OPINION

GILMAN, Circuit Judge.

Peter Kevin Langan was convicted by a jury of robbing two banks, in violation of 18 U.S.C. § 2113, and of using firearms and a destructive device in committing the robberies, in violation of 18 U.S.C. § 924(c). In a separate trial concerning the circumstances of his arrest on unrelated charges, Langan was convicted of assaulting federal officers, in violation of 18 U.S.C. § 111, and of further firearms offenses. Langan's convictions from both trials were consolidated for sentencing purposes and resulted in life imprisonment without the possibility of parole, plus 35 years.

On appeal, Langan claims that (1) the district court erred when it excluded a psychologist's expert testimony on the reliability of eyewitness identification, (2) the evidence was insufficient to establish that the device left at one of the banks was a "destructive device," and (3) none of his convictions under 18 U.S.C. § 924(c) should be considered a "second or subsequent" conviction that warrants a mandatory life sentence without parole. For the reasons set forth below, we AFFIRM the judgment of the district court.

I. BACKGROUND

Langan, who used the alias "Commander Pedro," led a small, white-supremacist group known as the Aryan Republican Army ("ARA") and, at times, as the Midwestern Bank Bandits. In 1991, he and his childhood friend, Richard "Wild Bill" Guthrie, founded the ARA after attending a Christian Identity meeting. The ARA was a copycat, neo-Nazi group inspired by the book The Silent Brotherhood, which detailed the exploits of an underground guerrilla bank robbery gang known as "The Order." In the mid-1990s, the ARA committed a number of bank robberies throughout the midwest to support their avowed purpose of committing terrorist acts against the United States government. A portion of the proceeds from the bank robberies was then funneled to similar neo-Nazi causes, while the remainder was used to finance future robberies. Before his trial, Langan informed a Pretrial Services Officer that he had been a "self-employed revolutionary" with the goal of "overthrowing the government."

In the fall of 1993, Langan went underground, breaking off all contact with his family and friends and staying in various motels. Langan, Guthrie, and an associate Shawn Kenny attempted to rob Society National Bank ("SNB") in Springdale, Ohio during this period. Armed and wearing bulletproof vests, they drove to the bank. They brought with them disguises, a mock explosive device, and a police scanner. For reasons not made clear in the record, they abandoned their plans to rob the bank on the day in question.

On June 8, 1994, however, SNB was robbed shortly after it opened. The robbery was carried out by two armed gunmen wearing Richard Nixon and Ronald Reagan masks. One of the robbers ordered everyone to the floor and removed money from the teller drawers while the other gunman stood watch in the lobby of the bank. The robbers stole $11,890 in federally insured funds that were attached to a dye pack. A dye pack is a safety device that is disguised as an ordinary, unused pack of currency. Soon after it passes through the magnetic field of the bank's doors, the dye pack is designed to explode, releasing red smoke, tear gas, and red dye.

While the robbers were driving away from SNB in a brown Chevy Citation, the dye pack exploded. The robbers threw some of the money out of the window as they fled the scene. They then abandoned the getaway car less than a block from the bank. The car was later discovered to be registered to "Ed McMahon," one of Guthrie's aliases. Inside the car, the police recovered a Mark-21 practice hand grenade, a $20 bill with red-dye stain, and a police scanner.

Later that month, Guthrie met with Kenny in Cincinnati and gave him approximately $200 to compensate him for his "earlier ventures," including the aborted attempt to rob SNB in the fall of 1993. The money was stained with red dye. Guthrie informed Kenny that he and Langan had been successful in robbing the bank on their second attempt, and that they had used a "takeover" command style as employed by "The Order" bandits in The Silent Brotherhood.

Four months later, on the morning of October 25, 1994, two masked men robbed the Columbus National Bank ("CNB") soon after it opened. Guthrie had purchased a.22-caliber rifle in the Columbus area on the previous day. During the CNB robbery, the gunmen wore construction overalls, ski masks, hard hats, sunglasses, and gloves. The men ordered everyone to the floor and shouted to one another using Spanish phrases such as "andale, andale" and "la bomba." At Langan's trial, Kenny testified that this diversionary technique was used by the white supremacists in The Silent Brotherhood in an attempt to divert the attention of authorities by passing themselves off as Hispanic. One of the robbers remained in the lobby to control the customers while the other drew his gun and jumped over the counter to collect the money. As he was taking the money from the teller drawers, he removed his ski mask, hard hat, and sunglasses and left them on a teller counter. The robbers stole $3,400 in federally insured funds from the bank.

