Date: 04-14-2014

Case Style: United States of America v. Sergio Velazquez

Case Number: 12-3992

Judge: Lipez

Court: United States Court of Appeals for the Third Circuit on appeal from the Eastern District of Pennsylvania (Philadelphia County)

Plaintiff's Attorney: Zane David Memeger, Esquire, United States Attorney, Robert A. Zauzmer, Esquire, Assistant United States Attorney, Bernadette McKeon, Esquire (Argued), Assistant United States Attorney, Joseph T. Labrum, III, Esq., Assistant United States Attorney, Office of United States Attorney, 615 Chestnut Street, Suite 1250, Philadelphia, PA 19106 Counsel for Appellee

Defendant's Attorney: Jerome Kaplan, Esquire (Argued), Kaplan, Kenegos & Kadin, 9150 Wilshire Boulevard, Suite 175, Beverly Hills, CA 90212, Counsel for Appellant

Description: We consider whether the right to a speedy trial guaranteed in the Sixth Amendment is violated when, after an initial effort to apprehend the defendant, the government’s effort for nearly five years consists only of running the defendant’s name a handful of times through the National Crime Information Center (“NCIC”) database, despite other available leads. Although the authorities in this case revived their efforts after the five-year lull, the defendant happened to be caught when he was arrested on an unrelated matter. Applying the four-factor test from Barker v. Wingo, 407 U.S. 514 (1972), and the Supreme Court’s elaboration of those factors in Doggett v. United States, 505 U.S. 647 (1992), we conclude that the government violated defendant’s speedy trial right. Thus we reverse the district court’s denial of defendant’s motion to dismiss on speedy trial grounds, and remand the case with instructions to dismiss the indictment with prejudice.

I.

We recount the basic facts of the investigation, drawing from the testimony and reports before the district

4

court detailing efforts to bring defendant Sergio Velazquez to trial.1

A. The drug investigation

The Drug Enforcement Administration (“DEA”) began investigating Velazquez in June 2005, after receiving a tip that he was interested in selling cocaine to a confidential informant. App. 149a. Velazquez traveled from California to Philadelphia to meet with co-defendant Pedro Curiel and the informant. App. 149a-50a. The informant wore a wire to that meeting, recording discussions of a plan to sell between five and ten kilograms of cocaine. App. 149a-53a. After the meeting, the DEA had Philadelphia police stop Curiel and Velazquez’s car to identify the men. App. 150a-51a. The DEA learned Velazquez’s name, that he had a California driver’s license, and that he listed his address as a postal box in the greater Los Angeles area (P.O. Box 2901, Bell Gardens, CA, hereafter “Box 2901”). App. 383a. Velazquez was not arrested. App. 152a. He returned to California.

The DEA monitored the informant’s calls with Velazquez. Id. On July 27, 2005, they tracked Curiel as he met with the informant and then met co-defendant Nelkis Gutierrez-Gainza at a truck stop. App. 152a-54a. Gutierrez-Gainza gave Curiel a sack and Curiel drove away. When agents then stopped Curiel they recovered the sack and determined it contained nine kilograms of cocaine. Id. They arrested the two co-defendants.

1 We borrow the general organization of the district court’s thorough factual recitation.

5

B. The initial search for Velazquez and Deputy Degan’s report

The co-defendants were indicted on August 2, 2005; a complaint and arrest warrant issued for Velazquez the next day. David Pedrini, a DEA special agent in Philadelphia, testified before the district court that he had a fellow agent from Los Angeles, Steve Pascoe, go to 6340 Woodward Avenue, Bell, California (“Woodward Avenue address”), an address that Pedrini had learned was “associated with” Velazquez. App. 155a-56a.2 According to Pedrini, Pascoe spoke to a man, a woman, and two children “to see if one of them was Mr. Velazquez and the results were negative, he was not at that address.” App. 155a. The DEA declared Velazquez a fugitive, turning over the task of apprehending him to the United States Marshals Service office in Philadelphia. App. 155a-56a. They gave the Marshals Service a personal history report on Velazquez, and told William Degan, the office’s deputy marshal assigned to the case, about the Woodward Avenue address. App. 157a.

In addition to searching for credit applications, department of motor vehicle reports, and records for any property or vehicle purchases by Velazquez, Degan entered the warrant into the NCIC database,3 and into a Marshals

2 It appears this link was based on registration paperwork for the postal box. See App. 388a (stating that the Woodward Avenue address was “[t]he address listed for” Velazquez’s postal box).

3 The NCIC is an “electronic clearinghouse of crime data that can be tapped into by virtually every criminal justice agency

6

Service information system called the Warrant Information Network. App. 277a-280a. Among the various pieces of information he entered were Velazquez’s name, aliases,4 his alien registration number, his Social Security number, a physical description, his driver’s license, and the Box 2901 address. Id. Entering Velazquez into the NCIC would allow any law enforcement agency, if it encountered Velazquez, to learn that he was wanted by checking his information against the database. The law enforcement agent could then take him into custody after verifying that the government intended to prosecute.

nationwide, 24 hours a day, 365 days a year.” FBI — National Crime Information Center, http://www.fbi.gov/about-us/cjis/ncic (last visited Jan. 23, 2014). The database is organized by “files” that contain sets of records. For example, Deputy Degan entered Velazquez’s information into the “Wanted Persons File,” which consists of records on individuals with outstanding warrants. FBI — NCIC Files, http://www.fbi.gov/about-us/cjis/ncic/ncic_files (last visited Jan. 23, 2014).

4 The report states that Velazquez was also known as Sergio Velazquez-Payan and Sergio Payan. App. 385a. These are permutations of his surname from his father (Velazquez) and from his mother (Payan). The government does not contend that these aliases are misleading. They are, in fact, a common result of using Spanish surnames in the United States. The report also uses “Velasquez,” but does not acknowledge this spelling change or label it an alias. See App. 386a.

7

Degan then prepared a collateral request, which is a request for investigative assistance from a Marshals Service office in another jurisdiction. App. 175a-76a. The request, dated October 7, 2005, sought help from the Marshals Service taskforce in the Los Angeles area, which consisted of marshals and local law enforcement officers. See App. 284a-85a, 385a. Degan’s request noted the unfruitful trip to the Woodward Avenue address, mentioned that DEA agents from Los Angeles “also checked a number of other addresses” — without identifying the addresses — and then offered a number of leads in the form of past addresses for Velazquez, as listed in databases:

- Velazquez’s mother’s name and a possible address for her in Pico Rivera, CA

- The Box 2901 address

- The address of a home in Bell Gardens, CA, that Velazquez appeared to have bought in 1999, and the name of a woman who bought it with him

- The phone number Velazquez used during the DEA investigation, with an indication that Velazquez’s calls may have been made from a calling card

- An address in Paramount, CA

- The Woodward Avenue address and its possible tie to Velazquez’s brother, Elias

- A Huntington Park, CA address associated with Cecilio Vasquez (relationship to Velazquez unknown)

- A further address in Bell, CA, associated with Velazquez’s mother App. 385-86a. The report concluded with this request, in relevant part: “Contact DEA [Special Agent] Scott Pascoe”

8

— the officer who had checked on the Woodward Avenue home — “regarding their efforts in Los Angeles to locate [Velazquez]” and “if all leads . . . are exhausted, please interview his parents” at the Pico Rivera address, “and his brother, Elias Velasquez.” App. 386a.

C. Deputy Degan’s testimony about the response to his report

Degan transferred from his position at the Marshals Service in the Eastern District of Pennsylvania in late November 2005, about seven weeks after he sent his report. App. 290a. At the district court’s hearing, Degan testified that, based on an exhibit showing that an officer at the Los Angeles Police Department ran a check for Velazquez through NCIC on October 31, 2005, it appeared that his collateral request was received. App. 289a, 403a. According to Degan, this NCIC check would have been “the first thing I would do before I’d go out and attempt to find him.” App. 289a. Degan agreed that the exhibit did “not indicate that they went out and talked to anybody.”5 App. 296a. Degan said that some communication between the requesting officer and the officer receiving the request would be by phone, but that it was standard practice to then make a written record of work done. App. 298a. He did not recall any

5 The district court stated that Degan “believed that work was underway on the request” because of the October 31 NCIC check, United States v. Velazquez, No. 2-05-cr-00432-003, 2012 WL 2094061, at *4 (E.D. Pa. June 11, 2012), but Degan’s testimony about the NCIC check only pertained to whether he thought the report was received, see App. 289a.

9

communications from the taskforce in Los Angeles about his request. App. 289a-90a, 300a-02a. D. Communication with Velazquez’s attorney;

superseding indictment

On November 1, 2005, an assistant U.S. attorney in Philadelphia sent a copy of the complaint and warrant to Velazquez’s counsel in California, Jerome Kaplan, and both sides stipulated below that Kaplan received these documents.6 App. 232a. The record does not explain how the assistant U.S. attorney knew Kaplan represented Velazquez, but Kaplan has represented that Velazquez hired him to contact the U.S. attorney to discuss a potential surrender. Appellant’s Reply Br. 11.7 Three weeks later, a superseding indictment was filed, charging Velazquez with one count of conspiracy to distribute and to possess with intent to distribute five kilograms or more of cocaine, and one count of knowingly and intentionally distributing, and aiding and abetting the distribution of, five kilograms of more of cocaine. See 21 U.S.C. §§ 841(a)(1), 841(b)(1)(A) (underlying drug statutes); 21 U.S.C. § 846 (conspiracy); 18 U.S.C. § 2 (aiding and abetting). The indictment was not sent to Kaplan. App. 244a. After the indictment, the government did not communicate

6 The parties did not explore whether Kaplan told Velazquez about these documents because they agreed such testimony would implicate the attorney-client privilege.

