COURT FILE NO.: 15-A12474
DATE: 2018/02/01
ONTARIO
SUPERIOR COURT OF JUSTICE
BETWEEN:
Her Majesty the Queen
- and -
Omr M. Mohamed
S. Dohoghue for the Crown
J. Addleman for the Accused
HEARD: January 29-31, 2018
DECISION RENDERED: February 1st, 2018 (from the Bench)
Introduction
[1] On the night of August 10, 2015, Kian Pezeshki was attacked while sitting in his car in a plaza parking lot at 1661 Hunt Club Road.
[2] Mr. Pezeshki and two friends, Tristan Larabie and Yusuf Abbadi, had played soccer earlier that evening, then spent time at Mr. Abbadi’s house, then stopped to fill Mr. Pezeshki’s car with gas and buy some cannabis at the Hunt Club plaza. By this time it was about 11:15 p.m. After these errands were taken care of, Mr. Pezeshki and Mr. Larabie sat in the car, waiting for Mr. Abbadi to return from the convenience store in the plaza.
[3] Without any warning, Mr. Pezeshki’s car was approached by six or seven people from a van parked nearby and a crossover utility vehicle (“CUV”) that just pulled up. One of the men from the CUV approached Mr. Pezeshki’s car window and began yelling at him in broken Arabic and English. He grabbed Mr. Pezeshki’s collar. Mr. Abbadi returned to the car and tried to intervene. Mr. Pezeshki saw his attacker’s arm swinging towards him, and covered his face with his forearms. He felt something hit his left arm.
[4] The attacker and the other men then left the scene. When Mr. Pezeshki got out of the car, he realized that he was bleeding from a serious cut in his left forearm. He was taken to hospital for surgery. He has been left with some residual nerve damage.
[5] Omr Mohamed was arrested a few weeks after the attack. He is charged with robbery, assault with a weapon, aggravated assault and possession of weapon for the purpose of committing an offence.
[6] Immediately after the attack, Mr. Pezeshki gave a general description of his assailant to the police. He said his attacker was an Arab male, around 20 years old, with a large body type and facial hair. He did not describe any particular identifying features.
[7] Six weeks after the attack, on September 24, 2015, Mr. Pezeshki identified Mr. Mohamed in a photo lineup.
[8] The description given by Mr. Pezeshki to police right after the attack would obviously not be enough on its own, to charge anyone. I furthermore cannot give any weight to Mr. Pezeshki’s identification of Mr. Mohamed in the courtroom at trial, since he had by then seen him at the preliminary inquiry.
[9] The Crown has produced no evidence of any connection between Mr. Mohamed and the other persons at the scene, or physical evidence linking him to the scene. Sgt. Donald, the Ottawa Police Services officer in charge of the investigation, suggested in his testimony that Mr. Mohamed’s vehicle resembles the attacker’s CUV seen in a video surveillance recording of the area where Mr. Pezeshki’s car was parked. Sgt. Donald admitted however that he is not an expert in vehicle identification and that the recording did not capture a licence plate or vehicle name. There is no persuasive evidence that the shape of Mr. Mohamed’s CUV is particularly distinctive.
[10] As a result, nothing aside from Mr. Pezeshki’s identification of Mr. Mohamed in the photo lineup establishes that he was the person who attacked Mr. Pezeshki on August 10, 2015. If I am not convinced beyond a reasonable doubt that Mr. Pezeshki identified the right person, I must dismiss the charges.
[11] In determining whether the Crown has proved its case, I will first review the proper approach to eyewitness identification evidence, and then apply this approach to the evidence in this case.
The proper approach to eyewitness identification evidence
[12] Courts have long recognized the danger of relying solely on eyewitness identification to convict someone. Although a witness may sincerely believe that they have a very good memory, eyewitness identification is the type of evidence most likely to result in a wrongful conviction (R. v. Miaponoose (1996), 30 O.R. (3d) 419, 1996 1268 (Ont. C.A.)). Witness credibility is usually not the issue. The risk is “an honest but inaccurate identification” (R. v. Jack, 2013 ONCA 80 at para. 14).
[13] Relying solely on eyewitness identification is dangerous because people remember events less accurately than they think they do. The frailties of eyewitness testimony were reviewed at length in a study paper published by the Law Reform Commission of Canada in 1983. The Commission noted that:
Since perception and memory are selective processes, viewers are inclined to fill in perceived events with other details, a process which enables them to create a logical sequence. The details people add to their actual perception of an event are largely governed by past experience and personal expectations. Thus the final recreation of the event in the observer’s mind may be quite different from reality. (…)
Thus, although most eyewitnesses are not dishonest, they may nevertheless be grossly mistaken in their identification.