CNB's assistant manager, Lisa Copley, later identified Langan as the robber. Copley testified at Langan's trial that she had seen his face clearly for about three seconds as he progressed down the teller line. When he took off the disguise, Langan was approximately four feet away from Copley, but he came within "touching distance" as he emptied the drawers. She saw him again when he returned to her area after attempting to steal money from the drive-up window. Following the incident, Copley described the robber as a white male, in his mid-thirties, weighing 165 pounds, 5 feet 8 inches tall, clean-shaven with dark hair and a medium build.

After the gunmen exited CNB, a black lunchbox containing a plastic pipe covered with wires was discovered behind a counter. The bomb was left at the scene to sow confusion and further divert police attention. A bomb squad was called to the scene, and the bank was evacuated. Upon arrival at the bank, bomb squad personnel photographed the device and then used a Nutrex Disruptor water cannon to render the pipe bomb safe. The Disruptor separated the components of the bomb without detonation.

These components were sent to the FBI laboratory for analysis. Steven Burmeister, an FBI toxicologist, conducted a chemical analysis on a powder sample from the device and concluded that it was composed of "nonperforated disk double-based smokeless powder" and a nonexplosive white powder consisting of calcium sulfate, calcium carbonate, silicon dioxide, and aluminum compounds. At Langan's trial, Burmeister was qualified as an expert in forensic chemistry, specializing in the analysis of explosives and explosive residues in general, and pipe bombs in particular. He testified that smokeless powder is an explosive that is sensitive to heat, shock, and friction, and although it would burn more slowly because of the presence of the non-explosive white powder, its explosive propensities were otherwise unaffected.

The other FBI expert to testify was Robert Heckman, a hazardous devices and explosives examiner, who addressed the engineering aspects of the device. He examined a photograph of the device before it was rendered safe, as well as the recovered debris. Relying upon Burmeister's characterization of the powder mixture as an explosive, Heckman determined that the device was capable of explosion. Furthermore, he determined that although it did not contain an "apparent classic initiator," it could explode by mishandling alone, such as by "dropping the pipe" on the ground. He also found that the device was equipped with an electric circuit hooked to a pager and powered by two nine-volt batteries, as well as a mercury switch. Based on these findings, he concluded that the device was "objectively a pipe bomb."

After being qualified as an expert in explosive and hazardous devices, Heckman testified that the pipe bomb contained the necessary components for a destructive device. He went on to state that, "properly assembled and initiated," the device "would explode, would cause property damage, personal injury, and possibly death." Heckman indicated, however, that he could not determine whether the mercury switch was operable before destruction because not enough fragments of the switch had been recovered to properly reconstruct the switch. He also testified that the wires appeared not to have been crimped properly for conduction, making it doubtful whether this method of detonation would have been successful or had even been intended.

In January of 1996, Langan was seized by a combined state-federal law enforcement team outside of a "safe house" in Columbus. The safe house was used by the ARA to store ammunition, firearms, and other bank robbery equipment. Guthrie had been captured three days earlier and had betrayed Langan, providing the authorities with critical details about their bank robbery exploits and with the location of their safe house. Based on Guthrie's information, law enforcement officials executed a raid on the property. The officers testified that they saw Langan brandishing a semi-automatic handgun from the back window of his van. They then riddled the van with more than 50 bullets, but Langan survived the gunfire with minor injuries and was placed under arrest. This dramatic confrontation was the subject of local television news stories. Copley was among those who saw the television coverage of Langan's capture. She testified at trial that when she saw the report, she instantly thought that Langan was the same man who had robbed CNB.

Law enforcement agents found several items in the safe house and the van that were connected to the SNB robbery. A Mark-21 grenade was discovered in a suitcase that was identified as belonging to Langan. The grenade had Langan's fingerprint on the hose clamp and was similar to the Mark-21 grenade found in the glove compartment of the getaway car used in the SNB robbery. Agents also recovered a pair of blue and white leather gloves with red-dye stain on them. An FBI chemist determined that the chemical make-up of the red dye was the same as the dye found on the $20 bill recovered from the SNB getaway car. A Ronald Reagan mask, similar to the masks worn during the SNB robbery, was found in Langan's van, as was a black lunch-box similar to the one containing the CNB pipe bomb. Two copies of The Silent Brotherhood were seized from the house. Officers also recovered hard hats, ski masks, and sunglasses that were identical to those left behind after the CNB robbery.