7 The record does not disclose when Velazquez hired Kaplan for this purpose. Kaplan is also Velazquez’s attorney on appeal.

10

again with Velazquez’s counsel. Neither Kaplan nor Velazquez appeared at the arraignment in Philadelphia in December 2005. Appointed local counsel entered an appearance on Velazquez’s behalf, but there is no indication that this counsel represented Velazquez in any further capacity, or told him of the indictment. The assistant U.S. attorney stated at the arraignment that “my understanding from my last conversation with Mr. Kaplan was that he was planning to surrender Mr. Velazquez, but that has not happened. And I have had no further contact with Mr. Kaplan . . . .” App. 8a-9a (alterations omitted).

E. Marshals Service and DEA efforts from

November 2005 to November 2010

From November 2005 until November 2010, authorities checked the NCIC eight times to see if any law enforcement agency had encountered Velazquez. App. 384a, 392a-403a. At the Marshals Service, someone in the Philadelphia office or the headquarters in Arlington, Virginia, checked the NCIC four of those eight times — in November 2007, January 2008, February 2008, and September 2009. Id. For his part, DEA Agent Pedrini in Philadelphia, one of the agents who had worked on the initial investigation, checked with the U.S. Attorney’s Office each year to make sure the office was still willing to prosecute. He also made sure the warrant was still active in the NCIC, and he contacted the Marshals Service to ask if they had any new information. App. 158a-61a.8 Records show DEA inquiries on the NCIC

8 The district court stated that the efforts of the Marshals Service and DEA from November 2005 to November 2010

11

in January 2007, September 2007, October 2008, and October 2010. App. 392a-403a. At some point, Pedrini put Velazquez on the “Most Wanted” section of the website for the DEA’s office in Philadelphia. App. 160.9 He testified that he believed “we could” have put Velazquez on the “Most Wanted” section for Los Angeles, where authorities believed he was living, but “it didn’t happen.” Id. In 2008, for reasons unclear from the record, the warrant for Velazquez was removed from the NCIC. Pedrini noticed the removal and contacted the Marshals Service, ensuring that the warrant was restored to the database. See App. 158a-61a.

included periodic checks of commercial databases (such as those that might reveal vehicle purchases). Velazquez, 2012 WL 2094061, at *4. This observation is not supported by the testimony from Deputy Marshal Enrico Ilagan that the court cited. See id.; App. 207a-08a (Ilagan clarifying that the example discussed happened in August 2005, as part of Deputy Degan’s work). In an example not cited by the district court, Ilagan testified that he thought another deputy ran a check in January 2008 that included a commercial database. App. 182a. Ilagan was testifying about an exhibit outlining the investigation, but the relevant descriptions for two database checks in January and February of 2008 state only “NCIC Criminal History ran” and “NCIC check/No new info.” App. 384a. The government conceded in its closing argument before the district court that these were only NCIC checks. App. 364a.

9 It is not clear at what point Pedrini took this step.

12

There was no testimony or documentation before the district court to show any further steps taken to find Velazquez in this five-year period. No law enforcement agency tried to visit the addresses Deputy Degan identified in his October 2005 collateral request for assistance; no one returned to the Woodward Avenue address where Velazquez’s brother lived; no one contacted his parents; no one contacted Agent Pascoe, the agent in Los Angeles who first checked the Woodward Avenue address, to find out what investigative steps had been taken; no one contacted Kaplan, Velazquez’s attorney, despite his conversation with a prosecutor before the indictment was issued; and no one conducted searches of commercial databases or other governmental databases beyond the NCIC. Those other databases could have been particularly helpful. They would have shown, for example, any new property records, traffic tickets, birth records, any records from the armed forces, or any immigration changes. See App. 164a-68a (admission of Agent Pedrini that such records and databases were available); see also App. 230a (Deputy Marshal Enrico Ilagan’s similar testimony with respect to automobile records).

Degan testified that it was “standard practice” for an officer working on a case to make a written record of steps taken, but that it was not unusual that he had received no written response from the Marshals taskforce in Los Angeles before he transferred. App. 298a-99a. He suggested that the absence of a written response could mean that someone was still working on the investigation. Id. Nevertheless, there is no evidence that anyone on the Marshals taskforce in Los Angeles wrote a response to Degan’s report after Degan’s departure in November 2005. It is not clear who had

13

responsibility for the investigation after Degan until November 2010, when Deputy Marshal Ilagan, also of the Philadelphia office of the Marshals Service, began working on the case. App. 173a. Degan testified that “these cases are considered a priority and it would have been reassigned to somebody.” App. 290a. Ilagan stated that he thought Deputy Marshal Cardinal, presumably in the same Philadelphia office,10 was working on the case at the time he started to work on it, but he did not know “how long she had the case,” and he could not say she was in charge over the interim five-year period. App. 339a-40a.

F. Efforts from November 2010 to June 2011; new collateral request

Ilagan’s first step was to run an NCIC check, along with a search of a Lexis-Nexis database that compiles information from public and commercial records. App. 174a-75a, 194a-95a. That search turned up Velazquez’s application to renew his California driver’s license, with a postal box address of Box 2037 (“Box 2037”) in Bell Gardens, CA, as well as a possible job for Velazquez at an auto repair shop in California. App. 174a.11 Ilagan recalled

10 It appears Deputy Marshal Cardinal was in the Philadelphia office, as there is no indication in the record that someone outside that office was ever in charge of the case.

11 Ilagan did not recall where in California the shop was located. App. 174a-75a. The record shows that Velazquez began using the Box 2037 address at least by 2010. See, e.g., App. 129a-31a, 136a. Velazquez has represented that Box 2037 and Box 2901 were in the same post office, Appellant’s

14

calling the garage and asking for Velazquez, but the business “didn’t know anything about him.” Id. Ilagan continued periodically to check databases. In June 2011, he came across a possible connection between a phone number listed on Velazquez’s paperwork for Box 2901 and a new address in Norwalk, CA. App. 175a, 186a; see also App. 388a.12 This possible connection led Ilagan to submit a new collateral request on June 22, 2011, for assistance from the Marshals Service taskforce in Los Angeles. App. 42a-43a. This request was the same type of request that Deputy Degan had written in 2005. Ilagan identified the Norwalk address, noted that Velazquez was receiving mail at Box 2901, and that the residential address listed for that box was the Woodward Avenue address, an “old address” that Velazquez “used in the past.” App. 388a. Ilagan noted that it was not clear if there was any connection between Velazquez and the two people living at the Norwalk address, but that it was possible he was living there because it “is only 7 miles away from [Velazquez’s] relatives that live [in the] Bell Gardens, CA area.” Id.

Br. 12 n.2, and the government does not contest this point. At some point Ilagan determined that both postal boxes were connected to the Woodward Avenue address that Velazquez at times listed for his home, and that turned out to be his brother’s house. App. 189a; see also App. 388a.

12 It appears that Ilagan found the phone number Velazquez provided on the registration form for his postal box, and then searched phone records for any residence connected to that phone number. See App. 388a. This phone was different from the one Velazquez was said to have used during the DEA investigation. Compare id. with App. 386a.

15

G. Investigative steps in the summer of 2011

Ilagan’s request was received by the Marshals Service taskforce and assigned to Deputy David Dominguez in Los Angeles.13 On July 7, 2011, Ilagan received an e-mail notifying him that it appeared Velazquez had applied to renew his permanent resident card because the FBI received fingerprint submissions from the application that matched Velazquez’s prints on file. App. 176a. Ilagan informed Dominguez and then tried to get United States Citizenship and Immigration Services (“USCIS”) to help apprehend Velazquez by arresting him when he went to the USCIS office to pick up his fingerprint card. App. 177a. The agency declined to help because of legal liability concerns. Id.

Ilagan then suggested to Dominguez that he “might have to sit and do surveillance” at the Woodward Avenue address. Id. Dominguez did not testify before the district court, but he did take notes of the steps he took. App. 314a-15a, 414a. Those notes reflect that he “sat surveillance” at the Woodward Avenue address on the afternoon of July 15, 2011, noted license plate numbers for two vehicles in the driveway, but did not find Velazquez. App. 414a. He contacted Velazquez’s post office and learned that employees there recognized Velazquez, that Velazquez came often to

13 The record is silent on which law enforcement agency employed Deputy Dominguez. This is the first time in the record that a Marshals Service taskforce officer in Los Angeles is identified by name as having responded to a collateral request for help in finding Velazquez.

16

collect his mail, but did not do so at a particular time. Id. At some point that summer Dominguez performed surveillance for half a day at the post office, without finding Velazquez. Id. He also surveilled the Norwalk address on July 20, 2011, and went to an address in Baldwin Park, CA, that he thought might be connected to Velazquez. Id. None of these steps taken in the summer of 2011 was fruitful, nor was a search for payroll tax records for Velazquez. App. 412a-13a, 323a-26a.

H. December 2011 apprehension and subsequent guilty plea

Velazquez was apprehended on December 9, 2011, after police in Glendale, CA, arrested him on an unrelated narcotics charge. App. 384a, 393a. The DEA in Philadelphia confirmed that the government remained willing to prosecute Velazquez for the 2005 charges. App. 161a-62a. Velazquez was then served with the arrest warrant and extradited to the Eastern District of Pennsylvania. Id.