[14] Some 40 years earlier, in R. v. Smierciak (1946), 1946 331 (ON CA), 87 C.C.C. 175, the Ontario Court of Appeal held at p. 177 :
The weight of evidence of identification of an accused person varies according to many circumstances. A witness called upon to identify another person may have been so well acquainted with him or her as to make the identification certain and safe. The person to be identified may possess such outstanding features or characteristics as to make an identification comparatively free from doubt. The conditions under which an observation is made, the care with which it is made, and the ability of the observer, affect the weight of the evidence. In addition to such matters, and of the utmost importance, is the method used to recall or refresh the recollections of a witness who is to be relied upon to identify a person suspected of wrongdoing or who is under arrest.
[15] A court must accordingly consider two elements: (1) the circumstances in which a witness got a chance to observe the perpetrator of a crime, and (2) the procedure by the police to get the witness to positively identify a suspect.
[16] With respect to the first element, the circumstances affecting a witness’ ability to reliably identify a perpetrator include:
• Whether the person was known to the witness. If the person was a stranger, they may be harder to identify accurately.
• How long the witness got to see the person. If the interaction was fleeting, the witness will have had less time to get a solid impression of the perpetrator’s features.
• The conditions in which the observations were made. If conditions obscured the witness’ ability to see, this may challenge the identification.
• The witness’ state of mind at the time. If the witness was in a state of stress, their memory may be less accurate.
• Whether, immediately following the event, the witness was able to provide a detailed description. If the witness was able only to give a generic description of a perpetrator, without mentioning any distinctive characteristics, this may reduce the weight given to the identification.
[17] These factors come up over and over again in the caselaw, and were summarized most recently by the Court of Appeal in R. v. Jack at paras. 15 and 16.
[18] With respect to the second element, the procedure by the police to get the witness to positively identify a suspect, we are dealing here with a photo lineup.
[19] In 2004, the Sophonow Inquiry established guidelines for photo lineups. These guidelines have been relied on in many cases, including a decision by this court in R. v. Gonsalves (2008), 2008 17559 (ON SC), 56 C.R. (6th) 379.
[20] One of the Sophonow guidelines concerns the selection of “decoy” photographs – that is, the photos of individuals who are not the suspect - in a photo lineup. The Inquiry recommended that:
The photos should resemble as closely as possible the eyewitnesses’ description. If that is not possible, the photos should be as close as possible to the suspect.
[21] This is again not a new idea. In its 1941 decision in R. v. Goldhar, 1941 311 (ON CA), 76 C.C.C. 270, the Ontario Court of Appeal instructed (at para. 6) that “it should appear that the selection of the other person to form the line-up has been made fairly, so that the suspect will not be conspicuously different from all the others in age or build, colour or complexion or costume or in any other particular”. Sixty years later, in R. v. Harvey (2001), 2001 24137 (ON CA), 57 O.R. (3d) 296, the Ontario Court of Appeal concluded that a photo lineup was fair in that the accused’s photo “did not stand out in any way and the individuals depicted in the other photographs were similar in appearance to the accused” (para. 14). Consistent with this principle, in R. v. Dorsey, 2003 26504, the Court of Appeal overturned a conviction where the decoy photos in a lineup were not sufficiently similar in appearance.
[22] In Justice C. Hill’s thorough review of the caselaw on photo lineups in Gonsalves, he noted that questionable identification procedures “may not be fatal to a finding of guilt” or render the identification evidence inadmissible (para. 44, cites omitted). He cautioned however that “[w]here cogent confirmatory evidence is absent, the circumstances of the identification may be sufficiently suspect that there exists a real risk of a miscarriage of justice” (para. 45, cites omitted).
Application of these legal principles on the evidence in this case
[23] In applying the principles I have just reviewed to the evidence in this case, I will consider first the circumstances in which Mr. Pezeshki interacted with his assailant on August 10, 2015, and second the photo lineup procedure on September 24, 2015.
The interaction between Mr. Pezeshki and his assailant
[24] The incident took place just before midnight in a side parking lot next to a drive-through MacDonalds.
[25] The lighting was poor in the area where Mr. Pezeshki’s car was parked. Sgt. Donald said there was a streetlight across the parking lot and at the entrance to the drive-through, as well as a light attached to the wall of building on the other side of the drive-through. He admitted that the incident took place in one of the darker areas of the plaza. This is confirmed on the surveillance video.
[26] It was raining. According to Constable Tataryn, who attended the scene about 10 minutes after it occurred, it was the proverbial “dark and stormy night”. It was raining hard enough that no blood or other trace evidence was found in the parking lot, even though Mr. Pezeshki’s arm was bleeding heavily when he got out of it and blood drops were found inside the MacDonalds where he was taken for first aid.