On February 15, 1996, a federal grand jury returned a seven-count indictment, charging Langan with the January assault on federal officers, in violation of 18 U.S.C. § 111, carrying and using firearms in connection to the assault, in violation of 18 U.S.C. §§ 2, 922(g), and 924(c), and failing to register a destructive device in violation of 26 U.S.C. § 5861(d). A superseding indictment was returned by the grand jury on March 16, 1996, incorporating the charges of the earlier indictment and alleging that Langan had committed two counts of armed bank robbery, in violation of 18 U.S.C. §§ 2 and 2113(a) (d), two additional counts of using and carrying a firearm during a crime of violence, in violation of 18 U.S.C. §§ 2 and 924(c), and one count of using and carrying an explosive device during and in relation to a crime of violence, in violation of 18 U.S.C. § 924(c). These later charges all related to the SNB and CNB robberies.

On March 18, 1996, Copley was interviewed by an FBI agent and shown a photo array. She informed the agent that she had seen Langan on television in connection with the January 1996 shootout. Copley went on to express her concern that "I hope I don't recognize this individual from T.V." When shown the photos, however, Copley did not hesitate in selecting Langan's. It was the first and only one she selected among the array.

Langan filed a motion to sever the bank robbery and related firearm charges of the superseding indictment from the original assault and firearms counts. The district court granted his motion, resulting in Langan being tried in two separate proceedings. Langan also moved to exclude Copley's testimony. The court denied the motion, as well as Langan's motion to present the expert testimony of David F. Ross, a psychologist at the University of Tennessee. Langan sought to offer Dr. Ross's testimony in order to undermine Copley's eyewitness identification.

The district court recognized Dr. Ross as an expert in psychology solely for the purpose of a suppression hearing regarding Copley's identification of Langan. It reserved ruling on whether Dr. Ross would be qualified to testify at trial in the subspecialty of eyewitness identification. At the suppression hearing, Dr. Ross testified regarding a variety of factors that had the potential of affecting the accuracy of Copley's identification. These factors included the 14 month delay between the CNB robbery and the photo array, the distractions created by the robber's gun and the stress of the situation, and her exposure to his capture on television news. Dr. Ross also criticized the identification procedures used by the FBI in the photo array. The district court determined, under the tests set forth in Neil v. Biggers, 409 U.S. 188, 199-200, 93 S.Ct. 375, 34 L.Ed.2d 401 (1972), that (1) Copley's testimony was sufficiently reliable to be presented to the jury, and (2) Langan had failed to meet his burden of proving that the photo array was impermissively suggestive.

Dr. Ross was also prepared to offer a more specific discussion concerning the possibility of "memory transference," or what Dr. Ross referred to as "conscious inference" or "conscious transference." This phenomenon is the misidentification of a familiar but innocent person, more commonly called " unconscious transference" (although Dr. Ross takes issue with this label, based on his theory of how the process occurs in the brain). Dr. Ross's testimony would have proposed an alternative explanation for Copley's identification, suggesting that Copley was not recognizing Langan from the CNB robbery, but instead from the more recent television news coverage, even though she may have believed that she was identifying Langan from both. He had co-authored an article on the subject, Unconscious Transference and Mistaken Identity: When a Witness Misidentifies a Familiar but Innocent Person, 79 Journal of Applied Psychology 918 (1994).

Following a Daubert hearing, the district court refused to allow Dr. Ross to testify at trial as an expert in eyewitness identification. The court determined that Dr. Ross's proposed testimony failed to meet the requirements of Rule 702 of the Federal Rules of Evidence as interpreted by Daubert and its progeny. Using the two-part test developed in Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579, 113 S.Ct. 2786, 125 L.Ed.2d 469 (1993), the district court concluded that Dr. Ross's proposed testimony on the factors influencing eyewitness accuracy would not have assisted the jury. Instead, the court determined that any testimony offered by Dr. Ross would have improperly invaded their province.

The district court also held that Dr. Ross's testimony concerning the transference theory was not sufficiently based on "scientific knowledge," because it failed to meet the reliability standards established by Daubert. Citing Dr. Ross's own 1994 article, the court noted that Dr. Ross had personally called the theory into question when commenting that the "literature provides mixed and somewhat weak support for unconscious transference" and that "the empirical evidence for the [theory's] existence is rather meager." The district court pointed out that even though Dr. Ross was given the opportunity at the suppression hearing to state that the theory was scientifically sound, he failed to effectively rebut the government's criticism of the theory. Moreover, the court found that Dr. Ross's methodologies were inadequate because he had never studied any victim or eyewitness of a bank robbery.

