THE PEOPLE OF THE STATE OF ILLINOIS v. TYRESE BELL (2026)

OPINION

¶ 1 Defendant Tyrese Bell was convicted of first-degree murder with a firearm and sentenced to 80 years in prison. We affirm his conviction over several claims of error—confrontation and discovery violations, improper “lay opinion identification testimony” from a police officer, and excessive other-crimes evidence—and correct the mittimus to reflect a single count of first-degree murder.

¶ 2 BACKGROUND

¶ 3 I

¶ 4 Shortly before 9 p.m. on January 8, 2015, several Chicago police officers responded to a report of a robbery at 1244 South Albany Avenue. Upon arriving, they discovered a murder. A taxi had crashed into a parked car. The driver's door was open, and the taxi driver, Chinedu Madu, was slumped over in his seat and nonresponsive. The medical examiner later confirmed that Madu died from multiple gunshot wounds.

¶ 5 A fresh set of footprints in the snow led from the rear passenger side of Madu's taxi to a nearby residence at 1250 South Albany Avenue. As Officer Cox approached the front porch, a man and a woman walked out of the building together. Officer Cox later identified the man as defendant.

¶ 6 Several officers converged on the second-floor apartment of Gabriella Mitchell, who consented to a search. In the rear bedroom, they found a Bersa .380 semiautomatic handgun on top of a storage bin. (And an intoxicated man named King Boone. He was tested for gunshot residue; none was detected.) The Bersa was loaded, with a live round in the chamber and five in the magazine. And the bottom of the magazine had moisture on it—as if it had recently been outside in the snow. A search of Madu's taxi yielded two fired cartridge cases. An Illinois State Police firearm analyst concluded that the two cartridge cases and the two bullets recovered from Madu's body were fired from the Bersa.

¶ 7 The police found two balled-up black knit gloves that appeared to match: one on the kitchen floor of the apartment and one on the floor in the rear passenger's side of the taxi.

¶ 8 Footage from a police observation device (POD) camera installed at the corner of the block showed the taxi driving down Albany Avenue. After the crash, someone exited from the passenger's side, approached the driver's side, walked to the sidewalk in front of 1250 South Albany Avenue, approached the taxi again, and then returned to the sidewalk. This individual cannot be identified from the POD footage alone.

¶ 9 Based on information he received from Mitchell (and another woman in the apartment), Detective Gorman asked the other officers to bring him a photo of defendant. Mitchell identified defendant from the photo and gave Detective Gorman a phone number, for which he obtained a “tower history.” The police arrested defendant the next day and seized his cell phone.

¶ 10 II

¶ 11 In due course, defendant was charged with the first-degree murder of Madu. Defendant did not testify or present any evidence at his jury trial. The defense argued a theory of reasonable doubt. There were no eyewitnesses, defendant did not confess, and no forensic evidence directly tied him to the murder, since he was excluded as a contributor to the major DNA profiles found on the gun and the taxi's door handles. The State's case rested on four key sources of evidence, as described below: security footage from the taxi, cell-site location information (CSLI) for defendant's phone, data extracted from defendant's phone, and other-crimes evidence.

¶ 12 A

¶ 13 First, Madu's taxi was equipped with security cameras that captured still images and some short video clips of the interior when certain triggers—such as the ignition, meter, or panic button—were activated. Someone, no doubt the shooter, tried to rip one of the security cameras out of the taxi and damaged it. But the State was able to recover the recorded images stored on the taxi's encrypted hard drive. A selection of images was published to the jury.

¶ 14 In sum, they show a Black man with facial hair and a hoodie, wearing a watch and dark gloves, sitting in the rear passenger's seat. At various times he is talking on a cell phone. Later, he brandishes a gun and points it directly at Madu's head and chest, consistent with the locations of the two entrance wounds observed by the medical examiner.

¶ 15 The State argued that the security images were strong evidence on the central question of identity. For one thing, the man in the images physically resembles defendant. For another, he wore the same kind of watch that defendant had on his wrist when he was arrested, and which also turned up in the browser history on his phone. More on these points shortly.

¶ 16 B

¶ 17 Second, FBI Special Agent Joseph Raschke analyzed the historical CSLI for the phone number. On its own, Special Agent Raschke's analysis did not connect this number—or more precisely, the associated phone—to defendant. But the CSLI analysis did connect the associated phone to Madu's taxi at the time of the murder.

¶ 18 The CSLI revealed that the associated phone made multiple calls between 8:15 and 8:56 p.m. on the night of the murder. (The security images from the taxi were timestamped and show the passenger on his phone at 8:19 and 8:34 p.m.) Special Agent Raschke mapped the locations of the cellular towers accessed by the phone during these calls and plotted the coordinates from the taxi's GPS device. Comparing the two, he concluded that the data was “consistent with” the phone and the taxi traveling “in the same times, in the same direction, from the same general location to the same general location”—that is, from the vicinity of 7 West Chicago Avenue to the vicinity of 13th Street and Albany Avenue.

¶ 19 C

¶ 20 Third, the police sent defendant's phone to the Chicago Regional Computer Forensic Laboratory (RCFL) for data extraction. We will circle back to the details of this procedure and the complicated story of how the evidence it generated came to be introduced at trial. For now, the important point is the evidentiary value of the extracted data itself.

¶ 21 Above all, the extracted data showed that the phone number analyzed by Special Agent Raschke was the number assigned to defendant's phone by his carrier. Together, the CSLI and the extracted number thus placed defendant's phone in Madu's taxi at the time of the murder.

¶ 22 The State highlighted a few other pieces of data extracted from the phone. For one, there were website entries associated with a Penguin branded watch in the browser history. As the State pointed out for the jury, the watch had the same model number as the watch defendant was wearing when he was arrested. And based on the specific pattern of blue circles on its face, the State argued that it was the same type of watch that can be seen in the security images from Madu's taxi.

¶ 23 An extracted multimedia (MMS) message included images of a gun—a mostly black handgun with a notable silver pattern that, as the State argued, marked it as a Bersa.

¶ 24 A user of the phone had visited employment and credit-score websites and entered the name “Tyrese Bell.” And several user accounts bookmarked on the phone—from services such as Gmail, Facebook, and others—were associated with the extracted telephone number.

¶ 25 D

¶ 26 Defendant's watch caught the detectives’ attention when he was arrested: it looked like the watch the shooter was wearing in the security images from the taxi. Believing that the watch would likely help identify defendant as the shooter, they decided to photograph it. But defendant mounted a violent resistance that injured two officers and resulted in charges of aggravated battery. The trial court allowed the State to offer testimony on this topic from three officers, along with video footage of the incident and photos of the detectives’ injuries, for the “limited purpose[s]” of “identification, lack of mistake, consciousness of guilt or knowledge.” The trial court gave appropriate limiting instructions to the jury.

¶ 27 Evidence technician Schicting was tasked with photographing defendant's watch and other personal effects. Sergeant Carr and Detectives Lutzow and Murphy were also present. Schicting testified, in sum, that defendant turned violent and had to be restrained by multiple officers, two of whom were injured in the altercation. Ultimately, she was able to photograph defendant's watch and her fellow officers’ injuries.

¶ 28 Detective Lutzow and Sergeant Carr filled out some finer points of detail. Defendant steadfastly refused to “produce” his arm for a photograph. So the officers grabbed his arm and forcibly pulled it into position, prompting defendant to bite Detective Murphy. Collectively, the officers wrestled defendant to the floor. Sergeant Carr's hands were bruised and swollen in the process. Detective Lutzow fared worse: he tore his right rotator cuff and labrum and missed several months of work while recovering from shoulder surgery. The State played video footage of the altercation twice, during Detective Lutzow's and Sergeant Carr's testimony.

¶ 29 ANALYSIS

¶ 30 I

¶ 31 Defendant claims that the State's use of the extracted cell-phone data and accompanying analyst report violated his sixth-amendment confrontation right. See U.S. Const., amend. VI.

¶ 32 The alleged confrontation violation arises from the State's introduction of this evidence through the testimony of John Dziedzic, the director of the RCFL. Dziedzic did not run the data extraction or write the accompanying report. Those tasks were performed by John Clisham Jr., a former forensic analyst who did not testify. Defendant thus contends that he had a right to cross-examine Clisham, specifically, about the testimonial statements in his work product. Dziedzic's “surrogate” testimony, says defendant, will not suffice for confrontation purposes, since Dziedzic did not participate in the data extraction and merely reviewed Clisham's work. We review defendant's (preserved) confrontation claim de novo. People v. Barner, 2015 IL 116949, ¶ 39.

¶ 33 The confrontation clause “ ‘applies only to testimonial hearsay.’ ” Smith v. Arizona, 602 U.S. 779, 784 (2024) (quoting Davis v. Washington, 547 U.S. 813, 823 (2006)). So defendant's confrontation claim can succeed only if an out-of-court testimonial statement made by Clisham was offered (through Dziedzic's testimony) for the truth of the matter it asserted.

¶ 34 Defendant's appellate arguments are largely focused on showing that Clisham's out-of-court statements were testimonial. But that focus ignores the real issue. The dispositive question here is not whether any of Clisham's (purported) statements were testimonial. The question is what statements Clisham made in the first place that were later offered, at trial, in his absence. And the answer is none.

¶ 35 Clisham's work product, as offered by the State, comprised two items: a roughly 1,400 page PDF document that compiled all of the data extracted from the cell phone, accompanied by a brief “Report of Examination,” written by Clisham. We begin with the extracted data itself, which surely posed the real problem for the defense.

¶ 36 Dziedzic explained the process by which RCFL analysts extract data from a cell phone. Briefly: after taking steps to disconnect the phone from the network, the phone is connected to a computer running two software programs made by the company Cellebrite. The first, Cellebrite 4PC, can bypass the phone's passcode lock and extract whatever categories of data are requested. The data, at this point, are in their “raw” or native form—that is, binary code consisting of long strings of zeros and ones that are indecipherable to a human being. So the second program, Physical Analyzer, converts the raw data into its recognizable forms—texts, browser histories, images, system settings, and the like—and outputs the translated data into a requested format, such as the PDF that was generated here.

¶ 37 Dziedzic thus described a data-extraction process that is performed largely by computer software, with minimal human intervention. Other courts have similarly noted that a Cellebrite extraction report is generated through an “automated” process that requires a human being to do little more than “plug the device into the forensic equipment and run the program.” United States v. Gonzalez, 144 F.4th 396, 407 (2d Cir. 2025); Baldwin v. Commonwealth, 723 S.W.3d 676, 686 (Ky. 2025). And as Dziedzic explained, the Cellebrite software “does not permit for any modification of data located on the device.” It simply removes whatever data is there to be found and prints it all out in a report that serves, so to speak, as a mirror of the phone's contents.

¶ 38 The State's principal argument is that the data compiled in the PDF document by the Cellebrite software does not contain any statements at all. For purposes of the confrontation clause, as for the hearsay rule, a statement is an assertion of fact, made by a human declarant, that can be either true or false. See Ill. R. Evid. 801(a)-(c) (eff. Oct. 15, 2015); People v. Price, 2021 IL App (4th) 190043, ¶¶ 140-41; Smith, 602 U.S. at 784 (confrontation clause “ ‘applies only to testimonial hearsay’ ” (quoting Davis, 547 U.S. at 823)).

¶ 39 There are several reasons, all variations on the same theme, why machine-generated data are not statements in the relevant sense. For one thing, uninterpreted data—the assigned number in a phone's settings, or the values that ring up in a lab analysis—are not assertions of fact. For another, a machine is not a declarant, much less a “witness against” anyone, as the confrontation clause requires. Unlike people, machines do not say or assert anything; they are neither sincere nor dishonest; and they cannot be cross-examined in any meaningful sense.

