Excerpted from Thinking, Hard and Soft
Conveying sensory impressions in words is difficult. Émile Peynaud (1912-2004), a bordelais legend among oenologists and wine-writers remarked that “we tasters feel to some extent betrayed by language.” How do writers convey information about colour, sound and taste to readers effectively? In other words: can you tell the difference between a robot and a human stock analyst?
Serious interest in wine notes began after the 1976 Judgement of Paris (not this one) pitted French against Californian wines in a legendary blind tasting. Nine judges rated wines on a scale from 1 to 20. Then the organiser of the event, Stephen Spurrier, arrived at the final ranking by “adding the judges marks and dividing this by nine (which I was told later was statistically meaningless)”. Motivated by the flawed nature of this procedure, economists Richard Quandt and Orley Aschenfelter showed how to use elementary statistics to produce a more meaningful ranking. Indeed, a software programme called “Winetaster” developed by Quandt codifies the underlying statistical methods. These have since been applied successfully in other high-profile blind tastings, most recently in the Judgement of Princeton in 2012.
In blind tastings conducted to rate quality, a group of tasters assess wines without knowing what is what. They may be told nothing about the wines that are being evaluated, or they may be told the names of the wines under assessment but not which wine they are tasting. We can think of these tasters as biased (and often unreliable) machines who gather information about a wine sample (input) before producing an assessment (output).
The output takes the form of a rating on a number scale backed up by tasting notes. To assess consistency and reliability of tasting output, there may be several rounds of tasting in which the order of the wines is varied. Aggregation of the rounds leads to a final relative ranking. The challenges associated with this process stem from the two natural facts that no expert is a consistent machine, and that the experts may disagree in their assessments.
Two tasting notes for the same wine may differ to such an extent that it is not clear they are both about the same wine. The same is true for analyst stock recommendations or user ratings on Amazon. It is not possible to compare ratings without translation to a common vocabulary.
We’d hope that professional writers and “critics” might be better than amateurs. Professionals use their vocabulary reliably and consistently. But amateurs, in the words of Émile Peynaud “speak in metaphors and allusions … inventiveness of their vocabulary conceals its vagueness”.
That may be true, but it is all too easy to poke holes in professional jargon. In an essay “On Wine Bullshit” (the term “bullshit” is of course! applied in a philosophical sense, see Frankfurt), Richard Quandt generates random wine notes from a lexicon of standard wine descriptors. You’d be forgiven for mistaking them for the real thing.
So where does that leave us? Be it a review of a plasma screen, a stock recommendation, a wine note or a theatre critique, the principle is the same. Tell the reader what your reference points are, make comparisons, explain the jargon you use. Or if you like to keep your reviews short, publish a categorized table of the descriptors you use as a reference point.
I look forward to the day when stock analysts publish guides to their “tasting vocabularies“.
(Adapted from the Introduction to: On the Information Content of Wine Notes: Some New Algorithms? Journal of Wine Economics 8 (03), 318-334)