers. The district court stated, `It would be a violation of your sworn duties as jurors to utilize the content of any of the transcripts in determining innocence or guilt. . . . Only the tape is evidence. The transcript is not evidence.'"). Additionally, the Court ruled that the jury could not take the transcripts back into the jury room during deliberations. Rather, the Court invited the jurors to notify the Court if they would like to listen to any portion of the tape in open court and follow along with the transcript. See United States v. King, 272 F.3d 366, 374 (6th Cir. 2001) (noting the district court's decision not to send the transcript back with the jury as one reason not to find an abuse of discretion); but see United States v. Scarborough, 43 F.3d 1021, 1024-25 (6th Cir. 1994) ("As long as the trial court instructs the jury that the tapes and not the transcripts are evidence it is not error to allow a jury to have transcripts in deliberations, even if the transcripts were not admitted into evidence."). Hence, the Court employed more than adequate procedural protections to ensure the accuracy of the transcripts and guard against any possible prejudice. The defendants' motions to strike the transcripts are denied.
The Court altered the instruction for foreign language transcripts to remove the invitation for the jurors to compare the text of the transcript to the audio themselves and decide whether the transcript was accurate. Such an instruction would invite bilingual jurors to translate the audio, and the Ninth Circuit has held this may be error. See United States v. Fuentes-Montijo, 68 F.3d 352, 255-56 (9th Cir. 1995).
"Members of the jury, the government will give you a transcript of the oral conversation which can be heard on the tape recording received in evidence. The transcript purports to identify the speakers engaged in the conversation. I have admitted the transcript for the limited and secondary purpose of aiding you in following the content of the conversation as you listen to the tape recording and also to aid you in identifying the speakers. The transcripts are not evidence and will not go to the jury room with you. Only the recordings are evidence.
You are specifically instructed that whether the transcript correctly or incorrectly reflects the content of the conversation or the identity of the speakers is entirely for you to determine, based upon your own evaluation of the testimony you have heard concerning the preparation of the transcript, the method of identifying the voices, and any evidence or testimony regarding the content of the recordings. If you should determine that the transcript is any respect incorrect or unreliable, you should disregard it to that extent."
"Members of the jury, the transcripts have been identified as a typewritten transcript of the oral conversation which can be heard on the tape recordings received in evidence as [exhibit numbers]. The transcripts also purport to identify the speakers engaged in the conversation. I have admitted the transcript for the limited and secondary purpose of aiding you in following the content of the conversation as you listen to the tape recording and also to aid you in identifying the speakers.
However, you are specifically instructed that whether the transcript correctly or incorrectly reflects the content of the conversation or the identity of the speakers is entirely for you to determine, based upon your own evaluation of the testimony you have heard concerning the preparation of the transcript and from your own examination of the transcript in relation to your hearing of the tape recording itself as the primary evidence of its own contents, and if you should determine that the transcript is any respect incorrect or unreliable, you should disregard it to that extent."
II. The government properly included voice identifications in its transcripts
The government followed the procedure for properly identifying speakers in its transcripts. The Sixth Circuit has twice held that a transcript's proponent can identify voices in transcripts if an identifying witness is available for cross examination. In United States v. Crane, the court affirmed the district court's decision to permit the identification of speakers in the body of a transcript and in accompanying cover letters because the identifying witness "was available for cross-examination." 632 F.2d 663, 664-65 (6th Cir. 1980) (per curiam). The Sixth Circuit recently came to the same conclusion in United States v. Ford, 187 F. App'x 496, 500-01 (6th Cir. 2006) ("The defendant also challenges the transcript on the ground that it identified him as one of the speakers in the recorded conversations. . . . [The identifying witnesses'] availability for cross-examination negates any prejudice to Ford.") (citation omitted). Other courts agree. Here, the government called a cooperating witness who purportedly worked with the defendants, Halil Batlak, to identify the speakers in nearly every recording based on his own familiarity with the speakers' voices. It also called Agents Madden and Falascino, who identified the speakers on several English language transcripts. Furthermore, case agent Hughes and the translator/transcribers explained how they originally identified each of the defendants' voices, grew familiar with them, and identified them in the transcripts.
See, e.g., United States v. Garcia, 334 F. App'x 609, 616-17 n. 31 (5th Cir. 2009); United States v. Scurry, No. 96-4249, 1997 WL 447188, at *2 (4th Cir. Aug. 7, 1997); United States v. Capers, 61 F.3d 1100, 1103 (4th Cir. 1995); Ulerio, 859 F.2d at 1146; United States v. Sease, No. 06-20304, 2009 WL 81324, *1-2 (W.D. Tenn. Jan. 12, 2009); United States v. Moskey, No. 89-cr-669, 1990 WL 77902, at *1-2 (N.D. Ill. May 17, 1990).
