Archive | December, 2012

Of what value are wine reviews?

A scandal and an unrelated Wine Spectator article prompt me to raise a question about what role wine reviews play in the marketplace. The scandal involves a prominent Canadian wine writer named Natalie MacLean who is accused by an online Canadian wine site of using wine reviews written by others without providing adequate attribution. She is further accused of demanding wineries subscribe to her newsletter before she will review their wine. Many wine bloggers, some of them rather noteworthy, have become extremely exercised about these matters. The WS piece, written by Matt Kramer, is an extraordinarily perceptive, and articulate opinion that flavor descriptors are of minimal consequence compared to characteristics such as complexity, texture, and balance.

I happen to agree with Matt Kramer, although probably for reasons different than his own. More importantly, I think the low regard in which he and I both hold flavor descriptors logically leads to a general disregard for wine reviews, and in particular for massive database aggregations of wine reviews. To me that’s the nub of the MacLean affair. Seeking to offer sortable databases containing tens of thousands of wine reviews is a priori a bad idea. The logistical difficulties alone lead to very questionable results. Garbage in; garbage out. Legal and ethical questions become almost superfluous. If most consumers shared my skepticism about wine reviews in any quantity over batches of about fifteen, the marketplace itself would sort out issues such as ‘pay for play,’ aggregation of copyrighted material, free samples, junkets for writers, etc. Lawyers would have no role to play (other than drinking wines just like the rest of us).

Perhaps unlike Matt Kramer, my problem with flavor descriptors in wine reviews is I think they are way too subjective and individualistic to be of much practical utility. If adroitly done, they can effectively convey a feeling, a mood about the wine. I think that is useful. It becomes harder and harder as one tries to compose unique moods for each wine when fifty to a hundred of them are included in a single sitting. Every human being has a different body chemistry, different taste experience, and different vocabulary. Expecting one person to accurately convey taste and smell impressions which another person would actually experience, using similes (“smells like Damson plums and Belgian chocolate”), is a very big stretch. Written notes with flavor descriptors are useful to help an individual recall their own organoleptic experience, but trying to get a consensus among a group of untrained tasters is really difficult.

This point is buttressed by experiments done at Stanford University by a linguist named Adrienne Lehrer. She would bring in two volunteers, and put an opaque screen between them. On one side would be five glasses of wine labeled A through E. The other subject would get the same five wines labeled 1 through 5. One subject would be asked to pick a wine, taste it, and describe it out loud. How frequently could the other subject pick out which wine was being described? Barely better than 20% of the time, i.e. about the same frequency as random chance. Now, rigorous training of both subjects can improve results quite a bit (as Ann Noble has shown at UC Davis), but how often are wine reviews read by someone who has received rigorous training (irrespective of whether the writer has)?

So when I’m at some big walk-around tasting, and the person pouring tells me I’m going to notice “cinnamon, mango, and just a hint of hibiscus,” I may subconsciously follow their lead. [I took 15 units of directed study on hypnosis in college. I firmly believe in suggestibility, and I’ve been scientifically measured as fairly high on the Suggestibility Scale.] Usually though, I take offense. “I’ll tell you what I notice, thank ya’ very much.”

The value of wine reviews is to narrow the field for consumers, and to prompt consumer interest in trying certain wines. That’s a good thing. I’m not inclined to bad-mouth the whole concept. I just want to imply that wine reviews need to be viewed with about the same degree of credibility one assigns to the political opinions of one’s in-laws. Don’t argue about them (degustibus non disputatum), but feel free to ignore large portions. And the more reviews written at one sitting, the less attention they deserve. Taste 12 to 20 wines (blind); write up 5 or 6. That’s my recommendation to achieve maximum value.

I’ve never met Natalie MacLean. I have read one of her books and several articles she’s written. I think she’s a damn good writer ~ a lot more talented and entertaining than some of the personalities now screaming for her scalp in Canada. Construction of something as dry and boring as a comprehensive wine review database is a serious misallocation of her abilities. There can be little argument, however, about the fervor of her self-promotion. And I suppose that, along with her success, has created a really severe level of vitriol. Vitriol and self-righteousness.

Alder Yarrow (Vinography) is about the only contributor to the comment thread cited above counseling restraint. And he has taken his share of drive-by abuse for doing so. One could be excused, after reading the comment thread, of assuming very few wine bloggers today accept wine samples, or trips, or meals from producers. Oh please! I went on a junket to Chile and Argentina six years ago put on by an American importer. There were about twenty of us, ostensibly all wine writers of some consequence. At one point I proposed to the group en masse that we purchase 15 bottles of wine at a retail store to do a little comparison tasting back at our Mendoza hotel outside the universe of producers the importer was taking us to see. Cost per person would have been about $25 to have a look at several of the best wines produced in Argentina. ZERO (none!) of these wine writers were interested. Maybe it’s my personality.

I think the place Natalie MacLean went wrong was embarking on a promotional path to offer a mobile app which would deliver wine reviews on 150,000 wines that users could access in stores and restaurants. That’s not a worthwhile product. It’s a software engineer’s wet dream. Consumers shouldn’t want it, and Natalie shouldn’t have agreed to try producing it.

Comments { 1 }