dBx Acoustics

Acoustics of the Oscar Pistorius Trial – What Happened Next

Those of you who have been paying attention will remember that earlier in the year I wrote a couple of blogs on the acoustic evidence presented in the Oscar Pistorius trial. The first blog was my “back of an envelope” opinion on the evidence presented early on, and later I updated with a bit of a rant about the evidence presented by Roger Dixon,

After that, I went pretty quiet on the subject. However there was a very good reason for that, although looking back and telling the story now, it sounds almost too far fetched to be true.

The morning after the blog before

Having linked to some testing that Alexander Jason carried out and reported on, I was thrilled when he got in touch, and said how interesting he had found the blog, as well as how I had understood and explained the limitations of the testing he had been able to carry out. A few days later, after a 6 hour stint at The Drowned Man and a few gin martinis, I woke up in London and checked my email.

There was one from someone whose name I didn’t recognise, saying he was from Oscar Pistorius’ legal team and wondered if I could give him a call to see if we could explore working together.

So I Googled the name and the law firm. It checked out…still, it had to be a wind up, right? Heads of law firms engaged in huge cases probably have secretaries who would make approaches like that? Apparently Alex had suggested they contact me – so I checked with him too. It turned out to be absolutely genuine.

Evidence, planning and proposal

At this point, life became incredibly strange, with a sequence of phone calls between here and South Africa, both with the lawyers and some of the people whose evidence had been presented (the famous gunshot vs cricket bat recordings). I’m not going to reveal too much detail of those conversations and the information that I was given, because clearly it isn’t my place to do so, rather I want to record the experience, and how on earth one goes about planning what to do in such circumstances.

The brief was to write a proposal as to what testing I would do, how I’d do it, and what I would charge. However, there were caveats; they’d need me to get all of the testing done, and be on the stand being grilled by Gerrie Nel, within the space of a fortnight. As you can imagine, at this point life went on hold, as I thought that at any moment I might be getting bundled onto a plane to SA.

I ended up with several sheets of A4 paper pinned to the wall as I considered exactly what tests I wanted to do, what I’d need in order to do them, and whether I’d have the time to process all of the data. After a long late night stint, and the QA skills of my colleague Sue (thanks, Sue!), a 15 page document setting out all of the details was ready to go.

The Scope of Work for the Oscar Pistorius Trial

Again without reproducing a 15 page document here, just a summary of what I was hoping to do:

Site testing

At the scene, to establish internal to external sound transmission from both the bathroom and bedroom, reverberation time within the bathroom, room to room sound insulation between the bathroom and toilet, sound transmission via open and closed windows to the nearest residences, external and internal ambient noise surveys. Also if possible testing internally and externally with gunshots and bat strikes (the bulk of the testing would have used pink noise).

Source Noise Levels

Determine frequency spectrum and levels of gunshot and bat strikes in an anechoic or semi anechoic environment to make the acoustic model as accurate as possible.

Acoustic Models

One model of the bedroom/bathroom, using the information gathered above, to look at how loud gunshot and bat strikes would have been and also to look at internal to external propagation.

A second model to follow on from this, looking at propagation over a distance to the ‘ear witness’ locations, to help determine what might have been heard, at what levels, audibility over background noise levels, and screening from intervening buildings.

Vocal Testing

Vocal analysis of Oscar Pistorius in various conditions (shouting, screaming, talking, etc), and of as many other people of both genders as possible, to examine whether in any condition his voice is similar to that of a female.

Clearly there was quite a bit more to the proposal than that – I was very detailed about times, equipment, assistance needed etc etc, but as you can see I tried to cover all of the outstanding questions I had talked about in the blog.

A Lost Opportunity?

So for a couple of weeks we went through a to- and- fro conversation, with me still on standby to travel. Unfortunately, however, it was decided that they couldn’t afford to use me, and ultimately they used a more local consultant. I was surprised; my fee would have cost less than three days of Barry Roux – however so be it.

I was pleased to see that Ivan Lin really did know his stuff about acoustics. And he did a great job in keeping his cool in the cross examination, presenting some quite technical information clearly. But at the same time, I was very curious that the evidence presented was extremely limited, essentially only looking at distance propagation and not going into the in depth testing or modelling I had hoped to do. Why the legal team chose to narrow the scope of acoustic evidence so much clearly has a number of possible interpretations, and I would be curious to know the real reason.

It would have been great exposure for dBx Acoustics to have worked on the trial, and an amazing experience for me personally, but on balance, I’ll get over it; the thought of going up against Gerrie Nel was a scary one. What I really regret, though, is that I didn’t get the chance to go and do all those tests and work out for myself the truth about the acoustic evidence.

That’ll teach me to mouth off on the internet. I plan to do it more often 🙂