¶ 40 Illinois case law has had little to say on this general point and nothing in the specific context of confrontation. The closest we can find is the holding from our supreme court that a printout of results from computerized telephone tracing equipment was not hearsay, as it was a mere record of the machine's operations. See People v. Holowko, 109 Ill. 2d 187, 193 (1985). But the federal circuits that have addressed the confrontation question all agree that, because machine-generated data are “not ‘statements’ in any useful sense,” United States v. Moon, 512 F.3d 359, 362 (7th Cir. 2008), they are “not hearsay statements as implicated by the Confrontation Clause.” United States v. Washington, 498 F.3d 225, 231 (4th Cir. 2007); see United States v. Arce, 49 F.4th 382, 392 (4th Cir. 2022).

¶ 41 The relevant case law from federal circuits and district courts, and the appellate courts of other states, is vast. But we will not belabor the general point about machine-generated data with overwhelming string citations. That general point is obvious enough, and no case that we have found disputes it. Nor does defendant dispute the point in his reply brief, even though it is front and center in the State's appellate argument.

¶ 42 Applying this general principle, several courts have specifically held that compilations of extracted data, like the PDF generated from defendant's phone here, are “not the statements of anyone,” as they “contain[ ] ‘only machine-generated results.’ ” Gonzalez, 144 F.4th at 407 (quoting Bullcoming v. New Mexico, 564 U.S. 647, 673 (2011) (Sotomayor, J., concurring in part)); Baldwin, 723 S.W.3d at 686; Pena v. State, 714 S.W.3d 214, 220-21 (Tex. Crim. App. 2024); Arce, 49 F.4th at 392.

¶ 43 These cited cases all arose within the context of the confrontation clause. In a state-law evidentiary context, we have held (in a nonprecedential decision) that testimony conveying the results of a data extraction, including the cell phone's assigned number and browser history, was not hearsay. Our reason: the data were “ ‘not statements made by a human declarant.’ ” People v. Perez, 2020 IL App (2d) 180073-U, ¶¶ 122-25 (quoting State v. Reynolds, 456 S.W.3d 101, 104 (Mo. Ct. App. 2015)).

¶ 44 Some other courts have followed the reasoning of United States v. Hill, 63 F.4th 335, 358-59 (5th Cir. 2023), which held that because cell-phone data extraction reports contain only “machine created data,” they do not qualify as “testimonial.” Fair enough, but the most basic reason they are not testimonial is because only statements can be testimonial—and they are not statements in the first place. See Moon, 512 F.3d at 362.

¶ 45 Whichever way you slice it, the result is the same. And every case to consider the issue, as far as we know, has agreed on this result: when an extraction merely “downloads” data from a cell phone, without human manipulation or interpretation, the data compiled in the report and offered at trial do not “implicate” the confrontation clause. Arce, 49 F.4th at 392.

¶ 46 Defendant does not dispute that extracted data “standing alone” are not statements for purposes of the confrontation clause. But here, says defendant, “the State carefully selected, curated, and highlighted” a few choice items (from the 1,400 or so pages of data) to prove “a link between [defendant] and the phone.” But the fact that the State highlighted a few non-statements does not magically convert them into statements. See Hill, 63 F.4th at 359. The State's use of the extracted data did not violate the confrontation clause.

¶ 47 Next, Clisham's “Report of Examination” was attached to the PDF of extracted data and introduced into evidence along with it. Reports written by forensic analysts are often the source of testimonial statements of two general varieties. If Clisham made any such statements, this is where we would find them.

¶ 48 First, such reports often set forth interpretations or conclusions that the analyst drew from the machine-generated data through the exercise of independent judgment. Bullcoming, 564 U.S. at 659-61. A readout from a gas chromatograph, or an image extracted from a cell phone, is just a piece of data. But an analyst's conclusion that the substance tested was cocaine, or that the phone contained child pornography, is a statement—one typically offered for its truth, and quite possibly testimonial, depending on the circumstances in which it was made. See Melendez-Diaz v. Massachusetts, 557 U.S. 305, 308-11 (2009); Arce, 49 F.4th at 392-93; People v. Holmes, 2025 IL App (2d) 240194, ¶ 131.

¶ 49 Second, an analyst's report may contain “testimonial certification[s]” that certain testing protocols were followed, that no conditions arose that might affect the integrity of the biological sample, or other such representations that speak to the validity or veracity of the results reported by the analyst. Bullcoming, 564 U.S. at 651-52, 659-61.

¶ 50 As forensic analyst reports go, there isn't much to Clisham's “Report of Examination.” It has some identifying information about the phone, the chain of custody, the categories of data that were extracted (messages, web history, etc.), and other such topics. It appears from the report that Clisham was not asked to draw any conclusions about the data; he was simply asked to extract “all retrievable information” from the phone. That is surely one reason why the report is relatively sparse. And here's another likely reason: cell phones are not biological samples, like DNA or blood, that are exquisitely vulnerable to contamination unless robust laboratory protocols are followed. So there was far less for Clisham to document.

¶ 51 But it is not for us to venture alone into the few details Clisham's report has to offer. The question is, apart from defendant's vague gestures toward the existence of a forensic report, what specific testimonial statements does he identify? The answer, again, is none. All defendant says is that “the contents of Clisham's report, presented through Dziedzic, were admitted for the truth of the matter asserted—specifically, that the extraction revealed that the phone number [in question] was connected to Tyrese Bell.”

¶ 52 If Clisham had concluded that this was defendant's phone, or that he was a user of the phone, or more vaguely, as defendant says, that the extracted phone number was “connected” to him, those would be likely candidates for testimonial hearsay statements. But Clisham never asserted any such thing. Nor did Dziedzic, or any other witness, for that matter.

