State v. LongAnnotate this Case
721 P.2d 483 (1986)
The STATE of Utah, Plaintiff and Respondent, v. Anthony L. LONG, Defendant and Appellant.
Supreme Court of Utah.
June 20, 1986.
*484 Karen Jennings, Salt Lake City, for defendant and appellant.
David L. Wilkinson, Atty. Gen., Dave B. Thompson, Asst. Atty. Gen., Salt Lake City, for plaintiff and respondent.
Defendant Anthony L. Long was convicted of aggravated assault, a third degree felony, and possession of a dangerous weapon by a restricted person, a second degree felony. Before this Court, Long claims error in the authentication of documents evidencing his earlier felony convictions, the refusal to sever trial of the two charges, and the refusal to give cautionary instructions about the eyewitness identification. We reverse and remand for a new trial.
On the evening of March 1, 1983, just after dark, Joe Rocha was approaching the front door of his home in Salt Lake City when he heard footsteps behind him and a voice say, "Don't move or I will blow your head off." He felt something cold and metallic on the back of his head. The voice ordered Rocha onto the front porch of his house and up against the wall next to the door. He faced the wall, unable to see his assailant. His girlfriend opened the door and stepped out of the house. She saw a black man holding a gun to Rocha's head and another black man coming toward the house. She ran inside, closed the door, and told Rocha's son, Jacob, what she had seen. She then called the police.
Meanwhile, Jacob took a .357 magnum revolver from a desk drawer and went to the front door. As he started to open the door, it was kicked in from the outside. A sawed-off shotgun was fired into the house. Jacob was hit and thrown back into the room. He fired a shot in return. Jacob testified that he saw the face of his assailant for about six seconds, during which he was crying and his vision was "glossy."
Joe Rocha testified that after the shots were fired he saw two men running away from the house. One of Joe Rocha's neighbors testified that he heard two shots fired. He went outside, saw a person approaching a bronze or tan Oldsmobile Cutlass or Seville with its motor running, and heard someone say, "Jessie, let's go." The person got in the passenger side of the car and it drove off.
Robin Lee, an acquaintance of defendant Long's, testified that she was with Long and Jessie Hobsun on the evening of March 1st, that they had parked their car in an alley, and that Long and Hobsun had then left the car. Hobsun returned shortly, got in the car, and told Lee to drive away. Long then approached the car and got in the back seat. He was wounded. Soon after, the three became involved in a high speed chase with the police. When they were finally stopped, all three were arrested. The arresting officers saw that Long had a large blotch of red on his shirt in the abdomen area. Long's coat had two bullet holes in it, one in the back where the bullet had entered and another high under the arm where it had come out.
Three days later, while Jacob Rocha was still in the hospital and on medication, a detective presented him with a photo array and asked him to identify his assailant. The photo array included a picture of Long. Jacob did not pick defendant Long's photo from the array. However, he did select two other photos, one of which was of Hobsun.
At trial, Jacob identified Long in person as the man who shot him, as he had previously done at two face-to-face encounters during preliminary hearings where Long was clearly identified. On cross-examination, however, Jacob did not recall being unable to pick Long's picture from the initial photo array. At the close of trial, Long's counsel requested cautionary instructions, patterned after those suggested in United States v. Telfaire, 469 F.2d 552 (D.C.Cir.1972), regarding Jacob's eyewitness identification. The court refused the instructions. Long was found guilty of aggravated assault and possession of a dangerous weapon by a restricted person. This appeal followed.
*485 Long first challenges his conviction of possessing a dangerous weapon on the ground that the documents used to prove that he was a convicted felon were not properly authenticated. The charge of possession of a dangerous weapon by a restricted person involves three elements: (1) possession of a dangerous weapon (2) by an individual who is on parole (3) from a felony conviction. U.C.A., 1953, § 76-10-503(2) (Repl.Vol. 8B, 1978). To establish that Long was a convicted felon, his Utah parole officer, Flint Mollner, was called as a witness. Through Mollner, the State offered into evidence copies of certified copies of documents from the Davis County clerk's office showing that Long had been twice convicted of felonies. The certified copies from which the copies introduced into evidence had been made were part of Mollner's parole file on Long. Long's counsel unsuccessfully objected to the admission of the copies of the certified copies in Mollner's files on the grounds that they were hearsay and were not within any of the exceptions in the Utah Rules of Evidence. On appeal, Long renews the argument made below.
The copies were admitted under Utah Rule of Evidence 63(17)(a), the official records exception to the hearsay rule. Under this provision, the content of an official record is admissible to prove the truth of the matter asserted if the requirements of both Rule 64 and Rule 68 of the Utah Rules of Evidence are met. Long argues that the prosecutor did not deliver copies of the documents to opposing counsel within a reasonable time before trial, as required by Rule 64, and that the copies were not adequately authenticated, as required by Rule 68(1).
Long's Rule 64 argument is without merit. Rule 64 specifically provides that even if a copy of the document sought to be admitted is not delivered to opposing counsel within a reasonable time before trial, it still may be admitted if "the judge finds that [the] adverse party has not been unfairly surprised by the failure to deliver such copy." The lower court's finding that there was no unfair surprise is supported by the facts. The documents related only to prior felony convictions, and Long and his counsel certainly knew that proof of at least one such conviction would be introduced at trial. Moreover, defense counsel's detailed and capably presented objection to the exhibit at trial belies any claim of unfair surprise. We find no abuse of discretion in the court's ruling.
There is merit, however, to Long's Rule 68(1) argument. That rule states: "An official record ... may be evidenced ... by a copy attested by the officer having the legal custody of the record, or by his deputy, and ... accompanied with a certificate that such officer has the custody." Long argues that Mollner was not "the officer having the legal custody of the record" because the originals of the documents at issue were official records of the Weber County clerk's office, not of the Utah Adult Probation and Parole section. The certification of a Utah parole officer, therefore, could not suffice to bring the copies within the ambit of Rule 68(1) or, consequently, Rule 63(17). Mollner had only copies of copies that had been certified by the Weber County clerk and was in no position to testify that the copies in his possession were copies of originals because he had never seen the originals. Long asserts that to accept Mollner's authentication would make the hearsay rule meaningless in cases involving copies of official records.
To support his position, Long relies on State v. Lamorie, 610 P.2d 342 (Utah 1980). In that case, the defendant was also charged with possession of a dangerous weapon by a restricted person. A parole officer introduced copies of court records, certified only by a notary public, to show the defendant's prior felony convictions. Because the certification was executed by someone who neither had legal custody of *486 the records nor was a deputy of the legal custodian, this Court held that the authentication was inadequate under Rule 68(1) and ordered a new trial.
The State seeks to distinguish Lamorie. It argues that in Lamorie the person certifying the copies of the judgment of conviction was only a notary public and not the custodian of the court records. In contrast, Mollner, who certified the copies, was legal custodian of Long's parole file. Therefore, the State asserts that as an official record of the Adult Probation and Parole section of the Division of Corrections, the contents of the file were admissible. In making this argument, the State relies on People v. Howard, 72 Cal. App. 561, 237 P. 780 (1925). Our reading of Howard, however, leads us to the opposite conclusion.
In Howard, the prosecution also attempted to prove the fact of a prior felony conviction by introducing a copy of a certified copy of a judgment and commitment to state prison. The original document was certified by the clerk of the court entering the judgment of conviction. The copy of this document was accompanied by a document executed by the warden of San Quentin and impressed with the state prison seal certifying that the introduced copy was a "true and correct" copy of the defendant's commitment papers. The California Supreme Court upheld the admission of the copy into evidence, stating that when a certified copy of a judgment of conviction is delivered with a convict to the state prison warden, as required by statute, that copy becomes an official document of the state prison, and a certified copy of the copy, authenticated by the warden, is admissible into evidence. Id. 237 P. at 781.
In the case before us, the copies introduced into evidence were analogous to those introduced in Howard. Under Howard, certification by the Utah state prison warden that the copies were copies of official documents of the Department of Corrections and that he was their custodian would have been sufficient to permit their introduction because he, like the California warden, is statutorily mandated to take possession of the documents when a prisoner is delivered to him for incarceration. See U.C.A., 1953, § 64-13-23 (2d Repl. Vol. 7A, 1978); § 77-19-2 (Repl. Vol. 8C, 1982). However, Mollner, not the Utah state prison warden, certified the copies. There is no evidence to show how the copies got in Mollner's file, that the copies in Mollner's file constituted official documents of the Division of Corrections, or that Mollner was their official custodian or deputy custodian. Absent this evidence, there is no basis in this case for applying the Howard rationale.
The State urges us to extend the Howard holding, in effect arguing that because the warden supervises the state prison, which is under the jurisdiction of the State Division of Corrections, and because the warden could be an official custodian of copies of judgments of conviction, any other branch of the Division of Corrections, including Adult Probation and Parole, should also have authority to authenticate copies of judgments. We decline the invitation.
The hearsay rule has as its declared purpose the exclusion of evidence not subject to cross-examination concerning the truthfulness of the matters asserted. J. Wigmore, Evidence § 1362 (Chadbourn rev. 1974). The exceptions to the rule have evolved to permit the admission of evidence that is deemed reliable notwithstanding its failure to satisfy the hearsay rule. Id. at § 1420; cf. Snyder v. Massachusetts, 291 U.S. 97, 107, 54 S. Ct. 330, 332, 78 L. Ed. 2d 674 (1934); United States v. Adams, 446 P.2d 681, 683 (9th Cir.1971); Kehm v. Procter & Gamble Manufacturing Co., 724 F.2d 613, 618-19 (8th Cir.1983). In the present case, the pertinent exception has not been satisfied, and we decline to broaden it because to do so would permit evidence to be admitted without any assurance as to the truth of the matters asserted. The State's position would open the door to the admission of any documents that appear to be copies of official records, *487 so long as they wend their way into some state functionary's file.
Because the trial court erred in admitting the copies, the State is left without proof of one essential element of the first charge against Longthat Long had been previously convicted of a felony. As a result, a new trial is necessary on the charge of possession of a dangerous weapon by a restricted person. See State v. Lamorie, 610 P.2d at 346-47.
Long next contends that his conviction for aggravated assault must be reversed because the trial court erred both in refusing to give a requested jury instruction concerning eyewitness identification and in refusing to instruct the jury that it could permissibly find that he was merely present at the scene of the crime. These instructions together covered Long's theory of the case, i.e., that Jacob did not clearly identify Long as the individual who fired the shot and that Long may have been merely present when the crime was committed.
The requested cautionary instructions regarding eyewitness identification were patterned on those suggested in United States v. Telfaire, 469 F.2d at 558-59. We have previously said that whether such instructions must be given in a particular case is a matter left largely to the discretion of the trial court. State v. Tucker, 709 P.2d 313, 316 (Utah 1985); State v. Reedy, 681 P.2d 1251, 1252 (Utah 1984); State v. Newton, 681 P.2d 833, 834 (Utah 1984); State v. Malmrose, 649 P.2d 56, 61-62 (Utah 1982); State v. Schaffer, 638 P.2d 1185, 1187 (Utah 1981). We have also indicated that the failure to give such an instruction may constitute an abuse of discretion when the circumstances surrounding the eyewitness identification raise serious questions of reliability. See, e.g., State v. Reedy, 681 P.2d at 1253-54. However, as both counsel noted at oral argument in this case, although we have reviewed many cases in which the trial court refused to give a cautionary instruction, we have never reversed a conviction for that reason. We also have been advised that this Court's de facto failure to ever require such an instruction has resulted in trial courts rarely, if ever, giving cautionary instructions. In a recent case raising the identical issue, the State suggested at oral argument that this Court either abandon any pretext of requiring a cautionary eyewitness instruction or make the requirement meaningful. State v. Quevedo, No. 19049, argued November 14, 1985. We have decided to follow the latter course, adopting the approach earlier articulated by Justice Stewart in his dissent in State v. Malmrose, 649 P.2d at 62-66.
On the facts of this case, it was plainly improper under Schaffer and its progeny for the trial court not to have given a cautionary instruction. The State's case hinged on the uncorroborated eyewitness testimony of a single witnessthe victim of the crime. The circumstances surrounding his identification raised grave concerns about its reliability. Jacob Rocha had an opportunity to view the face of his assailant for approximately six seconds during the assault. At the same time he was identifying his assailant, he was shot, was thrown back against the wall by the force of the blast, returned the fire, and experienced "glossy" vision. During the rest of the thirty or so seconds of observation, Rocha could see only the back or side of his assailant as he stood in the darkness outside the front door of Rocha's house. Further, Rocha failed to identify defendant from a six-photo array presented to him three days after the shooting; interestingly, he did select a photo of Jessie Hobsun, a man who was with defendant on the night of the crime but was not prosecuted. Rocha identified Long at trial and at two preliminary hearings; however, the record indicates that these identifications took *488 place not in formal lineups, but in courtroom proceedings during which Long was apparently the only black man present. Finally, a significant discrepancy exists between the victim's description of his assailant's clothing and the clothing Long was actually wearing when police stopped the car in which he was riding.
Considering the matters more generically, the circumstances surrounding Rocha's identification highlight the questionable wisdom of allowing the uncorroborated identification testimony of one eyewitness to serve as the linchpin of the prosecution's case, at least in the absence of an instruction to the jury focusing its attention on the well-documented factors that affect the reliability of eyewitness identifications. See F. Woocher, Did Your Eyes Deceive You? Expert Psychological Testimony on the Unreliability of Eyewitness Identification, 29 Stan.L.Rev. 969 (1977); J. Bibicoff, Seeing is Believing? The Need for Cautionary Jury Instructions on the Unreliability of Eyewitness Identification Testimony, 11 San Fernando Valley L.Rev. 95 (1983); R. Sanders, Helping the Jury Evaluate Eyewitness Testimony: The Need for Additional Safeguards, 12 Am.J. Crim.Law 189 (1984). The literature is replete with empirical studies documenting the unreliability of eyewitness identification. See generally P. Wall, Eyewitness Identification in Criminal Cases (1965); E. Loftus, Eyewitness Testimony (1979). There is no significant division of opinion on the issue. The studies all lead inexorably to the conclusion that human perception is inexact and that human memory is both limited and fallible. We therefore have concluded that a more rigorous approach to cautionary instructions than this court has heretofore followed is appropriate. See State v. Malmrose, 649 P.2d at 62-66 (Stewart, J., dissenting).
Some background is necessary. Anyone who stops to consider the matter will recognize that the process of perceiving events and remembering them is not as simple or as certain as turning on a camera and recording everything the camera sees on tape or film for later replay. What we perceive and remember is the result of a much more complex process, one that does not occur without involving the whole person, and one that is profoundly affected by who we are and what we bring to the event of perception. See R. Buckhout, Eyewitness Testimony, 15 Jurimetrics J. 171, 179 (1975) (reprinted from 231 Scientific American 23 (Dec.1974)).
Research on human memory has consistently shown that failures may occur and inaccuracies creep in at any stage of what is broadly referred to as the "memory process." This process includes the acquisition of information, its storage, and its retrieval and communication to others. These stages have all been extensively studied in recent years, and a wide variety of factors influencing each stage have been identified. See Loftus, supra, at chs. 3-5; Buckhout, supra, at 172-81.
During the first or acquisition stage, a wide array of factors has been found to affect the accuracy of an individual's perception. Some of these are rather obvious. For example, the circumstances of the observation are critical: the distance of the observer from the event, the length of time available to perceive the event, the amount of light available, and the amount of movement involved. Buckhout, supra, at 173. However, perhaps the more important factors affecting the accuracy of one's perception are those factors originating within the observer. One such limitation is the individual's physical condition, including both obvious infirmities as well as such factors as fatigue and drug or alcohol use. Another limitation which can affect perception *489 is the emotional state of the observer. Contrary to much accepted lore, when an observer is experiencing a marked degree of stress, perceptual abilities are known to decrease significantly. See, e.g., Woocher, supra, at 979 n. 29.
A far less obvious limitation of great importance arises from the fact that the human brain cannot receive and store all the stimuli simultaneously presented to it. This forces people to be selective in what they perceive of any given event. See Woocher, supra, at 976-77. To accomplish this selective perception successfully, over time each person develops unconscious strategies for determining what elements of an event are important enough to be selected out for perception. The rest of the stimuli created by the event are ignored by the brain. These unconscious strategies of selective perception work quite well in our day-to-day lives to provide us with only the most commonly useful information, but the strategies may result in the exclusion of information that will later prove important in a court proceeding. For example, the significance of the event to the witness at the time of perception is very important. Buckhout, supra, at 172-73. Thus, people usually remember with some detail and clarity their whereabouts at the time they learned of John F. Kennedy's assassination. Those same people, however, are generally less accurate in their descriptions of people, places, and events encountered only recently in the course of their daily routines. For instance, few of us can remember the color or make of the car that was in front of us at the last traffic signal where we waited for the light to turn green. An everyday situation such as this presents an excellent opportunity to observe, and yet, while such information may be a critical element in a criminal trial, our process of selective perception usually screens out such data completely. To the extent that court proceedings may focus on events that were not of particular importance to the observer at the time they occurred, then, the observer may have absolutely no memory of the facts simply because he or she failed to select the critical information for perception.
Another mechanism we all develop to compensate for our inability to perceive all aspects of an event at once is a series of logical inferences: if we see one thing, we assume, based on our past experience, that we also saw another that ordinarily follows. This way we can "perceive" a whole event in our mind's eye when we have actually seen or heard only portions of it. Id. at 980. The implications of this memory strategy for court proceedings are similar to those of selective perception.