Finally, the district court questioned Dr. Ross's qualifications in the specific field of "adult eyewitness identification." It noted that the great majority of Dr. Ross's work had focused on child eyewitness identification. The court also found it significant that Dr. Ross had been qualified to testify as an expert only once before, and in that instance his testimony related to a child eyewitness in a state court proceeding.

During the 22-day jury trial, the government presented more than 60 witnesses. The most important of these witnesses were two of Langan's coconspirators, Shawn Kenny and Kevin McCarthy. Kenny had been affiliated with the ARA from the time it began operations and had been part of the "dry run" at the SNB in 1993. Guthrie, Langan's chief cohort, had told Kenny that the second attempt at SNB had proved successful. As compensation for Kenny's efforts, Guthrie paid him with red-tinted money. McCarthy, an ARA member who was recruited shortly after the CNB robbery, testified that Langan told him that Langan had injured his knee while jumping over a teller counter and lost his mask during an earlier robbery. Guthrie was to have been a witness at the trial, but he hung himself in his prison cell in July of 1996. The government also presented substantial physical evidence recovered from the safe house that tied Langan to the robberies.

In February of 1997, the jury found Langan guilty on all counts related to the two bank robberies. Langan was found guilty by a second jury in October of 1997 on charges related to the assault on federal officers. For sentencing purposes, Langan's convictions were consolidated and resulted in a life sentence without the possibility of parole, plus 35 years. Langan was also ordered to pay restitution of $6,270 and a $400 fine. He filed this timely appeal in January of 1999.

II. ANALYSIS

A. The district court did not abuse its discretion when it excluded Dr. Ross's testimony regarding the reliability of eyewitness identification

Langan's first claim of error is that Dr. Ross's expert testimony was improperly excluded. Following Copley's courtroom identification of Langan, the district court refused to admit Dr. Ross's expert testimony regarding the reliability of Copley's eyewitness identification. After a lengthy Daubert hearing, the court concluded that Dr. Ross's proposed testimony (1) was not sufficiently based on "scientific knowledge" and (2) would not have assisted the jury. The district court instead opted to give the jury an eyewitness identification instruction that alerted the jury to the various factors to be considered when weighing Copley's testimony, and cautioned the jurors to carefully consider the shortcomings and trouble spots of the identification process.

We will not disturb a district court's evidentiary rulings unless we find an abuse of discretion. See United States v. Middleton, 246 F.3d 825, 836 (6th Cir. 2001). Our standard of review does "not categorically distinguish between rulings allowing expert testimony and rulings disallowing it," and are mindful of the "gatekeeper role of the trial judge in screening such evidence." Gen. Elec. Co. v. Joiner, 522 U.S. 136, 142, 118 S.Ct. 512, 139 L.Ed.2d 508 (1997) (internal quotation marks omitted).

Rule 702 of the Federal Rules of Evidence states that "[i]f scientific, technical, or other specialized knowledge will assist the trier of fact to understand the evidence or to determine a fact in issue, a witness qualified as an expert by knowledge, skill, experience, training, or education, may testify thereto in the form of an opinion or otherwise." In Daubert, the Supreme Court held that, when faced with a proffer of scientific expert testimony, Rule 702 requires a district court to determine whether the evidence "both rests on a reliable foundation and is relevant to the task at hand." Daubert, 509 U.S. at 597, 113 S.Ct. 2786. As part of its review, the court must assess "whether the reasoning or methodology underlying the testimony is scientifically valid and . . . whether that reasoning or methodology properly can be applied to the facts in issue." Id. at 592-93, 113 S.Ct. 2786. Such factors as testing, peer review, publication, error rates, the existence and maintenance of standards controlling the technique's operation, and general acceptance in the relevant scientific community should be considered in this review. See id. at 593-94, 113 S.Ct. 2786. This inquiry is a flexible one, with an overarching goal of assessing the "scientific validity and thus the evidentiary relevance and reliability" of the principles and methodology underlying the proposed expert testimony. See id. at 594-95, 113 S.Ct. 2786.

