« Animal experimentation: morally acceptable, or just the way things always have been? | Main | Setting The Stage »

November 18, 2008


I love this, having experienced this effect with wine myself! I agree with the paper that the 100-point system does a terrible disservice to consumers and should be discredited. Thus I read the original paper with glee.

When I got this to part: "These findings raise an interesting question: is the difference between the ratings of experts and non-experts due to an acquired taste? Or is it due to an innate ability, which is correlated with self-selection into wine training?" I stopped cold. I do not know the answer to this for wine, but I think I might for coffee, in which I have myself trained.

Why do you and everyone else persist in misreading this study? Cheap wines taste just as good to the untrained palate. Those with "wine training" can appreciate the difference. (As frelkins points out, what this means is an open question. But if we assume training rather than selection bias is at work . . .) if you think this is a complete case against fine wines, you must also believe that no acquired taste is worth acquiring.

There's a whole blog on this; AAWE has an excellent post on this very topic http://wine-econ.org/2008/06/20/test2.aspx authored by two of the economists who authored the study you reference.

In this post, they argue that non-experts enjoy cheaper wine slightly more, while experts enjoy more expensive wine slightly more (and, it should be noted, their preference for more expensive wine is more strongly correlated than is the non-experts' for cheaper wine).

So, Dave, hopefully that post will be more satisfying to you.

I found that I like looking at paintings a lot more after doing some research. Likewise, I enjoy classical music more after learning something about the history of it. Art is often about context, and it is no different with wine-making. Having some knowledge allows you to relate subtle differences in taste to the choices and skill of the wine-maker. When you can do this you start to find some wines more "interesting" than others rather than just tastier. Some of the interestingness doesn't even necessarily come from the taste itself, but the taste in context of the current trends in the craft in that region or with that grape. I'm sure there is a selection effect where people who care about that sort of thing choose to obtain a little knowledge. I am also sure that much of the pleasure from this comes from knowing you are detecting something that others can not, but also its just fun in and of itself to focus and detect (and it gets you drunk).

Also, inexpensive wines are good these days. They are well made and generally very fruity. Just like a good Bubblegum album, they are supposed to be pleasant to the senses.

Isn't it also the case that people do enjoy more expensive wine more when they know its more expensive? (I thought that was an ancillary point to this research). So, as a wine novice, am I just paying a premium to enjoy my wine/dinner/experience more when I don't have the palate training to achieve this without knowing that I've paid a premium?

This is funny - all the wine novices I know will buy cheaper wine because it tastes better, and then will buy more expensive wines when trying to develop a taste for it. I had no idea there was some sort of large-scale self-deception that more expensive wines actually taste better.

Can it be that knowledge of the price acts as a sort of placebo effect when it comes to evaluating the quality of wines? The placebo effect is borne of enhanced positive expectations. If you are sampling "fine" wine in the context of a fancy meal in nice surroundings, then expectations may already be heightened. But another thought about these experiments involves the implicit assumption that taste is or can be developed into something objective. That's not true, but it is also not equivalent to asserting that one cannot train individual perception. We understand memory to be a collection of things we sensed, that is, saw, felt, heard, smelled, or tasted, in a temporal context. Sad to say, there is a natural decline in the acuity of our senses after about age 30. Given memory bias, what we may be principally recalling is the context of drinking expensive wine, rather than the taste and smell of the stuff itself. Is anyone aware of studies done comparing beer? It would be a no-brainer to blind test the difference between, say, a Belgian trappist brew and mass market American swill. Is this because beer is less homogeneous than wine; or is it because the comparisons in wine studies don't involve dramtic quality and price grandients such as between say Pomerol and Night Train?


"the context of drinking expensive wine"

No. The limbic system can be trained; I have done so myself, as have the many professional coffee people I know at the NYBOT grading room. These people do blind tastings in a drab room all the time. Scent memory is quite strong and vivid, as Proust can tell you; surely this is an evolutionary artifact, so we don't poison ourselves.

Pro tasters in coffee, at least, can reliably and accurately tell the difference between various origins of coffee and in some cases, even the districts. This is why they are hired, to cup against the commodity coffee "C" contract in New York and resolve disputes. It's big business - many containers (37,500 lbs each) of coffee.

Fine coffee has a likewise strict standard for tasters, known as "cuppers." They must pass a rigorous sensory evaluation, in which they must be able to taste certain chemical concentrations, correctly arrange the concentrations to match, and in ascending order, and they also must be able to accurately identify samples against the industry standard chemical references offered by Jean Lenoir in his big chemistry set. We also have a big chemical reference book, written by Flament, which tells us what exact chemicals are responsible for all the aromas we detect in coffee.