Other important factors that affect the accuracy of a viewer's perception, and which are unique to each observer, include the expectations, personal experiences, biases, and prejudices brought by any individual to a given situation. Buckhout, supra, at 175-76. A good example of the effect of preconceptions on the accuracy of perception is the well-documented fact that identifications tend to be more accurate where the person observing and the one being observed are of the same race. Wells, Applied Eyewitness-Testimony Research: System Variables and Estimator Variables, 36 J. Personality & Social Psych. 1546, 1550 (1978); Note, Cross Racial Identification Errors in Criminal Cases, 69 Cornell L.Rev. 934 (1984); Bibicoff, supra, at 101.
The memory process is also subject to distortion in the second or retention stage, when information that may or may not have been accurately perceived is stored in the memory. Research demonstrates that both the length of time between the witness's experience and the recollection of that experience, and the occurrence of other events in the intervening time period, affect the accuracy and completeness of recall. Just as in the perception stage, *490 where the mind infers what occurred from what was selected for perception, in the retention stage people tend to add extraneous details and to fill in memory gaps over time, thereby unconsciously constructing more detailed, logical, and coherent recollections of their actual experiences. Thus, as eyewitnesses wend their way through the criminal justice process, their reports of what was seen and heard tend to become "more accurate, more complete and less ambiguous" in appearance. Buckhout, supra, at 179. The implications of this mental strategy for any criminal defendant whose conviction hinges on an eyewitness identification are obvious. See Woocher, supra, at 983 n. 53.
Research has also undermined the common notion that the confidence with which an individual makes an identification is a valid indicator of the accuracy of the recollection. K. Deffenbacher, Eyewitness Accuracy and Confidence: Can We Infer Anything About Their Relationship?, 4 Law and Human Behavior 243 (1980); Lindsay, Wells, Rumpel, Can People Detect Eyewitness-Identification Accuracy Within and Across Situations?, 66 J. Applied Psych. 79, 80-82 (1981); Bibicoff, supra, at 104 n. 35. In fact, the accuracy of an identification is, at times, inversely related to the confidence with which it is made. Buckhout, supra, at 184.
Finally, the retrieval stage of the memory processwhen the observer recalls the event and communicates that recollection to othersis also fraught with potential for distortion. For example, language imposes limits on the observer. Experience suggests that few individuals have such a mastery of language that they will not have some difficulty in communicating the details and nuances of the original event, and the greater the inadequacy, the greater the likelihood of miscommunication. An entirely independent problem arises when one who has accurately communicated his recollection in a narrative form is then asked questions in an attempt to elicit a more complete picture of the event described. Those asking such questions, by using a variety of subtle and perhaps unconscious questioning techniques, can significantly influence what a witness "remembers" in response to questioning. And as the witness is pressed for more details, his responses become increasingly inaccurate. See Loftus, Reconstructing Memory: The Incredible Eyewitness, 15 Jurimetrics J. 188 (1975). In addition, research has documented an entirely different set of no less significant problems that relate to the suggestiveness of police lineups, showups, and photo arrays. See, e.g., Buckhout, supra, at 179-87.
Although research has convincingly demonstrated the weaknesses inherent in eyewitness identification, jurors are, for the most part, unaware of these problems. People simply do not accurately understand the deleterious effects that certain variables can have on the accuracy of the memory processes of an honest eyewitness. See K. Deffenbacher & E. Loftus, Do Jurors Share a Common Understanding Concerning Eyewitness Behavior?, 6 Law and Human Behavior 15 (1982); J. Brigham, R. Bothwell, The Ability of Prospective Jurors to Estimate the Accuracy of Eye-witness Identification, 7 Law and Human Behavior 19 (1983). Moreover, the common knowledge that people do possess often runs contrary to documented research findings. See Loftus, supra, at 171-77.
Perhaps it is precisely because jurors do not appreciate the fallibility of eyewitness testimony that they give such testimony great weight. See Sanders, supra, at 189-90 n. 6; Loftus, supra, at 8-19. In one notable study involving a simulated criminal trial, 18% of the jurors voted to convict the defendant when there were no eyewitnesses to the crime. However, when a credible eyewitness was presented, 72% voted to convict. And, surprisingly, even when presented with an eyewitness who was quite thoroughly discredited by counsel, a full 68% still voted to convict. 15 Jurimetrics J. at 189-90. In one study which found a poor relationship between witness confidence and accuracy of identification, the researchers concluded, "[i]t is possible that the jurors' rate of belief is *491 around 80% irrespective of the actual rate of witness accuracy." G. Wells, R. Lindsay, T. Ferguson, Accuracy, Confidence, and Juror Perception in Eyewitness Identification, 64 J. Applied Psych. 440, 447 (1979).
The United States Supreme Court has acknowledged that "[t]he vagaries of eyewitness identification are well-known; the annals of criminal law are rife with instances of mistaken identification." United States v. Wade, 388 U.S. 218, 228, 87 S. Ct. 1926, 1932, 18 L. Ed. 2d 1149 (1967). That Court has long professed, as a fundamental value of our democratic society, that "it is far worse to convict an innocent man than to let a guilty man go free." In re Winship, 397 U.S. 358, 372, 90 S. Ct. 1068, 1076, 25 L. Ed. 2d 368 (1970) (Harlan, J., concurring). Yet despite judicial recognition of the documented unreliability of eyewitness identification, courts have been slow both to accord the problem the attention it deserves and to fashion ways of minimizing its potentially unjust effects. The fault probably lies with the narrowness of the vision of most lawyers and judges. We tend to comfortably rely upon settled legal precedent and practice, especially when long-settled technical rules are concerned, and to largely ignore the teachings of other disciplines, especially when they contradict long-accepted legal notions. I.D. Stewart, Jr., Perception, Memory, and Hearsay: A Criticism of Present Law and the Proposed Federal Rules of Evidence, 1970 Utah L.Rev. 1, 38; see also B. Clifford, The Relevance of Psychological Investigation to Legal Issues in Testimony and Identification, 1979 Crim.L.Rev. 153; State v. Warren, 230 Kan. 385, 635 P.2d 1236, 1241 (1981).
