An explosive murder confession – or a dodgy transcription?

17 March 2015

Listen to these two snippets of muttered self-talk, then read on to see how a transcript can prime journalists’ perception.

If you are among the few who have not already heard the media’s interpretation of this audio, you’ll find it useful if you write down what you hear now, before reading on – and if you have a moment, I would love to be told your perception – you can send a message here.

The sad thing is, if you have been aware of any media in the last 24 hours, you have probably already been ‘primed’ with an interpretation. Here’s the story as broadcast on ABC Radio National (3 minute excerpt from 20 minute broadcast of ‘AM’ on 17 March 2015):

Notice how Michael Vincent (an excellent journalist) gives the interpretation before playing the audio. No doubt that was how it was presented to him. Here’s a tiny fraction of the coverage by the US media.

The problem is, it looks like everyone ‘hears’ this confession – when really only one person has genuinely heard it. All the others have been primed. And we know how dangerous that can be: The crisis call experiment

Don’t get me wrong. I am not saying Robert Durst is not guilty. I have absolutely no opinion about that – I had never heard of him till this morning.

I am saying this transcription is not evidence of his guilt. Even if some of the words might be right (I have not examined this material in detail) the sentences as a whole are not reliable, and the overall interpretation of what might have been meant by the muttering is wide open.

At best, recordings like these should be taken as prompts to further investigation. I hope they do not appear in court as evidence. And if they do, I hope the trier of fact gets a chance to hear them genuinely unprimed – very difficult to achieve.

Sign up here for occasional news about Forensic Transcription (no spam!)