Cuppers must be able to taste positive qualities as well as defects accurately. Only then will they be certified. It is common to study for this test for 2-5 years(!) after several years of working full-time in the coffee industry. Once you understand what both the defects and positives taste like, it is relatively easy to identify "the good" from "the bad." It is scientific, actually.

@ Robin

"And is any added enjoyment just the pleasure of knowing you can distinguish something others cannot? "

I can say from personal experience - which may not hold for others - that by such training of the limbic system, I learned to distinguish each flavor individually. Thus there were more flavors packed into each sip of wine or coffee.

More flavors = more good sensations = more pleasure. So to your question, no, I just learned to detect in depth the true volume of pleasure in each cup or glass.

I was so happy with this deeper pleasure in coffee that for several years I ran around the world teaching other consumers to do the same, which was the origin of my work with the Specialty Coffee Association of America. I wouldn't ever have said that I derived snobbish, exclusionary "I can taste it, you can't, nyah nyah nyah" pleasure.

Rather I immediately began to preach and teach, as I do to this day. I truly believe that everyone can learn with practice to improve their appreciation of the most passionate and romantic beverage, coffee.

The lowest decile differed; bad wine is really bad to everyone so there is some evidence of crude discrimination going on. Are acquired tastes worth acquiring? No, I don't think so, for the majority of people anyway. Tastes should be immediate, instinctual, powerful, and without doubt or question. Only if a taste is very powerful should one look to refine it, but those are the connoisseurs. I think most people are willing to defer to them, even if they do not think so for themselves. Their taste is not so powerful that they wish to refine it, but are perfectly content that others may have more powerful reactions and are able and willing to concentrate their perceptions to do so.

Fortune, do people with better hearing that can distinguish closer tones or fainter sounds thereby feel more pleasure from hearing? Do people with better eyesight thereby feel more pleasure when they look? Do people who can distinguish finer gradations of temperature on their skin feel more pleasure thereby? Those effects seem pretty weak at best; so why would taste precision be any different?

I think it is a universal truth that it takes education to appreciate quality. It is true of fine arts, sports, science, you name it, and, yes, wine. Quality has nothing to do with taste.


"I truly believe that everyone can learn with practice to improve their appreciation of the most passionate and romantic beverage, coffee."

Ah, back to the glories of belief. One could make an equally compelling case advocating consumption of the most contemplative and mentally clarifying drink, imported shincha sencha green tea, or a bowl of matcha frothed at just the right temperature. There really is a noticeable difference between, for instance, daily sencha, gyokyro, kabusecha, matcha, etc. in terms of green teas, as there are gradations of quality and taste attributes among robusta and arabica beans, dark roasts, etc. in terms of coffees. (N.B. We are not even considering other preparations of tea plant camellia sinensis). But this would all undergird my almost embarassingly obvious argument, namely, that taste is inherently subjective.

At the risk of going even further far afield on the topic of potent potables, readers here may enjoy "A History of the World in Six Glasses." I don't disagree with your statement concerning the utility of the rhinencephalon in evolution, or even one's ability to consciously train the limbic system to some limited extent. But your belief in the rapturous quality of coffee is really (almost reminiscent of J.S. Bach's Kaffee Kantate) may be be not qualitatively different from my subjective proclivity for green tea, or someone else's predilection for a certain vintage wine. There may be a price signal correlation pertaining to quality of all three beverages, but I submit that disentangling subjective preferences above some low threshold is in the end, a quixotic endeavor.


"finer gradations of temperature on their skin"

The sensual world. Finally something I understand! Yes, Robin, as someone with good hearing still on one side, I love listening to my Bose headphones - to detect the small sounds of the music, to hear and catch a quick, soft grace note - yes, my delight in the music is enhanced. I feel more pleasure thereby.

When viewing a landscape with new, sharper contact lenses, I take great joy in being able to discern that small distant dot is a hawk whirling over the Gordon River in Tasmania. I feel more pleasure thereby.

When feeling the finest touch of friend, his smooth hands, the way different parts of his hand - say perhaps, fingers cooler, palm warmer - feel in mine - the fading trail of warmth and pressure as he ran his fingers up the inside of my forearm - oh yes, I derive great joy. I feel more pleasure thereby.

If those "effects" seem "weak" to anyone, I encourage re-engaging with senses. I think of Blade Runner, Rutger Hauer's "tears in rain" speech. Is it not precisely attention to these moments that give life meaning?

Point to frelkins.

In the non-sensual world, now OB has similarly spoiled me for Slashdot. At least they don't go around announcing there will be no more good wine!