On March 28, 2012, Velazquez filed a motion to dismiss the indictment on the basis of a speedy trial violation. App. 62a. The district court denied the motion, Velazquez, 2012 WL 2094061, finding that the government was reasonably diligent in pursuing Velazquez. Thus Velazquez had to show specific prejudice to his defense from the long delay between indictment and arrest, and the court held that he did not make this showing. Id. at *13 (citing Doggett v. United States, 505 U.S. 647, 656 (1992)).

To support its finding of reasonable government diligence, the court explained that the failed efforts to find Velazquez at the Woodward Avenue address in August 2005

17

and July of 2011 meant that “the government was unlikely to find defendant based on the information available to it as of November 2005,” and thus “the government reasonably elected to conserve its resources and wait for new information or a change in circumstances.” Id. at *11. The court also stated that the government’s reasonable efforts would have found Velazquez earlier if not for his “transient” lifestyle. Id. at *10 (citing United States v. Mundt, 29 F.3d 233, 236 (6th Cir. 1994)).

Velazquez pleaded guilty on June 29, 2012, pursuant to a plea agreement in which he reserved his right to appeal the speedy trial issue. He was sentenced in October 2012 to 80 months in prison, with five years of supervised release. Velazquez then filed this timely appeal.14 We recount further facts below as needed for our analysis. II. Barker established a four-factor test for evaluating whether the constitutional right to a speedy trial has been violated. The inquiry focuses on: (1) the length of the delay before trial; (2) the reason for the delay and, specifically, whether the government or the defendant is more to blame; (3) the extent to which the defendant asserted his speedy trial right; and (4) the prejudice suffered by the defendant. 407 U.S. at 530-31. “[N]o one factor is dispositive nor ‘talismanic.’” Hakeem v. Beyer, 990 F.2d 750, 770 (3d Cir. 1993) (quoting Barker, 407 U.S. at 533). We review de novo

14 The district court had jurisdiction over this case under 18 U.S.C. § 3231. We have jurisdiction under 28 U.S.C. § 1291.

18

the district court's legal conclusion that Velazquez failed to establish a violation of his constitutional right to a speedy trial; we apply clear error review to the factual findings underpinning that legal conclusion. United States v. Battis, 589 F.3d 673, 677 (3d Cir. 2009); Burkett v. Fulcomer, 951 F.2d 1431, 1437 (3d Cir. 1991).

A. Length of delay

The first factor is actually “a double enquiry,” Doggett, 505 U.S. at 652:

Simply to trigger a speedy trial analysis, an accused must allege that the interval between accusation and trial has crossed the threshold dividing ordinary from “presumptively prejudicial” delay, since, by definition, he cannot complain that the government has denied him a “speedy” trial if it has, in fact, prosecuted his case with customary promptness. If the accused makes this showing, the court must then consider, as one factor among several, the extent to which the delay stretches beyond the bare minimum needed to trigger judicial examination of the claim. This latter enquiry is significant to the speedy trial analysis because . . . the presumption that pretrial delay has prejudiced the accused intensifies over time.

Id. (citations omitted). In other words, a court first decides whether the delay is long enough that it should trigger analysis of other Barker factors. Id. If it is, the length of the delay is also separately weighed in the court’s analysis of the

19

remaining factors. Id. The Supreme Court has noted that “the lower courts have generally found ‘postaccusation delay’” long enough to trigger further review of Barker factors “at least as that delay approaches one year.” Id. at 652 n.1. This Circuit has concluded that a delay of fourteen months is “sufficient to trigger review of the remaining Barker factors,” Battis, 589 F.3d at 678 (citing Hakeem, 990 F.2d at 760), and once that threshold has been passed, “the state, not the prisoner, bears the burden to justify the delay,” Hakeem, 990 F.2d at 770.

The district court correctly found here that the delay between the November 2005 superseding indictment and Velazquez’s scheduled trial date of July 2012 crossed the threshold of prejudicial delay to justify analysis of the remaining Barker factors. The government concedes this point, Gov. Br. 25, and understandably so. The Supreme Court has characterized delays on this scale as “extraordinary.” See Barker, 407 U.S. at 533 (using that term to describe a delay between arrest and trial of over five years); see also Doggett, 505 U.S. at 652 (noting that “the extraordinary 8 ½ year lag between Doggett’s indictment and arrest clearly suffices to trigger the speedy trial enquiry”). Accordingly, we note the strength of Velazquez’s showing on this factor and move on to the heart of the appeal.

B. The reason for the delay

This factor, the “flag all litigants seek to capture,” United States v. Loud Hawk, 474 U.S. 302, 315 (1986), appropriately consumed the bulk of the district court’s analysis. Doggett described the range of government effort to pursue an accused as extending from “reasonable diligence”

20

to “bad-faith delay.” 505 U.S. at 656. With the former, the defendant's speedy trial claim will fail “however great the delay, so long as [the accused] could not show specific prejudice to his defense.” Id. Bad-faith delay, meanwhile, “would make relief virtually automatic.” Id. at 657. Between the two is a “more neutral reason such as negligence,” which weighs against the government, albeit “less heavily” than a deliberate or bad-faith delay. Barker, 407 U.S. at 531. Negligence “still falls on the wrong side of the divide between acceptable and unacceptable reasons for delaying a criminal prosecution once it has begun.” Doggett, 505 U.S. at 657. Just as the government has the burden to prosecute a case, it also has the burden to justify a delay once the Barker enquiry has been triggered. See Hakeem, 990 F.2d at 770 (citing Barker, 407 U.S. at 527). If “the defendant is not attempting to avoid detection and the government makes no serious effort to find him, the government is considered negligent in its pursuit.” United States v. Mendoza, 530 F.3d 758, 763 (9th Cir. 2008) (citing Doggett, 505 U.S. at 653).

Negligence over a sufficiently long period can establish a general presumption that the defendant’s ability to present a defense is impaired, meaning that a defendant can prevail on his claim despite not having shown specific prejudice. See Doggett, 505 U.S. at 658 (finding a speedy trial violation based on general prejudice where government’s negligence led to six-year delay). This general presumption applies because “impairment of one's defense is the most difficult form of speedy trial prejudice to prove because time's erosion of exculpatory evidence and testimony ‘can rarely be shown.’” Id. at 655 (quoting Barker, 407 U.S. at 532). “Thus, we generally have to recognize that excessive delay presumptively compromises the reliability of a trial in ways

21

that neither party can prove or, for that matter, identify.” Doggett, 505 U.S. at 655.

This general presumption may be rebutted, but it is difficult to do so. The prosecution is essentially put to the test of proving a negative — the absence of any prejudice to a defense from the passage of years. See id. at 658 n.4 (noting the prosecution “ably counter[ed] Doggett’s efforts to demonstrate particularized trial prejudice [but] it has not, and probably could not have, affirmatively proved that the delay left his ability to defend himself unimpaired”) (emphasis added). As Doggett further noted, such a test is demanding, but it should not surprise the government:

[T]he weight we assign to official negligence compounds over time as the presumption of evidentiary prejudice grows. Thus, our toleration of such negligence varies inversely with its protractedness, and its consequent threat to the fairness of the accused's trial. Condoning prolonged and unjustifiable delays in prosecution would both penalize many defendants for the state's fault and simply encourage the government to gamble with the interests of criminal suspects assigned a low prosecutorial priority. The Government, indeed, can hardly complain too loudly, for persistent neglect in concluding a criminal prosecution indicates an uncommonly feeble interest in bringing an accused to justice; the more weight the Government attaches to securing a conviction, the harder it will try to get it.

22

Id. at 657 (citation omitted).

With these teachings in mind, we examine the two primary factual justifications for the district court’s “reasonable diligence” finding: (1) that the unfruitful trip to the Woodward Avenue address in 2005 and the investigative efforts in 2011 show that government inaction in the intervening years was a reasonable choice to “conserve [] resources,” Velazquez, 2012 WL 2094061, at *11, and thus comports with the requirements of reasonable diligence; and (2) that Velazquez had an elusive, if not evasive, lifestyle and thus bears responsibility for the delay, id. at *10. We review the trial court’s determination that the government was not negligent with considerable deference. See Doggett, 505 U.S. at 652. The factual findings supporting that determination are reviewed for clear error. “A finding is clearly erroneous when, although there is evidence to support it, the reviewing body on the entire evidence is left with the definite and firm conviction that a mistake has been committed.” United States v. Grier, 475 F.3d 556, 570 (3d Cir. 2007) (en banc) (internal quotation marks omitted).

1. Reasonable choice

In its factual recitation, the district court stated that the Marshals Service and the DEA efforts to find Velazquez between November 2005 and November 2010 “were limited to periodic checks of the NCIC Wanted Persons database and the commercial databases.” Velazquez, 2012 WL 2094061, at *4. According to the court, this limited effort reflected a “choice” to pursue other leads from 2005 to 2010, given that earlier efforts initiated by Deputy Degan, and later work by deputies Ilagan and Dominguez, did not succeed. See id. at

23

*10 (finding “the government’s choice to pursue other leads was reasonable”); see also id. at *11 (“The failure of the DEA’s efforts in California in 2005 and the inferred failure of the [Marshal Service’s] October 2005 collateral request show that the government was unlikely to find defendant based on the information available to it as of November 2005. Accordingly, the government reasonably elected to conserve its resources and wait . . . .”) (emphasis added). In support of this conclusion, the court first stated that the October 31, 2005, NCIC search, performed by someone at the Los Angeles Police Department, supported the inference that “authorities in California exhausted the leads in Deputy Degan’s collateral request.” Id. at *11. It did not adopt this inference, however, finding that even if the Los Angeles Marshals Service taskforce made “only cursory efforts” to pursue those leads, the fact that investigative efforts in 2011 did not quickly lead to Velazquez meant that any effort to find Velazquez “would likely have been fruitless.” Id.

