Confirmation bias

The AFP has finally released a public version of its submission to the Clarke inquiry into Mohamed Haneef’s detention; The Oz has coverage, and the full submission is available [255kB PDF].

It’s a pretty good demonstration of how confirmation bias operates; the suspicion-arousing elements were fed into a theory that Haneef was connected to the crimes in the UK, while disconfirming evidence and alternative explanations were not given due weight. It seems to me that, if you look at the evidence described in the submission starting from a presumption that Haneef was involved with the terrorists, the selective attention and interpretation engaged in by the investigators is understandable. But it serves as an important reminder that investigators need to stop and consider how an innocent person might act in the same circumstances.

Two of the principal investigative findings that aroused suspicion about Haneef were:

• his travel plans, in particular his hasty departure on a one-way ticket and the chat room discussion with his brother about his planned departure; and
• in a new revelation (which was apparently not even put to Haneef during questioning), the finding of ‘jihadist’ material in his premises and on his computer.

The travel plans have been discussed previously. The investigators viewed it as suspicious that he was leaving so suddenly and without purchasing a (cost-effective) return ticket. Let’s now assume that Haneef had no prior knowledge of the terrorist attacks. His wife had delivered a baby by emergency Caesarean a week ago. Now people he knew had apparently committed crimes in the UK and the UK police were attempting to contact him. He was on his own in Australia.

These would seem to me to be pretty stressful and extraordinary circumstances. If I was in that situation, I imagine that I might be very keen to be with my family, particularly my wife and newborn child. Of course, I would also want to assist the police and attempt to make sure they knew I was not involved. So, while making plans to depart (arranging leave from work, buying a plane ticket, etc.), I would keep the contact details of the police officer who wanted to speak to me and attempt to make contact from wherever I was.

What’s more, given the uncertainty of the situation, I might not be confident that I would be ready to return in a week. I might even feel like I just want to be home and may not want to return to Australia at all. So I would probably opt for a one-way ticket – even though it would cost more if I did return – and take my important documents with me. And while I would know I am innocent, I would be a bit concerned that the police seem to think I am connected to the crimes and so might talk about it with my brother – but I wouldn’t want to go mentioning it to people at work who don’t know about my link to the terrorists.

All of these explanations are entirely consistent with Dr Haneef’s behaviour.

Now what about the jihadist material? The first thing the AFP found was a brochure that refers to “…the brutal invasion and occupation of Afghanistan and Iraq… the killing and murder of our brothers and sisters and the brutality of British and American foreign policy…”

If anyone who would entertain the idea that British and American foreign policy has been brutal and that the occupation of Afghanistan and Iraq is wrong is a terrorist suspect, then half of Western society should be investigated (starting with me). The AFP submission indicates that the brochure came from an international organisation that is considered a terrorist organisation in some countries, but gives no more specific information.

The second item of jihadist material was found on Haneef’s laptop – audio files of a lecture by an author “who has been linked to Al Qa’ida” who “expresses a militant view of jihad” and “goes on to say that undertaking jihad in the cause of Allah is the most virtuous act a Muslim can carry out” and “argues that no act of violence carried out by Muslims can ever reach the level of violence and evil committed against the Muslims.”

Naturally, these are positions I strongly disagree with, and I would have concerns about any person who endorses such opinions. But does the fact that Dr Haneef had them on his computer indicate that he agrees with the author? I read and listen to all sorts of material that I disagree with. Should anyone who has read Mein Kampf be regarded as a Nazi? I am assuming that some of the conservatives who comment on Teh Left might have read Marx – does that make them closet commies?

Haneef’s lawyers have noted that he was not even given an opportunity to explain this material. It seems that it was assumed that the explanation is that Haneef had sympathies for such a ‘jihadist’ philosophy. But there are other explanations. I would imagine that someone who is well-educated enough to be a medical doctor is interested in critically engaging with complex issues. This means seeking exposure to a range of opinions and arguments before forming one’s own view. It does not mean only reading and listening to people who say things you believe.

Although the AFP submission provides the specifics of some additional evidence that fed into their decisions, it does nothing to change my view of how the investigative process itself was flawed. While investigators must be sceptical and suspicious of any person of interest, they also need to weigh any evidence that implicates people against the potential alternative explanations for that evidence.