KANNE, Circuit Judge.
On August 15, 2000, two men unsuccessfully tried to rob a bank in Woodburn, Indiana. They fled before police could arrive, and, for years, they remained at large. Then, in 2008, new DNA tests cracked the case and tied defendant William Hagler to the crime. Hagler was indicted for attempted bank robbery, and a jury found him guilty. Hagler now appeals, arguing that the government waited too long to indict him, that the evidence was insufficient to convict him, and that new DNA testing entitles him to a new trial. We affirm.
I. Background
On August 15, 2000, George Townsend awoke in the morning and drove to work in Fort Wayne, Indiana — as he did almost every morning. When he arrived, he parked his white Pontiac Bonneville in a nearby lot and went inside. When he came back out for lunch, the car was gone. All that remained in its place was a pile of broken glass.
About fifteen miles away in Woodburn, Indiana, George Townsend’s Pontiac slowly came to a stop next to a branch of the National City Bank. Two armed men stepped out. Wearing gloves, dark clothes, and Halloween masks, they stormed inside the bank. “You know what this is,” one of the men told the frightened
customers and employees. The other crossed the room and made his way toward the vault. Their operation clearly had been planned in advance. But the two would-be Dillingers had a defect in their plan: they had no idea how to open the bank’s vault. When this obvious oversight dawned on them, their nerve failed; they fled without taking a dime and made their getaway before police could arrive.
When police made it to the scene, they took statements from the bank’s customers, the bank’s employees, and several other witnesses who had watched the attempted robbery from outside the bank. Witnesses also provided descriptions of the getaway car. Later that day, police found George Townsend’s Pontiac idling in a trailer park on the western edge of town. One of the back windows had been smashed in, and its steering column was broken open. Inside the automobile’s cabin, police found Halloween masks, yellow-orange gloves, a fleece jacket, and a gray sweatshirt. Townsend arrived later and identified the car. He also confirmed that the masks, gloves, and jackets did not belong to him. A woman who witnessed the robbery also came to see the Pontiac and identified it as the car used in the robbery.
Police swept the car for forensic evidence. Investigators recovered several latent prints from the automobile. However, for reasons not disclosed in the record, the fingerprints were not immediately analyzed. Investigators also recovered a human hair from the gloves found in the car and a DNA sample from one of the masks. An analysis of the hair detected some DNA but not enough for a full profile. The sample from the mask showed a “mixed,” partial profile, meaning the sample contained incomplete DNA sequences from at least two people. The profile was uploaded to Indiana’s DNA database on May 25, 2001 and immediately started registering multiple “hits” (i.e., automatic reports of potential matches) on unknown individuals. The profile remained on the database until 2007, when it was removed for collecting too many hits. The proceedings below did not determine how many “hits” the DNA profile collected before it was removed, but both sides represent, based on evidence presented in a companion case involving Hagler’s brother, that the number was about forty. Without a definitive DNA identification, the case went cold.
Things heated up again in 2008, when the Indiana State Police upgraded its equipment to allow for more sensitive DNA testing and received grant funding to revisit old cases. Investigators eventually worked their way back to this case. A DNA analyst retested the hair taken from the glove, and this time she was able to extract a complete DNA profile. When she uploaded it, the database “hit” on Hagler. Police picked him up and collected a fresh DNA sample to confirm the initial test. Hagler’s DNA indeed matched the DNA found in the getaway car. Investigators also retested the sample taken from the mask. This time, the test was able to distinguish a “major contributor” from a “minor contributor” in the mixed profile. The major contributor’s DNA profile was uploaded to Indiana’s DNA database and hit on William Hagler’s brother Shawn Hagler (We will refer to William Hagler as “Hagler” and Shawn Hagler as “Shawn” to avoid confusion). The minor contributor’s profile contained insufficient genetic information to allow for a match. Armed with this evidence, police also reviewed the latent fingerprints lifted from the stolen car. Of the seven recovered prints, only one was of sufficient quality for analysis. It matched Hagler.
Hagler and his brother Shawn were first indicted for one count of attempted robbery
(see
18 U.S.C. §§ 2
&
2113) on October 28, 2009, but the district court later dismissed the indictment without prejudice in light of
Bloate v. United States, 559
U.S. 196, 130 S.Ct. 1345, 176 L.Ed.2d 54 (2010). A second grand jury indicted both men again on July 28, 2010, giving rise to this case. Hagler moved to dismiss the indictment as barred by the statute of limitations and unreasonable delay, but the district court denied his motions. Because Shawn was not arrested until November 11, 2010, Hagler was tried first. Shawn’s case is still pending in the district court.