Even though the United States Supreme Court has recognized the fundamental problem posed by eyewitness testimony, its much-quoted articulation of how one should approach the evaluation of the credibility and admissibility of eyewitness identification is a fair example of the lag between the assumptions embodied in the law and the findings of other disciplines. The Court has said that:[T]he factors to be considered in evaluating the likelihood of misidentification include the opportunity of the witness to view the criminal at the time of the crime, the witness'[s] degree of attention, the accuracy of the witness'[s] prior description of the criminal, the level of certainty demonstrated by the witness at the confrontation, and the length of time between the crime and the confrontation.
Neil v. Biggers, 409 U.S. 188, 199, 93 S. Ct. 375, 382, 34 L. Ed. 2d 401 (1972). A careful reading of this statement will show that several of the criteria listed by the Court are based on assumptions that are flatly contradicted by well-respected and essentially unchallenged empirical studies. Although the law will always lag behind the sciences to some degree because of the need for solid scientific consensus before the law incorporates its teachings, we conclude that in the area of eyewitness identification, the time has come for a more empirically sound approach.
Some have proposed a radical approach to the problems of eyewitness identification. In Great Britain, for example, after two especially egregious instances of criminal convictions based upon mistaken identity, a committee chaired by Lord Devlin was appointed to study the matter. See G. Williams, Evidence of Identification: The Devlin Report, Crim.L.R. 407 (1976). After extensive work, the Devlin Committee recommended that trial judges should be required to instruct juries that an uncorroborated visual identification alone could not be a sufficient basis for convicting a defendant of a crime unless special circumstances were present. Williams, supra, at 412-13. These special circumstances suggested by the Devlin Report were, in essence, factors which would bolster the reliability of the visual identification, rather than provide independent corroborating evidence. Examples of such factors given in the Report include: (1) the witness's familiarity with the identified suspect; (2) the defendant's failure to deny he was a member of a small group, one member of which committed the crime; and (3) the defendant's *492 failure to counter the viewer's story. Id.
Perhaps because the Devlin Report's recommendations departed so substantially from the traditional heavy reliance of the police and the judiciary on eyewitness identification, they do not appear to have been adopted. They have, nonetheless, received some recognition here. See United States v. Butler, 636 F.2d 727, 735 (D.C.Cir.1980) (Bazelon, J., dissenting). While we, too, do not choose to modify the Utah common law as radically as the Devlin Report suggested, there is no question that credible evidence supports that approach. Such a bold departure will have to await further empirical evidence that less radical alternatives do not ameliorate the problem. However, we do consider ourselves compelled by the overwhelming weight of the empirical research to take steps to alleviate the difficulties inherent in any use of eyewitness identification testimony.
We are convinced that, at a minimum, additional judicial guidance to the jury in evaluating such testimony is warranted. We therefore today abandon our discretionary approach to cautionary jury instructions and direct that in cases tried from this date forward, trial courts shall give such an instruction whenever eyewitness identification is a central issue in a case and such an instruction is requested by the defense. Given the great weight jurors are likely to give eyewitness testimony, and the deep and generally unperceived flaws in it, to convict a defendant on such evidence without advising the jury of the factors that should be considered in evaluating it could well deny the defendant due process of law under article I, section 7 of the Utah Constitution.
It is true that some state courts have refused to give cautionary instructions on the ground that they constitute improper judicial comment on the evidence or suggest the weight that should be accorded certain testimony. See Annot., 23 A.L.R.4th 1089, 1110-11 (1983). We see little merit to this argument. A well-constructed cautionary instruction will not permit a judge to opine as to the credibility of the testimony. It will only pinpoint identification as a central issue and highlight the factors that bear on the reliability of that identification. This will do no more than apprise the jury of the inherent limitations of eyewitness identification. Such an instruction both "respect[s] the jury's function and strike[s] a reasonable balance between protecting the innocent and convicting the guilty." Sanders, supra, at 204. The approach we adopt today offers a defendant some protection from false conviction, while ensuring the efficacy of the jury system by providing jurors with the knowledge necessary for sound decision making.
Having decided that cautionary instructions should be given rather routinely, the question is whether this court should adopt one specific instruction as the only acceptable formulation, or whether we should grant trial court and counsel some latitude in formulating instructions. We have decided to opt for the latter approach, at least until experience shows that conferring such discretion on trial courts does not produce adequate instructions. To guide trial courts, we note that a proper instruction should sensitize the jury to the factors that empirical research have shown to be of importance in determining the accuracy of eyewitness identifications, especially those that laypersons most likely would not appreciate. These factors should include not only the externals, like the quality of the *493 lighting and the time available for observation, but also the internal or subjective factors, such as the likelihood of accurate perception, storage and retrieval of the information by a witness. For example, an instruction should address the following commonly accepted areas of concern: (1) the opportunity of the witness to view the actor during the event; (2) the witness's degree of attention to the actor at the time of the event; (3) the witness's capacity to observe the event, including his or her physical and mental acuity; (4) whether the witness's identification was made spontaneously and remained consistent thereafter, or whether it was the product of suggestion; and (5) the nature of the event being observed and the likelihood that the witness would perceive, remember and relate it correctly. This last area includes such factors as whether the event was an ordinary one in the mind of the observer during the time it was observed, and whether the race of the actor was the same as the observer's.
We have found numerous examples of cautionary instructions. Many are so short and superficial as to be of little utility in accomplishing our objectives. One that seems to satisfy most of the legitimate concerns about eyewitness identification is that set forth in Telfaire and cited with approval by Justice Stewart in his dissent in State v. Malmrose, 649 P.2d 56, 63-64 (Utah 1982). However, a critical examination *494 of the Telfaire instruction, proposed in 1972, reveals that it does not cover several factors which more recent empirical research have shown to be important, such as the limitations in our ability to perceive, store and retrieve information. It also incorporates some of the fallacious assumptions expressed by the Supreme Court in Neil v. Biggers. (See pp. 491-492, supra.) Be that as it may, the Telfaire instruction would seem to suffice as an adequate cautionary instruction under most circumstances. The state of the art in jury instructions is not so advanced that the failings of Telfaire can be said to disqualify it from use at this point.