The use of expert testimony in regard to eyewitness identification is a recurring and controversial subject. Trial courts have traditionally hesitated to admit expert testimony purporting to identify flaws in eyewitness identification. Among the reasons given to exclude such testimony are that the jury can decide the credibility issues itself, see United States v. Lumpkin, 192 F.3d 280, 289 (2d Cir. 1999) (determining that the "proposed testimony intrudes too much on the traditional province of the jury to assess witness credibility"); that experts in this area are not much help and largely offer rather obvious generalities, see United States v. Hudson, 884 F.2d 1016, 1024 (7th Cir. 1989) ("Such expert testimony will not aid the jury because it addresses an issue of which the jury already generally is aware, and it will not contribute to their understanding of the particular dispute."); that trials would be prolonged by a battle of experts, see United States v. Fosher, 590 F.2d 381, 383-84 (1st Cir. 1979) (adding "to the trial court's articulated concerns our own conviction that a trial court has the discretion to avoid imposing upon the parties the time and expense involved in a battle of experts"); and that such testimony creates undue opportunity for confusing and misleading the jury, see United States v. Rincon, 28 F.3d 921, 926 (9th Cir. 1994) ("Given the powerful nature of expert testimony, coupled with its potential to mislead the jury, we cannot say the district court erred in concluding that the proffered evidence would not assist the trier of fact and that it was likely to mislead the jury.").

Several courts, however, including our own, have suggested that such evidence warrants a more hospitable reception. See United States v. Smithers, 212 F.3d 306, 314 (6th Cir. 2000) (holding that "the district court abused its discretion in excluding [the eyewitness identification expert's] testimony, without first conducting a hearing pursuant to Daubert"); United States v. Smith, 156 F.3d 1046, 1053 (10th Cir. 1998) (agreeing that "expert testimony on eyewitness identification may properly be admitted under Daubert in certain circumstances"); United States v. Brien, 59 F.3d 274, 277 (1st Cir. 1995) (declining to adopt a blanket rule that qualified expert testimony on eyewitness identification must either be routinely admitted or excluded); United States v. Amador-Galvan, 9 F.3d 1414, 1417-18 (9th Cir. 1993) (declining to follow a per se rule excluding expert testimony regarding the credibility of eyewitness identification). Moreover, such testimony has been allowed in with increasing frequency where the circumstances include "cross-racial identification, identification after a long delay, identification after observation under stress, and [such] psychological phenomena as . . . unconscious transference." See United States v. Harris, 995 F.2d 532, 535 (4th Cir. 1993). Nonetheless, each court to examine this issue has held that the district court has broad discretion in, first, determining the reliability of the particular testimony, and, second, balancing its probative value against its prejudicial effect. See e.g., United States v. Hall, 165 F.3d 1095, 1101 (7th Cir. 1999); United States v. Kime, 99 F.3d 870, 883 (8th Cir. 1996).

The district court in the case before us did not automatically exclude Dr. Ross's testimony. Instead, it considered the matter in detail after conducting a lengthy Daubert hearing and specifically acknowledged the Sixth Circuit's willingness to accept psychological studies as a scientifically sound and proper subject of expert testimony. In fact, the court pointedly distinguished the circumstances presented in Langan's case from those in United States v. Smith, 736 F.2d 1103 (6th Cir. 1984), where this court noted that the science of eyewitness perception has achieved the level of "exactness, methodology, and reliability of any psychological research." Id. at 1106. The Smith court held that the district court had abused its discretion in excluding an expert in eyewitness identification because "[s]uch testimony might have been relevant to the exact facts before the court and not only might have assisted the jury, but might have refuted their otherwise common assumptions about the reliability of eyewitness identification," thus satisfying the "helpfulness test" of Rule 702. Id. at 1106 (emphasis in original).

In Smith, the government had conceded that the proposed expert was, in fact, an expert, and none of the eyewitnesses could identify Smith from a photo spread three weeks after the robbery in question, although they all identified Smith at an FBI line-up four months later. See id. at 1104-05. Despite the Smith court's holding, the district court in the case before us determined that Dr. Ross's testimony was "within the jury's own knowledge" as it related to the general psychological factors affecting memory and would therefore not be of assistance.

The district court also held that Langan had failed to establish the reliability of Dr. Ross's testimony regarding the "transference theory," even though Copley's identification presented a specific cognitive problem not common to all or even most eyewitness identifications and Copley herself had put the identification at issue. As a basis for this finding, the court pointed to Dr. Ross's only published article on the subject, concluding that "the plain language of the article is that the theory of unconscious transference is as yet empirically unproven and further research is necessary before the theory can be confirmed or disconfirmed."