At Hagler’s trial, the government presented several witnesses to the robbery. Although they disagreed on some of the details, the witnesses generally identified the perpetrators as two black males wearing dark clothing and Halloween masks. Four witnesses also described the getaway car as a white, four-door sedan consistent with George Townsend’s Pontiac, and one specifically identified Townsend’s car as the one used in the robbery. A resident of the trailer park testified that, on the morning of the attempted robbery, he saw two black males drive into the trailer park in a dark car, transfer some materials into a white car, and drive off. Stacy Gray, a physical security consultant for the bank, testified that the bank was FDIC insured at the time of the robbery. The government also presented testimony from Connie Evans Hanley, the mother of two children by Hagler. Hanley testified that, around December 1998, she and Hagler were talking in the Woodburn trailer park together when Hagler mentioned how easy it would be to rob a bank in Woodburn because the town was so lightly policed. Finally, the government introduced evidence that Hagler’s fingerprints were found on the getaway car and that his DNA was found inside it. Hagler elected not to present evidence of his own, and the jury convicted him of one count of attempted bank robbery. He now appeals, challenging the timeliness of his prosecution, the sufficiency of the evidence against him, and the district court’s decision not to grant him a new trial in light of additional DNA testing.
II. Analysis
A. Statute of Limitations
Hagler’s first argument is that he was indicted after the statute of limitations had run. The applicable statute of limitations provides that, “[ejxcept as otherwise expressly provided by law, no person shall be prosecuted, tried, or punished for any offense, not capital, unless the indictment is found or the information is instituted within five years next after such offense shall have been committed.” 18 U.S.C. § 3282(a). Here, the aborted robbery took place on August 15, 2000, and the operative indictment against Hagler did not issue until July 28, 2010, nearly ten years later. Thus, Hagler argues, his indictment was untimely, and his conviction cannot stand.
But, as both sides acknowledge, we cannot come to this conclusion so easily. Section 3282(a) specifically allows for exceptions to the general limitations period, and the government argues that one of these exceptions is in play here. Specifically, 18 U.S.C. § 3297 provides:
In a case in which DNA testing implicates an identified person in the commission of a felony, no statute of limitations that would otherwise preclude prosecution of the offense shall preclude such prosecution until a period of time following the implication of the person by DNA testing has elapsed that is equal to
the otherwise applicable limitation period.
Id.
The question, then, is when the limitations clock starts ticking under § 3297. We defer to the district court’s factual findings but ultimately decide the question
de novo. United States v. Hills,
618 F.3d 619, 634 (7th Cir.2010). The government argues that the clock did not start under § 3297 until a DNA test specifically identified a single, individual person. Since Hagler’s DNA in this case was not matched to the crime until 2008, the government reasons that the five-year clock did not start ticking until 2008 and therefore Hagler’s 2010 indictment was timely. Hagler, on the other hand, notes that a limited, partial DNA profile was uploaded to the DNA database at sometime in 2002. Once there, it started collecting roughly forty “hits” to unknown individuals before it was removed from the system in 2007. According to Hagler, this means that about forty people were “implicate[d]” in the crime beginning in 2002. As a result, he concludes, the statute of limitations expired in 2007, well before he was indicted.
We think that the government has the better argument. We begin with § 3297’s text, which, as discussed, provides that, where DNA testing “implicates an identified person in the commission of a felony, no statute of limitations that would otherwise preclude prosecution of the offense shall preclude such prosecution until a period of time following the implication of the person by DNA testing has elapsed that is equal to the otherwise applicable limitation period.” Hagler contends that the indefinite article “an” in the phrase “implicates an identified person” suggests that DNA evidence can implicate more than one person under the statute. Thus, he reasons, the multiple hits on the mixed DNA profile starting in 2002 were all “implications” that triggered § 3297.