A more complete instruction that remedies many of the problems of the Telfaire instruction has recently been proposed. It is also more understandable. However, although it appears to be a substantial improvement over Telfaire, it is even longer than the one it would replace. If used, it *495 would certainly satisfy our expressed concerns about the need for cautionary instructions.
Acceptable and perhaps shorter instructions other than the two discussed certainly could be constructed, especially for specific cases that raise only some of the concerns about accurate eyewitness identification. We trust that trial counsel and judges will be able to produce appropriate instructions that satisfy the concerns expressed here today. Perhaps, over time, the lessons of experience will demonstrate the inherent superiority of one type or form of cautionary instruction. But in the absence of such experience, we decline to dictate precisely what that instruction must say.
The final issuewhether the trial court erred in denying Long's tardy motion to severneed not be reached. However, in an effort to guide the trial court, which may well be faced with the same issue on remand, we make the following observation. On the second day of trial Long moved to sever the charge of possession of a dangerous weapon by a restricted person from the aggravated assault and attempted murder charges. He claims that denial of the motion was prejudicial because it put evidence of his prior conviction before the same jury that heard evidence on the aggravated assault and attempted murder charges.
Initially, we note that Rule 9 of the Utah Rules of Criminal Procedure justified the trial court's denial of the motion. "A defendant's right to severance of offenses... is waived if the motion is not made at least five days before trial." U.C.A., 1953, § 77-35-9 (Repl.Vol. 8C, 1982). Assuming that a timely motion is made on remand, the issue of severance will be presented in essentially the same posture as it was in State v. Saunders, 699 P.2d 738 (Utah 1985). There, we held that refusal to sever the possession of a dangerous weapon charge from the remaining charges was an abuse of discretion because of the unwarranted prejudice inherent in informing the jury that a defendant is a convicted felon. Here, too, there seems to be no compelling reason to present the evidence of prior offenses to the jury that is trying the assault charges.
The conviction is reversed and the case remanded for a new trial.
STEWART and DURHAM, JJ., concur.
HALL, Chief Justice (concurring and dissenting):
I join the Court in remanding for a new trial on the charge of possession of a dangerous weapon. However, I do not agree that because the trial court refused to give a cautionary instruction on the reliability of eyewitness testimony the defendant is also entitled to a new trial on the charge of aggravated assault.
*496 In State v. Green, the Court stated the law in Utah to be:In this jurisdiction the trial judge is not permitted to comment on the evidence.... It is the sole and exclusive province of the jury to determine the facts in all criminal cases, whether the evidence offered by the state is weak or strong, is in conflict or is not controverted.... These principles of law are so fundamental in our system of criminal procedure that we deem it unnecessary to cite cases and authorities in support thereof.
In the cases that have followed, the Court has adhered to the principle of law espoused in Green and in Utah R.Civ.P. 51 and Utah R.Crim.P. 19 that a trial judge is not permitted to comment on the quality or credibility of the evidence and may not indicate that the evidence is either weak or convincing. However, as was observed in State v. Sanders, the principle of law that precludes the trial court from commenting on the evidence does not preclude the court from including in its instructions general statements concerning certain types of evidence. The court is only enjoined from commenting on the quality or credibility of the evidence in such a way as to indicate that it favors the claims or position of either party. The enjoinder is necessary to prevent any intrusion upon the prerogatives of the jury to judge the credibility of the evidence and to determine the facts. Consistent with these concepts, the Court has concluded that the giving of special instructions on eyewitness testimony should be left to the sound discretion of the trial court. On appeal, this Court scrutinizes the instructions given to the jury to determine if, when viewed as a whole, the instructions adequately advised the jury on the law pertaining to the case. Specifically, the Court has noted that three general instructions cover the same substance as a cautionary instruction: (1) that the State has the burden to prove guilt beyond a reasonable doubt; (2) that the jury is the exclusive judge of the credibility of witnesses; and (3) that to find guilt the jury must find the defendant committed all of the elements of the offense.
I am not persuaded by the facts of this case that the trial court abused its discretion in declining to give a special cautionary instruction in addition to the general instructions on burden of proof, reasonable doubt, and witness credibility that were given.
The credibility of Jacob Rocha was tested on direct and on cross-examination, and defense counsel thoroughly explored the issue of his credibility in final argument to the jury.
The defendant's theory of the case was that he was merely present at the scene and that he was incorrectly identified as the gunman. However, the defendant rested at the close of the State's case in chief without calling any witnesses. Thus, the uncontroverted evidence placed the defendant at the doorway and not in some other area where a stray bullet might strike him. Furthermore, the evidence was that Rocha was able to observe the defendant standing alone in the doorway for some thirty seconds attempting to reload the shotgun and *497 that Rocha fired at the defendant and hit him.
The defendant makes no claim of insufficiency of the evidence, and indeed there is none. In fact, even if Rocha had been unable to identify the defendant as his assailant, the remaining undisputed evidence made out a prima facie case upon which the jury could deliberate guilt or innocence: the defendant was on the scene; the shotgun blast came from a lone person who appeared in the doorway; that person was fired upon by Rocha; and the defendant in fact was struck by the bullet fired by Rocha.
I would affirm the conviction of aggravated assault.
HOWE, Justice (concurring and dissenting):
I concur in remanding for a new trial on the charge of possession of a dangerous weapon.
