UNITED STATES of America, Plaintiff-Appellee, v. Lamar E. SANDERS, Defendant-Appellant, 708 F.3d 976

KANNE, Circuit Judge. In January 2008, Lamar E. Sanders and an accomplice abducted Timicka Nobles’s daughter, R.E. The reason: to induce Nobles to rob her own mother. Nobles attempted to comply — she left a bag of cash for Sanders’s accomplice to pick up— but law enforcement authorities were already apprised of the plot. They quickly arrested Sanders’s accomplice, and Sanders turned himself in shortly thereafter. Fortunately, no one was injured, and police recovered the money. After a five-day trial, a jury found Sanders guilty of kidnapping and extortion. He now appeals his conviction and sentence.

First, Sanders argues that the district court denied him due process by admitting Nobles’s three identifications of him. Second, Sanders claims that the district court ran afoul of the Confrontation Clause, or, alternatively, abused its discretion, by limiting his cross-examination of Nobles. Finally, Sanders contends that the district court applied the incorrect mandatory minimum sentence. Finding no error, we affirm both the conviction and sentence. I. Background Portage, Indiana. Saturday, January 5, 2008. 8:00 a.m.: Timicka Nobles has a busy morning. She has to be at work in Chicago soon.

Plus, along the way, she needs to stop by her mother’s house to drop off R.E., her ten-year-old daughter. Putting on shoes in the apartment entryway, Nobles and R.E. prepare to depart. As Nobles opens the front door, two men force their way inside. Pushing R.E. and Nobles back into the apartment, the men begin their ill-fated kidnapping operation. The first man, Ralph Scott, holds R.E. hostage in the living room, while the second man, Lamar Sanders, points a gun at Nobles and orders her into the bedroom. There, Sanders has Nobles face the wall as he lays out his demands.

Nobles must drive to her workplace in Chicago — a currency exchange owned by her mother. She will park her car nearby and leave it unlocked. Nobles will then enter the exchange as if nothing is wrong, as if it were any other day. Today, however, Nobles must empty the safe into a black garbage bag. She will take that bag, place it on the front seat of her car, and walk away. If she follows these instructions exactly, “things won’t get messed up.” (Trial Tr. at 390.) Nobles acquiesces.

As Sanders leads her back into the living room, she finds R.E. alone; Scott had left for Chicago minutes earlier in Sanders’s Dodge Magnum. Nobles gives her daughter a quick hug before Sanders orders R.E. to blindfold herself with her headband. Notably, it does not entirely cover R.E.’s eyes; she can still see above and below the band. Our antagonists did not learn from tales of countless foiled criminals never to leave a hostage unattended. As Sanders drove R.E. to Chicago in Scott’s Chevy Trailblazer (while remaining in frequent phone contact with Scott), he did not follow Nobles.

Realizing as much, Nobles stopped at a gas station and went into the attached convenience store. Concerned by Nobles’s apparent distress, the clerk allowed her to use the store’s phone. Nobles made a frantic phone call to her mother and warned her of the plot. Her mother alerted the security officers at the exchange, who in turn notified the Chicago Police. Thus, when Nobles arrived at the exchange, the authorities were prepared. Nobles did as Sanders ordered. She took the money from the safe, placed it in a garbage bag, and set the bag on the front seat of her car.

After Nobles walked away, Scott, who had parked Sanders’s Magnum near the scene, approached and removed the money bag. As he did, two exchange security officers and a Chicago Police sergeant ran towards him. Fleeing the scene, Scott ditched the bag in a bush. The officers quickly caught up, arrested Scott, and recovered the money. Observing Scott’s downfall from a block away, Sanders ordered R.E. out of the Trailblazer and sped away. When R.E. removed her headband, she recognized where she was and walked to the currency exchange, where she was reunited with her mother.

Just minutes after Scott’s arrest, Sanders called his mother. He then called his Arizona-based girlfriend, Carlena Williams. Sanders told Williams that his phone — the same phone on which he was making the call — had been stolen. Williams paid Sanders’s phone bill, so she promptly called Verizon and had service suspended on his phone (but she would reinstate the service later that same day). Back in Chicago, R.E. identified Scott as the man who had guarded her in the living room. The police also searched Scott’s pockets, where they found a key fob.

Taking the device in hand, an officer continuously pressed the unlock button while walking up and down nearby streets. When the fob activated Sanders’s Magnum, evidence technicians searched the car. Inside, they found Sanders’s driver’s license and seven photographs from a recent birthday party. In five of the images, Sanders appeared with various combinations of family and friends. An officer took these photographs back to the exchange and interrupted Nobles’s interview with a detective. The officer showed Nobles one or two photos and asked her if she recognized anyone.

Witnesses disagree about how many and which specific photos Nobles saw. She viewed at most two photographs. Of those, one depicted Sanders with two women, while the other depicted him with two other men: Scott and Sanders’s brother. All agree, however, that Nobles identified Sanders in at least one photograph as the second man in her apartment that morning. At this time, Nobles also gave an inaccurate verbal description of Sanders’s build that was off by about five inches and sixty pounds. This interview occurred within a couple hours of the kidnapping. R.E. was not shown the photographs found in the car.

Approximately two hours after Nobles’s first interview with law enforcement, officers drove her and R.E. to the Chicago Police Department. There, Nobles was shown a formal photo array. The array placed photos of Sanders alongside those of five other men with similar height, weight, and facial features. The other individuals in the photos were chosen based upon similarities to Sanders’s actual features, as opposed to the inaccurate verbal description that Nobles gave dui’ing her first interview. Nobles again identified Sanders. R.E. was independently shown a different array in another room.

