Right. I'll make one general observation, which is that there is so little research in this area around issues of effectiveness that what we actually know is very limited. Clearly, there are practitioners who have a lot of direct experience—and Mr. Shur has demonstrated that very clearly—but I suppose what tends to be missing is any independent evaluation and research into witness protection programs. I think that's why we're working with relatively little in terms of a secure, independent evidence base. I just make that as an initial statement.
The issue about who is best placed to make those decisions about inclusion and exclusion I think is a very difficult one. There are issues partly about the speed at which these decisions need to be taken, and also about the fact that often you need to monitor the way in which the threat to witnesses may rise and fall. The risk to witnesses tends to be a dynamic phenomenon; it isn't just a one-off thing. They might be more at risk at some times than others.
Certainly from the work we did in looking across Europe, there was some surprise that in some jurisdictions it was the police who were allowed to take this decision. It was felt they were perhaps too close to the whole investigation process and that you needed some people who had some distance from that, who perhaps could take a wider view as to whether that witness was essential to the prosecution and the investigation. I think there was a feeling in some cases that perhaps the police were too ready to take witnesses into protection programs, because it would allow the investigations to proceed more quickly. But clearly that was going to involve huge disruptions in those people's lives.
Certainly the opinions we heard tended to be of the view that it was perhaps better to take that decision out of the hands of the police and to allow a more independent assessment, perhaps, of the evidence the witness could provide, within a broader context.