¶ 53 Defendant's argument conflates the “matter asserted” by the out-of-court declarant, here Clisham, with an inference that the State asked the jury to draw from multiple pieces of evidence considered together. The CSLI placed a phone with a certain assigned number in Madu's taxi at the time of the murder. A phone was seized from defendant when he was arrested the next day. The data extraction showed that number in question was assigned to the seized phone. Thus, the number was “connected” to defendant. Or, stated without appellate counsel's circumspection: defendant's phone was in the taxi when Madu was killed. A compelling inference, given the evidence as a whole, but not an assertion that Clisham ever made.

¶ 54 Defendant has failed to identify any testimonial hearsay statement that the State offered into evidence. He has thus failed to establish any violation of the confrontation clause.

¶ 55 II

¶ 56 Confrontation aside, defendant objects that Dziedzic's previously undisclosed testimony, on the fifth and final day of trial, was an unfair surprise and thus a discovery violation. See Ill. S. Ct. R. 412(a)(i), (iv) (eff. Mar. 1, 2001). To put this claim in context, we must first explain why Dziedzic became a witness at the last minute. Much of this backstory was brought to light during the arguments that preceded his testimony. We rely here on the parties’ representations to the trial court during those extended colloquies.

¶ 57 A

¶ 58 Discovery receipts verified that the data extraction was disclosed to the defense in two installments, in May and June 2016. That was more than six years before trial. The State later disclosed Clisham as a potential witness in an amended answer filed in July 2019. Fast forward to the first day of trial, in September 2022: Clisham was not on the State's witness list. Nor was anyone else from the RCFL.

¶ 59 So it seems that the State did not anticipate using the data extraction. That might seem surprising, at first blush, given that the extracted data proved up defendant's phone number, a lynchpin of the State's case. But then again, a phone's assigned number is part of the subscriber information available from the carrier—in this case, Sprint—and is routinely proven up, often by stipulation, through the carrier's business records. That is what the State planned to do here. Until its plan fell through.

¶ 60 The parties did stipulate to some Sprint records. To this end, the State tendered a disk to defense counsel with two sets of records: call detail records and a tower list. These were the records that Special Agent Raschke used in his CSLI analysis. The defense agreed to stipulate “to the contents of that disk.” To be clear, the subscriber information for defendant's phone was not included on the disk. So defense counsel did not understand that information to fall within the scope of the agreed stipulation.

¶ 61 That was “Monday,” meaning the first day of trial. On Thursday evening, the day before Dziedzic testified, the State noticed the glaring omission and asked defense counsel to agree to an amended stipulation that included the subscriber information. Defense counsel said no.

¶ 62 Counsel understood that stipulating to this “key” or “very critical evidence” would have linked defendant's phone to the CSLI and, in turn, to Madu's taxi. The subscriber information for a given phone number includes the serial number of the phone to which it is assigned. And the serial number of defendant's phone was in evidence through Detective Gorman, who recorded it for inventory purposes. Rather than agree to an amended stipulation in the middle of trial—one that would all but seal defendant's fate, no less—counsel held the State to its burden of proving this pivotal fact through admissible evidence. As the defense had every right to do.

¶ 63 By allowing the intended stipulation to remain unresolved for the first few days of trial, the State left itself with few alternatives for proving up defendant's phone number. With enough lead time, the State easily could have subpoenaed Jill Rundquist, Sprint's keeper of the records, to testify in person. Rundquist was disclosed on the State's witness list. But getting her into court on Friday morning, after the intended stipulation fell through on Thursday evening, was difficult, if not impossible, as a practical matter. So the State “had to go the other route, the RCFL route.” Having gone that route, the natural witness to call was Clisham. But Clisham was “not available to testify, as he's retired and is out of state.”

¶ 64 Defendant had already objected to Dziedzic's testimony on confrontation grounds when the State made this representation. And if the State meant that Clisham was “unavailable” for purposes of the confrontation clause (or the hearsay rule), it was plainly wrong. See Crawford v. Washington, 541 U.S. 36, 53-54 (2004); People v. Torres, 2012 IL 111302, ¶¶ 53-54. An out-of-state witness is not immune from process and so is not necessarily “unavailable” in the “narrow” sixth-amendment sense. Barber v. Page, 390 U.S. 719, 724-25 (1968); Torres, 2012 IL 111302, ¶¶ 54-55 (deported witness not necessarily unavailable for confrontation purposes); see Ill. R. Evid. 804(a)(5) (eff. Jan. 1, 2011). As defense counsel noted, the State “could have subpoenaed” Clisham. Apparently it didn't. And for what it's worth, a subpoena may not have been necessary: Clisham may have agreed to testify, for all we know, if the State brought him to Chicago.

¶ 65 So we would caution the State not to rely on this cavalier view of a witness's availability for confrontation purposes, if that is what it really was. But perhaps the State simply meant that Clisham was “not available” in the same practical sense as Rundquist, meaning that the State could not get him into court on such short notice. And that much was probably true. Though as defense counsel noted, the State's bind was of its own making.

¶ 66 In any event, the State found a last-minute option: Dziedzic. Unlike Clisham, he was a local. And he was within the State's control, since he was serving as director of the RCFL on a “detail” from the Cook County Sheriff's Office, where he was employed as a sergeant. In court the next morning, the State informed defense counsel that Dziedzic would testify and tendered a copy of his curriculum vitae.

¶ 67 Counsel immediately objected that Dziedzic's testimony would give rise to a “discovery violation,” since the defense had “no idea what this witness is going to say,” “what this person even did in this case,” or “why this person is being called.” Counsel asked the trial court to bar Dziedzic's testimony on this basis. (And for the alleged confrontation violation discussed above.) It became clear, during the argument, that the principal reason for Dziedzic's testimony was to get defendant's cell-phone number into evidence, via the extraction report, after the intended stipulation failed. The trial court found that there was no discovery violation, since the extraction report was disclosed to the defense years ago.

¶ 68 The trial court granted the defense some time to prepare its cross-examination. When court reconvened a couple hours later, the defense moved for a two-week continuance. The defense hoped to subpoena RCFL's peer-review protocols and engage an expert to review them, along with any available “metadata,” to determine whether Dziedzic “actually did perform any review or any work on this case.”

