More on the Benefits of Wine Knowledge

wine geeks I’m still musing about how knowledge—of wine, cheese, landscapes, whatever—can influence our perceptions. In a previous post I wondered whether philosopher Kent Bach is right to argue that novice tasters and knowledgeable tasters have essentially the same sensory experience:

They may be in no position to know anything about the grape(s), the region and the vineyard, the producer, and the vintage, they may have no basis for comparing this great wine with similar but merely very good wines, and they may be unable to articulate what particular aromas and flavours they are experiencing or have any notion of what experienced tasters mean by balance, structure, and elegance of a great wine. Even so, it is not obvious that this wine does not taste as wonderful to them as it does to the expert. They may not be equipped to enjoy the cognitive pleasures that accompany tasting it, but that’s not to say they aren’t fully equipped to experience the sensory pleasure inherent in attentively drinking it.

My objection to Bach is that his account doesn’t explain my own experience as a novice taster. It wasn’t that I lacked the conceptual categories or language to talk about the flavors more experienced tasters detected. I wasn’t experiencing the flavor notes at all until I gained more knowledge of what to expect from particular wines. Or at least that is the way it seemed phenomenologically.

But as Jonathan Cohen pointed out in his helpful comments on the previous post, there is substantial evidence that, with regard to vision, which has been studied more extensively than taste, knowledge of categories and linguistic descriptors has no effect on test subjects’ ability to discriminate differences between color swatches when shown side-by-side. And although taste and aroma perception may differ from color vision in this regard, there may be no good reason to think so.

Assuming the science on vision is correct and can be extended to flavor discrimination, what would account for my subjective perception that knowledge of categories and wine descriptors has improved my ability to taste? Offhand, I can think of two hypotheses that would explain it.

(1) Improved ability to discriminate flavors happened concurrently but independently of the acquisition of knowledge. I have in mind some form of training that doesn’t require conceptualization analogous to learning to ride a bicycle.

(2) The acquisition of wine knowledge encouraged me to focus on detectable features of the wine I hadn’t noticed before.

I suppose (1) is plausible although I don’t know what the mechanism for such learning would be. But (2) strikes me as plausible as well for the following reason.

The overwhelmingly dominant flavor in wine is, unsurprisingly, a kind of generic “grapeness”. That is what novice wine tasters taste predominately, along with alcohol, acidity, and the tactile sensations of tannin in red wine. The signal to the brain coming from the taste buds and olfactory sensors indicating “grapeness” must be quite strong. My hypothesis (which was suggested by Jonathan’s comment) is that learning to discern the full aroma and flavor spectrum of a wine involves suppressing the strength of that signal so that other flavors can be more easily noted.

In other words, novice tasters are overwhelmed by the dominant flavors and can learn to taste only by learning to ignore them.

That is the role of wine knowledge. The expectation that I should expect to detect vanilla in some oaked Cabernets gets me to focus on the boundaries of the generic “grapeness” where it appears to shade off into something else only vaguely sensed. That focus has the effect of suppressing the strength of the signal signifying “grapeness” and perhaps amplifying the signal from other flavors and aromas, patterns of suppression and amplification that get reinforced with practice.

Just a hypothesis without evidence. But it would explain how conceptualization shapes sensory experience without modifying the genetically determined detection thresholds that govern what we taste and smell.

Wine knowledge, then, while strictly speaking not necessary for learning to taste well, is enormously helpful and efficient as a skill-building mechanism. I’m not sure whether Bach would find this congenial to his view or not. But it seems to me, on this hypothesis, conceptual knowledge is directly impacting phenomenological sensory discrimination thus belying any hard and fast distinction between intellectual and sensory pleasure.


  1. Hey Dwight:
    I agree that your hypothesis (2) seems more plausible than hypothesis (1). It’s not surprising to me that knowledge could direct attention on features that would otherwise be lost in the rich perceptual array.