[27] Mr. Pezeshki admitted that he was “dazed and confused” when people from the van and the CUV began to approach his car. When he reached Mr. Pezeshki’s window, the man who went on to attack him began yelling in a mixture of Arabic and broken English. Mr. Pezeshki did not understand what he was saying. He had never seen this person before. He could not recall if Mr. Abbadi said anything to his attacker because, again, he was “confused”.
[28] The entire incident took very little time. Based on the surveillance video, about 40 seconds passed from the time the attacker approached the car until his return to the CUV. Slightly over a minute passed from the time the CUV first drove into the parking area and the time it left.
[29] Despite these circumstances, Mr. Pezeshki insisted that he had a perfect memory of the entire incident. He had absolutely no doubt that he could identify his attacker.
[30] In his testimony, Mr. Pezeshki did not acknowledge any conditions that made it harder for him to see and identify his attacker. He denied that it was dark. He incorrectly stated that there was a streetlight directly overhead. Based on the timer on the surveillance video, he overestimated the length of time his attacker stood at his car window yelling before getting physically aggressive. Mr. Pezeshki maintained that he had a better memory two and a half years after the incident than he had a year earlier.
[31] Mr. Pezeshki also refused to admit that his position in the car limited his line of sight as his attacker approached from the back of the vehicle. Once the confrontation began, he insisted that he continued to have a clear view of his attacker’s face, even after he was grabbed by the collar, after Mr. Abbadi inserted his body between the attacker and the car, and after Mr. Pezeshki began fending off the physical attack that resulted in his arm being slashed.
[32] Mr. Pezeshki’s refusal to admit any limitations to his ability to see his attacker did not make his testimony more believable. Instead, it suggested to me that he was not clearly recalling all of the details of the incident.
[33] Mr. Pezeshki’s account also showed how a witness’ memory may be based on a reconstruction of events, as opposed to a reliable memory of the events themselves. There are two striking examples.
[34] First, Mr. Pezeshki testified that he remained in his car for up to a minute after the CUV left. He said that he and Mr. Larabie then got out, at which point Mr. Abbadi began to scream and Mr. Pezeshki realized that he was bleeding. He said that they then decided to drive to the hospital. Mr. Abbadi got into the driver’s seat, but drove only about two meters before Mr. Pezeshki told him that they should instead call an ambulance.
[35] As seen in the surveillance video, the aftermath of the attack was different. The video shows that Mr. Pezeshki’s car pulled out of its parking spot about 9 seconds after the CUV left. Mr. Pezeshki was still in the driver’s seat. This does not mean that Mr. Pezeshki was lying. Maybe he first moved the car to another spot in the plaza before getting out of the car and realizing he had been wounded. In that case, Mr. Pezeshki simply forgot part of what happened. Despite this, he recounted what he thought had occurred with great certainty.
[36] The second example involves Mr. Pezeshki’s testimony about why he had to borrow money from Mr. Abbadi to pay for gas. In direct examination, he said he forgot his debit card at home. In cross-examination, he explained that he deliberately left his entire wallet at home before going out to play soccer with his friends earlier that day. Defense counsel then pointed out to him that he had earlier testified that his wallet was in the car, and that his wallet had in fact been found in the backseat of the car by police. In response Mr. Pezeshki recalled that he simply discovered, once he had filled the car up, that he did not have his debit card. He further admitted that he did not know why he did not have it, and that it could have been lost or he could have taken it out of his wallet before leaving home that day. In other words, Mr. Pezeshki’s account of how he made a conscious decision to leave his wallet at home before going to play soccer was an on-the-spot reconstruction of events to explain why he had to borrow money from Mr. Abbadi.
[37] I mention these two examples not because I want to single out Mr. Pezeshki as a particularly unreliable witness. These examples simply show how a person can be sincerely convinced that they remember events perfectly, and yet omit or reconstruct important details. This is the very reason why court have to be so careful in relying on eyewitness identification.
[38] Mr. Pezeshki’s description of his attacker to police was generic rather than specific. He did not mention his attacker’s hair colour or eye colour or what he was wearing. He did not mention any distinguishing facial features.
[39] Moreover, some of Mr. Pezeshki’s description does not match Mr. Mohamed’s description. Mr. Pezeshki said his attacker was between 6’1” and 6’4”, about a head taller than Mr. Abbadi. The evidence showed that Mr. Mohamed and Mr. Abbadi are the same height, each just over six feet tall. According to Mr. Pezeshki, his attacker spoke only broken English. According to Sgt. Donald, Mr. Mohamed speaks “fine” English.
[40] I do not think that Mr. Pezeshki was being untruthful in his testimony. I find however that he had a very limited opportunity to see his attacker. Based on all of the evidence, I conclude that Mr. Pezeshki had a clear, unobstructed view of his attacker’s face for about 10 seconds. At all other times, his view was limited, either because the attacker was approaching the car from behind, or Mr. Abbadi was in the way, or Mr. Pezeshki was struggling with his assailant.
[41] In argument, the Crown relied on testimony by Constable Hebert, another officer who attended the scene immediately after the incident. Constable Hebert said that Mr. Pezeshki was excited, but also alert and coherent. This testimony does not persuade me that Mr. Pezeshki was not under a great deal of stress from the moment he saw his car being approached until his attacker fled the scene. This stress inevitably compromised his ability to focus on his attacker’s face, as did the darkness of the plaza, the lack of direct overhead light and the rain.
[42] Applying the factors that may affect a witness’ ability to identify a witness as set out by the Court of Appeal in R. v. Jack, I conclude that the circumstances of the incident were not conducive to an accurate eyewitness identification of Mr. Mohamed.
The photo lineup
[43] Based on the evidence of Sgt Donald and the video recording of the procedure, I am satisfied that the Ottawa Police Services conducted the photo lineup in a manner that was largely consistent with Sophonow guidelines. The officer who administered the lineup was not involved in the investigation and did not indicate, through words or body language, that Mr. Pezeshki should identify one particular suspect. He gave Mr. Pezeshki appropriate instructions, including a warning that the attacker’s photo might not be included in the lineup. Mr. Pezeshki clearly understood the instructions, as he identified one of the photos as someone he recognized, even though this person was not his attacker.
[44] Despite this, there was a shortcoming in the photo lineup. Mr. Pezeshki told police that his attacker had a large body type and facial hair. The school picture of Mr. Mohamed used in the lineup shows someone who is clearly overweight and who has facial hair. In the lineup, however, there were only two men who have, in the words of defense counsel, a “fat face”. One is Mr. Mohamed. The other is a man who has no facial hair.
[45] In assembling the decoy photos, Sgt Donald used a police databank of images and software that allowed him to set parameters for male, Middle Eastern, brown hair, facial hair, 18 years old plus or minus five years. This generated about 99 photographs. Sgt. Donald chose eleven of these photographs to use in the photo lineup on September 24, 2015. He testified that he was looking for images of men who had features similar to Mr. Mohamed, in that they had a low eyebrow line, facial hair, a darker complexion and a wide face.
[46] In cross-examination, Sgt. Donald admitted that he could have selected “heavy set” as a parameter. His explanation for not doing so was that this could limit the number of photos generated, because the descriptions of the photos sometimes do not accurately record body size. When asked why he had at least not tried to use this parameter, he justified his decision by saying that face size does not necessarily match body type, and that a person’s weight may change over time.
[47] Although I have no doubt that Sgt. Donald acted in good faith, I find that the photo of Mr. Mohamed stood out. He was clearly heavier than the other men in the lineup. I do not accept Sgt. Donald’s explanation that he wanted to account for possible changes in Mr. Mohamed’s size over time. The lineup took place only six weeks after the incident, which would not have allowed even a devoted dieter to significantly change his weight. This explanation is moreover inconsistent with Sgt. Donald’s insistence on using photos of men with facial hair, a feature that can be changed very quickly.
[48] As a result, I find that the photo lineup was flawed. This flaw could have been overcome, if there were other elements in this case that made Mr. Pezeshki’s identification of Mr. Mohamed more reliable. But, as stated by the Ontario Court of Appeal in R. v. Tat (1997), 1997 2234 (ON CA), 117 C.C.C. (3d) 481, a conviction based solely on eyewitness identification raises great concern “where the person identified is a stranger to the witness, the circumstances of the identification are not conducive to an accurate identification, pre-trial identification processes are flawed and … there is no other evidence tending to confirm or support the identification evidence” (para. 100). This accurately sums up the state of the evidence in this case. In these circumstances, justice would not be served by a conviction.
Verdict
[49] Mr. Mohamed, please stand up.
[50] For the reasons I have given I find you not guilty on all four counts.
Justice Sally Gomery
2018/02/01
COURT FILE NO.: 15-A12474
DATE: 2018/02/01
ONTARIO
SUPERIOR COURT OF JUSTICE
BETWEEN:
Her Majesty the Queen
- and -
Omr M. Mohamed
REASONS FOR JUDGMENT
Justice Sally Gomery
Released: 2018/02/01```