United States v. Carlos Mendiola

In the
United States Court of Appeals
For the Seventh Circuit
No. 10-1595
U NITED STATES OF A MERICA,
Plaintiff-Appellee,
v.
C ARLOS M ENDIOLA,
Defendant-Appellant.
Appeal from the United States District Court
for the Northern District of Illinois, Eastern Division.
No. 06 CR 825-2—Robert W. Gettleman, Judge.
A RGUED JANUARY 6, 2011—D ECIDED F EBRUARY 11, 2013
Before E ASTERBROOK, Chief Judge, and C UDAHY and
R OVNER, Circuit Judges.
R OVNER, Circuit Judge. A Spanish-speaking linguist
working for the Drug Enforcement Administration (DEA)
listened to recordings of Carlos Mendiola’s prison tele-
phone conversations prior to testifying before a jury
that Mendiola’s voice was likely the one on several
wiretapped calls in which Mendiola and others planned
a large-scale cocaine deal. Mendiola appeals his convic-
2 No. 10-1595
tion, arguing that the linguist’s testimony constituted
impermissible opinion testimony under the Federal
Rules of Evidence and violated the Best Evidence Rule
to boot. Finding neither of these arguments holds sway,
we affirm.
DEA agents suspicious of Alfredo Galindo Villalobos
(Galindo) began legally monitoring his telephone con-
versations in September 2006. Through these conversa-
tions, the agents learned that Galindo was trafficking
in drugs with someone whose name was Carlos, but
who went by the nickname “Pelon.” They also discovered
that Galindo and Carlos were expecting a large ship-
ment of cocaine from Mexico to be delivered to them
in Chicago. In the late morning of November 1, 2002,
agents observed a blue Ford Explorer linked to Galindo
in the vicinity of a bus station on the southwest side of
Chicago. Agents knew that a bus was scheduled to
arrive that morning from a border city in Texas. Galindo,
the driver, and his passenger, Mendiola, picked up
three men from the bus, two of whom were carrying
duffel bags. During a subsequent traffic stop, Galindo
and Mendiola each produced Florida driver’s licenses
with matching addresses. The back seat passengers were
later identified as Jose Valadez, Ricardo Mendoza, and
Juan Diaz-Casales. After all the occupants of the car
consented to a search, the police discovered approxi-
mately 5,000 grams of cocaine and three pairs of pants
containing hidden pockets filled with cocaine. In order
to continue their investigation and ferret out more par-
ticipants, the officers staged what they refer to as a
“rip.” They acted as though they were dirty cops con-
No. 10-1595 3
fiscating the cocaine they found for themselves in ex-
change for releasing the dealers. The plan worked, and
three weeks later the agents engaged in a nationwide
“takedown” in order to arrest members of the drug traf-
ficking organization, execute warrants, and seize drugs
and vehicles, including the blue Ford Explorer and a
white Lexus, both of which were found on the street
near Mendiola’s residence and both of which contained
hidden compartments secreting cocaine. After agents
gave Mendiola his Miranda warnings, he told officers
that the drugs in the car belonged to Galindo and that
Galindo had given the drugs to him because there was
“some kind of problem with the quality,” and that he
had forgotten about them because they had been in
the car for some time.
Galindo, Mendoza, and Valadez all pleaded guilty to
conspiracy and testified at Mendiola’s trial, implicating
him as an active participant in the conspiracy to import
and distribute multiple kilograms of cocaine. Rubiel
Mendiola, the defendant’s brother who was also ar-
rested and implicated, and Juan Diaz-Casales are
fugitives and have not been found. Each of the other
three co-conspirators testified that Mendiola recruited
Mendoza and Valadez to smuggle cocaine from Mexico
to Chicago, offering them $2,000 per kilogram trans-
ported. The details of the three co-conspirators’
stories were substantially consistent with each other
and with the DEA account. In his brief, Mendiola inven-
tories each inconsistency, and we acknowledge that
the testimony of the co-conspirators, like that of many
drug traffickers, was less than pristine.
