Court File and Parties
Court File No.: Toronto
Ontario Court of Justice
Between:
Her Majesty the Queen
Mr. T. Dimuzio, for the Crown
— And —
Yohannes Berhe
Mr. M. Kerbel, for the Accused
Heard: June 18, 19, 2013
Decision
NAKATSURU J.:
Overview of the Evidence
[1] The accused, Mr. Yohannes Berhe, is charged that on July 7, 2012, he committed the offences of indecent act in public and mischief to private property by masturbating on a Toronto subway car. Identity is a key issue at the trial.
[2] During the course of the trial, a voir dire regarding the admissibility of lay recognition opinion evidence was conducted. The Crown seeks to lead the opinions of two police officers to the effect they recognized the suspect in a photograph taken by the complainant while on the subway train as the accused before the court.
[3] I ruled that the evidence was not admissible. The following are the reasons explaining that decision.
Overview of the Evidence
[4] In the late evening hours of July 7, 2012, a young woman and her friends were taking the subway home from a music festival in the Beaches area of Toronto. While heading westbound on the subway, she observed a man standing with a black briefcase stuffed with newspapers. He was pretending to read a newspaper. As the subway train travelled along the westbound line, the complainant observed the man furtively masturbating while at the same time using the newspaper as a shield. He stopped when he saw her looking in his direction. The complainant described the man as brown skinned, about 6 feet tall with a black beard, perhaps in his forties, wearing black dress pants, shoes, black and blue shirt not tucked in and a black baseball cap on his head. Before she got off at her station, she discretely took a photograph of the man. The next day, she emailed this photograph to the Toronto Transit Commission (the "TTC") complaining about the incident.
[5] This photograph taken by the complainant was entered as exhibit #1 at the trial. Depicted in this photograph is a man leaning with his back against the doors of the subway car with a newspaper covering the front of his body. There appears to be a briefcase on the floor beside him. The man's face is visible but is partially obscured by a metal pole and a partition by the subway car door. The man has a black beard and his hairline appears covered by a cap. The subway car is well lit and the image, while not sharp, is clear enough to make out some facial features. The photograph is taken from a few short meters away.
[6] When asked at trial, the complainant could not identify anyone in court as the person she observed masturbating on the subway train on July 7, 2012.
[7] On July 13, 2012, P.C. Blonde and P.C. DeOliveira received a call to attend Ossington subway station regarding an incident. When they arrived, they were told that a TTC employee was following the suspect. They jogged to catch up to the men who were now walking west bound on Bloor Street West. The TTC employee spotted the police officers and pointed to the accused, Mr. Berhe, as the man they should investigate. The accused matched the description of the suspect put over the police radio: he had dark skin, a long beard, wore a baseball cap and a blue suit and carried a black briefcase and newspapers. He was detained and arrested. Both officers transported the accused to 14 Division where he was paraded and lodged in an interview room. P.C. DeOliveira had further dealings with Mr. Berhe as he needed to use the washroom and also had some complaints about chest pain. Both police officers were able to identify the accused before the court as the individual they arrested on July 13, 2012.
[8] Neither police officer had any prior dealings with the accused before July 13th. Both officers were asked whether they recognized the man in the photograph marked as exhibit #1. They both testified that this man looked like the same man they arrested on July 13th, Mr. Yohannes Berhe.
[9] A photograph taken of the accused on July 13th by the police when he was arrested, a "mug shot", was admitted into evidence as exhibit #3. It is a black and white photograph. It is a full frontal photograph of the face of the accused. It is clear and taken from close range.
The Law on the Admissibility of Recognition Opinion Evidence
[10] It is a well-established principle that a lay person can express an opinion about the identity of a person. Further, it is settled law that a lay person can give an opinion to the effect that he or she recognizes the image of a person seen in a relevant photograph or video based upon a prior connection or an acquaintance with that person. However, before such an opinion is receivable, the trial judge is tasked with a gatekeeping function to ensure only properly admissible evidence is heard by the trier of fact.
[11] It is important to keep distinct the threshold test for admissibility and the ultimate weight to be attributed to the opinion evidence once admitted. Similar factors may inform both but the two are legally distinct. Once satisfied by the evidence on the voir dire that the opinion is admissible, the ultimate weight to be given to the opinion will be up to the trier of fact.
[12] A voir dire regarding the admissibility of the opinion is necessary: see R. v. Leaney, [1989] 2 S.C.R. 393; R. v. Berhe (2012), 2012 ONCA 716, 292 C.C.C. (3d) 456 (Ont. C.A.). At this voir dire, the circumstances of the prior acquaintance can be fully explored without concern that prejudicial evidence may be divulged before the trier of fact. For instance, it is not uncommon that police officers who have had prior dealings with the accused are called by the Crown to give lay recognition opinion evidence. The details of those dealings may raise prior discreditable conduct or criminal offences by the accused that if presented before the jury could lead to a mistrial application. If ruled admissible, the trial judge can place limits on the questioning regarding the admission of the circumstances under which the witnesses came into contact with the accused in order to minimize the prejudice: see R. v. Brown (2006), 215 C.C.C. (3d) 330 (Ont. C.A.) at para. 36. In addition, if the frailties regarding the testimony of these witnesses cannot be understood without the prejudicial evidence being examined, the trial judge on the voir dire can also decide whether this evidence should be excluded on the basis that its prejudicial effect outweighs its probative value: see R. v. Brown, [1999] O.J. No. 4865 (S.C.) affirmed, supra, at para. 38.
[13] The substantive content of the test to be applied on the voir dire was stated by Blair J.A. in Berhe, supra, at paras. 20-21:
At the voir dire, the judge may or may not conclude that the potential witness is sufficiently familiar with the person whose identity is in question to be in a better position than the trier of fact to assist in making the identification…Both of these benchmarks are well established. The "prior acquaintance" branch of the Leaney/Brown test enables the trial judge, on a voir dire, to sort out whether the potential witness is sufficiently familiar with the person sought to be identified to have "some basis" for the opinion – or, "an articulated basis" as some have said – and the "better position" branch ensures that the evidence will only be admitted if it is helpful to the trier of fact because the potential witness has some advantage that can shed light on the evidence in question.
[14] As noted, both branches of the test must be met. At the same time, the two aspects of the test are closely interrelated. In assessing the evidence heard on the voir dire, I would suggest that the following analytical framework may prove to be of some use.
[15] In my opinion, once the witness testifies to some form of prior acquaintance or connection with the accused, this part of the test is established. This is a bright line that is easy to understand and apply. If the witness has no prior personal familiarity with the accused, any recognition opinion evidence is superfluous and a waste of time: see Leaney, supra, at para. 33. However, any prior acquaintance, whether particularly meaningful or not, should suffice. The only other additional requirement is that the acquaintance or familiarity be a personal one, not something gained through third hand sources such as the media or through investigation. Other than that, this threshold should not have any further quantitative or qualitative burden.
[16] Some authorities have held that the prior acquaintance requirement must achieve a certain sufficiency based upon the consideration of indicia such as the length of the prior relationship between the witness and the accused, the circumstances of the prior relationship between the witness and the accused, and the recency of the contact between the witness and the accused prior to the event where the witness recognizes the accused: see R. v. P.T.C., 2000 BCSC 342, [2000] B.C.J. No. 446 (S.C.) at paras. 66-67; R. v. Anderson, [2005] B.C.J. No. 305 (S.C.) at para. 25; R. v. Muncey, [2013] B.C.J. No. 793 (S.C.). At the same time, these authorities find that the second element of the test, "helpfulness", to be a low threshold.
[17] Respectfully, I am of the view that such an analysis is inverted. The focus should be on the question of whether the proffered opinion evidence will be of assistance to the trier in making the identification. This question must be answered contextually and on a case-by-case basis. In the course of determining that question, the afore-mentioned factors may be useful.
[18] I have come to this view because of the inherent difficulties in defining how much familiarity is enough familiarity for the purpose of the voir dire. Conceptually, prior acquaintance has import only when it is measured against some other factor. Therefore, investing the test of prior acquaintance with ill-defined standards such as "sufficient", "significant", or "meaningful" is unhelpful. A witness who has a fleeting encounter with the accused may not be able to give admissible recognition opinion evidence but this is not because the acquaintance was "fleeting". Rather, it is because an opinion about identity gained in such a circumstance is unlikely to assist the trier of fact.
[19] An easily applicable test of prior acquaintance of any kind can screen out recognition opinion evidence of the type that was found to be inadmissible in Leaney, supra. [1] In this manner, this facet of the gatekeeping function can be readily and swiftly accomplished. Under this branch of the test, it will be incumbent upon the party who seeks to introduce such evidence to have the witness testify about the circumstances of the prior acquaintance. To this extent, this branch of the test serves a limited but valuable purpose.
[20] In my opinion, this analysis is consistent with the judgment in Berhe, supra. Blair J.A. in Berhe, supra, clearly did not envision a high threshold when he referred to the prior acquaintance branch of the test as requiring "some basis" or "an articulated basis". The language referred to by Blair J.A. comes from the cases of R. v. Cuming (2001), 158 C.C.C. (3d) 433 (Ont. C.A.) at para. 21 and R. v. Brown, supra, at paras. 38-39. In Cuming four still photographs from the store security video camera capturing the robbery in progress were enhanced. The officer-in-charge of the investigation identified the appellant from these enhanced stills. However, there was no evidence as to how this officer had a better opportunity to make a visual comparison between the accused and the stills than the jury did. There was no evidence that he knew the appellant other than through his investigation of the robbery and no particulars were given regarding the extent of his dealings with the appellant. The Court of Appeal concluded that unless there was "some basis" given for his recognition opinion evidence, it had no weight. In my view, if this officer had testified to any prior personal dealings with the appellant, this would have met the standard. He would then have had some articulated basis on which the judge on the voir dire could assess admissibility and the trier of fact could assess the ultimate weight of the officer's opinion.
[21] Furthermore, in Berhe, supra, Blair J.A. rejected the adoption of the dissent of Harradence J.A. in R. v. Leaney (1987), 1987 ABCA 206, 81 A.R. 247 (C.A.) where Harradence J.A. added another layer to the test requiring the recognition evidence witness to have sufficient prior familiarity with the accused's unique features to enable the witness to describe the accused's idiosyncrasies as portrayed on the video. Blair J.A. held that the present test was sufficiently flexible to permit the trial judge to perform the gatekeeping role on the voir dire. Again, this supports my view that the test of prior acquaintance should not be given undue emphasis.
[22] That this being said, pivotally, it is not any and all prior acquaintance with the accused that will satisfy the second prong of the test. The witness must as a result of the prior acquaintance have an advantage or be in a better position to identify the person in the photo or video than the trier of fact. This is the key measure by which admissibility must be gauged.
[23] The starting point is the fact that the trier of fact will be able to compare the accused before the court and the image of the suspect on the photo or video to determine whether it is the accused who committed the offence. The trier of fact may determine identification solely on the basis of the photo or video: see R. v. Nikolovski, [1996] 3 S.C.R. 1197. The trier of fact undertakes this task by simple comparison of the photo or video with the accused before the court. Thus, in assessing whether the second branch of the test has been met, the witness in question must be able to offer the trier of fact some assistance beyond what the trier can do through such a comparison.
[24] It is in this regard that the factors mentioned in the authorities become relevant. The length and nature of the relationship between the witness and the accused may be important. It can be presumed in most cases that a close and long relationship would give the witness a greater ability to recognize a photo of the accused than a shorter and more distant one. A parent of the accused is more likely to have an advantage in recognizing the accused than a stranger to the accused. It is noteworthy to point out that the triers of fact are essentially just that, strangers to the accused. Someone who has known the accused closely for a long time is likely to be more familiar with traits such as subtle physical characteristics, idiosyncratic movement, posture, and facial expression than the trier of fact. On the other hand, depending on the case, a close and long relationship may not on its own put the witness in an advantageous position as compared to that of the trier of fact depending on other relevant factors. If the relationship was in the distant past and the appearance of the accused had changed substantially by the relevant time when the offence was committed, then this witness's opinion may not only be unhelpful but it could also be misleading. The trier may be overly persuaded by nature of the relationship without giving their own visual abilities in making a comparison sufficient credit.