In the rare instance Batlak was unable to corroborate a voice attributed to one of the defendants, the Court required that the attribution be redacted.
Despite their argument that including speaker identifications in translated transcripts is uniquely prejudicial, R. 452, Attach. 2 at 2, or should be held to a higher standard, R. 417 at 2, the defendants point to no authority to suggest a different standard should apply in this foreign language case. In fact, courts have held just the opposite. See, e.g., United States v. Ulerio, 859 F.2d 1144, 1146 (2d Cir. 1988); United States v. Nochez, No. 08-0730, 2010 WL 807447 (N.D. Cal. March 5, 2010). The Second Circuit specifically rejected an argument that more rigorous scrutiny should be applied to district courts' decisions to permit speaker identification in translated transcripts where the identifying witness testified before the jury. United States v. Moyhernandez, 17 F. App.'x 62, 70 (2d Cir. 2001).
The defendants appear to argue that, in addition to calling these witnesses, Rule 901 obligated the government to authenticate the speakers' identities on the transcripts to the Court's satisfaction before showing them to the jury. See Fed.R.Evid. 901(a) ("The requirement of authentication or identification as a condition precedent to admissibility is satisfied by evidence sufficient to support a finding that the matter in question is what its proponent claims."); see also R. 417 at 2 ("Under these circumstances, it is respectfully submitted that the Court must be extremely careful in assuring the reliability of everything in the transcript, specifically the identity of the speaker."), 3 ("[T]ranscripts must be authenticated[.]"); R. 425, Attach. 2 at 3. But this speaker authentication requirement applies to the tape recordings, not the transcripts. 2 McCORMICK ON EVIDENCE § 228 (6th ed. 2009) ("In such cases, if the recording is chosen as the mode of presenting the communication at trial, the recording itself must be authenticated as well as the voice or voices identified."); see also United States v. Simms, 351 F. App'x 64, 68 (6th Cir. 2009) (indicating that voice identification is part of authenticating audio recordings). The defendants did not object to the admittance of the recordings into evidence.
Even if the defendants were right, the government did "authenticate" the speaker identities in the transcripts before their introduction. Evidence is properly authenticated if "a reasonable jury could find in favor of authenticity or identification." United States v. Jones, 107 F.3d 1147, 1150 n. 1 (6th Cir. 1997) (quoting 5 JACK B. WEINSTEIN ET AL., WEINSTEIN'S EVIDENCE ¶ 901(a), at 901-19 (1996)). A party can authenticate a speaker's identity in two ways. First, a witness can identify the speaker based on familiarity gained at "any time" under circumstances connecting the voice with the alleged speaker. Simms, 351 F. App'x at 68 ("For voice identification, opinion testimony by someone who heard the voice `at any time under circumstances connecting it with the alleged speaker' can be used for authentication." (quoting Fed.R.Evid. 901(b)(5)). For example, in United States v. Mitchell, the Sixth Circuit held that a voice was properly identified under Rule 901(b)(5) because the witness testified that he "was very familiar with [the person's] voice because he monitored `dozens of phone calls coming out of the jail to his number at 1602 Hanauer." 104 F. App'x 544, 548 (6th Cir. 2004). Despite the defendants' arguments otherwise, R. 417 at 3; R. 420 at 2; R. 425, Attach 2 at 3, the identifying witness need not have first hand familiarity. Mitchell, 104 F. App'x at 548. Second, a litigant can identify a voice with circumstantial evidence. In United States v. Adams, the identifying witness "could not identify [the speaker] from his own familiarity with [the speaker's] voice." No. 85-1330, 85-1338, 1986 WL 17714, at*5 (6th Cir. Sept. 22, 1986). But the voice was nonetheless appropriately identified by other circumstantial evidence. The court held that it was true that "self-identification by the recorded speaker is not sufficient" to satisfy the authenticity requirement. But the speaker was sufficiently identified because he identified himself and the calls were made to numbers subscribed to by the speaker. Id.
Though, this was a sufficiency of the evidence challenge to voice identification, not a challenge to the voice's authentication.