Dr. Ross, who was questioned about this article at the suppression hearing, defended his position by stating that the transference phenomenon existed but should be renamed "conscious transference" or "conscious inference," because the eyewitness has a conscious recollection of the previous exposure to the innocent person in a lineup. His article, however, criticizes the overall theory by characterizing the empirical evidence as "rather meager." The article goes on to note that his experiments "have a number of limitations," including "limited external validity" and "limited generalizability." Moreover, "the theoretical notions described here need modification and further development, and they are proposed in terms of encouraging research in this area and not in terms of providing the definitive word on this fascinating and important topic."

The district court thus had justification for concluding that, by any name, the theory failed to meet the requisite reliability standards. At the Daubert hearing, Langan had the opportunity to establish whether Dr. Ross had conducted any further research to confirm his interpretation of the transference theory. Three years had passed since his article was published, yet Langan failed to demonstrate that the "limitations" on Dr. Ross's research no longer existed. The district court was therefore left with Dr. Ross's criticism of the general theory, including his own interpretation of conscious transference, without any showing that the theory had been subjected to further testing or been confirmed scientifically. Although Dr. Ross did not, as the government argues, "debunk" the phenomenon of misidentification, the district court did not abuse its discretion when it determined that Dr. Ross's proffered testimony lacked the requisite indicia of reliability. Such indicia would include empirical support for the theory, whether it had been subjected to peer review, and whether the theory enjoyed general acceptance within the field of eyewitness identification. See Daubert, 509 U.S. at 593-95, 113 S.Ct. 2786.

The district court also found it significant that Dr. Ross had never before been qualified as an expert in federal court, and only once before in state court as an expert in child eyewitness identification. This is in contrast with the expert in Smithers, whose "qualifications and scientific methods" had already been "praised" by the Sixth Circuit. See Smithers, 212 F.3d at 315. The district court further faulted Dr. Ross's methodology because he had never included a victim or eyewitness to a bank robbery in his studies. Although this fact appears to be more relevant to whether the opinion "fits" the case and is helpful, we cannot say that the district court erred in concluding that it undermined the scientific underpinnings of Dr. Ross's proffered opinion in this case.

Another of the district court's reasons for finding that Dr. Ross's testimony did not constitute "scientific knowledge" is more problematic. The district court questioned his ability to qualify as an expert in the subspeciality of "adult eyewitness identifications," because the bulk of Dr. Ross's research focused on child eyewitness testimony. Dr. Ross, however, had published articles, edited books, and given lectures dealing specifically with misidentification in adult witnesses. The fact that he has spent more time and effort studying children should not have disqualified him from testifying about the misidentification phenomenon in adults. To demand that there be a specific qualification in "adult eyewitness identification" is unjustified unless there exists some principled distinction between the methods, theory, and results used in the study of misidentification by adults as opposed to misidentification by children.

Despite our hesitation to adopt all of the district court's criticisms concerning Dr. Ross's proffered testimony under the "scientific knowledge" prong of Daubert, we agree that it failed to satisfy the second Daubert inquiry. This second step requires the district court to consider whether the proposed scientific evidence fits the issue to which the expert is testifying. See Daubert, 509 U.S. at 591, 113 S.Ct. 2786. In other words, a district court may admit the evidence only if such testimony will assist the trier of fact in understanding the evidence or in determining a fact at issue. The proposed testimony must thus "fit" the eyewitness identification in this case.

Dr. Ross's proposed testimony was being offered in order to impeach Copley's identification of Langan. His testimony, however, would not have been based on any personal knowledge of Copley, but rather on his general knowledge about the type of circumstances in which she had found herself. Dr. Ross would have testified to the limitations regarding Copley's ability to see and identify the bank robber during the robbery, her ability to identify the robber eighteen months after the robbery, the transference phenomenon related to her having seen Langan on television prior to the photo array, and the "psychologically-based weaknesses in the method used by the FBI at the photo array."

Although we recognize that expert testimony on eyewitness identification might inform the jury on all of the intricacies of perception, retention, and recall, we nevertheless agree with the district court that the hazards of eyewitness identification are within the ordinary knowledge of most lay jurors. With regard to the possibility that Copley was remembering Langan from television rather than from the CNB robbery, Langan's counsel had ample opportunity to cross-examine her on this point, and in fact did so at trial. The district court also was highly skeptical of this specific area of testimony because of the limitations that Dr. Ross had personally raised concerning the subject. If he had testified at trial, it is likely that an uninformative battle of experts would have occurred if the government had offered its own expert testimony in order to refute Dr. Ross's opinion, and the jury could have been unduly misled and confused.

