deachman: what i got wrong about siu investigations
6 minute read
the special investigations unit.
ashley fraser
/
postmedia
last september, a 55-year old ottawa woman was being arrested at home when — while handcuffed, intoxicated, and escorted by police— she fell down a flight of stairs and sent to the hospital.
if your first thought is “that shouldn’t happen — someone should look into that,” well, someone did.
and when ontario’s special investigations unit — the civilian watchdog tasked with investigating police-involved deaths, shootings, serious injuries and allegations of sexual assault — finished their report seven months later, the conclusion was a familiar one.
joseph martino, the director of ontario’s siu, issued a statement indicating that he “found no reasonable grounds” to believe that an ottawa police officer committed a criminal offence in connection with the woman’s broken right arm. “she simply lost her footing,” the report stated.
my reaction was also familiar: a foregone shrug. after all, isn’t that what the siu almost always says?
i came by my skepticism honestly. a steady childhood diet of protest songs by bob dylan and pete seeger, coupled with news stories during my formative teen years of rcmp barn-burnings, toronto police brutality and watergate, can teach you early on that power rarely surprises you by apologizing.
the numbers appear to back me up. i went through the siu’s own news releases from the 2025 calendar year and found that of the 255 investigations publicly closed by a director’s report — excluding cases “closed by memo,” or those deemed outside the siu’s jurisdiction — only six resulted in criminal charges, a rate of about 2.4 per cent.
over the five-year period ending in 2025, siu annual reports show yearly charge rates ranging from roughly three to 4.7 per cent.
at first glance, that looks damning. if an oversight body almost never lays charges, what, really, is it overseeing?
advertisement
but then i had to ask myself: what does my inner woody guthrie expect, and is that reasonable? i reached out to some experts and read annual reports and academic research to try to answer that.
what i found was an oversight body with a mandate that simply doesn’t match many people’s expectations of what it ought to do.
the siu’s responsibility is much narrower than i thought: to determine whether there are reasonable grounds to believe a police officer committed a criminal offence, and whether there is a reasonable prospect of conviction. that’s pretty much it.
and yet we keep asking it to solve other problems — problems it was never designed to solve — and then read the numbers as proof of its failure.
“the siu does not have a reform function,” cautions greg brown, a carleton university professor who served with the ottawa police from 1985 to 2018 and sat on the police services board. “they don’t have a branch that looks into whether there are more use-of-force occurrences this year and how to fix that.
“their function is strictly investigative — an initial determination on whether a charge should be laid. that’s the extent of what they do.”
in other words, the siu answers a legal question, while the public is often asking a moral one.
“justice doesn’t mean a conviction,” brown adds. “justice means that there’s a competent, thorough investigation and a determination. and if justice moves on to the next stage of a court, that means there’s a fair trial.
“if the definition of justice is that the officer is going to get convicted, that’s not justice.”
the distinction makes sense, even if it leaves a bad taste when police conduct causes serious harm without crossing the line into criminality.
raising charge rates would likely solve little. a 2021 study by professors at university of guelph and brock university followed siu cases in which charges were laid and found that roughly one-third were withdrawn by crown prosecutors, while acquittals are the most common outcome of cases that went to trial.
advertisement
ian d. scott saw this first-hand. as siu director from 2004 to 2009, he was responsible for deciding whether officers would be charged.
during his tenure, roughly 95 per cent of investigations ended without charges, a figure consistent with today’s numbers. the reason, he says, wasn’t due to sympathy for the police. it was evidence.
many cases involve chaotic arrests, intoxication, mental-health crises or medical emergencies. civilian witnesses may be unreliable or traumatized, or there may be none at all. subject officers — those being investigated — are legally entitled not to provide interviews or notes. even when video exists — a rarity when scott was director, it might clarify what happened without resolving intent.
“there can be serious harm,” he says, “without criminality.”
public perception still matters. scott welcomes the siu’s gradual shift toward hiring more investigators who are not former police, even as he and brown acknowledge that policing experience remains essential in many investigations.
“it’s a combination of public perception and independence,” he says. “if you have civilians involved, there’s a better chance of getting a perspective that isn’t overly police-oriented. there’s no guarantee, but there’s a higher probability.”
scott also points to body-worn cameras and a clearer statutory authority under the siu act as incremental but real gains.
it doesn’t help public confidence that subject officers are not required have to submit to interviews or provide notes, in keeping with charter protections against self-incrimination. neither scott nor brown believes subject officers should be compelled to participate in interviews — internal discipline processes can compel cooperation later — but the limitation remains a source of frustration.
advertisement
when andré marin headed the siu in the late 1990s, he argued that officers who refused to cooperate should face consequences. in his 2008 report as ontario’s ombudsman, marin criticized routine delays and resistance in officer interviews and called for stronger statutory tools to enforce cooperation. one of his recommendations — that directors’ reports explaining “no-charge” decisions be made public — was eventually adopted.
some of the mistrust was well-earned. in the siu’s early years, police services did not always notify the unit of incidents that should have triggered an investigation. those failures helped establish clearer statutory obligations. brown says that , at least among larger police services today, routine non-notification is no longer a problem.
brown and scott agree that the siu has improved. transparency is better, the mandate clearer, investigations more professional. but both also argue that the public’s frustration is aimed at the wrong institution.
brown suggests that systemic review, training reform and use-of-force policy belong elsewhere — upstream — not inside a criminal watchdog. he’s floated the idea moving charge decisions from a single director to a small panel to make the process feel less opaque. scott is unconvinced, arguing that charge decisions are legal judgments, not committee work, and that spreading responsibility would simply dilute accountability without changing outcomes.
scott does, however, support one structural change: removing the siu from the attorney general’s umbrella and making it an officer of the legislature, like the auditor general or ombudsman. that change, he argues, would strengthen independence and insulate directors from political pressure.
the siu was never designed to tell us whether policing is humane. it was designed to answer a narrower question. our frustration comes from confusing the two.
advertisement
skepticism toward power is healthy, and essential. we should probably listen to our inner woody guthries more often than we do. but i’ll stop treating every siu “no-charge” decision as evidence of indifference, or worse.
our website is your destination for up-to-the-minute news, so make sure to bookmark our homepage and sign up for our newsletters so we can keep you informed.
share story