    Fo what it is worth, I think it’s also interesting to ask what it is about pairwise discrimination that allows those same features to emerge even if you don’t have the knowledge. I conjecture that part of the story has to do with (several, various) mechanisms of perceptual contrast. Thus, the grapiness recedes under the condition of pairwise comparison because the pair of samples don’t differ (a lot) on that dimension. Whereas there IS a big contrast on other perceptible dimensions: one but not the other has tannins and vanilla flavors or whatever. And the usual story about lateral inhibition by itself predicts that neural systems are in general likely to give more pronounced responses for bigger contrasts rather little ones. I’m sure that’s not all there is to it (and even this is probably an interaction effect), but I suspect that’s part of the story about why we manage to pull out features under pairwise comparison that we otherwise need knowledge/conceptual categorization to pick out. No?

    1. Thanks Jonathan. That is really helpful. So the upshot is that knowledge is useful, perhaps even necessary, in the absence of an opportunity for pairwise comparison. I wonder, on the larger issue of cognitive penetrability, whether this research suggests that with effort, concentration, and a bit of practice, the influence of beliefs about price, producer, previous encounters with the wine, etc. can be minimized, since sensory mechanisms seem to be largely independent of at least higher order cognitive processes.

  2. Dwight: Jonathan Cohen just called my attention to your two very interesting and provocative posts prompted by an article of mine. I agree with much of what you say and also with his comments. There are many useful distinctions to bring in here, such as being able to tell differences and being able to tell what those differences are. For now, I just want to clarify what I was and what I was not claiming.

    I don’t think it’s fair to say, to quote from early in your 11/29 post, that “According to Bach, the only difference between a wine expert and a novice is the ability to describe and explain. As far as taste goes, the wine expert and the novice taste the same thing.” You can be an experienced taster rather than a novice without being a wine expert. To quote from my article, “Knowing how to taste obviously helps — there’s no substitute for experience at tasting — but how important is knowing about the wine, or knowing about wine in general?” Insofar as experience, and the knowledge it brings, enhances your ability to notice qualities, to categorize and remember those qualities, and to open yourself up to new possibilities, I think that (experiential) knowledge improves your ability to enjoy and appreciate different wines. I think factual knowledge, of the sort that adds up to most of being an “expert” on wine, is incidental to refined tasting ability and enjoyment, though fun to acquire, apply, and even flaunt in its own right. (Maybe some aspects of this knowledge, such as associating distinctive qualities with specific varietals/vineyards, may even play a direct role, by enhancing your ability to notice subtle differences.)

    So I heartily agree with your astute observation – in fact I’ll drink to it! – that “novice tasters are overwhelmed by the dominant flavors and can learn to taste only by learning to ignore them.” Excellent point! But whereas you go on to say that enabling this learning is “the role of wine knowledge,” I’d say that this learning is more the result of experience at tasting. But I gladly agree that tasting experience does yield a particular *kind* of wine knowledge.

    To finish with one simple thought: much of learning how to taste is learning how to pay attention to what you’re tasting. I’ll leave the more sophisticated thoughts to experts on perception and sensation, like you guys.

    1. Kent,

      Thanks for stopping by and commenting. I quite agree with you that some wine knowledge–how grapes are harvested, selected, modified in the winery, etc.–seems incidental to tasting. As you say, it is fun to acquire and deepens one’s knowledge of why one is tasting particular components of a wine but not crucial to the sensory experience. But knowledge of varietals, growing conditions, geography, oak programs, etc. can lead you to expect certain features in a wine. And what I’ve been thinking about, in conjunction with the larger issue of cognitive penetrability, is whether those expectations make an important contribution to our sensory experience.

      What puzzles me about tasting experience (understood as a purely sensory phenomenon) is how it becomes more refined if it is not making use of conceptual categories to organize it. As I pointed out in my posts regarding my own experience, the refinement wasn’t there to begin with. So what are the mechanisms by which it improves? Conceptual categorization and the refinement of those categories via acquiring beliefs would seem to be one plausible possibility. And I took you to be denying that.

      Jonathan has me persuaded that the key is the suppression of dominant flavor notes allowing the less dominant to emerge. It is intuitively easy to see how pairwise comparisons could accomplish that. But I suspect part of the story is that we then gain from those comparisons more refined conceptual categories which then become useful in assisting memory.

      At any rate, I am enjoying your work on this. It is sparking new insights I think.