Furthermore, as this court noted in Smithers, the numerous cases holding that the exclusion of expert testimony on eyewitness identification is not an abuse of discretion caution us to consider whether the testimony would have been helpful or confusing to the jury. See Smithers, 212 F.3d at 314. These "cases also discuss whether this type of testimony touched on the `ultimate issue' in the case and therefore usurped the jury's role; whether there was other evidence against the defendant; and whether the jury could more properly evaluate the reliability of eyewitness testimony through cross-examination." Id.

Our conclusion that the district court did not abuse its discretion in this case is further supported by the fact that Langan had the opportunity to thoroughly cross-examine Copley in order to cast doubt on her ability to identify him. As we have explained, weaknesses in eyewitness identification testimony ordinarily can be exposed through careful cross-examination of the eyewitness. Copley herself was the one who expressed initial concern that she was remembering Langan from his appearance on television. She brought up this possibility during her direct examination by the government, and she was thoroughly cross-examined by Langan's counsel on this issue. The jury was therefore able to consider this possible "transference" during its deliberations. In addition, the jury was cautioned about the "shortcomings and trouble spots" of eyewitness identification in the final jury instructions.

Moreover, Copley showed other indicia of having made a reliable identification. Her initial description of the suspect was consistent with Langan's own physical characteristics. She immediately recognized Langan as the robber when she saw him on television. Besides her responsibilities as an assistant manager at CNB, Copley also served as an aviation electronics technician with the Air Force National Guard. Her national guard training required her to participate on an annual basis in combat and hostile threat situations. During these exercises, she received training in identifying individuals attempting to enter unauthorized areas. She testified that a critical component to this training was the need to remain calm and focused during the attempted takeovers. In light of this specialized training, Dr. Ross's generalized testimony regarding such distracting factors as stress and the presence of a gun would not necessarily have helped the jury in evaluating Copley's identification of Langan.

Finally, a substantial amount of other evidence linked Langan to the CNB robbery. Two coconspirators, for example, testified regarding his involvement. When Langan was arrested, law enforcement officials recovered masks and hardhats similar to those found in the CNB getaway car. A silver semi-automatic pistol, like the one Copley identified as used by Langan at the robbery, was also recovered from the safe house. In addition, a white-plastic pipe bomb that closely resembled the one used in the CNB robbery was discovered in a suitcase belonging to Langan, and Langan's thumbprint was found on bomb-making equipment in the same suitcase. Thus abundant evidence to support the jury's verdict existed even without Copley's eyewitness identification.

Based on all of the considerations discussed above, we conclude that the district court did not abuse its discretion in precluding Dr. Ross from testifying at trial.

B. Sufficient evidence existed to show that the CNB bomb was a "destructive device"

Langan next maintains that there was insufficient evidence to support a verdict that the home-made pipe bomb left at CNB Bank constituted a "destructive device" as defined by 18 U.S.C. § 921. In order to convict Langan of using a destructive device, the government had to prove that the pipe bomb left at the robbery was "any explosive, incendiary, or poison gas bomb," 18 U.S.C. § 921(a)(4)(A)(i), or "any combination of parts either designed or intended for use in converting any device into any destructive device described in subparagraph (A) . . . and from which a destructive device may be readily assembled." 18 U.S.C. § 921(a)(4)(C). Langan argues that the government failed to prove that the device in question was operable or readily made operable, and therefore did not qualify as a "destructive device" under the relevant statutes.

When considering a challenge on direct appeal to the sufficiency of the evidence to sustain a conviction, we must determine "whether, after viewing the evidence in the light most favorable to the prosecution, any rational trier of fact could have found the essential elements of the crime beyond a reasonable doubt." Jackson v. Virginia, 443 U.S. 307, 319, 99 S.Ct. 2781, 61 L.Ed.2d 560 (1979) (emphasis in original).

To qualify under the statute, we do not require that the destructive device operate as intended. See United States v. Rushcamp, 526 F.2d 1380, 1382 (6th Cir. 1975). It is sufficient for the government to show that the device was "readily convertible" to a destructive device. See id.; United States v. Turczyk, No. 91-3489, 1992 WL 102499, *2-3 (6th Cir. April 29, 1992) (unpublished table decision) (finding that a pipe bomb with dubious initiator circuitry qualified as a destructive device, even though the defendant testified that he used an "inoperable circuit, bad switches, and dead battery" and had intended for it to be inoperable).

Moreover, the government need not establish that any particular component be present for a device to qualify as a destructive device. The only requirement is that the device be capable of exploding or be readily made to explode. See United States v. Wilson, No. 92-5075, 1992 WL 227472, * 2 (6th Cir. Sept.15, 1992) (unpublished table decision) ("[W]e do not read these cases as requiring the government to prove beyond a reasonable doubt that each component is present; instead, the government's burden is to prove beyond a reasonable doubt that the devices are destructive or could be readily made so.").