4 No. 10-1595
Galindo testified the most extensively about Mendiola’s
active role in the conspiracy, explaining how Mendiola
transported cocaine, collected money, arranged to send
narcotics proceeds back to Mexico, picked up couriers
who were transporting cocaine from Mexico, and pack-
aged money to be transported back to Mexico. He
also detailed Mendiola’s part in the November 1 incident
in which Mendiola and Galindo retrieved the couriers
and drugs at the bus station. The agents involved in
the initial seizure at the bus station and the eventual
arrest corroborated the testimony of the cooperating
defendants. The other evidence presented to the jury
included multiple stipulations and physical evidence
including the clothing with hidden pockets, drug
evidence, surveillance photographs, materials used to
package the drugs, a firearm, and the English transcripts
of the intercepted Title III wire intercepts involving
Mendiola and his co-conspirators.
Those intercepted calls played an important role in
the prosecution because they attributed particular acts
and responsibilities to Mendiola. Co-conspirator Galindo
identified Mendiola’s voice on several incriminating
recorded calls describing multiple aspects of the conspir-
acy. Those recordings, with Galindo’s identification
of Mendiola’s voice and nickname, provided evidence of
Mendiola’s participation in the conspiracy, including his
role in packaging the cocaine and money, arranging
cover loads to hide money sent back to Mexico, housing
the drug couriers, obtaining false identification cards,
and possessing firearms. At the end of the seven-day
trial, the jury found Mendiola guilty of three counts of
No. 10-1595 5
narcotics trafficking for which the district court judge
sentenced him to 151 months’ incarceration.
Fortunately for the prosecution, Mendiola’s trial did
not rise or fall on Galindo’s voice identification alone, for
Galindo, like many drug dealers hoping to secure
a better deal for themselves, was burdened by sig-
nificant credibility issues. DEA linguist, Georgina Nido
also identified Mendiola as the speaker on those inter-
cepted conversations. 1 Prior to trial, Mendiola stipulated
that a set of transcripts would be prepared for use at
trial and that the English translations of the Spanish-
language calls were authentic.2 Two days after the trial
began and the day before Galindo was due to testify,
Mendiola’s counsel informed the government that he
would stipulate to the translations of the transcripts
and the identities of the speakers for each transcript
with the exception of the identity of Mendiola himself.
The government then informed defense counsel that it
intended to call a DEA linguist to compare a known
voice exemplar of the defendant obtained from calls
recorded at the Metropolitan Correctional Center in
Chicago to the voices in selected calls offered into evi-
1
Nido actually worked for a private company, MVM Inc.,
which performs contract work for the DEA.
2
Common sense and our case law both dictate that juries
need transcripts of recorded conversations when those con-
versations take place in a foreign language and are admit-
ted into evidence before an English-speaking jury. United
States v. Cruz-Rea, 626 F.3d 929, 936 (7th Cir. 2010).
6 No. 10-1595
dence. Over objection from Mendiola, Nido testified
that the voice on four of the calls, “sounded very similar,
if not identical” to that on the voice exemplar of Mendiola.
App. R. 32-3, p. 699; D. Ct. R. 232, p. 151; Tr. 6/1/09, p. 509.
After the guilty verdict, Mendiola filed post-trial
motions requesting acquittal, or in the alternative, a
new trial, claiming, in part, that the district court erred
in admitting the DEA linguist’s voice authentication
testimony under Federal Rules of Evidence 701, 702,
and 1002. In rejecting the motion for acquittal or a new
trial, the district court determined that Nido had
sufficient familiarity with Mendiola’s voice and that the
prosecution did not tender Nido as an expert witness,
nor did it need to. Mendiola appeals to this court and
we affirm.
