R v Cetin — Rare instance of a directed verdict of acquittal

The result at a glance

Our client was acquitted of child luring, sexual assault and related offences on a directed verdict, before the defence called any evidence. In October 2026, Justice Kent Teskey of the Court of King's Bench of Alberta ruled that the Crown's case, even taken at its highest, could not reasonably support a finding that our client was the person responsible.

The challenge

Few allegations carry more stigma than a sexual offence against a child. The complainant, then 14, said he met a man through a dating app and a social-media account in December 2023, and that the sexual encounter happened in a vehicle. His identity is protected by a publication ban.

The Crown's case on identity was entirely circumstantial. It relied on:

• a selfie found among the complainant's messages, which he said in court was "exactly" the man he met;
• an enhanced still taken from a dark, low-quality video of the encounter;
• a second social-media account that briefly contacted the complainant, and an email address with a similar name found on our client's phone;
• the fact that the man said he was 31, our client's age at the time; and
• an address near the complainant's former group home.

The case had already been through one trial. In May 2026, after the defence shared alibi evidence, the Crown asked for a mistrial. It came back with a new theory and a different offence date. Our client's job, reputation and liberty depended on the second trial.

Preparation: knowing the file better than anyone in the room

The acquittal was won before trial, in some 18 months of work on the file. Disclosure came in 17 packages totalling more than 850 files. They included police notes, search warrant materials, the complainant's full social-media data export, and raw production-order returns from Snapchat and Google.

We did not rely on police summaries of that data. We went back to the source records and checked each claim against them.

• Rebuilt the digital timeline. We converted every message and media timestamp from UTC to Edmonton time and matched each to the complainant's account. That showed the encounter was arranged on a different account from the one the Crown linked to our client, and that this account had no connection to his phone.
• Traced the investigation step by step. We wrote more than 20 internal memos. They tracked how our client became a suspect: a geographic search around one address, then a resemblance to a profile photo. Facial recognition was run twice and never matched him.
• Tested the warrant materials. We compared the sworn applications for search warrants and production orders with the officers' notes, and catalogued the omissions and inconsistencies.
• Built a cross-examination binder for each Crown witness. Every question was pinned to a page of disclosure, with exhibit tabs drawn directly from the source records.
• Reacted within hours to mid-trial disclosure. After the retrial began, the Crown served a new expert report, an updated CV and a phone-extraction report. We analysed all three overnight and rebuilt the expert cross-examination before he took the stand.

At trial: excluding the Crown's forensic expert

The Crown called a police digital forensic examiner to give opinion evidence about our client's phone. Expert evidence is admissible only if the expert is impartial and understands that their duty is to the Court, not to the party that calls them (R v Mohan; White Burgess Langille Inman v Abbott and Wilcox, 2015 SCC 23). We challenged his qualification on that ground in a voir dire.

What the preparation uncovered. His latest report was written mid-trial, after a meeting at which the prosecution told him what it was looking for. He ran a single search term across the full extraction of our client's phone. The search found no trace of the social-media account the Crown relied on, but the report did not say so.

The report also said an email account was "first signed into this device" in October 2021. The phone was a model released in September 2024. The same database table showed records dating back to 2012, which pointed to data migrated from older devices. The report was not peer reviewed.

The cross-examination. We first had the examiner set out the standard he was required to meet:

• that he must be impartial and must not take a side;
• that he must report findings that help the defence as well as those that help the Crown; and
• that his evidence must be full, frank and fair.

He agreed with each point. We then took him through his own report, showing that it reported only what helped the Crown. He had not told the Court that the exhaustive search found nothing linking our client's phone to the account in question. When asked what his role was, he said he was given a job by the police and provided evidence of the offences.

The ruling. The Court excluded his evidence. The gap between the duties the examiner described and the work he actually did showed he did not properly understand his duty to the Court. Letting that evidence in would have risked an unfair trial.

The directed verdict: the Crown's case at its highest

When the Crown closed its case, we did not call evidence, though our client had an alibi. We applied for a directed verdict instead. That is rare: the judge must find that there is no evidence on which a properly instructed trier of fact could reasonably convict. The judge cannot weigh credibility, and must take the Crown's evidence at its highest.

The Court accepted that the complainant was credible. The issue was whether the evidence could reliably identify our client. Our cross-examinations had already shown that, piece by piece, it could not.

The selfie: No witness could say who sent it or when. The complainant was shown only that one photo, never a lineup, and could not otherwise describe or identify the man.

The in-court identification: "carried every danger that the photo lineup process is designed to avoid."

The video still: One enhanced frame from a dark, poor-quality video. Facial recognition never matched our client. The still could not support a finding of identity and added nothing.

The second Snapchat account: One message and one image, with no sexual content. The account on which the encounter was arranged and where the sexual material was sent had no connection to our client's phone.

Age: 31 - Many men share that age. The age did not distinguish our client as a perpetrator.

Location: The complainant could not recall the address. Police supplied the map image, and a geographic search flagged 52 people. This could at most be consistent with our client, but not capable of identifying him.

The Court summed up the Crown's case at its highest as showing only that Mr. Cetin was a 31-year-old man living in the same area. That could not support a conviction. The directed verdict was granted and our client was acquitted.

The skills behind the result

• Command of the evidence. We read every page of disclosure and every row of the raw data, not just the summaries. The most important facts were absences: what was not on our client's phone, and what the reports left out.
• Technical fluency. Exposing the expert took a working knowledge of phone extractions, iOS account databases, timestamps and platform data exports. Without it, his errors would have gone unnoticed.
• Disciplined cross-examination. We had each witness commit to the standard first, then confronted them with their own documents. We asked only questions whose answers we could prove.
• Knowing the law of expert evidence. Gatekeeping under Mohan and White Burgess is where unreliable or partial opinion evidence is stopped. We fought there, instead of leaving it as a question of weight.
• Judgment about when to stop. Moving for a directed verdict rather than calling a defence is a high-stakes call. It worked because cross-examination had already taken the Crown's case apart.

Identification cases carry a real risk of wrongful conviction. Once investigators settle on a suspect, confirmation bias can shape everything that follows. This case shows why every assumption must be tested against the evidence.

This case study describes one matter and is not legal advice. Every case turns on its own facts and evidence, and past results do not guarantee a similar outcome. The complainant's identity is protected by a court-ordered publication ban.

‍