Langan maintains that the government failed to show that the device contained an explosive main charge. The government counters that the testimony provided by forensic chemist Burmeister and hazardous devices and explosives examiner Heckman established that the CNB pipe bomb had the requisite elements of a "destructive device," including an explosive main charge, a container, and a method of initiation.

Burmeister testified that, based on his forensic analysis, the powder contained within the CNB device was "nonperforated disk double-based smokeless powder . . . [and] a composition of calcium sulfate, calcium carbonate, silicon dioxide[,] as well as an aluminum containing material." When heat, shock, or friction is applied to nonperforated disk double-based smokeless powder contained within a plastic pipe, Burmeister opined that it is capable of exploding. Langan faults Burmeister for failing to subject the powder to further testing based on its containing a white inert mixture in addition to smokeless powder. At trial, however, Burmeister testified that although the additional materials were not explosives, they would not affect the explosive propensities of the smokeless powder.

The government then presented Heckman, who addressed the engineering aspects of the device. Heckman stated that he had relied upon Burmeister's chemical analysis of the smokeless powder, and then determined that the explosive had been confined in the plastic pipe enclosed in the lunchbox. As such, he testified that the device was capable of explosion by being thrown or dropped, even if it did not contain an "apparent classic initiator." He also found that the device was equipped with an electric circuit hooked to a pager and powered by two nine-volt batteries, as well as a mercury switch. Based on these findings, Heckman concluded that the device was, in fact, a pipe bomb, which is defined as the purposeful containment of an explosive mixture in order to achieve detonation. See 18 U.S.C. § 921(a)(4); 26 U.S.C. 5845(f).

Langan next contends that the device was not capable of being readily made operable because its wires were insulated and had not been soldered to the pager. Heckman, however, testified that it would only take a few seconds to recrimp the wires and strip them properly. The CNB device therefore bears significant similarities to the device in Turczyk. As in the case before us, the putative pipe bomb in Turczyk had been fragmented by the bomb squad, and government experts were unable to ascertain whether the circuitry would have achieved detonation. See Turczyk, 1992 WL 102499 at *3. Explaining that the statute did not demand that the device operate as initially intended, but that it was enough that it be readily convertible into a destructive device, the Turczyk court pointed out that there had been testimony that "even if the circuitry were removed, the device could function by merely connecting the two wire leads to any electrical source on a car." Id. This conversion would have required considerably more effort than the conversion Heckman proposed could have been performed on the CNB device. Given the above evidence, it was reasonable for the jury to have found beyond a reasonable doubt that the CNB pipe bomb was indeed a destructive device, which contained an explosive mixture and was capable of explosion.

C. Langan's life sentence under 18 U.S.C. § 924(c)

As his final claim of error, Langan maintains that the district court improperly enhanced his sentence under 18 U.S.C. § 924(c)(1). Specifically, Langan objects to the treatment of Counts 2, 4, and 5 as "second or subsequent convictions," arguing in "good faith" that his life sentence should be reversed. The district court's application of 18 U.S.C. § 924(c) is a question of law which we review de novo. See United States v. Deal, 954 F.2d 262, 262-63 (5th Cir. 1992), aff'd, 508 U.S. 129, 113 S.Ct. 1993, 124 L.Ed.2d 44 (1993).

Langan frankly acknowledges that the basis of his objection is in direct conflict with the Supreme Court's holding in Deal v. United States, 508 U.S. 129, 113 S.Ct. 1993, 124 L.Ed.2d 44 (1993). As the Supreme Court explained, if a defendant is charged with and convicted of separate offenses to which § 924(c) applies, the separate convictions on the associated § 924(c) counts can be used to determine previous and subsequent convictions. See Deal, 508 U.S. at 132-34, 113 S.Ct. 1993 (1993). Because Langan was convicted of the two separate and distinct armed robbery violations, his § 924(c) convictions relating to those underlying violations qualify as a previous and subsequent conviction. Nevertheless, Langan argues that Deal should be overruled. We are of course without authority to disregard controlling Supreme Court precedent. Langan's challenge to his sentence is therefore rejected. Accordingly, the district court did not err in enhancing Langan's sentence for the second § 924(c)(1) conviction.

III. CONCLUSION

For all of the reasons set forth above, we AFFIRM the judgment of the district court.