We review a district court’s evidentiary rulings for
abuse of discretion. United States v. Stadfeld, 689 F.3d 705,
712 (7th Cir. 2012). Mendiola bandies about the de novo
standard, but as our discussion will reveal, this was a
simple evidentiary ruling about whether Nido met the
requirements for identifying a voice or not. The dis-
trict court did not have to interpret the Federal Rules
of Evidence. This is just one of the ways in which
Mendiola’s 13,939-word, single-issue brief creates com-
plexities where there are none.3
In fact, there is but a single issue presented on appeal:
Whether the district court erred in admitting Nido’s
3
The limit for an appellate brief—even one in which the
parties must address multiple complex issues is 14,000
words. Fed. R. App. p. 28.1(e)(2)(A)(1).
No. 10-1595 7
voice identification. According to Mendiola, the initial
issue we need to address is what exactly Nido was
doing when she identified Mendiola as the speaker on
the recording. Mendiola argues that Nido was a wolf
in sheep’s clothing—or rather an expert in a lay wit-
ness’s clothing—trying to squeak in evidence as a
lay witness to avoid the more stringent qualification re-
quirements for expert testimony. See Fed. R. Evid. 701, 702.
It is Federal Rule of Evidence 901 (b), however,
which enunciates the amount and quality of evidence
sufficient to satisfy the requirement of voice identifica-
tion. It states that the following is sufficient evidence
of voice identification: “an opinion identifying a person’s
voice—whether heard firsthand or through mechanical
or electronic transmission or recording—based on
hearing the voice at any time under circumstances that
connect it with the alleged speaker.” Id. The ac-
companying notes state that “aural voice identification
is not a subject of expert testimony.” Fed. R. Evid. 901
advisory committee’s note to subdivision (b), example (5).
This Circuit has long agreed. United States v. Recendiz,
557 F.3d 511, 527 (7th Cir. 2009) (“In light of Rule 901, [the]
contention that the court erred in admitting [an agent’s
voice] identification because he was not qualified as an
expert is wholly meritless.”), United States v. Magana,
118 F.3d 1173, 1208 (7th Cir. 1997); United States v.
Degaglia, 913 F.2d 372, 375-76 (7th Cir. 1990). In short,
Mendiola did not need to be vetted as an expert prior
to identifying Mendiola’s voice.
Mendiola’s point seems to be that using a person who
is an “expert” in some tangential field (in this case, a
8 No. 10-1595
linguist who translates live wiretap conversations
from Spanish to English) as a lay witness for voice iden-
tification can confuse the jury into thinking that the
person is an expert in voice identification. Experts in
other areas of law enforcement, however, are routinely
used as lay voice identification witnesses, as they are
the ones who have often heard the wiretap, or had
an interview with a suspect. For example, in Mansoori,
the FBI language specialist who prepared the English
translations on the recorded conversations also identi-
fied the voices of the recordings as belonging to the
defendant and his brother after hearing the brothers
speak at a court proceeding. United States v. Mansoori, 304
F.3d 635, 665 (7th Cir. 2002). See also United States v. Cruz-
Rea, 626 F.3d 929, 935 (7th Cir. 2010) (DEA agent who
listened to voice exemplar 50-60 times was able to
identify speaker in recorded conversations); Recendiz,
557 F.3d at 527 (special agent participating in arrest and
interview identified voice as the same as one in wire-
tapped call); United States v. Ceballos, 385 F.3d 1120, 1124
(7th Cir. 2004) (Spanish language translator identified
the voice); Degaglia, 913 F.2d at 376 (voice identification
by DEA agent). Not surprisingly, prosecutors frequently
enlist language interpreters and translators to identify
voices in court as they are the ones who have listened
intently to the recorded or overheard conversations.
See, e.g., Ceballos, 385 F.3d at 1124 (Spanish interpreter
identified voice); United States v. Pulido, 69 F.3d 192, 197
(7th Cir. 1995) (FBI translator identified voice); United
States v. Garcia, 413 F.3d 201, 207 (2d Cir. 2005) (the prose-
cution did not present the interpreter as an expert in
No. 10-1595 9
voice identification, but as a lay witness who had
acquired considerable familiarity with the intercepted
voices from her work monitoring the wiretap); United
States. v. Rrapi, 175 F.3d 742, 751 (9th Cir. 1999) (The
FBI translator identified defendant’s voice discussing
the crime.).