There are two related findings from the district court here: first, that there was a tactical choice not to pursue Velazquez; and second, even if there was no tactical choice, the inaction was effectively “harmless” because later investigation shows that any earlier effort would have been fruitless. As an initial matter, we question the pertinence of either finding to the “reasonable diligence” inquiry here. The Supreme Court observed in Doggett that even if law enforcement inaction “may have reflected no more than [defendant's] relative unimportance in the world of drug trafficking, it was still findable negligence.” 505 U.S. at 653; see also id. at 657 (“Condoning prolonged and unjustifiable delays in prosecution would both penalize many defendants for the state’s fault and simply encourage the government to

24

gamble with the interests of criminal suspects assigned a low prosecutorial priority.”). We read this language to say that law enforcement priorities have little role to play in the negligence calculus. If authorities choose to ignore available leads about a suspect’s whereabouts in favor of other tasks, they may nonetheless be found negligent within the context of the speedy trial right.

Furthermore, we find no evidence in the record to support a finding that investigators made an actual “choice” not to pursue Velazquez or that the failure to pursue him was in any event harmless. Importantly, the government did not offer the tactical choice justification in the district court. See App. 94a-95a (contending that Velazquez was evasive and fled). The district court cited Deputy Ilagan’s testimony that his database checks in 2010 “came up with the same things that we already had” as the apparent reason the Marshals Service took no action from November 2005 to November 2010 beyond checks in the NCIC. Velazquez, 2012 WL 2094061, at *5 (citing App. 226a). But Ilagan’s testimony on what he did in 2010 does not shed light on possible tactical decisions made between 2005 and 2010.

With respect to that five-year period from November 2005 to November 2010, Ilagan testified only that he worked with another deputy (Cardinal) when he took over. But he did not know how long Cardinal had been handling the case or if anyone else had been in charge for any period since Degan. Degan testified that “these cases are considered a priority and it would have been reassigned to somebody,” App. 290a, but the record does not support a finding that any particular individual — including Deputy Cardinal — was that somebody. Consequently, there is no evidence of a

25

decision by anyone to forgo pursuit of Velazquez to conserve resources. Indeed, given the priority status that Degan said would ordinarily have been given to such a case, the five years of inaction are more indicative of inattention than an affirmative judgment about resources.15

15 Our dissenting colleague would affirm the district court because the court “heard the law enforcement witnesses and evidently believed that the limited investigative activity undertaken from 2005 to 2010 was the product of informed discretion.” The district court did not hear from any witness who worked on the case during the five-year lull, nor did it hear from any witness who even claimed to be able to speak to decisions taken during that period. Indeed, the government never advanced a conservation-of-resources argument before the district court, contending instead that “at no point did the Government cease in its attempts to locate” Velazquez, App. 247a. Just as it did for the “transient lifestyle” argument, see Section II.B.2, the government now adopts the district court’s reasoning on appeal. Our colleague further contends that “[i]t is not beyond the pale to believe that the person responsible for the case was making decisions about how to work it.” Respectfully, we hold the view that such a belief requires some foundation in the evidence. To the extent our colleague credits a possibility that Deputy Cardinal was the decision maker, this is an unsupported possibility that the district court did not even mention in its opinion. To affirm the reasonable diligence finding of the district court on the facts of this case would effectively ignore the burden on the government to justify the lengthy delay, and would reduce clear error review to a mere formality. Deferential review is still review.

26

The district court’s finding that attentive pursuit would have been fruitless was similarly in error. Degan’s collateral request to the Los Angeles Marshals Service taskforce in 2005 set out multiple avenues for investigation that Degan considered promising at that time. These included contacting the local DEA agent familiar with investigative steps already taken, exhausting the available leads, and interviewing Velazquez’s parents. None of these actions was taken during the nearly six years that followed. In the summer of 2011, however, Deputy Dominguez followed up on the new collateral request filed by Deputy Ilagan. Among other leads, Dominguez learned that Velazquez came often to collect mail from at a postal office in Bell Gardens, and Dominguez conducted a half day of surveillance there. Although Velazquez did not appear at the post office during those few hours, it does not follow that nothing would have come of a longer effort, or the same surveillance years earlier. Moreover, the post office lead might have been productive in 2005 if combined with other steps urged by Degan in his request, such as interviewing family members in an attempt to gain insight into Velazquez’s daily habits at that time.

We thus cannot uphold the district court’s “reasonable diligence” conclusion based either on its finding that the law enforcement authorities made a reasonable tactical choice to limit their efforts to find Velazquez or its finding that the authorities’ limited effort was sufficient because more ambitious pursuit also would have been unsuccessful. We therefore consider whether the court’s other articulated rationale — Velazquez’s lifestyle — adequately supports the finding of reasonable diligence during the period from 2005 to 2011.

27

2. Defendant’s conduct

The government seeks to attribute the delay in apprehending Velazquez to his own conduct — which it characterizes as evasive — rather than to the lack of effort by law enforcement authorities. At one point in its analysis, the district court appeared to agree that the evidence would support a finding of evasion. The court reasoned that the government’s inability to find Velazquez at the Woodward Avenue address, his lack of employment history, and his “ability to avoid apprehension until his arrest on unrelated charges, even after the government intensified its efforts in 2011” all “strongly supports the inference that defendant did hide.” Velazquez, 2012 WL 2094061 at *10. That inference would weigh heavily against Velazquez in assessing the government’s negligence. See Hakeem, 990 F.2d at 766.

The court refrained from actually finding “evasion,” however, in favor of characterizing Velazquez’s lifestyle as “transient.” Velazquez, 2012 WL 2094061 at *10. In the court’s view, even if Velazquez did not deliberately conceal his whereabouts, his lack of permanence nonetheless was responsible for the government’s inability to find him sooner: “[W]hether or not defendant was intentionally evading authorities, his lifestyle made it difficult for authorities to track him down. If defendant had not been so transient and if he had lived at his mailing address instead of using post office boxes, he would have been found much earlier as the [government] used conventional search methods in a reasonably diligent manner.”

28

Id. (quoting Mundt, 29 F.3d at 236 (second alteration in original)). The government did not frame its argument before the district court in terms of a “transient” lifestyle, see Gov. Response (App. 94a-96a), but it now adopts the district court’s reasoning as a separate ground for affirmance in addition to deliberate evasion, see Appellee’s Br. 31-34.

Although the district court ultimately refrained from finding “evasion,” we nonetheless believe we should assess that characterization because of its centrality to the government’s position and the court’s favorable view of it. Hence, we consider both whether the record supports a finding of deliberate evasion by Velazquez and whether the government can justify its meager search by blaming Velazquez’s “transient” lifestyle. As we explain, neither inquiry favors the government.

With respect to evasion, the factors cited by the district court do not support the inference of bad motive that the government urged and the court almost drew. A lack of verifiable employment, without more, does not signify an attempt to evade capture, and an individual’s choice to receive mail at a post office box or a relative’s home does not fill the gap. Velazquez had used his brother’s address and Box 2901 for years before the DEA investigation began, including when he sought to enlist in the United States Naval Reserve in 2002-03. Indeed, he used the same addresses in his 1998 application for a replacement Alien Registration Card. App. 75a. The duration of this usage negates any inference that these alternatives to a traditional home address were designed to protect him from a law enforcement manhunt.

29

Other evidence also undermines the inference of a furtive life. Various documents from 2010 onward show that Velazquez consistently listed Box 2037 as his address, and he went to retrieve his mail at the post office so frequently that postal employees recognized him. See, e.g., App. 130a (motorcycle title); App. 387a (driver’s license renewal). Velazquez listed the long-used Woodward Avenue address in paperwork for this postal box. See App. 388a. In addition, other public documents submitted during the period he was being sought also contained Velazquez’s name and identifying information. See, e.g., App. 122a-27a (application in 2011 to replace alien registration card); App. 78a-82a (birth records for three children born between October 2005 and February 2007). In brief, we see no correlation between Velazquez’s lifestyle and an intention to hide. Most significantly, there were no changes in his behavior over time that could be attributed to a deliberate effort after 2005 to evade detection.

We note that, as the government argues, the record would support a finding that Velazquez was aware that he was being sought in connection with the July 27, 2005 drug transaction in Philadelphia. Indeed, his lawyer was in touch with the United States Attorney’s Office in the fall of 2005 — before Velazquez’s indictment — to discuss surrender. Velazquez had no duty to bring on his own trial, however, and his lawyer’s inquiry does not diminish any governmental negligence in failing to pursue him, or to even contact his lawyer again. See Barker, 407 U.S. at 527 (“A defendant has no duty to bring himself to trial; the State has that duty.”) (citing Dickey v. Florida, 398 U.S. 30, 37-38 (1970)); see also Dickey, 398 U.S. at 50 (Brennan, J., concurring) (“The accused has no duty to bring on his trial.”); Mendoza, 530

30

F.3d at 763 (noting that “it was not [defendant’s] responsibility to contact the government during the investigation”). Although knowledge by Velazquez that he was being sought could contribute to an inference of evasion, it does not make the conduct we describe above more suggestive of hiding. Absent evidence of evasive conduct, Velazquez’s knowledge does not aid the government’s argument.