"But this would all undergird my almost embarassingly obvious argument, namely, that taste is inherently subjective"

People often say this, but I dispute this common error. Think of evolution. Evolution has given us all the ability to discern - with greater or lesser untrained clarity depending - certain key flavors, probably so we don't kill ourselves by eating bad stuff. All normal human beings share the taste receptors for the common tastes: sweet, sour, salty, bitter, umami. This is basic direct chemoreception.

No normal human being puts salt on their tongue and says "um, sweet!" Normal humans all easily detect and identify vanilla in the parts-per-billion range. No normal person has even put a drop of real vanilla on their tongue and said "pepper!" Our tongues have adapted to actually receive the underlying chemicals in these substances.

Whether you like or dislike the sensation of this chemoreception is indeed subjective - thus the real meaning of the old saw you mention - but the process of it, and its ability to be trained is not in doubt.

You may mock me for "belief" if you wish, but I have taught hundreds of people in several countries basic coffee cupping. That's it: at the next OB meetup, I'm serving the coffee and giving everyone a copy of the SCAA Flavor Wheel.

@Robin, Alan

"Fortune, do people with better hearing that can distinguish closer tones or fainter sounds thereby feel more pleasure from hearing?"

Much like good wine running down an untutored palate, I think numerous fine distinctions are getting lost in this thread!

To answer Robin's question, yes, increased *sensory resolving power* increases your capacity for pleasure. But although this might be due partly to the mere pleasure of sensation, that is unlikely to be the main effect. Rather it's because -- in the cases of music, painting, wine, etc. -- we are applying that increased resolving power to artefacts that have been designed with an aesthetic structure that requires high sensory resolution to appreciate. For instance, if we reduced a piece of music to a sequence of loud/soft, high/low tones, then obviously we would be stripping away all of the aesthetic structure that lies in the distinction between loud, very loud, soft, very soft, etc..

Second, there is also a matter of taste. Even if the senses detect a distinction, the mind must be able to detect its aesthetic implications. And just as one can train oneself to an increased sensory resolving power (training the 'limbic system' as frelkins puts it), so to can one train onself to appreciate and experience difference in aesthetic quality. This is just called taste. (You might call it, *aesthetic resolving power*, and say it supervenes on sensory qualities.) Taste is obviously in part subjective, since two people can have extremely refined taste that is nevertheless defined in different ways, and is in disagreement. Alan points this out. But taste is still objective, in the sense that there is still a clear distinction between refined taste and unrefined taste. For instance, I can hear the differences between Eminem and Jay-Z, but it means nothing to me aesthetically since I have never studied hiphop and don't know anything about it.

But I think the most interesting question here is on a third point, what I would call *the returns to connoisseurship*. For instance, learning more about wine lets you enjoy good wine more, but it makes you enjoy poor wine less. Since most wine served will be poor wine, learning about wine will reduce your expected future enjoyment. That is, there are negative returns to increased connoisseurship in wine, as Robin suggests.

What's interesting is then to ask, what areas have the best returns to connoisseurship? In what areas of perception is it profitable to refine one's taste? Is it taste in food? taste in clothes? taste in ideas? taste in people? taste in places? taste in images? The returns of connoisseurship depend both on the availability of quality and on the pleasure premium earned by quality. So my thinking is that you should cultivate your taste in things that are easily available and where small quality differences produce large pleasure differences.

I'd guess good candidates are ideas (freely available, widely variable), the visual world (freely available -- just open your eyes), human beings (good ones are scarce, but incomparable), and common foodstuffs (bread, cheese, etc..). Ironically, I'd say good objects to avoid are product like wine, which are the object of a widespread status competition in connoisseurship itself. That competition will have driven up the average quality level sufficiently that the superiority of the maximum quality items will be marginal and the price of the maximum quality items will be prohibitive.

(Of course this ignores the pleasure of the connoisseurship status competition itself, which is a separate issue.)

Could anybody explain this to me: the title says "Cheap wines taste just as good as expensive ones". Then one goes on reading and discovers the claim to be "Individuals who are unaware of the price do not derive more enjoyment from more expensive wine". Are you all assuming that the "tasting better" relation can be reduced to some "producing more enjoyment among x-kind-of-people" formula? I have doubts about this equation even if you substitute "x-kind-of-people" with "connoisseur". All the experiment talks about is a fact about enjoyment. I can see no result regarding taste.

This all may be true, but so what? All classical music sounds the same, unless you have a trained ear, and the same with rock or rap music. There is still immense satisfaction from listening to music with a trained ear. Yes, there are signaling effects of course - "I can understand this music and you cant", but that is not all of it.

The comments to this entry are closed.

Less Wrong (sister site)

May 2009

Sun Mon Tue Wed Thu Fri Sat
          1 2
3 4 5 6 7 8 9
10 11 12 13 14 15 16
17 18 19 20 21 22 23
24 25 26 27 28 29 30