I would affirm the conviction of aggravated assault because it does not rest entirely on the identification made by Jacob Rocha. It is also supported by strong circumstantial evidence. Jacob testified that a man (whom he identified as the defendant) fired a sawed-off shotgun through the door as it was forced open. Jacob was then four to five feet away from the door. He "flew back" twelve feet and from that location looked out the doorway and saw the man with the shotgun apparently reloading it. Jacob picked up his handgun and fired it at the man, who was eighteen to twenty feet away. At no time did he see anyone else through the doorway other than the man who had shot him. According to the testimony of Robin Lee, when the defendant and Hobsun returned to their car, it was Long and not Hobsun who was wounded. The arresting officers also saw that Long had been shot. As pointed out by the Chief Justice, these facts fully support the conclusion that it had to be the defendant and not someone else who was Jacob's assailant, quite independently of Jacob's eyewitness identification of the defendant.
I would continue to adhere to our present rule that the giving of a cautionary instruction on eyewitness identification is discretionary with the trial court. State v. Malmrose, 649 P.2d 56 (Utah 1982). I believe that this rule has worked well. That we have not reversed a trial court for refusing a defendant's request, as pointed out by the majority opinion, fails to prove otherwise. We have usually, if not always, found evidence corroborating the eyewitness identification. When a cautionary instruction is given, I would favor a brief instruction, couched in general cautionary terms. The instruction suggested in the majority opinion (and it is only suggested) is in my opinion much too long, repetitious, and leads the jury through a detailed checklist. By its sheer length and detail, it overshadows the other instructions and is heavily slanted in favor of the defendant. It is not clear whether the jury can rely on the identification if they have reasonable doubt about any of the twenty considerations which are listed.
The suggested instruction is also objectionable to me because it incorporates conclusions from the articles relied upon in the majority opinion which I am not prepared to embrace. For example, in factor (3), the suggested instruction states as a fact that "[e]ven if the witness had adequate opportunity and capacity to observe the criminal actor, the witness may not have done so unless he was aware that a crime was being committed." While I agree as a general proposition that an observer would give more attention to a scene where he thought a crime was being committed than he might ordinarily otherwise do, it by no means follows that because the observer does not know that a crime is being committed, his identification and perception will be less accurate or faulty. Some people are keenly observant, and their identification is very reliable if they have an adequate opportunity to observe. This would be true even though the setting was business, social, or casual. The fact that a crime was not being committed would make no difference.NOTES
 At the time of Long's trial, Rules 63(17), 64 and 68 of the Utah Rules of Evidence were still in effect. These rules have since been superseded by scattered provisions in the new Utah Rules of Evidence, which became effective September 1, 1983.
 Mollner's testimony that he supervised Long on parole is proof of that fact only. It cannot be inferred from that fact alone that the conviction was for a felony. Because the Board of Pardons has authority to parole those committed to county jails as well as to state prisons, it is possible that Long might have been on parole as the result of a misdemeanor conviction. See U.C.A., 1953, § 77-27-11 (Repl.Vol. 8C, 1982, and Supp.1985).
 It might be argued that the police officer's testimony that defendant was suffering from a bullet wound when his car was stopped corroborates the identification of defendant as the victim's assailant. However, two men participated in the incident, and nothing in the testimony suggests that the man wielding the shotgun, rather than the other man, was the one hit by Rocha's fire. Thus, the testimony going to defendant's wound can be fairly construed to indicate only that defendant was present at the scene of the crime.
 For a critique of this dominant line of research and an alternate approach to applied eyewitness testimony research, see G. Wells, Applied Eyewitness-Testimony Research: System Variables and Estimator Variables, 36 J. Personality and Soc. Psych. 1546 (1978).
 A cautionary instruction plainly is not a panacea. See D. Starkman, The Use of Eyewitness Identification Evidence in Criminal Trials, 21 Crim.L.Qtrly. 361, 375-77 (1978-79). The only study evaluating the Telfaire instruction that has come to our attention indicated that the model Telfaire instruction had little effect on jurors' sensitivity to the factors affecting the reliability of eyewitness identification. When the instruction was coupled with a weak identification, however, it did produce less juror reliance on the eyewitness identification. Similarly, when it was given in conjunction with strong eyewitness testimony, it bolstered jurors' beliefs in the correctness of the identification. See Sanders, supra, at 212 n. 178 and at 217 n. 221. Full evaluation of the efficacy of cautionary instructions must await further experience.
 See, e.g., United States v. Telfaire, 469 F.2d 552 (D.C.Cir.1972); State v. Green, 86 N.J. 281, 430 A.2d 914 (1981); Smith v. United States, 343 A.2d 40 (D.C.App.1975); State v. Calia, 15 Or. App. 110, 514 P.2d 1354 (1973), cert. denied, 417 U.S. 917, 94 S. Ct. 2621, 41 L. Ed. 2d 222 (1974); People v. Guzman, 121 Cal. Rptr. 69, 47 Cal. App. 3d 380 (1975); State v. Warren, 230 Kan. 385, 635 P.2d 1236 (1981), and Annot., 23 A.L.R.4th 1070 (1983); see also R. Sanders, Helping the Jury Evaluate Eyewitness Testimony: The Need for Additional Safeguards, 12 Am.J.Crim.L. 189, 222-24 (1984).
 The instruction suggested by the Telfaire court reads as follows:
One of the most important issues in this case is the identification of the defendant as the perpetrator of the crime. The Government has the burden of [proving] identity, beyond a reasonable doubt. It is not essential that the witness himself be free from doubt as to the correctness of his statement. However, you, the jury, must be satisfied beyond a reasonable doubt of the accuracy of the identification of the defendant before you may convict him. If you are not convinced beyond a reasonable doubt that the defendant was the person who committed the crime, you must find the defendant not guilty.
Identification testimony is an expression of belief or impression by the witness. Its value depends on the opportunity the witness had to observe the offender at the time of the offense and to make a reliable identification later.
In appraising the identification testimony of a witness, you should consider the following:
(1) Are you convinced that the witness had the capacity and an adequate opportunity to observe the offender?
Whether the witness had an adequate opportunity to observe the offender at the time of the offense will be affected by such matters as how long or short a time was available, how far or close the witness was, how good were lighting conditions, whether the witness had had occasion to see or know the person in the past.
[In general, a witness bases any identification he makes on his perception through the use of his senses. Usually the witness identifies an offender by the sense of sightbut this is not necessarily so, and he may use other senses.]
(2) Are you satisfied that the identification made by the witness subsequent to the offense was the product of his own recollection? You may take into account both the strength of the identification, and the circumstances under which the identification was made.
If the identification by the witness may have been influenced by the circumstances under which the defendant was presented to him for identification, you should scrutinize the identification with great care. You may also consider the length of time that lapsed between the occurrence of the crime and the next opportunity of the witness to see defendant, as a factor bearing on the reliability of the identification.
[You may also take into account that an identification made by picking the defendant out of a group of similar individuals is generally more reliable than one which results from the presentation of the defendant alone to the witness.]
[(3) You [may] take into account any occasions in which the witness failed to make an identification of defendant, or made an identification that was inconsistent with his identification at trial.]
(4) Finally, you must consider the credibility of each identification witness in the same way as any other witness, consider whether he is truthful, and consider whether he had the capacity and opportunity to make a reliable observation on the matter covered in his testimony.
I again emphasize that the burden of proof on the prosecutor extends to every element of the crime charged, and this specifically includes the burden of proving beyond a reasonable doubt the identity of the defendant as the perpetrator of the crime with which he stands charged. If after examining the testimony, you have a reasonable doubt as to the accuracy of the identification, you must find the defendant not guilty.
469 F.2d 552, 558-59 (D.C.Cir.1972). (Bracketed portions are optional, depending upon specific facts and circumstances of the case.)
 The proposed instruction reads:
One of the most important questions [The only important question] in this case is the identification of the defendant as the person who committed the crime. The prosecution has the burden of proving beyond a reasonable doubt, not only that the crime was committed, but also that the defendant was the person who committed the crime. If, after considering the evidence you have heard from both sides, you are not convinced beyond a reasonable doubt that the defendant is the person who committed the crime, you must find the defendant not guilty.
The identification testimony that you have heard was an expression of belief or impression by the witness. To find the defendant not guilty, you need not believe that the identification witness was insincere, but merely that [the witness] was mistaken in his [her] belief or impression.
Many factors affect the accuracy of identification. In considering whether the prosecution has proved beyond a reasonable doubt that the defendant is the person who committed the crime, you should consider the following:
1) Did the witness have an adequate opportunity to observe the criminal actor?
In answering this question, you should consider:
a) the length of time the witnesses observed the actor;
b) the distance between the witness and the actor;
c) the extent to which the actor's features were visible and undisguised;
d) the light or lack of light at the place and time of observation;
e) the presence [or] absence of distracting noises or activity during the observation;
f) any other circumstances affecting the witness' opportunity to observe the person committing the crime.
2) Did the witness have the capacity to observe the person committing the crime?
In answering this question, you should consider whether the witness' capacity was impaired by:
a) stress or fright at the time of observation;
b) personal motivations, biases or prejudices;
c) uncorrected visual defects;
d) fatigue or injury;
e) drugs or alcohol.
[You should also consider whether the witness is of a different race than the criminal actor. Identification by a person of a different race may be less reliable than identification by a person of the same race.]
[3) Was the witness sufficiently attentive to the criminal actor at the time of the crime?
In answering this question, you should consider whether the witness knew that a crime was taking place during the time he [she] observed the actor. Even if the witness had adequate opportunity and capacity to observe the criminal actor, he [she] may not have done so unless he [she] was aware that a crime was being committed.]
4) Was the witness' identification of the defendant completely the product of his [her] own memory?
In answering this question, you should consider:
a) the length of time that passed between the witness' original observation and his [her] identification of the defendant;
b) the witness' [mental] capacity and state of mind at the time of the identification;
c) the witness' exposure to opinions, descriptions or identifications given by other witnesses, to photographs or newspaper accounts, or to any other information or influence that may have affected the independence of his [her] identification;
[d) any instances when the witness, or any eyewitness to the crime, failed to identify the defendant;]
[e) any instances when the witness, or any eyewitness to the crime, gave a description of the actor that is inconsistent with the defendant's appearance;]
f) the circumstances under which the defendant was presented to the witness for identification.
[You may take into account that an identification made by picking the defendant from a group of similar individuals is generally more reliable than an identification made from the defendant being presented alone to the witness.]
[You may also take into account that identifications made from seeing the person are generally more reliable than identifications made from a photograph.]
I again emphasize that the burden of proving that the defendant is the person who committed the crime is on the prosecution. If, after considering the evidence you have heard from the prosecution and from the defense, and after evaluating the eyewitness testimony in light of the considerations listed above, you have a reasonable doubt about whether the defendant is the person who committed the crime, you must find him not guilty.
R. Sanders, Helping the Jury Evaluate Eyewitness Testimony: The Need for Additional Safeguards, 12 Am.J.Crim.L. 189, 222-24 (1984). (Bracketed portions of the instruction should be used when appropriate to the facts of the case.)
 78 Utah 580, 589-91, 6 P.2d 177, 181 (1931). See also Utah R.Civ.P. 51; Utah R.Crim.P. 19 (U.C.A., 1953, § 77-35-19 (Repl.Vol. 8C, 1982 ed.)) (providing that the court shall not comment on the evidence in the case, and if the court refers to any of the evidence, it shall instruct the jury that they are the exclusive judges of all questions of fact).
 27 Utah 2d at 362, 496 P.2d at 275.
 State v. Booker, 709 P.2d 342, 346 (Utah 1985); State v. Tucker, 709 P.2d 313, 316 (Utah 1985); State v. Bingham, 684 P.2d 43, 45 (Utah 1984); State v. Reedy, 681 P.2d 1251, 1252-53 (Utah 1984); State v. Newton, 681 P.2d 833, 834 (Utah 1984). See also State v. Schaffer, 638 P.2d 1185, 1187 (Utah 1981); State v. McCumber, 622 P.2d 353, 359 (Utah 1980).