[25] Recency in time to the offence can be an important factor. This is so especially when the accused's appearance from the time of the offence to the trial has indeed changed. Many personal characteristics such as hair style and facial hair are easy to alter. Other personal characteristics, while more difficult, such as weight and build, can also change if the trial is delayed and these changes can dramatically alter a person's features. In such circumstances, a witness who has familiarity with the accused closer in time to the offence would be in a superior position to the trier of fact in his or her ability to recognize and identify the suspect in the photo or video. This may be so even when the length and nature of the witness's relationship with the accused is not particularly significant. Witnesses such as police officers may find themselves in this position. This factor has uniformly been accepted as important in deciding the issue of admissibility: see Brown, supra, at para. 39; R. v. John, 2010 ONSC 6085, [2010] O.J. No. 4738 (S.C.).
[26] The opportunity to observe the accused in the prior interaction may also be a significant consideration when the relationship with the accused is not close. As previously mentioned, a witness who has had a "fleeting" or "momentary" acquaintance with the accused on a prior occasion may be of no real assistance to the trier of fact who will have a greater opportunity to make a comparison in court. Furthermore, if the prior interaction was not one in which the witness could be expected to have made any notable observations of the accused or to recall his appearance, then the witness may not be in any greater advantaged position than the trier of fact. It is easy to think of numerous personal, occupational, and commercial interactions, potential witnesses could have with an accused as he or she goes through daily life. Some of those potential witnesses, such as a shopkeeper or a server at a coffee shop that an accused frequents regularly, may even have a long-term relationship with the accused. However, given how limited their interactions are, though it may even be on a daily basis, it cannot simply be assumed that this prior relationship would put the potential witness in a better position to identify a suspect in a photo or video.
[27] On the other hand, even a single "fleeting" interaction may put the witness in a better position than the trier of fact if there are other counterbalancing factors in play. For example, if the potential witness had a brief encounter with an accused that was a memorable one and the accused had a relatively unique characteristic such as in his gait, this may suffice. If the suspect in the video is seen with a similar gait, this witness despite the apparent limitations in his or her prior observations could be in a better position to make an identification than the trier of fact. The latter's ability to make a comparison would be limited to a visual observation of the accused who would be in a static position in the artificial and formal courtroom setting.
[28] Finally, the form of the visually captured out-of-court depiction of the suspect may be an important consideration. A witness with a prior acquaintance with the accused may be in a better position than the trier of fact to recognize someone in a video than a still photograph. Such a witness may have had the opportunity to see the accused move and interact with others that a trier of fact cannot in the courtroom environment. If the depiction of the suspect in the video makes this opportunity meaningful and important, then the witness's opinion can offer help to the trier. Conversely, if there is only a still photograph of the suspect, then the witness's advantageous position may not have much tangible value. The trier of fact could as readily make an identification based upon a still photograph and the accused before the court as the proffered witness. In such situations, there is a real danger that the witness is effectively doing no more than the trier of fact. In other words, rather than using the witness's prior knowledge and familiarity with the accused in offering the recognition opinion, the witness is merely making a visual comparison between the photograph and the accused before the court. This danger is compounded by the fact that this mental process that is being used may not even be fully appreciated by the witness.
[29] Conversely, although a witness is largely making a comparison between the photograph and the appearance of the accused before the court, if the circumstances of the prior acquaintance put the witness in close quarters with the accused, this may, in the right case, make a difference. If the witness had an opportunity and did make close observations of certain facial features, for example, that a trier of fact cannot, given the distance between the position of the accused seated in the courtroom and the trier, this may place the witness in a superior position even though the witness is mainly engaged in a comparative exercise.
[30] In the final analysis, the question of admissibility will depend on the facts of the case. As it is often the situation, it is a matter of balance; weakness in one set of factors may be offset by strength in others. The key question that must be answered is whether the proposed recognition opinion will be helpful to the trier of fact, not in general or because of some presumed advantage due a prior relationship, but on the particular realities of the individual case.
[31] Now I appreciate that much of the inquiry comes close to the issue of the ultimate reliability or the weight of the recognition opinion. This issue, of course, must be left to the trier of fact. However, the gatekeeping function cannot be simply fulfilled by a superficial assessment of whether the witness has a better position. I say this because of the policy concerns behind permitting such evidence.