Thus Nido’s qualifications in another area do not dis-
qualify her as a lay witness for voice identification pur-
poses provided she otherwise met the qualifications for
the latter role. Federal Rule of Evidence 901(b) permits
a witness to identify a voice “based upon hearing the
voice at any time under circumstances connecting it
with the alleged speaker.” The bar for familiarity is not
a high one. This court has held that hearing a de-
fendant’s voice once during a court proceeding satisfies
the minimal familiarity requirement. Mansoori, 304 F.3d
at 665; see also Recendiz, 557 F.3d at 527 (DEA agent
who listened to a recorded phone conversation between
defendant and another speaker and then spoke with
him during his arrest and post-arrest interview was
qualified to testify that defendant’s voice at the hearing
was the same one from recorded call); United States v.
Jones, 600 F.3d 847, 857-58 (7th Cir. 2010) (voice iden-
tification legally sufficient where a detective never
personally spoke with defendant, but on four or
five occasions heard defendant speak in a courtroom
uttering as little as two or three sentences each time);
United States v. Khorrami, 895 F.2d 1186, 1194 (7th Cir.
1990) (recorded phone conversations properly admitted
based in part on testimony of a lay witness who identi-
fied the defendant’s voice after making one call to the
10 No. 10-1595
defendant’s residence and comparing the voice of the
person who answered to the voice on the recordings);
United States v. Saulter, 60 F.3d 270, 276 (7th Cir. 1995) (two
short conversations during a drug purchase satisfied
the minimal familiarity requirement); United States v.
Grier, 866 F.2d 908, 921 (7th Cir. 1989) (recordings
properly admitted where FBI agent who had spoken
once with the defendants identified their voices on the
tapes). Questions concerning the accuracy of the iden-
tification in light of the amount of familiarity go to the
weight the jury accords to the identification, not its admis-
sibility. Mansoori, 304 F.3d at 665; Jones, 600 F.3d at 858.
Mendiola’s argument that Nido’s testimony was not
based on personal knowledge is a red herring, first
because Nido herself listened to both the recorded con-
versations and the exemplar recording, and second
because Nido met the “minimal familiarity test” for voice
identification under Federal Rule of Evidence 901(b)(5).
Although the single issue presented in this case is
whether the court erred by allowing Nido to identify
Mendiola’s voice on the recordings, and the federal rule
of evidence that addresses the requirements for evi-
dence identifying a person’s voice is Rule 901(b)(5),
the defendant fails to mention this rule even once in
forty-four pages of his opening briefing. Instead,
Mendiola spends the bulk of his brief discussing
Federal Rules of Evidence 701 and 702, which discuss
requirements for lay and expert witnesses respectively.
This is akin to discussing only the qualifications for
getting a driver’s license rather than the rules for
making a right turn on red, in a case where the sole
No. 10-1595 11
issue is whether the defendant was permitted to make
a right turn on red.
To be certain, Nido’s testimony must meet the require-
ments of both Rule 901 and Rule 701, but the 701 require-
ments are readily met in this case.4 Pursuant to Rule 701,
a witness who is not an expert may offer an opinion
when it is: “(a) rationally based on the witness’s percep-
tion; (b) helpful to clearly understanding the witness’s
testimony or to determining a fact in issue; and (c) not
based on scientific, technical, or other specialized knowl-
edge within the scope of Rule 702.” Fed. R. Evid. 701.
Requirement (c) is easily met. Nido was not proffered
as a witness nor did she need to be. Rule 901(b)(5) and
its commentary make it clear that expert testimony on
voice identification is not required.
Mendiola is simply incorrect that the requirement in
subsection (a) that her opinion be rationally based on
her own perception means that Nido must have had
“personal interaction with the defendant.” (Mendiola’s
opening brief at 16). Rule 701 does not require that
the witness actually have participated in the recorded
conversations. Saulter, 60 F.3d at 276. The requirement
4
If for some reason there was some conflict between Rule 901
and Rule 701, the specific language of Rule 901 would
control over the general language of Rule 701. United States
v. Kuecker, 740 F.2d 496, 502 (7th Cir. 1984). In this case
there is no conflict and the requirements of both rules
have been satisfied.