¶ 69 Alternatively, the defense moved to limit Dziedzic's testimony to the extracted phone number. This was, after all, the “principal reason” for his testimony: “[T]o cure that little issue from yesterday that was missed.” But based on the selections from the extraction report tendered that day, counsel believed that the State now intended to introduce other extracted data, including pictures of the gun and watch, that it did not previously intend to use. Preventing the State from engaging in this opportunistic behavior, as counsel painted it, would at least “partially cure” the prejudice to the defense.

¶ 70 The trial court adhered to its earlier ruling that there was no discovery violation, since the defense had the data extraction in hand for years, and denied both defense motions. The defense moved for a mistrial and then, as a last resort, to postpone closing arguments until Monday, to give counsel the weekend to retool the defense argument in light of Dziedzic's testimony. That last motion was also denied. With that, Dziedzic took the stand and was qualified as an expert in “computer forensics.” We will circle back to the details of his testimony.

¶ 71 B

¶ 72 Defendant argues that the last-minute disclosure of Dziedzic violated Rule 412, which requires the State to disclose the names (and any statements) of its intended witnesses and the qualifications of any proposed expert “as soon as practicable.” Ill. S. Ct. R. 412(a)(i), (iv), (d) (eff. Mar. 1, 2001). The State maintains that it complied with its “[c]ontinuing [d]uty to [d]isclose” under Rule 415(b), as soon as the need to call Dziedzic as a witness arose “during trial.” Ill. S. Ct. R. 415(b) (eff. Oct. 1, 1971). Thus, says the State, there was no discovery violation, and even if there was, the violation did not prejudice defendant and did not warrant sanctions in the trial court, much less reversal on appeal. See People v. Smith, 2025 IL 130067, ¶¶ 45-46.

¶ 73 We review de novo the purely legal question of whether the State violated its discovery obligations; and if it did, we review the trial court's choice of sanctions, if any, for an abuse of discretion. People v. Cunningham, 2018 IL App (1st) 153367, ¶ 51.

¶ 74 One preliminary point. The trial court ruled that there was no discovery violation, as the State disclosed the extraction report more than six years before trial. True, all the data the State went on to use—the phone number, the photos of the gun and the watch, and everything else—was timely disclosed; the State's use of the extraction report itself was not a discovery violation.

¶ 75 But defendant does not frame the alleged discovery violation in terms of the extraction report, as the trial court did. He frames it in terms of Dziedzic's testimony. That is a different allegation. And given the circumstances of Dziedzic's testimony, we cannot agree that the State had free reign to call him without prior disclosure, even if the extraction report introduced through his testimony was itself fair game (as far as discovery rules are concerned).

¶ 76 The State's position, in a nutshell, is that Dziedzic's testimony was the result of defendant's refusal to accept the amended stipulation in the middle of trial. That is wildly wrong. The defense was under no obligation to stipulate to anything. Dziedzic's testimony—and the State's sudden need for it—was the result of the State's own lack of diligence and nothing else.

¶ 77 It was the State that wanted a stipulation to the subscriber information, for the benefit of its own case, so it was the State's responsibility to make a diligent and timely effort to secure it. But the State went to trial without a finalized stipulation in hand and then negligently omitted the subscriber information. Rundquist could have authenticated that evidence, but the State seemed unprepared to get her into court. And the State made no effort to bring Clisham to court, either—in part, no doubt, because it was counting on the stipulation; and in part, perhaps, because it mistakenly believed that it was under no obligation to secure the appearance of a witness who retired and moved to another state.

¶ 78 Spelled out more fully, the State's position is that it complied with its continuing duty to disclose. Rule 415(b) contemplates that “contingencies” (the State's phrase) may arise “during trial” (the rule's phrase), thus giving rise to new disclosure obligations as “strategies evolve” (the State's phrase) in light of these developments. Ill. S. Ct. R. 415(b) (eff. Oct. 1, 1971); People v. Hendricks, 325 Ill. App. 3d 1097, 1103 (2001).

¶ 79 The State's references to “contingencies” and “evolv[ing]” “strategies” are too broad and forgiving. Rule 415(b) provides, more narrowly, for mid-trial disclosure of “additional material or information [that] is discovered during trial.” Ill. S. Ct. R. 415(b) (eff. Oct. 1, 1971). The State did not first “discover[ ]” Dziedzic during trial. It discovered a gap in its case, created by its own lack of diligence—its tardiness and negligence in preparing its hoped-for stipulation and its failure to arrange for its disclosed witnesses to testify if necessary. The State's need to close that gap did not justify disclosure of a new witness and his expert qualifications on the morning of his testimony, five days into trial. Rule 415(b) is not meant to bail the State out of self-inflicted predicaments like this. Granted, the State did not deliberately withhold Dziedzic from the defense, but his last-minute disclosure and testimony was a discovery violation all the same.

¶ 80 Still, not every discovery violation warrants reversal. Defendant must show that the violation caused him prejudice, and that the trial court failed to remedy that prejudice through appropriate sanctions. Smith, 2025 IL 130067, ¶ 41. Relevant factors include (1) the closeness of the evidence, (2) the strength of the undisclosed evidence, (3) the likelihood that prior notice could have helped the defense discredit the evidence, (4) the State's willfulness in failing to disclose the evidence, and (5) the remedies, including any continuances, sought by the defense in the trial court. Id. ¶ 46; People v. Weaver, 92 Ill. 2d 545, 559-60 (1982).

¶ 81 The second and third factors identified in Weaver are particularly relevant here, and they lead us to conclude that Dziedzic's testimony did not prejudice defendant. For reasons we will soon explain, his limited testimony can hardly be called strong evidence against defendant. There was little to it, and thus little if anything for the defense to discredit—and equally little chance that timely disclosure (or a continuance) would have helped the defense discredit it.

¶ 82 Here's what we mean. The “strength” of the evidence that came in through Dziedzic derived from the extraction report, not anything that Dziedzic had to say. As noted above, the extraction report was timely disclosed and therefore admissible—subject to foundational testimony. And that is where Dziedzic came in. The purpose of Dziedzic's testimony, in the first instance, was to authenticate the extraction report and thus lay the necessary foundation for its admission into evidence. Dziedzic thus testified, in sum, that the extraction report tendered to him on the witness stand was the report that Clisham submitted, pursuant to a Chicago Police Department request to extract all retrievable data from a cell phone sent to the RCFL under a certain inventory number.

¶ 83 Defendant has never raised any foundational objection to the extraction report, much less a specific objection that Dziedzic, for whatever reason, could not lay the necessary foundation. And what would that objection be? Dziedzic was the director of the RCFL. He was Clisham's direct supervisor. For good measure, he also conducted both a “second peer review” and an “administrative review” of Clisham's work on this matter, though that was hardly necessary for the limited purpose at hand. Surely he could authenticate the extraction report. Defendant had a vanishingly small chance of ever proving otherwise, even with a timely pretrial disclosure or the two-week continuance that counsel requested. Dziedzic's limited foundational testimony, for an otherwise plainly admissible piece of evidence, did not prejudice defendant.

¶ 84 Dziedzic's testimony also had a second, and even more limited, purpose. As an expert in computer forensics, he explained how a Cellebrite data extraction works. Because this process is largely rote and almost entirely automated (see Gonzalez, 144 F.4th at 407; Baldwin, 723 S.W.3d at 686), it is a fair question whether an expert is even necessary for this purpose. But that is a question for another day.

¶ 85 The point here is that this explanation did not prejudice defendant. To be clear, Dziedzic did not claim that he personally observed what Clisham did in this particular matter; he simply explained the data extraction process in general, to help the jury understand the evidence before it. And beyond that, Dziedzic merely served as a mechanism for the State to pick out, and thus draw the jury's attention to, the few items in the voluminous extraction report that provided probative evidence in this case. All told, there was nothing prejudicial in Dziedzic's testimony.

¶ 86 As expert testimony in a field of so-called forensic analysis goes, Dziedzic's testimony was rather thin. Of course, we mean no criticism in saying this; the point reflects the limited purposes for which the State needed his testimony. The discovery violation that arose from his testimony was thus minimal, a mere technical matter.

¶ 87 Even so, it bears emphasis that the State's continuing-disclosure theory is egregiously wrong. On a slightly different set of facts, it could easily lead to a prejudicial violation.

¶ 88 Consider one of many possible examples. Say the State calls a surrogate forensic analyst to testify in the same circumstances as Dziedzic—a hoped-for stipulation falls through at the last minute, the State falls back on a forensic analysis, but (for whatever reason) the State fails to arrange for the actual analyst to testify. So the State calls the analyst's peer reviewer, who has never been disclosed as a potential witness until now. So far, this is our case, in all essentials.

¶ 89 But unlike our case, imagine that the hypothetical surrogate testifies, on claimed personal knowledge, that all the proper lab protocols were followed and that the results were correctly and accurately recorded. In other words, the surrogate bolsters the credibility of the forensic evidence by offering all the same (testimonial) certifications that the actual analyst would have offered. See Bullcoming, 564 U.S. at 651, 659-61.

¶ 90 We would like to think the State knows better than to try this. But its continuing-disclosure theory leaves us with no such confidence. At a minimum, the obvious discovery violation in our hypothetical would entitle the defense to a meaningful continuance (not just an hour or two) for exactly the reasons counsel requested one here—to vet the surrogate analyst and the lab's procedures, with an eye toward attacking, if possible, the surrogate's certifications about the process and its reported results. Springing undisclosed testimony like this on the defense would surely court reversal.

¶ 91 Here, though, Dziedzic offered no such certifications. Nor did he draw any conclusions or interpret any results. Put simply, he did not say anything that could be called substantive evidence. Notably, neither did Clisham. And that does not strike us as happenstance. In no small part, it is because the so-called forensic analysis at issue was a mere data extraction.

¶ 92 Certifications or other substantive input from the analyst witness are far more important, and likely, for a forensic analysis in the fuller (and perhaps more proper) sense of the term—say, a DNA analysis that requires skilled interpretation and can be skewed by even slight mishandling of a sensitive biological sample. But as we have said, a data extraction is a rote and automated process, one that produces a machine-generated record of a phone's contents that largely speaks for itself. And that is ultimately why the State needed little more from Dziedzic than to authenticate the extraction report. With most other types of forensic analysts, called to testify in similar circumstances, opportunities for prejudicial discovery violations would abound.

¶ 93 Dziedzic's last-minute disclosure was a clear violation of Rule 415(b). But it did not prejudice defendant and does not warrant reversal.

¶ 94 III

¶ 95 Detective Lutzow testified that “during the course of the investigation we observed prior photos of Mr. Bell wearing that watch [that is, the one he was wearing when he was arrested] inside the cab where the victim was driving him.” Later, Detective Lutzow identified People's exhibit No. 78, a physical watch, as “the watch that was on Tyrese Bell's hand in the taxicab the night the victim was shot and killed.” Here too, Detective Lutzow based his testimony on “the video or camera stills [from] the taxi.”

¶ 96 Detective Lutzow thus identified defendant and his watch in the security images from Madu's taxi. Defendant argues that these two instances of “lay opinion identification testimony” were improper and amounted to reversible error, especially since the trial court failed to employ the “precautionary procedures” that are necessary when testimony of this kind is elicited from a law-enforcement officer. People v. Thompson, 2016 IL 118667, ¶¶ 39, 59. Some of these errors were preserved at trial, others forfeited. As for the forfeited errors, defendant argues that plain error occurred or, in the alternative, that counsel's failure to raise them at trial constituted ineffective assistance of counsel. See Strickland v. Washington, 466 U.S. 668, 687-88 (1984) (ineffectiveness); People v. Herron, 215 Ill. 2d 167, 178-79 (2005) (plain error).