[32] These concerns were explored in the seminal case of R. v. Graat, [1982] 2 S.C.R. 819. In Graat, Dickson J. (as he then was) held that lay opinion regarding intoxication and the impairment of a person's ability to drive due to alcohol was admissible. In returning to first principles, Dickson J. found that lay persons may offer their opinions based upon personal observations when that opinion is a compendious mode of speaking about facts that the witness could not as accurately and adequately narrate separately. In turning to those first principles of whether the evidence should be admissible, Dickson J. first asked himself whether the evidence sought to be admitted was relevant. After relevance was established, the question which had to be answered was whether, though probative, the evidence should be excluded by a clear ground of policy or law. In assessing this question, Dickson J. held that the probative value of the lay opinion evidence was not outweighed by such policy considerations as the danger of confusing the issues or misleading the jury. It did not unfairly surprise a party who had not had reasonable grounds to anticipate such evidence being offered and the adducing of the evidence did not necessitate undue consumption of time: see Graat, supra, at p. 836.
[33] In Graat the police officers who testified that the appellant's ability to drive was impaired by alcohol had personally observed him on the night in question. They were in an understandably superior position to the trier of fact since the trier was not present on that night to make those observations. The trier of fact very much required the help of these witnesses in resolving this issue. This is to be contrasted to the situation in the case at bar. Recognition witnesses are not per se in a better position to the trier because of a knowledge or ability not held by the trier. The trier does through the existence of the photograph or video have some ability to make an identification. It is a question of degree. In my opinion, when the litmus test of helpfulness is applied in this situation, as compared to Graat, one cannot as freely dismiss the policy concerns that may argue for the exclusion of the opinion evidence.
[34] A number of the policy considerations mentioned in Graat have greater prominence when applied to lay recognition opinion. Here, there is a potential that a jury could be misled and confused. The frailties of identification evidence in general are well known. Confident expressions of identification are not always reliable and probative identifications. There have been miscarriages of justice due to mistaken identifications. Care is required before permitting recognition opinions to be heard by the trier of fact given that without a rigorous gatekeeping function being exercised, the dangers of misidentification can be compounded with each lay opinion admitted. In addition, there is a potential that an undue amount of time will be taken up with such recognition evidence. There may be numerous witnesses who have known the accused in the past and who may feel assured enough in proffering an opinion about whether the person in the video or photo is or is not the accused. This applies to both potential Crown and defence witnesses. If such witnesses can because of their advantageous position truly help the trier of fact, then any time spent in the examination or cross-examination of those witnesses cannot be said to be unwarranted. At the same time, the prospect of a seemingly endless parade of witnesses being shown a photo or a video and being asked to try and recognize a figure within, should be avoided unless there is a real gain in the pursuit of truth at the particular criminal trial.
[35] Whether the probative value of the opinion evidence is outweighed by the policy concerns is a justifiable consideration in determining the admissibility of this type of opinion evidence. It is recognized that trial judges exercise by necessity a large measure of discretion in deciding whether the opinion is admissible: see Graat, supra, at pp. 840-41.
Application of the Law to the Case at Bar
[36] The Crown submitted that the opinions of P.C. Blonde and P.C. DeOliveira should be admitted. Both officers had close face-to-face encounters with the accused on July 13th as they investigated and interacted with Mr. Berhe. Furthermore, it was more recent in time to the July 7th offence date. Finally, the accused had changed his appearance considerably from that time. He has shaved off his beard, has grown out the hair at the sides of his head, and is wearing eyeglasses.
[37] Having giving careful consideration to the evidence and the submissions of counsel, I find that I am in as good a position to make the identification as the two police officers. I find that their opinions based upon this one encounter with the accused are of no assistance to me given the particular circumstances of this case. I have decided so for the following reasons.
[38] First of all, close scrutiny of the testimony of both officers do not support a conclusion that by this one encounter they have a significant advantage in making the identification. I appreciate that they both had close face-to-face encounters with the accused at a time closer to the relevant event in question. However, their recognition opinions are far from being definitive. While I am mindful that the ultimate weight to be given their opinions is a matter that must be decided after the voir dire, some attention must be paid to the issue of whether the witness has actually purported to recognize the person.