She also identified Sanders. Following these identifications, the government issued a criminal complaint, and Sanders turned himself in shortly thereafter. As the case proceeded to trial, Sanders moved to suppress Nobles’s identifications of him. Sanders had three theories behind this motion. First, he argued that showing Nobles the birthday party photographs was so unnecessarily suggestive as to violate the Due Process Clause. Second, he asserted that the photo array was imper-missibly suggestive because only he appeared in both the photos on the scene and in the subsequent array.

Finally, Sanders claimed that any in-court identification by Nobles could only be the product of these previous, allegedly tainted, identifications. The district court denied Sanders’s motion on all three grounds. Also prior to trial, the government moved to limit cross-examination of Nobles based on her previous convictions. In 2001, when working at a different currency exchange in Chicago (one not owned by her mother), Nobles forged and delivered at least six fraudulent checks. She was subsequently convicted for these crimes and was sentenced to both boot camp and three years in prison.

In its motion, the prosecution sought to limit the admission of details surrounding these convictions. The government conceded that Sanders should be allowed to introduce the fact that Nobles was convicted of theft and forgery, the dates of those crimes, and her sentence. The government, however, argued that Sanders should not be allowed to elicit any further details about the crimes, including the fact that they occurred at a currency exchange. The district court agreed and imposed the requested limitations. At a five-day jury trial, the government presented a strong case.

Nobles identified Sanders as the second man in her apartment the morning of the kidnapping. So did R.E. The government also presented cell phone records showing that Sanders’s phone was in frequent contact with Scott’s phone throughout the morning of the crime. Expert witnesses traced the cell towers used during these calls to show that the phones traveled the approximate path of the kidnappers. Although the phone records could not directly verify that Sanders had his phone, other evidence spoke to that question.

The morning of the kidnapping, Sanders called Carlena Williams and told her that his phone was stolen. Records showed a corresponding call from Sanders’s phone to Williams’s phone, made from the vicinity of the currency exchange, approximately ten minutes after Scott’s arrest. The records also showed that this call was made from Sanders’s own phone — the same one he was claiming was stolen. Just after that call, Williams had the service on Sanders’s phone suspended, although she reinstated it later that evening. Defense counsel criticized Nobles’s identifications and tried to implicate Nobles herself.

Nobles remained romantically involved with Vincent E., R.E.’s father, who was also a fellow gang member of Sanders and Scott. Scott, who signed a plea agreement with the government, testified that Vincent had planned the whole plot and that Nobles was complicit in the scheme. The jury, however, did not believe Sanders’s defense. On January 24, 2011, he was found guilty of one count of kidnapping under 18 U.S.C. § 1201 and one count of extortion under 18 U.S.C. § 1951. The district court sentenced Sanders on September 28. Two mandatory mínimums apply to kidnapping: 18 U.S.C. § 1201(

g) requires twenty years and 18 U.S.C. § 8559(f)(2) requires twenty-five years. The district court concluded that the higher penalty applied and accordingly sentenced Sanders to concurrent sentences of twenty-five years on the kidnapping count and twenty years on the extortion count. The court also sentenced Sanders to five years of supervised release. Sanders timely appealed on October 7, 2011. II. Analysis Sanders makes several arguments. He first contends that the district court violated the Due Process Clause by admitting into evidence each of Nobles’s three identifications of him.

Second, he challenges the district court’s decision to limit cross-examination on Nobles’s prior convictions. Finally, he claims that the district court should have applied the lower of the two applicable mandatory minimum sentences. We address each of these arguments in turn. A. Identification Testimony Our Constitution protects against “conviction based on evidence of questionable reliability.” Perry v. New Hampshire, — U.S. -, 132 S.Ct. 716, 723, 181 L.Ed.2d 694 (2012). Despite the importance of this right, the admission of evidence rarely implicates due process. See id.

Rather, courts typically rely on other means to ensure reliable evidence — state and federal rules, as well as different constitutional guarantees, such as the Sixth Amendment rights to counsel and confrontation. Id. Yet, “when evidence ‘is so extremely unfair that its admission violates fundamental conceptions of justice,’ ” due process, like the sleeping giant, awakens. Id. (quoting Dowling v. United States, 493 U.S. 342, 352, 110 S.Ct. 668, 107 L.Ed.2d 708 (1990)). In those situations, other protections have proven insufficient, and courts must step in to prevent injustice.

Unduly suggestive identification procedures represent one example of those fundamentally unfair situations. A procedure becomes so flawed as to implicate due process when it creates a “very substantial likelihood of irreparable misidentification.” Neil v. Biggers, 409 U.S. 188, 198, 93 S.Ct. 375, 34 L.Ed.2d 401 (1972) (quoting Simmons v. United States, 390 U.S. 377, 384, 88 S.Ct. 967, 19 L.Ed.2d 1247 (1968)). In such cases, the identification must be suppressed. Perry, 132 S.Ct. at 724-25. To decide whether a situation has risen to that level, we follow a two-pronged approach.

First, we consider whether the identification procedure used by law enforcement was “both suggestive and unnecessary.” Id. at 724; accord United States v. Gallo-Moreno, 584 F.3d 751, 757 (7th Cir.2009). Second, we examine the “totality of the circumstances” to determine whether other indicia of reliability “outweigh[ ] ... the corrupting effect of law enforcement suggestion.” Perry, 132 S.Ct. at 725 (internal quotation marks omitted); accord Gallo-Moreno, 584 F.3d at 757.

As the Supreme Court recently reiterated, courts will only consider the second prong if a challenged procedure does not pass muster under the first. See Perry, 132 S.Ct. at 730. To fail the first prong, however, even a “suggestive” procedure must also be “unnecessary.” Id. at 724. In other words, the situation must have involved “improper state conduct” — one in which the circumstances did not justify law enforcement’s suggestive behavior. Id. at 728. As these descriptions show, both prongs are highly situation-dependent, which may seem to blend the two inquires. Yet, they are distinct.

The first prong focuses on police conduct — its suggestiveness and necessity in the specific situation at hand. In contrast, the second prong focuses on the identifying witness and her knowledge of the suspect absent the suggestive procedure. Perhaps, for example, the witness saw the suspect for several minutes in broad daylight. See United States v. Kimbrough, 528 F.2d 1242, 1246-47 (7th Cir.1976). Such considerations could lead us to conclude that an unduly suggestive identification was nonetheless reliable, such that its admission would not violate the Due Process Clause. See id.

We will therefore begin by applying this dual-pronged standard to Nobles’s first two identifications of Sanders. The first identification occurred shortly after the crime, when Chicago Police officers showed Nobles the birthday party photographs removed from Sanders’s car. The second identification occurred a few hours later, when Nobles took

part in a formal photo array. After addressing those instances, we can examine Nobles’s third identification, made during trial. If either of the first two procedures was unnecessarily suggestive, then the in-court identification must demonstrate an independent basis of reliability to be admissible. See United States v. Rogers, 387 F.3d 925, 937-38 (7th Cir.2004); see also Cossel v. Miller, 229 F.3d 649, 655 (7th Cir.2000). As we consider each of these three questions, our review is de novo with “due deference to the trial court’s findings of historical fact.” United States v.

Benabe, 654 F.3d 753, 774 (7th Cir.2011) (internal quotation marks omitted). 1. Identification in the Birthday Party Photographs Under the first prong of our inquiry, we now analyze whether showing Nobles the photographs found in Sanders’s car was “both suggestive and unnecessary.” Perry, 132 S.Ct. at 724. a. Suggestiveness According to Sanders, the police conducted a “show up” when they asked Nobles about the birthday party photographs. In a show up, the police present only one suspect to the identifying witness. United States v. Funches, 84 F.3d 249, 254 (7th Cir.1996).

Consequently, show ups are “inherently suggestive.” United States v. Hawkins, 499 F.3d 703, 707 (7th Cir.2007). Yet, it remains unclear whether this identification procedure actually was a show up, as defined by our case law. We have most often used that term to describe situations in which law enforcement have apprehended a suspect and then physically shown that person to a witness. See, e.g., United States v. Newman, 144 F.3d 531, 535 (7th Cir.1998); Abrams v. Barnett, 121 F.3d 1036, 1040-41 (7th Cir.1997); Funch-es, 84 F.3d at 254.

Here, however, the police showed Nobles one (or potentially two) photographs of Sanders. Our cases leave unsettled whether we also consider it a “show up” when a witness is presented only with the suspect’s photograph. Compare Cosset, 229 F.3d at 655 (describing a photographic “show-up”), with Hawkins, 499 F.3d at 708 (describing photographic identification as “akin to a showup”). An added wrinkle stems from witnesses’ disagreement over which photographs Nobles saw. Again, a show up involves presenting a witness with only one suspect.

Here, however, one of the photos allegedly shown to Nobles would have given her the opportunity to identify another male of similar features to Sanders (his brother). That said, the police showed Nobles at most two photographs, and those photos presented, at most, one other possible suspect for identification. For that reason, the procedure was closer to a show up than other photographic identification techniques, such as a line up, in which several suspects are presented. Cf. United States v.

Clark, 989 F.2d 1490, 1495 n. 2 (7th Cir.1993) (showing a witness two arrested suspects was more analogous to a show up than a line up). Still, for the sake of consistency, we will not refer to the procedure as a “show up.” When Sanders argues that this procedure was suggestive, he ignores a key trend in our case law. In recent years, we have noted proliferating social science data on the reliability of eyewitness testimony. See United States v. Ford, 683 F.3d 761, 766 (7th Cir.2012) (collecting articles).

Accordingly, we have held that scientific sources should generally accompany an argument that a particular procedure was unnecessarily suggestive. United States v. Acox, 595 F.3d 729, 730 (7th Cir.2010) (“Lawyers’ assertions that the effects of a photo spread are ‘clear’ or ‘obvious’ are no substitute for evidence”); see also United States v. Williams, 522 F.3d 809, 812 (7th Cir.2008). Sanders has not properly presented us with such data here. Although his counsel submitted several sources after oral argument, in accordance with Federal Rule of Appellate Procedure 28(j), this attempt is too little, too late.

These sources raise complicated points that Sanders should have addressed in his briefs. A Rule 28(

j) letter is not the appropriate forum to make new, complex arguments. Spiegla v. Hull, 481 F.3d 961, 965 (7th Cir.2007). Yet, we have not made social science data a strict requirement for us to determine whether a procedure was unnecessarily suggestive; “[o]ften the right disposition will be evident with or without the aid of social science.” Williams, 522 F.3d at 812. Such is the case here. As stated earlier, the procedure employed by law enforcement, with the paucity of suspects presented to Nobles, was similar to a show up. We have previously found show ups “inherently suggestive.” Hawkins, 499 F.3d at 707.

Therefore, it seems likely that this procedure was also suggestive. But since we do not have data to help us resolve that question, we think it best to set it aside for now. Rather, we can simply assume, for current purposes, that it was suggestive, because the disposition is nevertheless clear: Sanders’s claim falters because he cannot prove that the procedure, even if suggestive, was also unnecessary. b. Necessity As discussed, a procedure that is suggestive — even when inherently so — may still be necessary. Perry, 132 S.Ct. at 724; accord United States v. Recendiz, 557 F.3d 511, 525 (7th Cir.2009).

As in all aspects of life, context matters. Thus, the circumstances surrounding an investigation can justify even a show up. Hawkins, 499 F.3d at 707-08. Citing United States v. Funches, Sanders argues that show ups are only acceptable to “allow identification before the suspect has altered his appearance” or to “permit the quick release of innocent persons” if witnesses cannot identify the apprehended individual. 84 F.3d at 254. Sanders then argues that the situation here does not present one of those exceptions, thus rendering evidence derived from the procedure impermissible.

Funches, however, merely listed “example[s]” of acceptable reasons for a show up; it did not claim to be exhaustive. Id. Moreover, Funches involved a physical show up, not a single-photograph identification procedure. Id. at 251-52. Photographic identification techniques, while perhaps similar in some ways to physical show ups, are also different in important ways. When police conduct a physical show up, they already have the suspect in custody. Therefore, there is not the same exigency to catch a criminal on the loose, which, depending on the situation, could justify the suggestive procedure.

For that reason, Simmons v. United States, rather than physical show up cases, presents the key precedent. 390 U.S. 377, 88 S.Ct. 967. In that case, Simmons committed an armed robbery of a bank with an accomplice. Id. at 379-80, 88 S.Ct. 967. The next morning, police obtained a few photographs that depicted Simmons with the man they suspected of being his accomplice. Id. at 380, 88 S.Ct. 967. Later that day, officers showed the photographs to five bank employees. Id. Every witness identified Simmons as one of the robbers. Id.

After being convicted, Simmons argued that the photographic identification procedure violated the Due Process Clause. Id. at 381-82, 88 S.Ct. 967. The Supreme Court, however, disagreed, writing: [a] serious felony had been committed. The perpetrators were still at large. The inconclusive clues which law enforcement officials possessed led to ... Simmons. It was essential for the FBI agents swiftly to determine whether they were on the right track, so that they could properly deploy their forces ... and, if necessary, alert officials in other cities. Id. at 384-85, 88 S.Ct. 967.

These circumstances, the Court held, justified the identification procedure used. Id. at 386, 88 S.Ct. 967. This case presents clear parallels to Simmons. Here, a serious felony had also been committed: someone had kidnapped R.E. to induce Nobles to rob her own mother’s currency exchange. The police also had clues pointing to a suspect: Sanders. Although law enforcement officers had arrested Scott, they knew a second man was involved, and their best clues were the photos found in the car Scott drove. Finally, with an armed felon still on the loose, the police needed to act quickly.

Showing Nobles the photos was the best way to proceed. In fact, the situation in this case presented even greater necessity than in Simmons. Here, police showed Nobles the photographs within a couple hours of the crime. In Simmons, on the other hand, the Supreme Court upheld a procedure in which the police did not show the photos to witnesses until the next day. With the crime much closer at hand here, the rationale for upholding the procedure as necessary is even more pressing. Sanders’s attempts to distinguish Simmons do not persuade.

First, he argues that Simmons only challenged the witnesses’ in-court identifications. Sanders does not explain why that distinction matters, but, in any event, his assertion is incorrect; Simmons also challenged the pretrial identifications that resulted from the procedures described above, just as Sanders does here. Simmons, 390 U.S. at 381-82, 88 S.Ct. 967. Second, Sanders contends that the witnesses in Simmons were shown mostly group photographs, thereby making the identifications more reliable than the one here.

But that argument does not negate any of our necessity analysis, and finding a suggestive procedure unnecessary is a prerequisite to considering other indicia of reliability. See Perry, 132 S.Ct. at 730. Thus, because Sanders has not effectively distinguished Simmons, we need not turn to such considerations under the second prong. Our own cases involving unnecessarily suggestive photographic identifications are not to the contrary. For example, in United States v. Kimbrough, the police showed the witness a composite sketch of the suspect followed by photos of only the defendant. 528 F.2d at 1244.

Yet, the authorities had no good reason for failing to use more images; they could have easily produced an array using other available photos. Id. at 1244-45. Similarly, in Israel v. Odom, the police showed a rape victim only an image of her suspected assailant. 521 F.2d 1370, 1372 (7th Cir.1975). Before doing so, the police had left the scene of the crime, gone back to the station, pulled up a stored image of the suspect, and returned to the victim’s home. Id.

Given that sequence of events, “[n]o appreciable time would have been lost” by pulling a few extra files at the station so the victim could have viewed more potential suspects. Id. at 1375. Those cases are distinguishable. Here, the police obtained the photographs on the scene of the crime itself, while an out-of-town victim was still present, and her memory was at its freshest. It would have taken significantly more time for the police to leave the scene, go to the station house, locate photos similar to those found in the car, and return.

A dangerous suspect could have used that extra time to facilitate his escape. Thus, unlike in Kim-brough and Israel, the police in this case could not have produced a significantly less suggestive procedure without sacrificing critical time. In this quickly developing situation, showing Nobles the photographs was the most responsible way to proceed with the early stages of investigation. Law enforcement’s procedure may have been suggestive to some degree, but it was also necessary. Because we find that the procedure was necessary, we need not address the reliability prong of the analysis.

See Perry, 132 S.Ct. at 730. Rather, we can rely on our criminal procedure to ensure due process. After we are convinced that no constitutional violation occurred, “the jury, not the judge, ... determines the reliability of evidence.” Id. at 728. c. Harmless Error Although we do not think there was any error in admitting Nobles’s first identification of Sanders, even if there was, we could alternatively resolve the issue as harmless error. Constitutional errors (or, as here, potential constitutional errors) divide into two categories for determining their amenability to harmless error review. United States v.

Gonzalez-Lopez, 548 U.S. 140, 148, 126 S.Ct. 2557, 165 L.Ed.2d 409 (2006). The first is “structural defects,” which “affect the framework within which the trial proceeds.” Id. (internal brackets omitted). Examples of these grave errors, which are not subject to harmless error review, include denial of the right to counsel or denial of self-representation. Id. at 148-49, 126 S.Ct. 2557. In contrast, “trial error[s]” merely affect the “presentation of the case to the jury” and are thus subject to harmless error review. Id. at 148, 126 S.Ct. 2557. This case falls into that latter category.

See Rogers, 387 F.3d at 939 (acknowledging- that admission of identification testimony can be harmless). Therefore, even if admitting Nobles’s identification proved erroneous, Sanders’s conviction will stand if the error was “harmless beyond a reasonable doubt.” Gonzalez-Lopez, 548 U.S. at 148, 126 S.Ct. 2557; accord United States v. Garcia, 528 F.3d 481, 485 (7th Cir.2008).

In making that determination, we consider factors such as “(1) the importance of the witness’s testimony in the prosecution’s case; (2) whether the testimony was cumulative; (3) whether other evidence corroborated or contradicted the witness’s material testimony; and (4) the overall strength' of the prosecution’s case.” United States v. Ochoa, 229 F.3d 631, 640 (7th Cir.2000) (citing Delaware v. Van Arsdall, 475 U.S. 673, 684, 106 S.Ct. 1431, 89 L.Ed.2d 674 (1986)). These factors lead us to conclude that any potential error in admitting Nobles’s identification was harmless beyond a reasonable doubt.

First, R.E. independently identified Sanders in a formal photo array. The Chicago Police did not show R.E. the birthday party photos found in the car, so Sanders cannot criticize her array as tainted. R.E. also had greater opportunity to observe Sanders, having spent nearly an hour in the car with him (albeit with her eyes partially covered). Finally, the cell phone records foreclose any lingering doubt.

Sanders’s cell phone was in frequent contact with Scott’s throughout the morning of the kidnapping, and records show that these phones traveled the approximate path of the kidnappers at the same time as the crime. Sanders argues that someone else had his cell phone — just as someone else had his car and wallet. That argument cannot withstand the other evidence. Sanders’s own mother testified that he frequently swapped cars with others, including Scott. In addition, just a few minutes after authorities arrested Scott, Sanders called both his mother and his girlfriend, Carlena Williams.

Sanders told Williams that his phone had been stolen. Yet, according to the cell phone records introduced into evidence, that call to Williams was made on Sanders’s phone — the same one he claimed was stolen. Williams promptly suspended service on Sanders’s phone, just a few minutes after Scott’s arrest. Later that same day, however, Williams reinstated the service on Sanders’s phone,

an act further undercutting the claim it was stolen. In light of this overwhelming evidence, any potential error in admitting Nobles’s initial identification was harmless beyond a reasonable doubt. 2. Identification in the Photo Array Sanders next challenges Nobles’s second identification of him, which occurred during a photo array conducted a few hours after her initial interview with law enforcement. The concern with this identification is repetition: Sanders was the only person whose photographs appeared both in Nobles’s initial interview and in the array.

To start, we note that “there is nothing per se impermissible about placing the same suspect in two different identification procedures.” United States v. Griffin, 493 F.3d 856, 865 (7th Cir.2007). Yet, when upholding repetitive procedures in the past, we have often focused on the mitigating effect of elapsed time between the identifications. See, e.g., id. at 865-66 (two months between line up and photo array); United States v. Carter, 410 F.3d 942, 949 (7th Cir.2005) (three months between photo arrays); United States v.

Harris, 281 F.3d 667, 670-71 (7th Cir.2002) (six months between photo array and line up); Stewart v. Duckworth, 93 F.3d 262, 265 (7th Cir.1996) (eleven days between photo arrays); United States v. Cord, 654 F.2d 490, 492-93 (7th Cir.1981) (two weeks between photo arrays). Here, however, a mere two hours passed between showing Nobles the birthday party photos and conducting the array. Regardless, we need not decide the more difficult constitutional question today. Instead, we can resolve the issue using the same harmless error analysis discussed above.

The same reasoning applies with equal force to this identification. The government’s evidence was strong, and Sander’s case weak. As a result, any error in admitting the second identification was also harmless beyond a reasonable doubt. 3. In-Court Identification Third, Sanders challenges Nobles’s in-court identification, which he claims lacked an independent basis of reliability. Since we have already concluded that the first photo identification was constitutional, we need not be concerned with potential taint from that procedure.

Therefore, the only remaining concern arises out of any potential taint from the photo array identification, for which we have reserved the constitutional question. Even if the photo array proved unnecessarily suggestive, the district court would have nonetheless properly admitted Nobles’s in-court identification if “clear and convincing evidence [shows] that [it] was based upon observations ... other than at the prior, illegal identification, or, alternatively, ... that the error complained of was harmless beyond a reasonable doubt.” Cossel, 229 F.3d at 655.

We believe that any error here was also harmless, for the same reasons discussed above. A Cumulative Error As his final due process argument, Sanders claims that any potential errors described above, taken together, present an issue of cumulative error. Such would be the case if the errors “could possibly have influenced the jury to reach an improper result.” United States v. Rogers, 89 F.3d 1326, 1338 (7th Cir.1996). That is not the case here. For the reasons already discussed, Nobles’s identifications were largely cumulative, given R.E.’s testimony.

Furthermore, defense counsel had ample opportunity to discredit Nobles’s identifications through cross-examination and hammered home that point during closing argument. The district court also instructed the jury about weighing the reliability of identifications and considering the circumstances under which they were made. Most importantly, Nobles’s identifications—separately or in tandem—were simply a drop in the bucket compared to the overwhelming evidence against Sanders. Even if admitting all three identifications was erroneous, the impact was harmless beyond a reasonable doubt.

We thus find no due process violation. B. Limitation on Cross-Examination Next, we address two separate challenges to a limitation imposed by the district court on Sanders’s cross-examination of Nobles. In 2001, Nobles was convicted of theft and forgery for making and delivering at least six forged checks in connection with her employment. At the time, Nobles worked for a different currency exchange in Chicago. Nobles was sentenced to both boot camp and three years in prison. Prior to Sanders’s trial, the district court granted the government’s motion to limit testimony regarding these convictions.

The court ruled that Nobles’s convictions could come into evidence, along with the type of crime, date, and sentence; but, the defense could not probe any details of the offenses, including the fact that they involved a currency exchange. Sanders, however, had hoped to use Nobles’s past to implicate her. Nobles remained romantically involved with Vincent E., R.E.’s father. Scott testified that Vincent had planned the whole plot and that Nobles was complicit.

This information, Sanders argues, would have made Nobles a prime suspect to the police; then, their suspicions would have grown when they learned that Nobles had previously committed crimes at a currency exchange. Thus, Sanders claims that Nobles needed to divert police attention away from herself — which she did by falsely implicating him during the identification procedures. Sanders also argues that this theory makes it less likely he committed the crime. When the district court disallowed details of Nobles’s past convictions, the court allegedly thwarted this defense theory.

For that reason, Sanders now claims that the limitation violated his right to confront witnesses against him. Alternatively, he argues that the district court abused its discretion in denying admission of the evidence. 1. Confrontation Clause The Confrontation Clause assures the ability of the accused in criminal prosecutions “to be confronted with the witnesses against him.” U.S. Const, amend. VI. Sanders contends that the district court violated this right by limiting his cross-examination of Nobles.

When considering a district court’s restraint on cross-examination, our standard of review depends upon whether the limit “directly implicate[d] the core values of the Confrontation Clause.” Recendiz, 557 F.3d at 530 (internal quotation marks omitted). If it did, we review de novo; “otherwise, we review for abuse of discretion.” Id. “[E]xposing witness bias” lies within the protected “core” of the Confrontation Clause, United States v. Manske, 186 F.3d 770, 778 (7th Cir.1999), which may lead one to believe our review here is de novo.

Yet, “the Confrontation Clause guarantees an opportunity for effective cross-examination, not cross-examination that is effective in whatever way, and to whatever extent, the defense might wish.” Delaware v. Fensterer, 474 U.S. 15, 20, 106 S.Ct. 292, 88 L.Ed.2d 15 (1985). Thus, a limitation on cross-examination implicates the core of the Confrontation Clause when “the defense is completely forbidden from exposing the witness’s bias.” Manske, 186 F.3d at 778.

But after the defendant reveals a witness’s motive to lie, “how much opportunity defense counsel gets to hammer that point down to the jury” becomes “peripheral” for Sixth Amendment purposes. Id.; accord United States v. Scott, 145 F.3d 878, 888 (7th Cir.1998) (citing Van Arsdall, 475 U.S. at 682, 106 S.Ct. 1431). In other words, merely having the chance to present a motive to lie is sufficient to satisfy the core values of the confrontation right. See, e.g., Recendiz, 557 F.3d at 530; Manske, 186 F.3d at 778. Here, the district court did not foreclose Sanders’s ability to establish Nobles’s bias.

Scott testified that Vincent had planned the kidnapping and that Nobles was complicit in the scheme. Then, through cross-examining Nobles, Sanders established that she remained romantically involved with Vincent during and after the kidnapping. Defense counsel also asked Nobles about a time when Vincent told her that if “anyone confronted [her] or robbed [her], that [she] should just do exactly what they said.” (Trial Tr. at 425.) Immediately after eliciting that testimony, Sanders’s attorney asked Nobles about her prior crimes.

Nobles admitted to her convictions and sentences for theft and forgery. (Id. at 426-27.) Those pieces of testimony collectively establish Sanders’s defense theory: because Nobles and Vincent were involved in the plot, and Nobles had a criminal history, she had a motive to implicate Sanders and turn attention away from herself. This theory may have been ever-so-slightly more compelling if the jury knew Nobles’s prior crimes involved a currency exchange. But the defense still presented the theory to the jury — and that is the critical point.

After that threshold has been crossed, it is not important, for Sixth Amendment purposes, how much counsel was able to drive the point home. The district court’s limitation on cross-examination thus did not implicate the core values of the Confrontation Clause. Consequently, we review the limit for abuse of discretion. See Recendiz, 557 F.3d at 530. In so doing, we must determine “whether the jury had sufficient information to make a discriminating appraisal of the witness’s motives and biases.” Id. (internal quotation marks omitted).

In light of the previous discussion, we cannot find that the district court abused its discretion. Sanders presented the jury with his entire theory of Nobles’s motive to lie. The fact that the prior convictions involved crimes at another currency exchange would not have given the jury any further material information in appraising her credibility. The jury might not have possessed all the information Sanders wanted it to have, but it certainly had sufficient information to evaluate Nobles’s testimony. 2.

Reverse WMb) Evidence Sanders next argues that the district court did not understand his reasons for introducing details about Nobles’s prior convictions and thus used the wrong approach in analyzing their admissibility. As mentioned earlier, Sanders had two related reasons for cross-examining Nobles about her prior crimes: introducing her motive to lie and casting doubt on Sanders’s own guilt. We have already addressed the attempt to show bias in the Confrontation Clause analysis above and concluded that the district court did not abuse its discretion in prohibiting the testimony with that purpose in mind.

Now, we turn to Sanders’s second reason for offering the testimony — making his guilt less likely. Sanders could attempt to do so by arguing that Nobles’s past offenses and this kidnapping were so alike that, if Sanders did not commit the previous crimes, then it is less likely he committed this one. Given that purpose, Sanders argues that the district court should have used Federal Rule of Evidence 404(

b) to analyze whether the details of Nobles’s convictions were admissible. Rule 404(

b) provides that, “Evidence of a [past] crime ... is not admissible to prove a person’s character in order to show that on a particular occasion the person acted in accordance with the character.” Fed.R.Evid. 404(b)(1). Yet, such evidence “may be admissible for another purpose.” Fed. R.Evid. 404(b)(2). For example, as Sanders hoped to do, a defendant can seek to introduce evidence of a government witness’s prior bad acts if that evidence tends to negate the defendant’s guilt. United States v. Alayeto, 628 F.3d 917, 921 (7th Cir.2010).

Colloquially (at least among lawyers), such evidence is referred to as “reverse 404(b)” evidence. United States v. Savage, 505 F.3d 754, 761 (7th Cir.2007); United States v. Wilson, 307 F.3d 596, 601 (7th Cir.2002). When deciding the admissibility of reverse 404(

b) evidence, the district court must determine whether the information’s probative value is outweighed by other considerations, such as undue prejudice, confusion of the issues, or delay. Savage, 505 F.3d at 761; Wilson, 307 F.3d at 601. We review the district court’s findings on this non-constitutional question for abuse of discretion. Wilson, 307 F.3d at 599. Sanders’s argument relies on United States v. Murray, in which we said, “[c]on-cern with the poisonous effect on the jury of propensity evidence is minimal” when a defendant attempts to employ reverse 404(

b) evidence. 474 F.3d 938, 939 (7th Cir.2007). Sanders thus claims that, given the minimal risks associated with admitting such evidence, the district court abused its discretion in excluding it. The key follow-up to the quotation cited by Sanders, however, comes in the next paragraph of that opinion.

The most “serious objection to [reverse 404(b) ] evidence is that its probative value is slight”; in other words, “unless the other crime and the present crime are sufficiently alike to make it likely that the same person committed both crimes, so that if the defendant did not commit the other crime he probably did not commit this one, the evidence will flunk.” Id. This case demonstrates that precise concern.

Nobles’s prior convictions were not at all similar to the case at hand; a world of difference separates forging a check and plotting to kidnap your own daughter at gunpoint before robbing your mother’s business. Yes, both crimes involved currency exchanges, but the similarity ends there. With only a gossamer thread connecting Nobles’s prior convictions to the current crime, further details of her past offenses had minuscule probative value. More importantly, the other considerations referenced earlier clearly outweighed that value.

Like the district judge, we are skeptical that these details were offered for any reason beyond attempting to show conformity with prior unlawful conduct. As such, the district court did not abuse its discretion in finding the evidence inadmissible. C. Mandatory Minimum Sentence Sanders last challenges his sentence. The district court sentenced Sanders to the twenty-five-year mandatory minimum imposed by 18 U.S.C. § 3559(f)(2). Sanders contends that he should have instead received the twenty-year minimum imposed by 18 U.S.C. § 1201(g). Because statutory

interpretation presents a question of law, we review de novo. United States v. Rosenbohm, 564 F.3d 820, 822 (7th Cir.2009). Federal criminal law defines kidnapping in 18 U.S.C. § 1201(a). The version of the offense relevant here occurs when a person “unlawfully seizes, confines, inveigles, decoys, kidnaps, abducts, or carries away and holds for ransom ... any person ... when ... the person is willfully transported in interstate or foreign commerce ... or the offender travels in interstate or foreign commerce.” 18 U.S.C. § 1201(a).

Starting in 2008, Congress imposed a new heightened penalty for kidnappings involving minors. See Prosecutorial Remedies and Other Tools to End the Exploitation of Children Today Act of 2008, Pub.L. No. 108-21, § 104(b), 117 Stat. 650, 653 (2003) (codified at 18 U.S.C. § 1201(g)). Under the new provision, if the victim was under the age of eighteen, and the kidnapper did not belong to an enumerated list of relatives, the offender had to be sentenced to “not less than 20 years” in prison. 18 U.S.C. § 1201(g).

Then, in 2006, Congress enacted another law that enhanced the mandatory minimums applicable to those who commit certain “erime[s] of violence” against minors. Adam Walsh Child Protection and Safety Act of 2006, Pub.L. No. 109-248, § 202, 120 Stat. 587, 612 (2006) (codified at 18’ U.S.C. § 3559(f)). Specifically at issue here, “if the crime of violence is kidnapping (as defined in [18 U.S.C. § 1201])” the offender shall “be imprisoned for life or any term of years not less than 25.” 18 U.S.C. § 3559(f)(2).

This heightened penalty applies “unless a greater mandatory minimum sentence of imprisonment is otherwise provided by law and regardless of any maximum term of imprisonment otherwise provided for the offense.” 18 U.S.C. § 3559(f). Sanders argues using a well-known canon of statutory

interpretation: courts interpret statutes to avoid rendering portions of them “superfluous” or “pointless.” Circuit City Stores, Inc, v. Adams, 532 U.S. 105, 113, 121 S.Ct. 1302, 149 L.Ed.2d 234 (2001). Sanders claims that reading the two mandatory mínimums as applying to the exact same crime renders the lower minimum meaningless. In other words, if the offenses covered by the two provisions completely overlap, courts would only apply the higher minimum, thereby eliminating the lower penalty through a disfavored repeal by implication. See Granholm v.

Heald, 544 U.S. 460, 483, 125 S.Ct. 1885, 161 L.Ed.2d 796 (2005). For that reason, Sanders seeks to find a version of kidnapping that does not qualify for the higher penalty. Given the language of “crime of violence” in § 3559(f), Sanders argues that Congress intended the enhanced mandatory minimum only to apply when the kidnapping involved “actual violence.” We disagree. The phrase “crime of violence” is a term of art defined in 18 U.S.C. § 16.

Under the statutory definition, such crimes do not require “actual violence.” Rather, a “crime of violence” only must have “as an element the use, attempted use, or threatened use of physical force ... or ... by its nature, involve[ ] a substantial risk that physical force against the person ... may be used in the course of committing the offense.” Id. Although Congress did not specifically cite this definition in § 3559(f) (as it did for § 1201), another “longstanding” canon of statutory

interpretation is “construing statutes in pari material’ Crawford Fitting Co. v. J.T. Gibbons, Inc., 482 U.S. 437, 445, 107 S.Ct. 2494, 96 L.Ed.2d 385 (1987); see also Wachovia Bank, N.A. v. Schmidt, 546 U.S. 303, 315-16, 126 S.Ct. 941, 163 L.Ed.2d 797 (2006) (“under the in pari materia canon of statutory construction, statutes addressing the same subject matter generally should be read as if they were one law”) (internal quotation marks omitted). We have no reason to think that Congress intended the term “crime of violence” to mean anything other than how it is defined at the beginning of the very same

chapter of the United States Code. Indeed, we believe Congress intended the higher penalty to apply. The twenty-five year minimum was passed later in time. The

chapter of the act that included the new minimum was titled “Federal Criminal Law Enhancements Needed to Protect Children from Sexual Attacks and Other Violent Crimes.” Pub.L. No. 109-248 ch. 2, 120 Stat. 587, 588 (2006) (emphasis added). Congress knew it was increasing the penalty from what was previously established. In addition, the statute says courts should impose these higher sentences “unless a greater mandatory minimum sentence” applies and “regardless of any maximum term of imprisonment otherwise provided.” 18 U.S.C. § 3559(f) (emphasis added).

In other words, Congress wanted courts to use the higher provisions unless something even greater applied and notwithstanding conflicting maximum terms. Thus, as the name of the act demonstrates, Congress was focused on criminals receiving higher sentences, not lower ones. Sanders argues that this reading renders § 1201(

g) a nullity, but “the rule against redundancy does not necessarily have the strength to turn a tide of good cause to come out the other way.” Gutierrez v. Ada, 528 U.S. 250, 258, 120 S.Ct. 740, 145 L.Ed.2d 747 (2000). That is the case here, to the extent that we deprive § 1201(

g) of force. In any event, we do not believe we have rendered § 1201(

g) meaningless. Applying the higher minimum complies with both statutes, since § 1201(

g) requires merely that the offender be sentenced to “not less than 20 years.” 18 U.S.C. § 1201(g) (emphasis added). A twenty-five-year sentence is “not less” than twenty years. Thus, we do not need to deviate from the standard definition of “crime of violence,” as Sanders urges us to do, to give meaning to the statute. The fact that two statutes mandate “different penalties for essentially the same conduct is no justification for taking liberties with unequivocal statutory language.” United States v. Batchelder, 442 U.S. 114, 121-22, 99 S.Ct. 2198, 60 L.Ed.2d 755 (1979).

As such, the district court correctly interpreted the twenty-five-year sentence of § 3559(f)(2) as the applicable minimum. III. Conclusion For the foregoing reasons, we find all of Sanders’s arguments unavailing. We therefore Affirm both his conviction and sentence. . Even if we considered Sanders’s new sources, some of them do not even support his position. One book chapter, for example, compares the reliability of physical show ups with photographic ones (the book uses the term "show-up” to include single-photograph identification techniques). Jennifer E. Dysart & R.C.L.

Lindsay, Show-up Identifications: Suggestive Technique or Reliable Method?, in 2 The Handbook of Eyewitness Psychology: Memory for People 137, 142-43 (R.C.L. Lindsay et al. eds., 2007). In so doing, the authors explain that, "[w]hen the identification procedure is conducted with the use of photographs, there is no significant difference in correct identification rates between show-ups and line-ups.” Id. at 143.