As for justifying the government’s inaction based on Velazquez’s “transient” life, we have serious doubts that this characterization is helpful in the reasonable diligence inquiry. To be sure, the government can only pursue reasonably available leads, and the absence of a paper trail for a defendant might leave the government with fewer avenues for investigation. Our focus is not the type of life a suspect leads, however, but whether the government has diligently used the information available to it. Describing a defendant as “transient” adds little to this analysis. Indeed, it carries the risk that the label will be used as a substitute for a detailed factual assessment of the government’s investigation, diluting the “serious effort” that Doggett requires of law enforcement authorities. 505 U.S. at 652. Moreover, this risk would likely be felt disproportionately by those in more limited economic circumstances, unfairly lessening the Sixth Amendment speedy-trial protection for those who are not so fortunate as to be well-rooted in society.16

16 We note that the “transient” lifestyle comment in Mundt on which the district court relied added little to the analysis in that case. The government periodically checked for the defendant at two motels and had verified that he occasionally

31

We thus conclude that neither premise for the district court’s finding that the government did not act negligently — a supposed alternative allocation of resources and the defendant’s way of life — withstands scrutiny. Hence, the court clearly erred to the extent it relied on those findings to hold that the government satisfied the “reasonable diligence” prong of the Barker test, and we therefore afford no deference to the court’s determination on negligence. Nevertheless, we must still consider whether the record supports the adequacy of the government’s efforts. See Burkett, 951 F.2d at 1441 (noting that, where district court erroneously attributed delay to the defendant and, hence, did not make a prejudice finding, appeals court “must exercise our power of plenary review to determine if the testimony establishes sufficient qualitative prejudice to weigh this factor in [defendant’s] favor”). In so doing, we continue to defer to the district court’s findings of fact as to the underlying events that are not clearly erroneous.

3. The reasonable diligence determination

To satisfy the requirement of reasonable diligence, the government does not need to make “heroic efforts” to pursue a suspect, Rayborn v. Scully, 858 F.2d 84, 90 (2d Cir. 1988), but it must at least make a “serious effort,” Doggett, 505 U.S. at 652. If “the defendant is not attempting to avoid detection and the government makes no serious effort to find him, the

stayed at each one. See 29 F.3d at 235. Investigators also tracked records in two states, leading to surveillance at his credit union until he showed up and was arrested. Id. The government thus had shown reasonable diligence in pursuing the leads it had available.

32

government is considered negligent in its pursuit.” Mendoza, 530 F.3d at 763 (citing Doggett, 505 U.S. at 653). Here, it is essentially undisputed that the government made only minimal attempts to find Velazquez in the five years between Degan’s transfer in November 2005 and Ilagan’s assumption of control over the investigation in November 2010. The sum total of its activity was limited to checking the NCIC eight times and, perhaps, placing Velazquez on the “Most Wanted” list for the Philadelphia DEA office.

Although we doubt that such negligible effort could be deemed “serious” in any circumstances, we need not speculate in this case about how far short of the mark it fell. Deputy Degan’s collateral request in 2005 detailed the measures he thought advisable in tracking down Velazquez, and we believe his prescription for the investigation is a helpful guidepost in assessing whether the government met the standard of reasonable diligence.17 In addition to recommending follow-up with DEA Agent Pascoe, who had visited the Woodward Avenue home at the request of the 17 Our dissenting colleague disagrees with our use of Deputy Degan’s collateral request as guidance in the reasonable diligence analysis, contending that we turn each suggestion into a “requirement[].” We do no such thing. The pertinent point here is not that the taskforce missed a particular suggestion on Degan’s list, but rather that — as the government explicitly acknowledges — there is no evidence that anyone from the Los Angeles Marshals taskforce visited any of the addresses linked to Velazquez’s close relatives that were identified in Degan’s 2005 request. See Appellee’s Br. 43.

33

Philadelphia DEA office, Degan listed a number of addresses that the Marshals Service taskforce could explore if DEA agents had not already done so. He expressly asked that Velazquez’s parents and brother be interviewed if all other leads had been exhausted. None of this was done.

Nor did authorities attempt to reach Velazquez during this period. They could have sent mail to his PO box or the Woodward Avenue address or sought out a relative to relay a request that Velazquez turn himself in. Cf. Mendoza, 530 F.3d at 763 (“Based on its previous success in contacting [defendant], the government was negligent when it failed to attempt to inform [him] of the indictment by calling [his family].”) Additionally, no contact was made with his attorney throughout the five years.

On this record, we think it plain that the government was not reasonably diligent in its pursuit of Velazquez. For the reasons we have explained, the district court’s contrary determination was clearly erroneous.18 The second Barker 18 We recognize, as did the district court, that speedy-trial cases are intensely fact-bound and thus of limited value as precedent. Velazquez, 2012 WL 2094061, at *11; see also Barker, 407 U.S. at 530 (“A balancing test necessarily compels courts to approach speedy trial cases on an ad hoc basis.”). Nevertheless, we have reviewed United States v. Spaulding, 322 Fed. App’x 942 (11th Cir. 2009), and United States v. Walker, 92 F.3d 714 (8th Cir. 1996), the cases relied upon by the district court and government respectively. Both support the finding that the government is responsible for the delay in finding Velazquez. Spaulding involved government

34

factor — the reason for the delay — squarely favors Velazquez.19

effort far more extensive and consistent than was the case here, including months of government surveillance, efforts to trace the defendant’s mail from his postal box, and calls to him to ask him to return. 322 F. App’x at 944. The defendant in Walker, unlike Velazquez, had fled after posting bail on state drug charges, and was on the run at the time he was indicted for federal drug charges. 92 F.3d at 715. He also used a false identity, making his evasiveness obvious. Id. at 716.

19 Courts have also examined speedy trial claims by assigning responsibility for specific periods of the delay, and then weighing the delay attributable to the government. See Battis, 589 F.3d at 680; see also Doggett, 505 U.S. at 657-658 (identifying six years of the delay from indictment to arrest as attributable to the government and using that portion to find general presumption of trial prejudice to the defendant). We have concluded that the government was not reasonably diligent for the five-year period from November 2005 until November 2010. See Battis, 589 F.3d at 678 (“[D]elay is measured from the date of arrest or indictment, whichever is earlier, until the start of trial[.]”). We can assume the government was diligent from November 2010 until the arrest in December 2011, a period of about thirteen months. The remaining delay from arrest to the projected trial date is neither side’s fault. Thus, to find the government diligent here would require finding that thirteen months of diligence outweighed a period of negligence that was more than four times as long. We decline to so hold.

35

C. Defendant’s assertion of the right

The third Barker factor requires a court to examine “[w]hether and how a defendant asserts his [speedy-trial] right,” 407 U.S. at 531, including “the frequency and force” of such assertions, id. at 529. The district court found it unnecessary to make any findings on this factor because it had already rejected Velazquez’s speedy trial motion on the basis that the government was reasonably diligent and Velazquez could not show specific prejudice. Hence, the court simply assumed that Velazquez timely asserted his right by filing a motion to dismiss within four months of his arrest. Given this posture of the case, we resolve this factor based on the undisputed facts in the record.

The Supreme Court noted in Barker that the fundamental “right to a speedy trial is unique in its uncertainty as to when and under what circumstances it must be asserted or may be deemed waived.” Id. at 529. Barker attempted to plot a path through this uncertainty by rejecting a rigid rule that a defendant waives his speedy-trial right “for any period prior to which he has not demanded a trial.” Id. at 525. The Court observed that such an approach presumed the “waiver of a fundamental right from inaction,” id. (footnote omitted), which is inconsistent with the definition of waiver as “an intentional relinquishment or abandonment of a known right or privilege,” id. (internal quotation marks omitted). Given such a high standard for establishing waiver, “[c]ourts should indulge every reasonable presumption against waiver . . . [and] not presume acquiescence in the loss of fundamental rights.” Id. at 525-26 (internal quotation marks omitted).

36

The Barker Court also rejected the notion that “the defendant has no responsibility to assert his right.” Id. at 528. Instead, it held that the issue of waiver — “the defendant’s assertion of or failure to assert his right to a speedy trial” — would be one of the factors to consider on a case-by-case basis in balancing the defendant’s and government’s conduct. Id. The Court emphasized that while the defendant bears “some responsibility to assert a speedy trial claim,” the prosecution retains the burden to show the knowing and voluntary waiver of a fundamental right. Id. at 529. Thus a defendant’s limited responsibility to assert his speedy trial right exists alongside the government’s overarching burden to prove waiver of that fundamental right. Applying its rule in Barker, the Court considered the defendant’s failure to object to eleven continuances lasting a total of more than three-and-a-half years, followed by the defendant’s motion to dismiss and his opposition to several further continuances. See 407 U.S. at 517-18. In that setting, where the defendant had ample opportunity to object in court, the defendant raised a “close” case, id. at 533, but ultimately a losing one.

Barker’s teachings necessarily left uncertainty about when a defendant would be obligated to assert his speedy-trial right if the defendant is at large. In considering this issue in Doggett, the Court observed that an at-large defendant’s knowledge of a pending indictment could weigh heavily against him on this Barker factor, even if he was not evading capture. 505 U.S. at 653. In Velazquez’s case, the government has argued that knowledge of charges, rather than the subsequent indictment, should weigh heavily against him. Although the Doggett Court did not explicitly announce that the defendant’s awareness of the indictment — rather than knowledge of earlier events, such as an investigation or an

37

arrest warrant — was the critical measuring point, such a rule is consistent with longstanding principles governing a defendant’s speedy trial rights.