[39] P.C. Blonde was shown the photograph taken by the complainant in the subway. P.C. Blonde paused for a considerable length of time after being shown it. She testified that it looked like accused to her. When asked why, she referred to his beard, skin coloring, face, and stature. She pointed out the newspaper the man was carrying but immediately self-censored herself acknowledging it would be inappropriate to use that as an identifying characteristic. In cross, she agreed a person's height can change depending on the camera angle. She admitted she dealt with the accused face-to-face at the scene of his arrest for about 2 to 3 minutes. She did not look at him during the transport to the station. Back at the police station, she could not recall how long the parade before the officer-in-charge of the station took place but in her experience such parades usually took about 5 to 10 minutes. The overall state of P.C. Blonde's testimony, as recognized by the Crown in submissions, is far from being unequivocal in her identification of the accused as the person in the photograph.
[40] On the other hand, P.C. DeOliviera did not hesitate when she was shown exhibit #1. She immediately stated that the man in the photograph was the accused. He had the same facial hair and skin color, the height was consistent, and he was wearing dark colored clothes, collared shirt and a baseball cap. In sum, she testified that she had about forty minutes with the accused not counting the time in the scout car when she was not looking at him.
[41] The problem I have with this officer's opinion is the answers she gave on cross-examination. She admitted that by looking at that photograph for a few moments did not put her in any better position than I, the judge. She further admitted that in the photograph the man's face was a bit obstructed by the pole and that the man "resembled" the accused before the court.
[42] I agree with the Crown that it is a legal determination whether recognition opinion evidence would be helpful to the trier of fact and the police officer's testimony is not conclusive on the issue. Nevertheless, I find that the officer's admission indicates that her present ability to offer an opinion on the photograph marked as exhibit #1 is no greater than that held by the trier of fact. In other words, she acknowledged that whatever familiarity or knowledge she gained on July 13th about the accused which may be helpful in making an identification, did not give her any greater advantage than me.
[43] Had this been the only factor, I may have been persuaded that these frailties in the officers' evidence were matters of weight that could be explored before the trier of fact and that their opinions should be admitted. However, there are other factors that must be considered.
[44] One of the strongest arguments in favour of admissibility would have been the fact that the accused had significantly changed his appearance since July 7th. Both police officers had dealings with the accused much closer in time and before he had made such a marked change in his appearance. However, it is the photograph taken on July 13th when arrested by P.C. Blonde and P.C. DeOliveira marked as exhibit #3 that extinguishes the force of this argument. I am permitted to compare this photograph to the person depicted in exhibit #1 to determine the question of identity. The principle in Nikolovski, supra, also applies when comparing a known image of the accused to the image of the perpetrator/suspect: see R. v. Panghali, 2012 BCCA 407, [2012] B.C.J. 2115 (C.A.) at para. 77.
[45] This "mug shot" is very clear and detailed. It shows Mr. Berhe as he was on the date of his arrest in a close full facial shot. By comparing this "mug shot" to the photograph taken by the complainant on the subway, I am in some ways in a superior position to that of the police officers in making the identification. Their recollection of what the accused looked like on the date of his arrest suffers from the passage of time and the limited opportunity to observe the accused. I have a fixed image of the accused taken from the same date prior to any change in his appearance to scrutinize and compare without the limitation of time or the frailty of diminished memory. I am mindful that the mere fact that I am in a position to make an identification from a photograph or, in this case, two photographs, does not mean that lay recognition opinions are not helpful and inadmissible but it remains an important consideration.
[46] Finally, in this case, exhibit #1 is a two dimensional still photograph and not a video or an audio/video depiction. From this one photograph, one cannot meaningfully gain any information about the suspect's height, stature, posture, manner of movement, facial expression, voice, or physical manifestation of personality. Given this, the fact that the officers had personal interaction with the accused at close quarters does not provide them any advantage that I do not have by simply making the comparison between the "mug shot" and exhibit #1.
[47] In conclusion, I find that the lay recognition opinion evidence of the two police officers is not admissible. In the particular circumstances of this case, they are not in a better position than me based upon their one prior encounter with the accused on his arrest and I do not find their opinions to be of any assistance on the crucial issue of identity.
Released: July 4, 2013
Signed: Nakatsuru J.
[1] Four officers who stated that they recognized the appellant in the video were unacquainted with him.