We are not persuaded. It is true that the indefinite article “an” generally implies the possibility of a larger number than just one,
see, e.g., United States v. Jain,
174 F.3d 892, 898 (7th Cir.1999), but that alone does not resolve the issue. The term “identified person” is singular, suggesting one person. And the rest of the statute is written using definite articles; it refers to “a” case and “a” period of time following “the” implication of “the” person. Taken together, these words all suggest that the DNA evidence in question must be much more specific in its identifications than Hagler’s reading allows.
More importantly, to “implicate” someone of a crime is to strongly tie that person to wrongdoing. To adopt Hagler’s reading, we would have to hold that, when his partial DNA profile hit on some forty different people, every one of those people was “implicated” in the commission of the a crime. That seems implausible to us. We do not think that the other people on the list are fairly described this way. Perhaps they comprised a list of potential suspects, but they were not all “implicated” in the commission of a crime. Being implicated in a crime is a serious thing. Surely it requires more than just a one-in-forty chance.
Of course, statutory interpretation also “depends upon reading the whole statutory text, considering the purpose and context of the statute, and consulting any precedents or authorities that inform the analysis.”
Dolan v. U.S. Postal Serv.,
546 U.S. 481, 486, 126 S.Ct. 1252, 163 L.Ed.2d 1079 (2006). And we think that § 3297’s evident purpose also supports our reading. It should come as no surprise these days that DNA evidence is special. “Modern DNA testing can provide powerful new
evidence unlike anything known before.”
Dist. Attorney’s Office for the Third Judicial Dist. v. Osborne,
557 U.S. 52, 62, 129 S.Ct. 2308, 174 L.Ed.2d 38 (2009). “DNA is the most reliable evidence of identification — stronger even than fingerprints or photographs,”
Green v. Berge,
354 F.3d 675, 679 (7th Cir.2004), and “there is no technology comparable to DNA testing for matching tissues when such evidence is at issue,”
Osborne,
557 U.S. at 62, 129 S.Ct. 2308.
This unique reliability explains why Congress enacted § 3297. Statutes of limitations exist, in part, to protect people from having to defend against charges where “the basic facts may have become obscured by the passage of time.”
Toussie v. United States,
397 U.S. 112, 114-15, 90 S.Ct. 858, 25 L.Ed.2d 156 (1970);
see also United States v. Daniels,
387 F.3d 636, 643 (7th Cir.2004). But properly stored DNA evidence, unlike most other kinds of evidence, can maintain its reliability for decades.
See, e.g., Banks v. Workman,
692 F.3d 1133, 1137-38, 1151 (10th Cir.2012) (affirming denial of habeas corpus relief where petitioner was convicted using DNA testing performed “[njearly two decades” after the crime). Thus, when DNA evidence is available, the reasons for having a statute of limitations are significantly attenuated. Or, put another way, § 3297 extends the limitations period in DNA cases because DNA evidence is uniquely precise. It would be odd, then, to apply § 3297 where DNA profile lacks this precision, as was the case with the first DNA test here. Indeed, it turned out that this particular sample did not identify Hagler
at all;
it later matched conclusively to his brother Shawn. Given the high level of precision that § 3297 is premised on, we think that, in the vast majority of cases, DNA evidence will “implicate” someone only when, as here, it matches to a single, identified person.
At oral argument, the government urged us to adopt an even brighter line: that DNA evidence “implicates” someone under § 3297
only
if it matches to a single, identified person. We sympathize with the government’s desire for certainty. Nevertheless, we are apparently the first appellate panel in the country to interpret § 3297, and we decline to adopt such a broad holding. While forty matches is surely too many to “implicate” someone, we can still imagine unusual cases where DNA evidence might be said to “implicate” more than one person. Suppose, for example, the defendant had an identical twin. In such a case, any DNA evidence would necessarily match two people, but still might be said to “implicate” them both. We expect that such exceptions will occur rarely. But we do not want to rule out the possibility of them ever occurring at all, particularly so early in the game.
One final point to address before moving on: Hagler argues in the alternative that, even if we interpret § 3297 against him, we should still reverse his conviction because § 3297 is void for vagueness. This argument appears only in a footnote, and that fact alone would justify declining to address it.
See Long v. Teachers’ Ret. Sys. of Ill., 585
F.3d 344, 349 (7th Cir.2009). In any event, we do not think that § 3297 is vague. As our previous discussion indicates, § 3297 is perfectly susceptible to reasoned interpretation. Accordingly, we also reject Hagler’s alternative argument.
B. Excessive Pretrial Delay
Hagler next argues that the preindictment delay in his case was unconstitutionally excessive. The district court denied Hagler’s request to dismiss the case for unreasonable pre-indictment delay, and
we review that decision for abuse of discretion.
United States v. McMutuary,
217 F.3d 477, 481 (7th Cir.2000).
The primary safeguard against unreasonable prosecutorial delay is the statute of limitations, not the Constitution.
Id.
Nevertheless, “ ‘we have also noted that the Fifth Amendment’s due process clause plays a limited role in assuring that the government does not subject a defendant to oppressive delay.’ ”
Id. (quoting United States v. Spears,
159 F.3d 1081, 1084 (7th Cir.1998)). Hagler bears the burden of demonstrating that the delay caused actual and substantial prejudice to his right to a fair trial.
See id.
at 481-82. This burden “is an exacting one; the showing must rest upon more than mere speculative harm,”
id.
at 482, and Hagler must present “facts that are specific, concrete, and supported by evidence.”
United States v. Henderson,
337 F.3d 914, 920 (7th Cir.2003). If he were to make this showing, the burden would shift to the government “to demonstrate that the ‘purpose of the delay was not to gain a tactical advantage over the defendant or for some other impermissible reason.’ ”
Id. (quoting McMutuary,
217 F.3d at 482). The court then balances the government’s reasons and the defendant’s prejudice to determine whether the defendant was denied due process.
Id.
Hagler has not cleared the first hurdle. Hagler identifies three general ways in which he believes the preindictment delay prejudiced him, but we do not think that any of them cause him any actual and substantial disadvantage. First, Hagler notes that some of the physical evidence had degraded by the time of trial; specifically, one of the Halloween masks had “degraded quite a bit” and become “brittle.” But Hagler does not even suggest, much less demonstrate, how this impeded his ability to defend himself. Similarly, Hagler complains that he was not able to present the jury with the list of “hits” generated by his initial, mixed DNA profile because the list was deleted in 2007. But another, apparently more-accurate DNA test was later performed on the same genetic material, and that test matched the material to Shawn, not to Hagler. Given that Hagler later received the results of a second, more-accurate test, we do not believe that he was prejudiced by missing out on the results of the first test.
Third and more generally, Hagler argues that he was prejudiced because the witnesses’ memories have faded over time, and, as a result, they disagreed on various details at trial. But the mere fact that memories have faded is not enough to establish excessive delay.
Id.
at 919;
see also United States v. Baker,
40 F.3d 154, 157 (7th Cir.1994) (statutes of limitations reflect “a legislative judgment that so long as prosecutions are brought within the designated timeframe, then, notwithstanding the possible loss of crucial evidence or failure of memory, a defendant will be able to adequately defend himself’);
United States v. Koller,
956 F.2d 1408, 1414 (7th Cir.1992) (rejecting defendant’s speedy trial claim because his “general allegation that his witnesses’ memories faded during the delay does not rise to the level of specificity required to show actual prejudice”). Here, some of the government’s witnesses failed to remember specific details or contradicted each other on insignificant factual matters, but all of the witnesses told the same basic story. Moreover, Hagler was able to bring all of these issues to the jury’s attention through argument and cross-examination. And, most importantly, none of the witnesses’ faded recollections calls into question the significant DNA and fingerprint evidence against him. Accordingly, Ha
gler has not shown that the pre-indictment delay prejudiced him, and the district court did not abuse its discretion by declining to dismiss the case.
C. Sufficiency of the Evidence
Hagler’s next claim is that the government did not prove that National City Bank was federally protected at the time of the attempted robbery. The federally protected status of the victim bank is an essential element of bank robbery. 18 U.S.C. § 2113(a), (f);
United States v. Locklear,
97 F.3d 196, 199 (7th Cir.1996). Here, the government argued that the federal protection took the form of FDIC insurance. Thus, the government had to provide enough evidence such that a reasonable jury could find that the bank was FDIC insured at the time of the robbery.
Locklear,
97 F.3d at 199;
United States v. Higgans,
507 F.2d 808, 813 (7th Cir.1974). The government did so with two pieces of evidence: (1) the bank’s FDIC certifícate; and (2) the testimony of one of the bank’s employees.
Hagler contends that this evidence is insufficient because an FDIC certificate alone is not enough to prove federal protection and because the employee’s testimony was not supported with enough detail. We disagree. True, we have previously held that an FDIC certificate, taken alone, did not demonstrate that a bank was FDIC insured because it did not establish that the bank was insured
at the time of the robbery. See United States v. Shively,
715 F.2d 260, 265 (7th Cir.1983). But we have also held that an FDIC certificate, together with a bank employee’s testimony based on personal knowledge, are sufficient to support a conviction.
See, e.g., United States v. Hampton,
464 F.3d 687, 688 (7th Cir.2006) (“The bank employees who testified about their banks’ insured status testified that the banks were currently insured, and the jury was entitled to believe their testimony.”);
Higgans,
507 F.2d at 813 (testimony of bank vice president and FDIC certificate sufficient to establish FDIC insurance);
cf. United States v. Taylor,
728 F.2d 930, 933 (7th Cir.1984) (holding that “uncontroverted” testimony by bank’s vice president, even without FDIC certificate, was sufficient to establish FDIC insurance). Here, Stacy Gray, a physical security consultant for the bank, specifically testified that the bank was FDIC insured on the day of the robbery and that its FDIC certificate had been maintained in the ordinary course of the bank’s business. Because a jury could reasonably find that the bank was FDIC insured based on this testimony,
see Hampton,
464 F.3d at 688;
Taylor,
728 F.2d at 933;
Higgans,
507 F.2d at 813, we reject Hagler’s argument.
D. Declining to Grant Hagler’s Motion for a New Trial
Finally, Hagler contends that the trial court abused its discretion when it declined to grant his motion for a new trial. Hagler notes that a gray sweatshirt and a black jacket were found inside the getaway car. Neither was DNA tested before Hagler’s trial. But, in February 2011, both were tested in preparation for Shawn’s trial. The testing report indicated that the DNA obtained from the gray sweatshirt “demonstrated the presence of a mixture from which at least two individuals cannot be excluded as possible contributors.” However, the report also found that “[n]o conclusion can be drawn as to whether William Hagler ... or Shawn Hagler ... could be possible contributors to the mixture.” As for the samples taken from the black jacket, -the report found that the samples “failed to demonstrate a
sufficient quality of DNA for further analysis.”
Hagler argues that these inconclusive tests would have created grave doubt of his guilt had they been introduced to the jury, but we are not persuaded. Federal Rule of Criminal Procedure 38(a) gives district courts the discretion to grant a new trial “if the interest of justice so requires.” The rule is “reserved for only the most extreme eases,”
United States v. Linwood,
142 F.3d 418, 422 (7th Cir.1998) (internal quotation marks omitted), and we “approach such motions with great caution and are wary of second-guessing the determinations of both judge and jury,”
United States v. McGee, 408 F.3d 966, 979
(7th Cir.2005). To show that the interest of justice requires a new trial, a defendant must provide evidence that (1) came to his knowledge only after trial; (2) could not have been discovered sooner through the exercise of due diligence; (3) is material and not merely impeaching or cumulative; and (4) would probably lead to an acquittal in the event of a retrial.
Id.
We need not address all four requirements here; the fourth will suffice. Hagler’s fingerprint was found outside the getaway car, and his DNA was found inside it. Both of these facts were powerful evidence against him. And, contrary to Hagler’s suggestion, the introduction of further, inconclusive DNA testing would not have undermined this evidence. Suppose that instead of DNA evidence, the government offered five eyewitnesses who all saw the perpetrator’s face during the robbery and identified him at trial. Then suppose that, after the jury found him guilty, it came to light that two other witnesses saw the robbery, but neither could identify the perpetrator because they did not get a good look at his face. The testimony of those new witnesses would not
disprove
the testimony of the trial witnesses; it would simply fail to lend it further support. As such, it would be evidence of innocence only in the most attenuated sense. The absence of proof (or, more accurately here, the absence of further proof) is not proof of absence.
Cf. Denson v. United States,
574 F.3d 1318, 1343 (11th Cir.2009).
Hagler’s DNA and fingerprints were strong proof of his guilt. The mere fact that his DNA was not found in
greater abundance
does little to undermine this proof. Thus, we do not think that it would probably result in an acquittal if presented to a jury, and we certainly do not think that the district court abused its discretion in so holding.
III. Conclusion
We Affirm Hagler’s conviction.