This proposition is evident from the Court’s examination of the scope of a Sixth Amendment speedy trial claim in United States v. Marion, 404 U.S. 307 (1971). In rejecting the defendants’ contention that their Sixth Amendment right had been violated by a three-year delay between the end of their criminal activity and the return of the indictment, the Court emphasized that “the Sixth Amendment speedy trial provision has no application until the putative defendant in some way becomes an ‘accused,’” through indictment or arrest. Id. at 313. The Court noted that the “public act” of an arrest “may seriously interfere with the defendant’s liberty, whether he is free on bail or not, and . . . may disrupt his employment, drain his financial resources, curtail his associations, subject him to public obloquy, and create anxiety in him, his family and his friends.” Id. at 320. With this impact in mind, the Court observed, “it is readily understandable that it is either a formal indictment or information or else the actual restraints imposed by arrest and holding to answer a criminal charge that engage the particular protections of the speedy trial provision of the Sixth Amendment.” Id. Given the concerns arising from an arrest, “[i]nvocation of the speedy trial provision . . . need not wait indictment, information, or other formal charge,” but the Court refused to extend the “reach of the amendment to the period prior to arrest.” Id. at 321.

It is thus well established that the constitutional speedy trial clock does not start for an individual who has not yet been arrested until an indictment has issued. See id. at 320;

38

Battis, 589 F.3d at 678 (“In general, delay is measured from the date of arrest or indictment, whichever is earlier, until the start of trial.”). The Speedy Trial Act also looks to arrest or indictment to start the clock. See 18 U.S.C. § 3161(b) (requiring that a defendant be charged within thirty days of the date he was arrested or served with a summons); id. § 3161(c)(1) (requiring that trial begin within seventy days of indictment/information or first court appearance, whichever occurs later). It would thus be odd to conclude that the defendant has a duty to assert his speedy trial right before either indictment or arrest — i.e., at a time when he does not yet have such a right. Hence, the Court’s focus in Doggett on the indictment, without an explicit statement that it is the pivotal event for an at-large defendant’s assertion of his speedy trial right, undoubtedly stems from the previously recognized importance of that formal charge.

We note that, in its discussion of the underlying events in Doggett, the Supreme Court observed that there was no evidence the defendant was aware of the pre-indictment charges against him. See 505 U.S. at 653. Although that statement has led some courts to look to knowledge of the charges apart from knowledge of the indictment, see, e.g., United States v. Tchibassa, 452 F.3d 918, 926 n.8 (D.C. Cir. 2006) (noting that “[t]he Doggett Court appeared concerned generally with Doggett’s awareness vel non that charges were pending against him rather than with his specific knowledge that a formal indictment had been filed”), we see the Court’s observation as compatible with a focus on the indictment. If the defendant did not even know of the charges against him, he necessarily did not know of the later-rendered formal indictment that started the speedy trial clock. Indeed, the Court twice pointed to knowledge of the “indictment” as

39

pertinent, and we cannot conclude that the Court meant to modify this measuring point through its discussion of the evidence.20 Other circuits also have treated knowledge of the indictment as the necessary inquiry. See, e.g., United States v. Molina-Solorio, 577 F.3d 300, 306 (5th Cir. 2009) (focusing on defendant’s knowledge of the indictment pending against him); Mendoza, 530 F.3d at 763 (same).

Having identified knowledge of the indictment as the appropriate measure for the timely assertion of the speedy trial right, we turn to the evidence in the record, keeping in mind the government’s burden to demonstrate Velazquez’s knowledge. See Barker, 407 U.S. at 529. Although Velazquez’s attorney contacted the U.S. Attorney’s Office to discuss a possible surrender, and the attorney stipulated that he received a copy of the arrest warrant and the pre-indictment charges, there is no evidence that either he or

20 However, as we previously noted, supra at 24-25, a defendant’s knowledge of the investigation or charges, apart from knowledge of the indictment, might bear on the second Barker factor, the reason for the lengthy delay, as this knowledge might inform a finding that the defendant was evading authorities. This overlap has been noted in the case law. Justice Brennan combined what later became the second and third Barker factors in a concurrence in Dickey, reasoning that consideration of the defendant’s assertion of the speedy-trial right overlaps with consideration of who was responsible for the delay. 398 U.S. at 48 n.12 (Brennan, J., concurring). The Court in Barker noted that “there is little difference between [Justice Brennan’s] approach and the one” it adopted. 407 U.S. at 530 n.30.

40

Velazquez was later told of the indictment.21 The indictment was not sent to the attorney, App. 244a, and the government did not communicate with him after the indictment was issued. On this record, the most we can say is that Velazquez learned of the indictment at the time of his arrest. See Molina-Solorio, 577 F.3d at 306-07 (construing a lack of evidence that defendant knew of the indictment as weighing in the defendant’s favor). Velazquez brought his speedy-trial motion within four months of his arrest, and thus we count this factor in his favor.

D. Prejudice suffered by the defendant

As noted, the district court found that the government was reasonably diligent in pursuing Velazquez, and it thus required Velazquez to show specific prejudice to his defense from the lengthy delay before trial, or as it happens here, his conditional guilty plea. Velazquez, 2012 WL 2094061, at *13 (citing Doggett, 505 U.S. at 656). Because the court erred when it found reasonable diligence, its prejudice analysis compounded that error. Instead of analyzing whether the government could overcome the general presumption that “excessive delay . . . compromises the reliability of a trial in ways that neither party can prove or, for that matter, identify,” Doggett, 505 U.S. at 655,22 it found that “defendant failed to

21 The government argued that he must have known of the charges. The court, however, made no findings on knowledge of the indictment.

22 Doggett identified three types of harm caused by “unreasonable delay between formal accusation and trial”: oppressive incarceration, the accused’s increased anxiety, and

41

identify any specific witness or piece of evidence that he now cannot access,” Velazquez, 2012 WL 2094061, at *14.

To “warrant granting relief, negligence unaccompanied by particularized trial prejudice must have lasted longer than negligence demonstrably causing such prejudice.” Doggett, 505 U.S. at 657. The Court in Doggett found that the durational requirement for relief without specific prejudice was met where the delay attributable to the government’s negligence was six years, an amount that “far exceeds the [one-year] threshold needed to state a speedy trial claim.” Id. at 658. The defendant was entitled to relief because the presumption of general prejudice was not “persuasively rebutted.” Id.

The government contends that, even if the general presumption of prejudice applies, it has met its burden of rebuttal by affirmatively showing that Velazquez’s defense is unimpaired. The government notes that the critical meetings and phones calls in the case were all recorded and that Velazquez’s co-defendants have previously testified about the pertinent events — leaving the evidence intact, and without the risk of fabrication, despite the trial delay. App. 163a. The district court accepted the government’s contentions and

“the possibility that the accused’s defense will be impaired by dimming memories and loss of exculpatory evidence.” 505 U.S. at 654 (internal quotation marks omitted). Here only the last harm is at issue. It is “the most serious . . . because the inability of a defendant adequately to prepare his case skews the fairness of the entire system.” Id. (internal quotation marks omitted).

42

held, as part of its discussion of specific prejudice, that the government had rebutted any showing of general prejudice. Velazquez, 2012 WL 2094061, at *13.

The government’s reasoning is flawed in two ways. First, it equates the preservation of evidence the government would rely on with the materials Velazquez might need to challenge the government’s case. The Court in Doggett recognized that “impairment of one’s defense is the most difficult form of speedy trial prejudice to prove because time’s erosion of exculpatory evidence and testimony can rarely be shown.” 505 U.S. at 655 (internal quotation marks omitted). Not only may Velazquez plausibly claim that his own recollection of the events at issue has eroded, but he also has argued that the passage of time makes it harder to investigate an entrapment defense or find witnesses to “his whereabouts and involvement.” Velazquez, 2012 WL 2094061, at *13. Forecasting how faded memories could harm him is precisely the sort of difficult-to-obtain proof that supports the finding of general prejudice in a case of extraordinary delay. Second, the government’s argument in effect turns its burden to disprove general prejudice on its head by suggesting that its rebuttal effort must be found successful unless Velazquez can identify prejudice. In Doggett, the Court noted that the government “ably counter[ed] Doggett’s efforts to demonstrate particularized trial prejudice [but] it has not, and probably could not have, affirmatively proved that the delay left his ability to defend himself unimpaired.” 505 U.S. at 658 n.4 (emphasis added) (citing H. Richard Uviller, Barker v. Wingo: Speedy Trial Gets A Fast Shuffle, 72 Colum. L. Rev. 1376, 1394-1395 (1972)). The Court thus

43

indicated that the government faces a high, and potentially insurmountable, hurdle in seeking to disprove general prejudice where the period of delay is extraordinarily long.23

23 In the passage of Professor Uviller’s article that the Doggett Court cited as sufficiently analogous to lend support, Uviller recognized that this burden on the prosecution to prove a negative could be “unfair since it may require proof by facts inaccessible to the state.” Uviller, supra at 1394. He nonetheless argued for such a burden: How can the state prove that no evidence for the defense was lost or impaired? . . . Proof of a negative is always difficult and in this instance, it may be contended, the task is impossible since the critical facts are known only to the defendant. Further, the argument would run, where the prosecutor must demonstrate harmless error at trial, he may do so from the record; but since he cannot show lack of prejudice by record citations, a parallel burden would be inappropriate to impose.