12 No. 10-1595
that lay opinion be based on the perception of the
witness imports into Rule 701 the personal knowledge
standard of Rule 602. United States v. Bush, 405 F.3d 909,
916, n. 2 (10th Cir. 2005). And the knowledge required
by Rule 602 is not absolute or unlimited knowledge but
simply that awareness of objects or events that begins
with sensory perception of them, a comprehension of
them, and an ability to testify at trial about them.
29 Charles Wright & Victor Gold, Federal Practice and
Procedure § 6254 (1st ed. 1997). See also Payne v. Pauley,
337 F.3d 767, 772 (7th Cir. 2003) (personal knowledge
may include reasonable inferences as long as those in-
ferences are grounded in observation or other first-
hand personal experience). Moreover, the specific rule
governing voice identification—Rule 901(b)(5)—clearly
contemplates that the interaction leading to identifica-
tion might not be in person: “[i]dentification of a voice,
whether heard firsthand or through mechanical or electronic
transmission or recording—based on hearing the voice at
any time under circumstances that connect it with the
alleged speaker.” Fed. R. Evid. 901(b)(5) (emphasis ours).
Nido listened to the recordings, compared them to
the exemplar, and was able to report her perceptions to
the jury. To reiterate, if there was a question as to the
quality of the perception, that went to the weight
the jury attributed to her comparison.
Mendiola also argues that Nido’s testimony fails
part (b) of Rule 701’s requirement that lay opinion testi-
mony be “helpful to clearly understanding the wit-
ness’s testimony or to determining a fact in issue.” Fed.
No. 10-1595 13
R. Evid. 701(b). Mendiola argues that Nido’s testimony
was not helpful because the jury could have listened
to the recordings and made its own determination as to
whether the voices matched. We have recently noted
that although Rule 701 requires that evidence be helpful,
the fact that a jury might have the same opinion as
the testifying witness does not negate the helpfulness of
the testimony. Cruz-Rea, 626 F.3d at 935; see also United
States v. Towns, 913 F.2d 434, 445 (7th Cir. 1990). More-
over, the recordings were in Spanish whereas the jury
members were not, we presume, fluent in Spanish (and
even if one were, it would not be proper for that juror
to act as translator for the group). One of the ways in
which we identify a person’s voice is through the
idiosyncrasies of his or her speech. Often regional
idioms or dialects give us away. Did the speaker say
“soda” or “pop?” Is the second person plural uttered as
“yous,” “you guys,” “y’all,” or “yinz?” Hearing a distinc-
tive New York accent in Chicago, or a Boston accent
in Birmingham for example, would help a jury or lay
witness identify a voice. It is highly unlikely that a non-
native Spanish speaker would be able to hear or identify
these regional idioms and dialectal differences. In fact,
recorded conversations in foreign languages present
unique issues for juries. To address these challenges, a
district court judge has wide discretion in determining
whether to allow juries to use written transcripts as aids
in listening to audiotape recordings. United States v.
Breland, 356 F.3d 787, 794 (7th Cir. 2004). We have em-
phasized, however, that in cases in which the recording
is in a language foreign to the jury, transcripts (along
with the proper admonishments about their use) are a
14 No. 10-1595
“virtual necessity.” Cruz-Rea, 626 F.3d at 936. It is simply
common sense that an English-speaking jury cannot
adequately identify voices in languages in which they
are not familiar or even fluent. This is why defendant
Mendiola’s myriad references to opinions in which
courts held that juries could compare photographs or
surveillance video themselves are inapt. (See Mendiola’s
brief at 17-22). In many instances, juries are indeed capable
of making comparisons between pictures, videos, and
human likenesses without any experience or particular
knowledge, in a way that an English-speaking jury listen-
ing to a Spanish language conversation cannot. This
is not to say that juries never need assistance from a
lay witness for visual identification. “Generally, a lay
witness may testify regarding the identity of a person
depicted in a surveillance photograph ‘if there is some
basis for concluding that the witness is more likely to
correctly identify the defendant from the photograph
than is the jury.’ ” United States v. White, 639 F.3d 331,
336 (7th Cir. 2011) (citing Towns, 913 F.2d at 445).
