Wednesday 19 April 2017

Stradivarian nonsense

Recently, I was doing my regular Sunday boys-night-in with Pat the Salt. The old fellow decided that there was nothing new under the sun on the telly and stomped off to bed. He was/is right on one level because the first things I caught on BBC after bedtime was a repeat of a 2011 BBC Horizon programme called Global Weirding. Old video material can often be caught on youtube but not this one so I had to find it on DailyMotion and here it is. We in Ireland are cultural slaves and chattels of the Yankee-dogs and I was not aware of >!Vive La France!< Daily Motion until just now - it was founded by Benjamin Bejbaum and Olivier Poitrey in Paris in 2005 and has got to be richer and less derivative in content than youtube, its much larger rival. ANNyway, back to Old Horizons: Global Weirding is a jumpy cut / cut /cut overview of the history and hazards of global warming. The title itself dates it a bit - global warming is no longer quite The Thing - we tend to call it climate change nowadays to acknowledge the extra cold freezing winters that are one consequence of poor old Gaia being on her last Cheyne-Stokesy gasp.

In one of the snippets, they show a woman playing a Stradivarius and saying the the unique quality of sound ["nearest to the human voice" the violinist intones], is due to the fine grained wood that the maker had at his disposal. My crap-detector immediately tuned up a notch and I Googled "blind testing Stradivarius" which popped up two studies - Indianapolis 2011 and Paris 2014 which showed that, when confirmation bias was taken off the register, the expert violinists didn't rate the Strads as better or even distinctively different from quality modern instruments. In one test the expert players were blindfolded and asked to rate fiddles as antient or modern. They were right in 31 cases <win> but wrong 33 times <oooops> but were only humble enough to confess that they didn't know on 5 instruments. This is so close to the expected result of tossing unbiassed coins, that you might even cite it as being too accurately on the random button to be credible. Like RA Fisher's [prev] suspicions that Gregor Mendel [prev] had fudged his results. This is the sort of evidence assertion that works in the fiddling world. More Stradiworship, gets a little tiresome in the light of the evidence above.

One deeply concerning aspect of the above results is the certainty in the assertion of the expert even when they were bzzzzzt wrong. As scientists it should be okay to say more often "heck, I just don't know the answer to that . . . I'll endeavour to find out." rather than push out another weak paper with insufficient power to answer aNNy question. Mark Twain said it better “It ain't what you don't know that gets you into trouble. It's what you know for sure that just ain't so.

We have seen the exposure of wine experts as charlatan poseurs in a formal blind tasting. Where the experts were unable to read the goddamned label they were no better than thrown dice at comparing different wines. Ditto pretty much any male who gets hold of the wine-list at a fancy restaurant. The French, bless 'em, talk large about terroir  to explain the unique quality inflated prices of Premier cru wine and they may be right but you can use evidence based instrumentation to improve quality . . . even in California. And Doris Lessing famously showed that experts in the publishing world could only recognise talent in someone who had been previously published. I blew a gasket recently about designers slagging off comic sans.  It doesn't have to be like this. If you didn't obey my instructions and watch Steve Mould and Matt Parker bending over backwards to eliminate bias in testing their hypothesis that Canadian plastic money smells of maple syrup, you can do so now.

Don't go way thinking this Expert culture is not harming you; and bitching when someone makes a career out of pulling off such a coup is probably sour-grapes. The phenomenon is not limited to rarified marginal fields like wine-tasting and high-end Booker Prizey publishing. It is everywhere there are people in charge of other people. In Ireland definitely, and in your country probably, there is a managerial class who claim a fat salary and possibly an annual bonus because they are The Managers.  And they do manage . . . so long as nothing strange or different or unexpected happens. In good times, organisations trundle along and could be successfully managed by CEO's goldfish. But when the glass is falling and there's a dark-cloud roiling up and over the horizon then they start making terrible decisions and/or abandon ship . . . and it turns out that HR has arranged a contract that requires the crew to lob a huge severance package into the departing Captain's gig.

Now I'm all cross, I'll have to consume a nice cup of tea and listen to a bit of Bruch - not on a Strad though, because Stradivarius is a bill of goods.

No comments:

Post a Comment