The government used both approaches to authenticate the transcript voice identifications in this case. First, it called cooperating defendant Halil Batlak and asked him to listen to an excerpt of nearly every call the government planned to play for the jury. Strikingly, every identification Batlak made was consistent with all the identifications listed in the transcripts. Other witnesses authenticated the voices in smaller collections of transcripts. For recordings taken from a body wire placed on a cooperating witness at the early stages of the investigation, Agent Madden testified to the identity of the speakers. At the very least, Madden was familiar with the voice of his own cooperating witness — which is all that is required. See United States v. Degaglia, 913 F.2d 372, 376 (7th Cir. 1990) (holding that a tape was admissible where the identifying witness "positively identified one of the voices on the tape as belonging to" the defendant because "[n]othing more was required" by Rule 901). Officer Nick Falascino also authenticated the voices on a series of wiretap recordings taken during an investigation in Pittsburgh, as well as the attendant transcripts. He identified Omer Dugalic in the calls because his interlocutor identified him by name, physical surveillance of Omer Dugalic was consistent with the content of the calls, and Officer Falascino gained familiarity with his voice while listening to "thousands" of calls.
For this reason, Omer Dugalic's unsupported claim that an identifying witness must identify all voices on the tape is incorrect.
Second, even though it called these witnesses to identify the speakers based on their personal familiarity with the voices, the government still presented circumstantial evidence to corroborate its speaker identifications before introducing transcripts stemming from the six wiretaps in this case. At pretrial and at trial, Special Agents Dalrypple and Hughes explained how the investigators determined the identities of the participants in the recorded conversations. At trial, Special Agent Hughes offered the following account:
(1) First, they identified Kemal Dugalic as a participant in the conversations. After a cooperating witness purchased drugs from him while wearing a wire, the investigators secured a wiretap on a phone subscribed to Kemal Dugalic by name. Mr. Dugalic identified himself on the wiretapped calls. And the investigating agents, while conducting surveillance, observed Kemal Dugalic behave consistently with what was on the calls. The cooperating witness and Agent Hughes both reviewed recordings of the wiretap and confirmed Dugalic's voice.
(2) Next, the investigators identified Omer Dugalic as another participant in the calls. Kemal Dugalic called Omer Dugalic, his brother, and referred to him by name. Omer Dugalic's phone was subscribed in his name.
(3) Emir Dadanovic followed. The investigators heard reference to a person named "Emir" and "Dida" in the calls. Then, "Emir" or "Dida" himself participated in calls subject to wiretap. The police, conducting surveillance of Emir Dadanovic, then observed him behaving in a manner consistent with what "Emir" and "Dida" discussed in the calls. When Dadanovic was arrested, the police found "Dida" tattooed on his back.
(4) Investigators identified Jerdin Ovidio Yanes as a participant in the calls when they observed him engaging in conduct consistent with that discussed by a person named "Gordo" on the calls.
(5) Finally, investigators identified a speaker referred to as "Donta" in the calls as Donta Hamilton. "Donta's" phone was listed under the name "Donta Hamilton." The phone was also subscribed to a house owned by Donta Hamilton. Further, the speaker "Donta" made reference to a traffic stop — just as Donta Hamilton was stopped and searched by the police.
Having identified the speakers on the calls, the investigators grew familiar with the voices they associated with each of the defendants. They then advised the transcribers which voices matched which participants. The transcribers were also able to identify speakers when they referred to one another by name. Hence, the government also authenticated the identities of the speakers through circumstantial evidence. And they did a good job. No defendant has pointed to any inaccuracies in the transcript speaker identifications.
At least one defendant argued before trial that it is uniquely prejudicial to identify speakers in foreign language transcripts before the government has proven the speakers' identities. R. 425, Attach. 2 at 2. Whatever merit this contention had was lost when the government called Batlak, Special Agents Hughes and Dalyrpple, Detective Madden, the translator/transcribers, and Agent Faliscino to identify the speakers in all of the transcripts before their introduction.
To guard against any lingering prejudice, the Court took several prophylactic measures. The Court required the government to redact any identifications of any of the defendants in the transcripts that were not specifically corroborated by witnesses. Next, the Court admonished the jury — repeatedly — that they must be the judge as to whether the voice attributions were correct. See, e.g, United States v. Nixon, 918 F.2d 895, 901 (11th Cir. 1990); United States v. Sease, No. 06-20304, 2009 WL 81324, at *2 (W.D. Tenn. Jan. 12, 2009). And the Court did not permit the jury to take the transcripts into the jury room for deliberations. United States v. Berry, 92 F.3d 597, 600 (7th Cir. 1996). Accordingly, the government's motion for leave to identity the speakers in transcripts was appropriately granted.
CONCLUSION
For the foregoing reasons, it is ORDERED:(1) The defendants' motions objecting to the transcripts, R. 299, R. 315, R. 370, R. 371, are DENIED.
(2) The government's motion for leave to identify speakers in transcripts, R. 384, is GRANTED.
This the 13th day of September, 2010.