Mendiola argues that because Galindo had already
authenticated the recording, Nido’s identification
was not helpful and thus not allowed. As we noted
above, we have never held that testimony is unhelpful
merely because a jury might have the same opinion as
the testifying witness; Cruz-Rea, 626 F.3d at 935; nor
would it be unhelpful merely because another witness
has offered the same identification. It is true that
either Galindo or Nido could have authenticated the re-
cording—that is, made a prima facie showing that the
evidence was what the government purported it to be—
No. 10-1595 15
a wiretap recording on which Mendiola was speaking.
Authentication, however, does not require the pro-
ponent to prove beyond a reasonable doubt that the
evidence is what it purports to be. “The task of deciding
the evidence’s true authenticity and probative value is
left to the jury.” United States v. Fluker, 698 F.3d 988, 999
(7th Cir. 2012); United States v. Harvey, 117 F.3d 1044,
1049 (7th Cir. 1997). The government was entitled, there-
fore to put on as much evidence as the court would
tolerate to prove the true authenticity and fortify the
probative value.5
Galindo and Nido’s testimony each contributed dis-
tinctly to the government’s case. Although Galindo
had severe credibility issues, the testimony of other co-
conspirator drug dealers can be quite helpful in identi-
fying voices and describing conversations, as those co-
conspirators have knowledge of the group’s nicknames
and the terms being used in drug dealings in general,
and in these drug dealings specifically, as well as
having familiarity with the defendant’s tone and
method of speech. See Saulter, 60 F.3d at 276. Nido, on
the other hand, had less familiarity with Mendiola’s
voice, but far greater credibility and more experience
in listening to wiretap recordings.
As for Mendiola’s argument that Nido’s testimony
violated the Best Evidence Rule, one need only read the
5
The defendant never raised any issues about whether
the content of that recording was admissible, i.e., whether it
contained hearsay evidence, privileged communication, or
the like.
16 No. 10-1595
twenty-three words of the Best Evidence Rule to see
why it is inapplicable here:
An original writing, recording, or photograph is
required in order to prove its content unless these
rules or a federal statute provide otherwise.
Fed. R. Evid. 1002 (emphasis added). Despite Mendiola’s
confusion as to what constitutes the content of a re-
cording, Nido’s voice identification had no role in
proving the content of the recording. A person’s voice is
an identifying physical characteristic and does not con-
stitute the content of a communication. See Gilbert v.
California, 388 U.S. 263, 267 (1967), United States v.
Dionisio, 410 U.S. 1, 7; Hubanks v. Frank, 392 F.3d 926, 932
(7th Cir. 2004) (voice exemplars are not testimonial, but
merely demonstrate a physical property of a defendant).
Nido’s testimony was offered only to identify the
speaker on the recordings, not to prove, for example,
whether the quantity of drugs discussed was the actual
quantity of drugs involved in the transaction, whether
Valadez had flown directly from Mexico City or Laredo,
Texas, or whether Galindo was borrowing the truck
to transport cocaine or to move from one house to
another (see e.g., Mendiola’s brief at 5, 7). More impor-
tantly, Mendiola never requested that the court
submit this “best evidence”—the actual recording—to
the jury. See App. R. 32-3, pp. 35-36; D. Ct. R. 226, pp. 7-8;
Tr. 10/14/09 pp. 7-8. See also oral argument at 00:45-1:04.
We thus hold that the district court properly admitted
Ms. Nido’s voice identification testimony. As no error
occurred, we need not delve into the arguments
No. 10-1595 17
on harmless error. The judgment of the district court
is affirmed.
2-11-13