¶ 97 We agree with many of defendant's claims of error. But given the strength of the evidence, these errors, preserved or otherwise, were not so prejudicial to warrant reversal. To focus our analysis, we begin with a brief overview of defendant's claims and the applicable law.

¶ 98 A

¶ 99 “Lay opinion identification testimony,” in which a witness identifies the defendant (or other person) in a photograph or video, is “helpful” to the jury and thus admissible under Illinois Rule of Evidence 701 (eff. Jan. 1, 2011), “where there is some basis for concluding that the witness is more likely to correctly identify the defendant from the photograph [or video] than is the jury.” Thompson, 2016 IL 118667, ¶¶ 41, 50 (internal quotation marks omitted).

¶ 100 As for defendant's watch, we have held that Thompson’s standard for admissibility under Rule 701 also applies to “a lay-witness's opinion testimony as to the identity of an object” and not just the identity of “a person depicted in a [photograph or] *** video.” People v. Gharrett, 2016 IL App (4th) 140315, ¶ 75 (emphases in original).

¶ 101 But lay opinion identification testimony from a law-enforcement officer necessitates special “precautionary procedures.” Thompson, 2016 IL 118667, ¶ 59. First, when the State seeks to introduce such testimony, the trial court should allow the defense to examine the officer without the jury present. Id. The purpose of this precautionary examination is to create a proper record for the trial court's application of Rule 701 and subsequent Rule 403 balancing—by exploring such topics as the officer's level of contact and familiarity with the defendant and any possible sources of bias or prejudice—without revealing potentially prejudicial information to the jury. Id.

¶ 102 A failure to conduct this precautionary examination is error, even when “there is some basis to conclude,” from the record as it stands, that the officer “was more likely to correctly identify [the] defendant than the jury”—that is, even when the testimony appears to be admissible under Rule 701. Id. ¶ 62; see also People v. Stitts, 2020 IL App (1st) 171723, ¶ 28.

¶ 103 As far as we know, no case has expressly decided whether a precautionary examination is required when the officer will identify an object, such as defendant's watch, as opposed to the defendant himself. The closest case on point, Gharrett, 2016 IL App (4th) 140315, ¶ 75, does not answer that question, as the witness there was not law enforcement. Defendant offers little by way of analysis. The State, for its part, flatly asserts that a precautionary examination is not required for the identification of an object, only of the defendant, thus limiting Thompson to its facts in this respect. Skeptical as we are of the State's position, we need not settle the issue. As we already indicated, and as we will explain below, we do not find any error in this regard sufficiently prejudicial to warrant relief, so we leave that question for another day.

¶ 104 Second, when the trial court allows lay opinion identification testimony from a law-enforcement officer, it should instruct the jury (before the testimony and in the final charge) that it need not give the testimony special weight or any weight at all, nor should it draw any adverse inference from the fact that the witness is a law-enforcement officer. Thompson, 2016 IL 118667, ¶ 59. These instructions are now codified in Illinois Pattern Jury Instructions, Criminal, No. 3.15B (approved Jan. 26, 2018) (hereinafter IPI Criminal No. 3.15B) (“It is for you to determine what weight, if any, should be given to that evidence. In determining the weight to be given to this evidence, you should not draw any inference from the fact that the witness is a law enforcement officer.”). And by its terms, this instruction applies when the officer identifies either a person or an “object” from a video or photo. Id.

¶ 105 So defendant has established multiple errors. For one, no precautionary examination of Detective Lutzow took place. For this reason alone, as the State concedes, his identification of defendant in the security images was error. See Thompson, 2016 IL 118667, ¶ 62. Second, the trial court did not give the jury IPI Criminal No. 3.15B, which was error twice over, as the instruction, by its terms, applies to the detective's identification of defendant and his watch. And third, the trial court did not give this cautionary instruction before Lutzow's testimony. Id. ¶ 59.

¶ 106 An error in admitting lay opinion identification testimony—including one that results from a failure to conduct a precautionary examination of a law-enforcement officer—may be “harmless” if the evidence is sufficiently strong and/or the jury is given the proper cautionary instruction. Thompson, 2016 IL 118667, ¶¶ 62-71. A failure to properly instruct the jury with IPI Criminal No. 3.15B is also error in its own right, as well as a factor in a harmless-error or prejudice analysis for the underlying testimony. And subject to limited exceptions not relevant here, a jury instruction error is subject to harmless-error analysis. People v. Hartfield, 2022 IL 126729, ¶ 42.

¶ 107 To the extent these errors are unpreserved, defendant claims, among other things, that they are reversible as second-prong plain errors. But “ ‘second-prong plain error can be invoked only for *** errors that are not subject to harmless error analysis.’ ” People v. Chambliss, 2026 IL 130585, ¶ 67 (quoting People v. Jackson, 2022 IL 127256, ¶ 49)). None of the errors under review requires automatic reversal, so they cannot be second-prong plain errors.

¶ 108 Which means that reversal is only warranted if the errors were prejudicial to defendant—under a harmless-error standard for preserved errors or under Strickland, for those that counsel failed to raise. (Defendant does not argue first-prong plain error.) Sorting the errors along these lines would needlessly complicate the issue, since our confidence in the jury's verdict would remain intact under either standard, for the reasons we explain below.

¶ 109 B

¶ 110 As defendant sees it, Detective Lutzow's improper lay opinion identification testimony “invaded the province of the jury on the ultimate question”—whether defendant was “the shooter depicted” in the security images from Madu's taxi. And without a proper cautionary instruction, there was an “impermissible risk” that the jury would give undue deference to the opinion of a detective on this critical question. All of which, says defendant, adds up to an appreciable risk that the detective's improper testimony swayed the verdict.

¶ 111 Notice what defendant's prejudice argument does not address: the strength of the State's case. And we perceive no serious risk that the jury convicted defendant on the basis of anything other than the strong case for his guilt made by the evidence as a whole. The jury was not left to decide whether defendant was “the shooter depicted” just by viewing the security images and considering Detective Lutzow's opinion on the matter. Other evidence spoke to this question, some of it perhaps more clearly than the security images themselves.

¶ 112 To return to a familiar theme: the real lynchpin of the State's case was the combination of the extracted phone number and the CSLI analysis. This evidence made it all but undeniable that defendant's phone was in Madu's taxi at the time of the murder. The CSLI and (time-stamped) security images further establish, beyond any dispute, that the same person who put a gun to Madu's head also made multiple calls from defendant's phone during the same taxi ride. And in case there is any doubt, Dziedzic confirmed that defendant's phone had a passcode, one that the Cellebrite software had to unlock.

¶ 113 So even without considering the physical appearance of “the shooter depicted,” or the watch he was wearing, the jury knew that the shooter not only possessed but used defendant's phone in the taxi. Whoever that person was, he was alone with Madu; there were no other passengers. After the shooting, whoever that person was stashed the murder weapon in Michell's apartment at 1250 South Albany Avenue. Officer Cox identified defendant walking out of that building only moments later. And Mitchell was clearly no stranger to defendant; if nothing else, she identified a photograph of him for the detectives. As for the murder weapon, a Bersa with a distinctive silver pattern, defendant had pictures of this same kind of gun attached to messages in his phone.

¶ 114 If all of this evidence still leaves one unsure whether defendant is “the shooter depicted,” here is perhaps the final nail in the coffin. Detective Lutzow's opinion aside, the jury viewed the security images for itself. Like nearly all surveillance footage, it has its limitations, and a positive identification on this basis alone would be tenuous. That said, as the State argued to the jury, the physical resemblance is there for all to see.

¶ 115 For good measure, though it is hardly necessary to make the point, we will add that the shooter also wore a watch that looked like the one defendant was wearing when he was arrested. Here, too, the jury could judge for itself. And defendant violently resisted the police when they tried to photograph that watch as evidence linking him to the murder. As defendant concedes (in the next issue), this conduct evinced a consciousness of guilt.

¶ 116 To deny that defendant is the shooter, one would have to believe, among other things, that someone who looked like defendant, and had defendant's phone and passcode, shot Madu and then stashed the murder weapon in the apartment of one of defendant's acquaintances, in the exact building that defendant walked out of only minutes after the murder. We do not believe that a rational jury would come to that conclusion. Detective Lutzow's testimony does little, if anything, to explain the jury's verdict.

¶ 117 Two final points. First, it bears emphasis that the jury saw the security images for itself, so this is not a matter of blind faith in an officer's opinion about matters outside the jury's own perception. And this is in part why Thompson, 2016 IL 118667, ¶¶ 54, 70, says that lay opinion identification testimony does not, as defendant claims, invade the province of the jury. The jury was free to make up its own mind, after viewing the security images with its own eyes and in conjunction with other evidence that pointed to defendant as the shooter. We are thus confident that the jury's verdict was the product of its own factfinding, and that the jury would have found as it did, even without Detective Lutzow's lay opinion identification testimony at all.

¶ 118 Which brings us to our final point. IPI Criminal No. 3.15B would have reminded the jury (1) that it was free to reject Detective Lutzow's testimony altogether and (2) to make nothing of the fact that the testimony came from a law-enforcement officer. This instruction should have been given. But if Detective Lutzow's improper testimony was not prejudicial, neither was the lack of a cautionary instruction. The risk of undue deference to the detective's opinion does no more to explain the verdict than the opinion testimony itself. Given the strong evidence of guilt, the errors arising from Detective Lutzow's testimony do not warrant reversal.

¶ 119 IV

¶ 120 When the police tried to photograph defendant's watch, he mounted a violent resistance that injured two detectives and resulted in charges of aggravated battery. As described above, the State offered testimony on this topic from three police officers, video footage of the incident that it played twice, and photographs depicting the detectives’ injuries.

¶ 121 Defendant concedes that this evidence was generally admissible to show consciousness of guilt. See, e.g., People v. Abernathy, 402 Ill. App. 3d 736, 754-55 (2010) (attempt to destroy evidence); People v. Shanklin, 367 Ill. App. 3d 569, 578 (2006) (refusal to participate in lineup); People v. Hart, 214 Ill. 2d 490, 519 (2005) (resisting officers upon apprehension).

¶ 122 But the amount of evidence, says defendant, was excessive—far more than the legitimate probative value of this topic warranted—and thus unfairly prejudicial. The State certainly could have made its point with less evidence, but we need not decide whether the trial court thus abused its admittedly wide discretion in evidentiary matters. Any surplus evidence on this topic was clearly harmless, for two reasons. First, the evidence was generally admissible and not intrinsically improper, which mitigates any prejudicial effect. And more importantly, as explained above, the State presented a strong case for defendant's guilt, even without considering any of this other-crimes evidence.

¶ 123 V

¶ 124 Defendant was found guilty of the first-degree murder of Madu, on alternative theories of intentional (Count 3) and strong-probability (Count 6) murder. The death of a single victim can only support a single murder conviction. People v. Simmons, 2016 IL App (1st) 131300, ¶ 175. The trial court thus merged the knowing-murder count into the more serious intentional-murder count for sentencing purposes. See People v. Smith, 233 Ill. 2d 1, 20-21 (2009). But the mittimus erroneously lists two convictions, on Counts 3 and 6. The State concedes, and we agree, that the mittimus should reflect a single conviction for first-degree murder, entered on Count 3.

¶ 125 Pursuant to Illinois Supreme Court Rule 615(b)(1) (eff. Jan. 1, 1967), we direct the clerk of the circuit court to correct the mittimus accordingly. We need not remand the case for this ministerial task. People v. Smith, 2016 IL App (1st) 140039, ¶ 19.

¶ 126 CONCLUSION

¶ 127 We affirm the judgment of the circuit court and correct the mittimus.

¶ 128 Affirmed; mittimus corrected.

JUSTICE ELLIS delivered the judgment of the court, with opinion.