The argument is not without merit. Realistically, prejudice lies beyond the capacity of either side to prove or disprove, except in the rare instance where a known defense witness of known competence actually disappears or reports a recent impairment of memory, and no prior testimony from him is available. Therefore, the shift of burden actually permits the presumption of prejudice to prevail on the issue. Since that presumption is well-founded,

44

Perhaps the government could make that showing in another case, but it has not done so here. The presumption of general prejudice, triggered by the government’s extraordinary delay in bringing Velazquez to trial, continues to favor the defendant in the application of the Barker speedy trial test.

III.

We recognize the significance of our decision. A defendant who pleaded guilty to serious drug charges will no longer have to answer those charges. But we accept such rare outcomes as the necessary cost for the protection of the speedy trial right set forth in the Constitution. Here, with respect to the first factor in the Barker analysis, the length of the delay in bringing Velazquez to trial was extraordinary by any measure. Contrary to the conclusion of the district court, the government was not reasonably diligent in pursuing the defendant. Indeed, its pursuit of the defendant was strikingly inattentive for five years. Hence, the reason for the delay, the

however, justice is served. The establishment of prejudice, albeit presumptively, does not end the inquiry; it merely focuses attention on other elements wherein impropriety or justification may be more meaningfully discerned. Id. at 1394-95 (emphasis added). We need not resolve in this case whether general prejudice is irrebutable when the period of delay is extraordinarily long. We simply note that, in citing the above passage for support, the Doggett Court was keenly aware of the practical difficulties for the prosecution in making such a rebuttal.

45

second and most important factor in the speedy trial analysis, strongly favors the defendant. The third factor — the timely assertion of the speedy-trial right — benefits the defendant. As to the fourth factor, the government did not overcome the presumption of general prejudice that applies with considerable force in a case of such extraordinary delay. Under these circumstances, all of the Barker factors support the defendant's claim of a violation of his speedy trial right. We must therefore reverse the district court's judgment of conviction and the related sentence. The indictment against Velazquez must be dismissed with prejudice. We remand for that purpose.

1

JORDAN, Circuit Judge, dissenting

Whether Sergio Velazquez’s constitutional right to a speedy trial was violated is a close question, and, as my colleagues in the Majority point out, the answer hinges largely on the reasons for the delay in bringing him to trial. The primary issue is whether the government exercised reasonable diligence to find him, or, to phrase it differently, whether the government was negligent in its investigation. My colleagues reject the District Court’s determination that the government was not negligent, although that Court held an evidentiary hearing, carefully considered all of the evidence, and thoughtfully explored the factual and legal disputes. I think the Majority is mistaken. While the government’s investigative efforts fell well short of praiseworthy, they were not so lacking that, on this record, the District Court’s decision should be seen as reversible error. Given that conclusion, and the District Court’s finding that Velazquez did not demonstrate specific prejudice from the delay, we should affirm. I therefore respectfully dissent from giving Velazquez a pass for dealing in multiple kilograms of cocaine.

I.

Considering the government’s obligation to exercise reasonable diligence in bringing a defendant to trial is an intensely fact-specific inquiry. “[T]he precise amount of effort that is required is apt to vary depending on the circumstances of the case,” Rayborn v. Scully, 858 F.2d 84, 90 (2d Cir. 1988), so understanding the facts is particularly important. That, of course, is something that district courts, with their fact-finding capacity, are best suited to accomplish,

2

and we are thus obliged to view a district court’s determination of reasonable diligence with, as the Supreme Court has put it, “considerable deference.” Doggett v. United States, 505 U.S. 647, 652 (1992). We must review the District Court’s underlying factual findings regarding a speedy trial claim for clear error, United States v. Battis, 589 F.3d 673, 677 (3d Cir. 2009), and would be well-advised to think of “considerable deference” in the same terms, given the context.1 A finding of fact is, as my colleagues note, clearly erroneous “only if we are left with a definite and firm conviction that a mistake has been committed.” United States

1 There is a persuasive argument that the “considerable deference” standard for the reasonable diligence determination is simply another way of saying “clearly erroneous review.” In referencing “considerable deference,” the Supreme Court in Doggett, 505 U.S. at 652, cited Cooter & Gell v. Hartmarx Corp., 496 U.S. 384, 401-02 (1990) (“The [Supreme] Court has long noted the difficulty of distinguishing between legal and factual issues. … The considerations involved in the Rule 11 context are similar to those involved in determining negligence, which is generally reviewed deferentially.”), and McAllister v. United States, 348 U.S. 19, 20-22 (1954) (finding, in an admiralty case, that a district court’s findings of negligence were not clearly erroneous). The Doggett Court also cited a section of Wright and Miller’s Federal Practice and Procedure discussing negligence cases and stating that the “natural” reading of McAllister is that “a determination of negligence is reviewed under the ‘clearly erroneous’ rule.” 9 Charles Alan Wright & Arthur R. Miller, Federal Practice and Procedure § 2590 (1st ed. 1971); see also Doggett, 505 U.S. at 652.

3

v. Lessner, 498 F.3d 185, 199 (3d Cir. 2007). It is not enough to be left with lingering questions.2 The Majority takes issue with two conclusions of the District Court: that the few investigative steps taken by the government between 2005 and 2010 may be excused in light of the more intense but fruitless efforts that preceded and followed that period, and that Velazquez’s off-the-grid lifestyle contributed to the delay. While I acknowledge that fair-minded people can differ about inferences and conclusions to be drawn from the record, I do not have a definite and firm conviction that the District Court erred in its handling of those essentially factual issues or in its ultimate conclusion of reasonable diligence.

My colleagues begin by asserting that “law enforcement priorities have little role to play in the

2 We review legal conclusions regarding a speedy trial claim, including the balancing of the Barker factors, de novo. See Battis, 589 F.3d at 677; Hakeem v. Beyer, 990 F.2d 750, 771 (3d Cir. 1993). The Supreme Court has held that no single factor is “either a necessary or sufficient condition,” and the factors “must be considered together with such other circumstances as may be relevant.” Barker v. Wingo, 407 U.S. 514, 533 (1972). Even accepting my colleagues’ conclusions regarding the length of delay in this case and Velazquez’s assertion of his right to a speedy trial, however, a defendant’s speedy trial claim will typically fail when the government has demonstrated reasonable diligence and the defendant has failed to show specific prejudice. See Doggett, 505 U.S. at 656.

4 negligence calculus.” (Maj. Op. at 24.) I disagree and think it plainly relevant to consider the likelihood that an investigative step will bear fruit when considering what actions constitute reasonable diligence. The concept of “reasonableness” is itself dependent upon circumstances. “Reasonable diligence” necessarily incorporates the notion that specific circumstances, including but not limited to the constraints operating on the government, factor into what constitutes a reasonable investigative effort. The Majority relies on the Supreme Court’s statement in Doggett that “even if law enforcement inaction ‘may have reflected no more than [defendant’s] relative unimportance in the world of drug trafficking, it was still findable negligence.’” (Id. at 23 (quoting Doggett, 505 U.S. at 653).) That statement, however, does not indicate that prosecutorial judgment about how best to deploy law enforcement resources should be given little or no weight. It was made in conjunction with the observation that, had government agents in that case tested “their progressively more questionable assumption that Doggett was living abroad, … they could have found him within minutes.” Doggett, 505 U.S. at 652-53. In other words, under the specific circumstances of that case, the lower prosecutorial priority assigned to the defendant did not offset the government’s utter failure to act on developments indicating his whereabouts.3 Id. It bears repeating that “the

3 I recognize the Supreme Court’s concern that “[c]ondoning prolonged and unjustifiable delays in prosecution would both penalize many defendants for the state’s fault and simply encourage the government to gamble with the interests of criminal suspects assigned a low prosecutorial priority.” Doggett, 505 U.S. at 657. However, for the reasons discussed herein, the delay here was not

5

precise amount of effort that is required is apt to vary depending on the circumstances of the case.” Rayborn, 858 F.2d at 90.

The Majority claims that there is “no evidence in the record to support a finding that investigators made an actual ‘choice’ not to pursue Velazquez.” (Maj. Op. at 24.) That is not entirely true. Notably, the District Court found that, after Deputy Degan’s departure, someone continued to work the case, and the Majority acknowledges that that person may have been Deputy Cardinal. It is not beyond the pale to believe that the person responsible for the case was making decisions about how to work it. To the extent my colleagues demand a clearer record that some “particular individual” made the decision “to forgo pursuit of Velazquez” (id. at 24-25), they miss the mark. Government decision-making at its best is the product of a deliberative process in which costs and benefits are weighed and reflected in a well-kept record. But decision-making is not always as carefully done or as clearly preserved as we would like, and yet we do not assume that government actions are random. The District Court

unjustifiable. While I do not advocate blind deference to executive decisions about the allocation of law enforcement resources, I do believe that requiring constant government activity in fugitive cases, regardless of the prospects that such activity will bear fruit, may rightly be perceived as unwelcome and unnecessary judicial meddling in the executive sphere. There is ample room to disagree with executive decisions on resource allocation before one arrives at the point where a constitutional violation should be declared.

6

heard the law enforcement witnesses and evidently believed that the limited investigative activity undertaken from 2005 to 2010 was the product of informed discretion. I am not prepared to say that was clearly erroneous.4 Nor am I willing to fault the District Court for thinking that later, more intensive but equally fruitless investigation could be seen as some vindication of an earlier government decision to do less. It is true that one cannot say that later failure proves – in a deductive sense – that earlier efforts would have failed, but

4 Although, as my colleagues point out, the District Court did not hear directly from any law enforcement officer who worked on the case between 2005 and 2010, there was certainly circumstantial evidence that somebody worked on it after Deputy Degan’s departure. For example, Deputy Degan testified that someone in the Marshals Service would have been assigned to the case, and somebody in the Marshals Service did indeed run period NCIC checks between 2005 and 2010. In addition, DEA Agent Pedrini made sure that Velazquez’s warrant remained active in the NCIC during that time and was in contact with the Marshals Service to see if there was any new information on the case. The Majority seems to think that the level of deference I believe should be given to the District Court’s “reasonable diligence” ruling amounts to no review at all. (See Maj. Op. at 25 n.15.) That is a basic disagreement. I believe a reasonable inference that someone was assigned and working the case can be drawn from the record, and, again, under clear error review, I am not left with a definite and firm conviction that, in light of the totality of the circumstances, the District Court erred in concluding that the work done was sufficient to constitute reasonable diligence.

7

those points of failure may add up inductively to allow a reasonable mind to determine that earlier decisions to do less were grounded in something better than sheer laziness.

The Majority makes too much of the fact that there is no record of the government taking all of the investigative steps suggested by Deputy Degan in 2005. The District Court was well aware that there was “no direct evidence that authorities in California exhausted the leads in Deputy Degan’s collateral request.” (App. at 20a.) However, it found that the additional leads were “far more speculative than [Velazquez’s] connection to the Woodward Avenue address” (id.), and it credited Deputy Degan’s “belie[f] that work was underway on [his] request” (id. at 7a). In other words, the Court accepted the deputy’s understanding of how the Marshals Service works, and it was entitled to do so.5 5 I note parenthetically my disagreement with any implication in the Majority opinion that Deputy Degan’s suggested list of investigative steps should be viewed as the measure of reasonable diligence. (Maj. Op. at 32.) Though my colleagues disclaim relying on it (see id. at 32 n.17), a reader might nonetheless conclude that the references to that list are meant to give it weight. One officer’s investigative suggestions to another on the opposite side of the country, however, may vary from what the receiving officer’s local experience tells him will and will not be worthwhile. That difference of opinion does not make the receiving officer a slacker. Moreover, we risk building a perverse incentive into the system if we turn suggestions into requirements. There may be fewer suggestions committed to paper if deputy marshals believe that courts will turn unfollowed leads into “stay out of jail free” cards for fugitives.

8

There is no dispute that, between August and November 2005, the government entered information about the charges against Velazquez into NCIC,6 searched other databases for information, sent the 2005 collateral request to authorities in California to seek help, and visited the Woodward Avenue address, all to no avail. Between November 2005 and November 2010, the government periodically ran Velazquez’s name through NCIC and made sure that the warrant was still in the system.7 No additional leads surfaced until a database check in November 2010

6 “NCIC,” as the Majority notes, is the acronym for the National Crime Information Center, a database of criminal justice information.

7 Velazquez relies on United States v. Fernandes, 618 F. Supp. 2d 62 (D.D.C. 2009), and United States v. Mendoza, 530 F.3d 758 (9th Cir. 2008), for the proposition that merely entering names into a database – as he characterizes the government’s efforts between November 2005 and November 2010 – is insufficient to meet reasonable diligence. As the District Court noted, however, both cases are distinguishable because they are extradition cases involving fugitive defendants who had left the country. Periodically entering a fugitive’s name into a national warrant database when the individual is believed to still be in the country, as is the case here, provides some chance that the fugitive might be found by authorities somewhere within the United States. It cannot be compared to doing the same for a fugitive believed to be outside of the country, in which case extradition is “the most obvious step” to bring that person promptly back for trial. Fernandes, 618 F. Supp. 2d at 71.

9

revealed a new place of employment for Velazquez. Given those circumstances, it was not clearly erroneous for the District Court to conclude that deputies could “infer[] failure” would have dogged further pre-2010 efforts in California to find Velazquez. (Id. at 21a.) It was, in turn, not clearly erroneous to decide that “the government reasonably elected to conserve its resources and wait for new information or a change in circumstances.” (Id.) My colleagues in the Majority also reject the District Court’s determination that, purposefully or not, Velazquez’s decisions made it difficult for the government to find him. The Majority does acknowledge that “the absence of a paper trail for a defendant might leave the government with fewer avenues for investigation” (Maj. Op. at 30), but it gives that fact short shrift and instead expresses concern that taking account of a fugitive’s “transient” lifestyle “would likely be felt disproportionately by those in more limited economic circumstances” (id.). That concern is more a matter of speculation than proof at this point, but, assuming it is true, that does not address the investigative reality confronting the government both generally and in this case specifically. First, as a general matter, law enforcement decisions made under budgetary constraints and without any hint of improper motivation should not be overturned because of a vague concern that being hard to find is peculiar to the poor.8 More specifically, however, in this case there was significant

8 There may be significant challenges tracking people at the other end of the economic scale too, since a person of means who is constantly traveling or moving among multiple addresses may be as difficult to find as someone who is not well-rooted in society.

10

evidence indicating that Velazquez was quite deliberately hard to find, which is understandable since, as the Majority notes, “the record would support a finding that Velazquez was aware that he was being sought in connection with the [underlying] drug transaction.” (Id. at 29.) The District Court thought it unnecessary to take the final step of concluding that Velazquez was deliberately evading authorities, though it did note that the evidence “strongly supports the inference that [Velazquez] did hide.” (App. at 20a.) That evidence, and the District Court’s comment on it, should make us leery of overturning the Court’s determination that the government’s investigative efforts were reasonably diligent under the circumstances.

In sum, while law enforcement officers certainly could have done more to search for Velazquez, particularly between 2005 and late 2010, they were not obligated to take every possible action and chase every lead. See Rayborn, 858 F.2d at 90 (noting that the government must exercise only “due diligence” and not “heroic efforts”). Given the considerable deference that we must give the District Court’s finding of reasonable diligence, I would accept it and turn attention to the question of prejudice.

II.

Because the government sufficiently demonstrated reasonable diligence, the District Court required Velazquez to show that he would suffer specific prejudice, not just general prejudice, from the passage of time in order to prevail on his speedy trial claim. That decision was in keeping with the Supreme Court’s instruction in Doggett, in which the Court held that, when the government has been negligent in its

11

investigation and the delay is excessively long, “consideration of prejudice is not limited to the specifically demonstrable,” and defendants may claim prejudice without providing “affirmative proof of particularized prejudice.” Doggett, 505 U.S. at 655. Conversely, if the government can show that it “pursued [a defendant] with reasonable diligence from his indictment to his arrest” – as the District Court concluded the government did here – then the defendant’s right to a speedy trial claim will fail, regardless of the length of delay, unless the defendant can show specific prejudice to his defense. Id. at 656.

The Majority rightly notes that in general there are three types of prejudice that can result from delay: (1) oppressive pretrial incarceration; (2) the defendant’s anxiety and concern over the outcome of the litigation; and (3) impairment of the defense. Battis, 589 F.3d at 682. The prejudice that Velazquez asserts is impairment of his defense. “[T]he possibility that the … defense will be impaired by dimming memories and loss of exculpatory evidence” is the most important form of prejudice faced by a defendant when his right to speedy trial is denied. Doggett, 505 U.S. at 654 (internal quotation marks omitted); see also Barker v. Wingo, 407 U.S. 514, 532 (1972) (“[I]nability of a defendant adequately to prepare his case skews the fairness of the entire system.”). Generally, the burden of showing prejudice is on the individual claiming the violation, see Hakeem v. Beyer, 990 F.2d 750, 760 (3d Cir. 1993), and the mere “possibility of prejudice is not sufficient to support [the] position that … speedy trial rights were violated.” United States v. Loud Hawk, 474 U.S. 302, 315 (1986) (emphasis added). In Hakeem v. Beyer, one of the defendants argued that he would

12

suffer prejudice because the delay prevented him from calling witnesses to corroborate his alibi defense to robbery. 990 F.2d at 763. We rejected that argument, holding that, absent extreme circumstances, “[g]eneral allegations that witnesses’ memories have faded are insufficient to create prejudice.” Id.

As the District Court observed, Velazquez “relies almost exclusively on the general assertion that he is prejudiced by the passage of time because witnesses to his whereabouts and involvements may be impossible to locate and those witnesses that are available will have impaired memories.” (App. at 25a (internal quotation marks omitted).) I agree with the District Court that Velazquez’s claims in this regard are “too general and too speculative” to demonstrate specific prejudice. (Id.)

Velazquez also argues that his ability to investigate an entrapment defense has been impaired. As the Court pointed out, however, the informant’s conversations with Velazquez on July 3, 2005 and those that took place over the phone were recorded, and the testimony from co-defendants Pedro Curiel and Nelson Gutierrez-Gainza at Gutierrez-Gainza’s trial is available. Velazquez “failed to identify any specific witness or piece of evidence that he now cannot access.” (Id. at 26a.)

Because Velazquez did not carry his burden of proving particularized prejudice, the District Court correctly determined that, under the circumstances of the case, he did not demonstrate a right to relief.

III.

This case presents a serious question regarding the unusual delay between indictment and trial, but I believe the District Court handled the matter wisely and well, and I would therefore affirm its decision that Velazquez’s constitutional right to a speedy trial was not violated. I thus respectfully dissent.

Outcome: The indictment against Velazquez must be dismissed with prejudice. We remand for that purpose.

Plaintiff's Experts:

Defendant's Experts:

Comments: