Public Safety Committee on May 31st, 2007

[The Chair, Conservative, Garry Breitkreuz]

That's an interesting comment.

Mr. Shur, do you have any reply to that?

[Gerald Shur]

Yes.

First, of the 8,000 witnesses who have been relocated—and these are federal witnesses, as opposed to state and city witnesses—there was a recidivism rate of less than 18%. I can't be more precise, because each person I called to check on the recidivism rate told me a different number. The highest was 18%. The lowest number—from the happiest person I talked to—was 11%. So there's recidivism of somewhere between 11% and 18% among the 8,000 people.

That may come about in part because of the procedure we follow in entering the witnesses into the program. First, we have a separate office in the United States Department of Justice called the Office of Enforcement Operations, which has a witness security unit that determines who shall or shall not enter into the federal witness protection program.

The judgment is based upon a significant amount of information they receive from a variety of people, one of whom is a psychologist, who examines each potential witness recommended by the federal investigative agency, and every member of the family over the age of 18, to determine whether or not the witness is likely to commit a violent act, how well they would fit within the program, whether they'd be able to follow the rules given to them, what sort of employment they would need, and what their skills are and such. So a thorough psychological examination is given.

There is some difficulty as to whether or not you can truly predict violence in an individual through psychological testing. On occasion, I felt that my own gut reaction was as good as the test.

Also, we receive the previous arrest record of the individual. We also receive a report from the United States attorney. The United States has 93 United States attorneys, who are the chief federal law enforcement officers in their districts. So we get a report from them about the case and the witness and his or her background.

Then we have the United States Marshals Service do the actual relocating of the witness. Because of their role, they have the opportunity to interview the witness and his family beforehand, in what we call a preliminary interview, and they give us their opinion as to whether or not this witness will function properly after relocation. So we also have the benefit of that.

We have the benefit of resident experience, the people in the Office of Enforcement Operations who have worked so often in this that they have reactions to it and such.

Of those factors, I think the screening factor is a very important factor in determining what happens afterwards.

Finally, we notify the Federal Bureau of Investigation of every witness we relocate, so there is a record. And if any local law enforcement agency were to make an inquiry about a person under a new name, they would immediately know from the FBI that the person is someone else. I think that is a deterrent for the witness.

[The Chair, Conservative, Garry Breitkreuz]

Thank you very much.

As the committee can see, our normal routine has been disrupted here and we're not following the normal time constraints.

Mr. Comartin, you are next. Just give a brief introduction of who you are and then go ahead with your question.

[Joe Comartin, NDP, Windsor—Tecumseh, ON]

Thank you both for being here.

I'm Joe Comartin. I'm a member of Parliament for Windsor—Tecumseh, a member of the New Democratic Party, and I'm the justice critic as well as the public safety critic for my party.

I want to go back, Mr. Shur, to a point you made about the screening. Can you give us any numbers or percentages of how many people are considered for the program but then denied access to it?

[Gerald Shur]

I can tell you that about 150 to 165 witnesses enter the program each year. Of those, the largest percentage by far are entering the program by going into prison. We have protective custody units, which are prisons within prisons, buildings within prisons, where we have only witnesses--approximately 70 cooperating prisoner witnesses in each unit--so they're not circulating out in our society.

As far as the rejection number goes, that's a difficult one to give because there are two ways of rejecting. One is informal, which is the most popular way, where the investigator or the U.S. attorney calls on the phone and says, “I have a case I'd like to present to you formally, but before I do that, would you accept this type of case?” and you say “No”. That's not counted anywhere, so we don't have that. Then we have the formal applications. I would say the formal applications probably run at about 10% rejection.

Again, the total number each year is 150 to 165, with a substantial percentage of them going to prison and their families being relocated. So when I say 150 or 160 people are going to prison, that doesn't mean there's no other job. The marshalls relocate them.

[Joe Comartin, NDP, Windsor—Tecumseh, ON]

I'm sorry, is it Professor Fyfe?

[Prof. Nick Fyfe]

That's right, yes.

[Joe Comartin, NDP, Windsor—Tecumseh, ON]

Professor, do you have any sense of what the rejection rate is in the United Kingdom?

[Prof. Nick Fyfe]

No, unfortunately. I certainly don't have access to that kind of data. Since part of the difficulty in the U.K. is that witness protection remains a local responsibility, all that kind of information is held by individual police forces that run witness protection programs. There's no one central source of information about witnesses who have been taken into these programs.

So no, I certainly don't have access to that kind of data.

[Joe Comartin, NDP, Windsor—Tecumseh, ON]

Mr. Shur, you've been quoted in some news articles here.

[Gerald Shur]

I regret that.

[Joe Comartin, NDP, Windsor—Tecumseh, ON]

Well, I'm going to ask you about it anyway.

As I think you both know, the inquiry we're conducting right now was prompted by two individuals who were in the program, one most recently having subsequently committed murder while in our witness protection program. In one case, the family very much wants to know about the individual who perpetrated the murder on their family member.

Mr. Shur, you indicated again in the news articles that in the U.S., in a situation like that there would be more information shared with the family, or perhaps with the public generally.

Can you explain how that process works in the United States?

[Gerald Shur]

First, I have to make a correction from the introduction, because I conferred with someone yesterday who is doing the work I used to do. I'm retired. I don't want him getting angry thinking I'm reclaiming his job.

I talked to him about that issue, as a matter of fact, to see that the procedures were still the same, and they are. We have a great deal of flexibility in our program resident in the Office of Enforcement Operations to make determinations such as whether we should disclose or not disclose, when it is appropriate, what the rules would be, and so on.

When a person commits a crime in the United States who's been in the program--the person has a new name and commits a crime--it generally becomes public because they're going to trial. Our trials are very public. He is prosecuted under his new name. Frequently his old name might be mentioned in the course of the indictment. It might be Joe New Name and Joe Old Name, so that the prosecutor feels he has it absolutely right.

[Joe Comartin, NDP, Windsor—Tecumseh, ON]

Can I interrupt you here?

[Gerald Shur]

Sure.

[Joe Comartin, NDP, Windsor—Tecumseh, ON]

If the old name is disclosed in the indictment, that is not a breach of the legislation?

[Gerald Shur]

That would not be a breach of the legislation.

Also, we might notify local law enforcement of the person's old name when we relocate them, so that they know who they have in their area.

In the specific case, I'm not entirely familiar with the case, but with the assumption that the person commits a crime and the relatives of the victim wish to know who that person was, under our statute we have a victims compensation requirement—that is, we must offer to the family of any victim who is killed up to $25,000, I think, to cover medical expenses or funeral expenses, and so on. They certainly would have a right to know who that person really was who had killed their relative.

The one complicated area is that if disclosing that information would compromise an ongoing investigation, we might delay it for a bit. But that would happen so rarely. I can't recall it happening, as a matter of fact; it's just a potential.

By disclosing the name to the family, you give them some peace. The cost to the United States government to do that simply means relocating again the family of the witness who had committed the murder, so that the family of the victim has the peace of closure and the family of the witness has the safety of being relocated again, and the loss comes in a money sense to the United States government.

[Joe Comartin, NDP, Windsor—Tecumseh, ON]

Professor Fyfe, do you understand the legislation in the U.K. to permit that kind of disclosure in those circumstances?

[Prof. Nick Fyfe]

I'm fairly confident that there is no specific legislation that deals with that issue.

[Joe Comartin, NDP, Windsor—Tecumseh, ON]

Mr. Shur, you said that in the United States the recidivism rate, at the high end, was about 18%. Are there official statistics kept on this? Is there an analysis done on an ongoing basis?

[Gerald Shur]

Yes, there have been analyses done, some by sampling and some by agencies other than our own that have examined the records. I think the 18% figure is an outside figure and a safe figure.

[Joe Comartin, NDP, Windsor—Tecumseh, ON]

Right.

Do you know, within that 18%, whether there have been any murders committed?

[Gerald Shur]

Yes, there have been.

[Joe Comartin, NDP, Windsor—Tecumseh, ON]

Do you have any sense of what proportion or how many of, say, the 150 or 160 who go in each year?

[Gerald Shur]

Well, it's not of them; it's 18% of the 8,000 witnesses.