      1. Thanks, Dwight. Yes, knowledge of varietals, growing conditions, etc. can certainly give you ideas of what to expect when you taste a wine. But why, and in what way(s), is that good? Knowing what to “look” for might help you notice it, especially if it’s a subtle feature, but this might also make it harder to notice (or value) other equally subtle but atypical features. I can envisage doing controlled experiments (to the limited extent that the relevant variables can be controlled) to test for this sort of thing. In fact, experiments along such lines have been done, where people are given true, false, or no information about some attribute of the wines they are to taste, describe, and compare. But, so far as I know, these experiments have had highly circumscribed aims.

        One complication here is that we need to distinguish effects on what you taste from effects on what you think you taste. It’s a notorious fact that knowledge of price or producer can affect what you think you taste (and what you think of what you taste).

        I’m not sure what you mean by “conceptual categories” for flavor features. I hope you don’t mean labels for them, because labels don’t do them justice, even if the labels do have such benefits as enabling one to group wines and to facilitate meaningful comparisons. Maybe you mean what some philosophers call “recognitional concepts.” Fine. But what about the je ne sais quoi qualities that distinguish great wines and even merely good but distinctively interesting wines? To be sure, some wines are interesting because they defy expectations of what they’re supposed to be like.

  3. Kent,
    Sorry for the delay in responding. I’ve been swamped with end-of-the-semester duties.

    The problem of distinguishing “what you taste” from “what you think you taste” (essentially, confirmation bias) is a real one, not just in properly marking the distinction philosophically but in the practice of wine tasting. I find that there are two ways of practically reducing error in this regard: (1) critically assessing a judgment using knowledge of what other flavor notes could plausibly be present in order to find a better match, and (2) returning again and again to the wine to reconfirm my original assessment. Neither is foolproof, certainly.

    Furthermore, I think having a conceptual grasp of what one is tasting helps rather hinders the discovery of new or unexpected components. Recognition of difference is sharper if I know the contrast introduced by the unexpected component. As to the “je ne sais quoi” qualities that define great wines, you are right–they defy our expectations. But I think it helps to have the expectations (and the conceptual categories) in place in order for the defiance to stand out. This is what Kant had in mind with his notion of the subime–sensory experience that cannot be accomodated by our concepts.

    I’m not sure what to say about “flavor concepts”. By “recognitional concepts” I take it you mean what we have when we perceive “another one of those” without specifying further what the demonstrative refers to. I’m not sure that these are properly called concepts. Although this is not my area of expertise, I’ve been mucking about in the literature on non-conceptual content and I am persuaded by those who argue that our perceptual capacities outstrip our conceptual categories. (Wine tasting would seem to be a good example of the “fineness of grain” argument) That would seem to lend weight to your view that sensory experience is independent of at least certain kinds of cognition.

    But my view is not that rich and comprehensive sensory experience without knowledge is apriori impossible. Your connoisseur who lacks expert knowledge is not logically impossible; just difficult to achieve. The view I’m defending is that knowledge is an aide to sensory experience and helps to refine it more quickly.

    1. Good points, Dwight. I’ll reply with one point per paragraph of yours.

      In distinguishing “what you taste” from “what you think you taste,” I wasn’t so much concerned with the possibility of error or bias about what you’re tasting as with the fact that you can be perceptually aware of a certain quality without thinking, right or wrong, about what quality it is.

      As for defying or otherwise going beyond expectations, yes, conceptually contrasting novel qualities with familiar ones can put them into a coherent space of possibilities and thereby enable you to nail down what is novel about them. But I’m not sure that that’s necessary for discovering these novel qualities or noticing their novelty. Maybe a feeling of unfamiliarity will do.

      I wouldn’t put too much weight on “concept” in the notion of recognitional concepts. You’re right — “another one of those” captures the idea. Compare recognizing someone you see from time to time whom you know nothing about, aside from how they look, other than that you see them from time to time.

      It would be interesting, though difficult and expensive, to put your claim about expert knowledge to the experimental test. Anyway, I’m not sure just what sorts of knowledge you think aid and help refine sensory experience, at least in the case of wine tasting. How does wine tasting differ in that respect from experiencing special qualities in sunsets, desert landscapes, sex, or roller coasters?

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.