Vol. 80, No. 3Ask an expert

Male police officer standing in front of computer screen.

Looking for leads

Cold-case investigator turns to social media

Cpl. Kerry Shima, who investigates historic murders in Alberta, uses Twitter to raise more awareness about his cases. Credit: Courtesy of Cpl. Kerry Shima

RCMP Cpl. Kerry Shima investigates cold-case murders in Alberta. Some are decades old while others are relatively young. All have stumped police. The job involves traditional legwork, but this year Shima turned to social media to leverage more information about hard-to-solve crimes. Paul Northcott spoke to Shima about his work.

How do you define a historical crime and what is your role with the unit?

Historical is kind of misleading. Typically we look at any homicide investigation that is no longer fresh. The leads have, for lack of a better word, gone cold. It could be a year old or, for example, it could be one that is 35 years old. Everyone in the unit is an investigator. I take on different roles. Sometimes I'm the primary investigator; sometimes I'm in the field and will go out and collect evidence.

When do you decide a case is cold?

We have a major crimes unit that investigates murders. When that analysis is over the unit will either carry it through court, or, if they don't come to a resolution and it's still deemed an unsolved homicide, we can take it over. We also look at files that are very old, that have been sitting on a shelf somewhere, and we review them and find that maybe forensic leads, for example, can be reviewed and resubmitted. Or maybe people can be re-interviewed or there's people who haven't been talked too yet for a variety of reasons.

Why tackle these cases when current ones may be more pressing?

Every homicide case is important. We investigate them for the victims and for the families, who obviously will never forget what happened. There's never an unsolved homicide case that's closed. But we are aware resources are pressing and we do have to select cases based on viability. Our oldest case is from the 1930s. Common sense would say it's never going to be solved but you don't really know when that golden nugget is going to come up that warrants reopening cases. That's why we keep track of all of them and none of them are actually ever closed.

What triggers re-opening a cold case?

Forensic evidence that can be resubmitted for analysis is always the biggest thing. When we review a file, the first thing we look for is DNA. Things that may have been overlooked. Science evolves daily and one of our best friends is the forensic lab and we talk to them and we try to get as much evidence to be reviewed as possible. But you really have to understand the old file, what the investigators did and what role the witness played at the time.

What challenges come with investigating cold cases?

Time is a burden and people forget things. Some other people think they remember things or they have manufactured memory. I mean sometimes you speak to a witness now and their statement seems to be much more detailed than it was from the offence date. That's because people naturally fill in the blanks. Also, people die, we lose witnesses, and people move. Some people just don't want anything to do with the investigation anymore. But it works the other way as well. In some cold cases, witnesses who were once living high-risk lifestyles were unwilling to talk. But years later they may be at a different place in their lives and are prepared to talk.

Why did you start your Twitter account?

I watched a presentation from a Toronto police homicide detective who opened a Twitter account to generate tips for unsolved homicides. I touched based with him and got some ideas and advice because that's what I wanted to do for a case I was working on. It's garnered a lot of attention but it's still a small facet of a very large investigation. I think having the face of the investigation out there and an instant mechanism to get a hold of investigators via Twitter probably led to people reaching out in a fashion they may not have otherwise.

Date modified: