Wine Tasting and Objectivity

by Dwight Furrow

Wine judgingThe vexed question of wine tasting and objectivity popped up last week on the Internet when wine writer Jamie Goode interviewed philosopher Barry Smith on the topic. Smith, co-director of CenSes – Center for the Study of the Senses at University of London's Institute of Philosophy, works on flavor and taste perception and is a wine lover as well. He is a prominent defender of the view that at least some aesthetic judgments about wine can aspire to a kind of objectivity. His arguments are worth considering since, I think, only something like Smith's view can make sense of our wine tasting practices.

The question is whether flavors are “in the wine” or “in the mind”. On the one hand, there are objectively measurable chemical compounds in wine that reliably affect our taste and olfactory mechanisms—pyrazines cause bell pepper aromas in Cabernet Sauvignon, malic acid explains apple aromas in Chardonnay, tannins cause a puckering response, etc. But we know that human beings differ quite substantially in how they perceive wine flavors. Even trained and experienced wine critics disagree about what they are tasting and how to evaluate wine. This disagreement among experts leads many to claim that wine tasting is therefore purely subjective, just a matter of individual opinion. According to subjectivism, each person's response is utterly unique and there is no reason to think that when I taste something, someone else ought to taste the same thing. Statements about wine flavor are statements about one's subjective states, not about the wine. Thus, there are no standards for evaluating wine quality.

The problem with the subjectivist's view is that no one connected to wine really believes it. Everyone from consumers to wine shop owners, to wine critics, to winemakers are in the business of distinguishing good wine from bad wine and communicating those distinctions to others. If wine quality were purely subjective there would be no reason to listen to anyone about wine quality–wine education would be an oxymoron. In fact our lives are full of discourse about aesthetic opinion. The ubiquity of reviews, guides, and like buttons on social media presupposes that judgments concerning aesthetic value are meaningful and have authority even if enjoyment and appreciation are subjective. In such cases we are not just submitting to authority but we view others as a source of evidence about where aesthetic value is to be found. Wine tasting is no different despite attempts by the media to discredit wine expertise. So how do we accommodate the obvious points that there are differences in wine quality, as well as objective features of wines that can be measured, with the vast disagreements we find even among experts?

The first important distinction to make is between perception and preferences. As Smith points out:

I think when critics say it is all subjective they are saying your preferences are subjective. But there must be a difference between preferences and perception. For example, I don't see why critics couldn't be very good at saying this is a very fine example of a Gruner Veltliner, or this is one of the best examples of a medium dry Riesling, but it is not for me. Why can't they distinguish judgments of quality from judgments of individual liking? It seems to me you could. You know what this is expected of this wine and what it is trying to do: is it achieving it? Yes, but it's not to your taste.

This is important but all too often goes unremarked. Wine experts disagree in their verdicts about a wine and in the scores they assign. But if you read their tasting notes closely you will often find they agree substantially about the features of the wine while disagreeing about whether they like them or not.

For example, consider these two tasting notes regarding the 1990 Chateau Margaux from two top wine critics:

JAMES LAUBE: “A tight, hard-edged and unyielding young wine. Some cedar and currant flavors attempt a coup on the finish, but they're tightly wrapped in tannin. 86 points.”

JAMES SUCKLING: “Slightly dumb now. Ripe, almost raisiny aromas and flavors that develop a minty, menthol accent. Full-bodied and rich with loads of tannins. Needs time. Better after 2005. 90 points.” [Thanks to Bob Henry in a comment thread for these notes.]

If you look at their scores and their overall verdict it would seem they disagree substantially. But in fact they largely agree about the features of the wine and differ only in their preferences.”Tight” and “dumb” mean essentially the same thing–-the wine is not very expressive. “Tightly wrapped in tannins” and “loads of tannin” again have similar meanings. Suckling mentions over-ripe qualities which appear early in the taste experience; Laube focuses on the finish. They are clearly focusing on different aspects of the wine. But the essential descriptions “closed” and “excessively tannic” are shared by both critics. Both agree that the wine needs more time, Laube calling it “young” and Suckling saying “needs time”. What they differ about is how much to discount the scores given these factors. Suckling is more forgiving than Laube. The disagreement is over a preference for wines that are ready to drink vs. wines that need age.

No doubt preferences are subjective; but it doesn't follow that perceptions are. Of course, critics sometimes disagree about what they perceive as well, but those disagreements are less extreme than some of the commentary would have you believe.

There is an important philosophical question here: Can you separate how something tastes from whether you like it? It seems that we can. As Smith points out, if we could not separate them we could never acquire new preferences. If you hated broccoli as a child but like it as an adult you must have been able to separate taste from liking at some point. That is a persuasive argument although it could be argued that there was just something about broccoli you didn't notice as a kid that has now come into focus. Did broccoli as a child taste the same as broccoli as an adult, the difference being you now like what you hated in your youth? Or does the broccoli taste differently as an adult? The question is not settled but is important to answer. If we can separate what we taste from whether we like it, then we can view wine criticism as involving two aspects– description which legitimately aspires to something more or less objective, and a verdict which will rely much more on personal taste. A good critic then should be able to keep the two tasks distinct and communicate that distinction to readers.

In the second part of his interview Smith briefly lays out a model for how to think about objectivity in wine tasting. But first some background to set up the problem he's trying to solve. Smith doesn't define what he means by “objectivity” but I suspect he has in mind something like this: An objective judgment is a judgment that accurately tracks features of the external world. Thus, an objective judgment satisfies conditions of correctness, where evidence points to reality to help explain how things appear to us.

The question is whether our judgments about wine quality are at least sometimes objective in that sense.

On the one hand, flavor depends on molecules, and taste would seem to be a matter of our sensory mechanisms accurately tracking these compounds in a wine. To the extent a wine taster is accurately tracking chemical constituents of the wine she is tasting objectively. But those who think wine is subjective seem to think our sensory mechanisms aren't reliably connected to objects in the world. How things seem to us individually is just how things are. There is no right answer to what a wine tastes like and no standards for judging wine quality according to subjectivists.

It would seem that in order to answer this question about objectivity we need a scientific inquiry that would connect the molecules in wine to the subjective impressions of tasters. But given what we know about the mechanisms of taste, this is a hopeless task. Taste is not a direct perception of a compound but involves very complex mental processes that link several perceptual modalities into a unified impression of flavor. Taste, smell, and tactile impressions combine with visual and auditory stimuli as well as input from emotions, beliefs, and mood to give us an impression of what something tastes like. It thus seems like flavor is formed in the mind and is subject to a variety of influences unique to each individual. Moreover, the subjective impressions reported by tasters are all over the map and produce contradictory results not only between individuals but for individual persons from one time to the next. There is room for individual variation because people have different thresholds for detecting molecules and well as different histories, associations, and environments that influence what they taste.

The idea that there could be laws that connect molecules in wine to these unstable individual experiences seems implausible. Yet, in the absence of these laws, it would seem there is nothing else for wine tasting to be about except how individuals form their own subjective flavor perceptions. Individual sensory perceptions seem so unstable and widely variable that it looks like the subjectivists have won this debate.

But not so fast.

Smith argues that we should think of flavors as intermediaries between compounds in the wine and our individual reactions to them.

“What I say is, you need an intermediate level. We need a level in between the chemistry and the variable perceptions, and this is flavour. Flavours are emergent properties: they depend on but are not reducible to the chemistry. Then these flavours are things which our varying and variable perceptions try to latch onto. Each flavour perception is a snapshot of that flavour. We don't even want to think of it as static: we want to think of a flavour profile: something which itself evolves and changes over time. As a professional taster you are taking snapshots in each of your tastings and trying to figure out what the flavour properties of that wine are that will continue to endure and alter as the wine ages. How would it taste if it was a degree or two colder or warmer? You make predictions and then you can go back and sample it later and say, I was right: I figured that it needed another hour in the glass and needed to be one degree warmer and it would change like this. The thing about which you are making the predictions is flavour. This is what depends on but is not reducible to chemistry. Now you have two tough jobs instead of one, with this intermediate level. One task is to say, what is the relationship between the chemistry and the flavours that emerge. The second task is what is the relationship between individual flavour perceptions and flavours? These two jobs need to be done independently, but they have to reach the same terminus. Having this intermediate level gives you the job of saying how does my individual experience as a taster lock on to flavour, and how does the chemistry give rise to flavour. Don't try going from the chemistry to perception, you need that middle level.”

Smith is claiming that in addition to the objectively-observable, chemical constituents of the wine and the private, individual taste experiences of tasters, there is a third, intermediate object that is just as real as the chemical properties of the wine. This intermediate object is a dynamic flavor profile that each taster is trying to identify, yet is independent of our individual psychological experiences. These intermediaries are caused by the chemical compounds in the wine but there is no purely chemical description of them since they are emergent properties that depend on human perception. They are the intersection of chemical properties and human perceptions but not reducible to either. Thus, our individual perceptions are tracking something—this intermediary object called flavor that is relatively stable and not dependent on individual perceptions.

Smith seems to have in mind something like this. When the chemical properties of wine interact with human perceptual mechanisms flavor emerges. That flavor is not dependent on my perceptions or your perceptions. If you or I were to cease existing, the flavor of a wine would still exist as long as human beings can taste wine. In that respect flavor is like a dollar bill. A dollar bill is real, as real as a tree or rock. But its value as a medium of exchange depends on human practices. If human beings didn't exist it would just be a piece of paper. Yet its value is not reducible to the chemical constituents of the paper nor is it reducible to your individual willingness or my individual willingness to accept it as payment. It is what John Searle calls a social fact which is as objective as any other fact.

The upshot is that individual variations in taste experiences, whether between individuals or between taste experiences by the same person, are different ways of perceiving the same thing. Each taste experience is a snapshot of a larger whole that is the wine's flavor. No one ever experiences the wine as a whole; it's never present all at once but transcends each individual taste experience. When I taste apricot or perceive balance, I'm sensing a larger whole that is given to me via those impressions.

Wines contain many different flavors and tasters may differ as to which flavors are prominent. I might think a Cabernet smells of smoke; you might say it smells of vanilla. We may both be right in that the wine has the power to cause impressions of smoke in me and impressions of vanilla in you given our individual differences. Different descriptions are not necessarily incompatible descriptions; there is no reason why a wine could not smell of both smoke and vanilla. Even more complex properties such as balance or finesse can admit of variation. There are several ways in which a wine might be in balance or exhibit finesse or depth, some more accessible to individual tasters than others.

Of course, the question that still lurks is whether someone can be wrong about what they perceive. If there are multiple ways of perceiving a wine's flavor in what sense can someone be accused of making a mistake? The fact that there are multiple ways of perceiving a wine does not entail that anything goes and that all judgments are correct. How then do we decide which critic to listen to or which judgment to follow? How do we know whether our individual perceptions are tracking this intermediate object?

These are complex questions. What we need in order to answer them is a conception of what counts as appropriate evidence for an aesthetic judgment that can be confirmed through experience and that can help correct us if we've made a mistake.

Smith's suggestion is that we can make predictions about the wine and how it might be perceived under a variety of circumstances. If those predictions are confirmed that is an indicator that we are tracking the wine's flavor. I think this is in the right direction although I worry that predicting how wines will change is subject to too many unforeseeable variables to be reliable. However, there is more to be said about the proper approach to any aesthetic object that Smith's account helps us to understand. It will not be via laws, principles, or rules that we can confirm judgments but through proper critical practice.

In providing judgments that we expect others to accept and endorse, we are only bluffing if we are not part of a publically accessible dialogue with a background of shared normative assumptions giving our claims relevance and persuasive force. This means critics must have a commitment to allowing the wine (or the painting, or piece of music) itself to pull us out of our limited, subjective points of view and strive to occupy a shared aesthetic universe if only for a moment. In other words, critics in striving for objectivity must listen to each other and strive to understand alien points of view. Only such a practice will tell us if our views are too parochial, if there is some dimension of flavor (in Smith's sense) that we are missing. Our ability to occupy multiple points of view reinforces our pursuit of objectivity.

If our palates (or aesthetic sensibilities) are as sensitive as we claim, this exploration of flavor will yield new insights. Beyond mere enjoyment critics should be seeking a greater capacity to appreciate beauty and aesthetic value in all its forms. Although there are some biological limitations and historical constraints on what each individual can taste, these limitations and constraints are not immutable or immune to education. Of course, in the end our judgments may not converge with those of other critics. There may be some flavor profiles I can't grasp, certain aesthetic sensibilities that are inaccessible to me. In the end there are judgments that will fall well outside any attempt to achieve critical consensus because, from first-hand experience, most critics cannot make sense of them. But to make these critical judgments we do not need to suppose there is only one way to experience something nor must we assume that all judgments are equally valid. Both of these assumptions are toxic to a vibrant critical culture.

However, what we must assume is that our critical judgments are reaching for something out there, that there is value yet to be discovered of which our subjective preferences have not yet taken measure. This is what Smith's model accomplishes. It provides grounds for rejecting subjectivism about wine tasting by clarifying what wine tasting is about. Wine flavor is not something that happens only within each individual taster. It is an objective feature of a world in which human perceptual mechanisms function in a particular way. When those perceptual mechanisms are functioning properly and sources of bias are reduced that interfere with their functioning, there is no reason to think human beings are incapable of accurately recognizing wine flavor despite substantial individual variation. It is difficult to do so and we make many mistakes. All individual judgments are defeasible and judgments about personal preference have no claim on objectivity (which includes wine scores). But it leaves much we can say about which flavors are present, whether a wine is in balance, has complexity, a long finish, finesse, is tannic or soft, etc. Thus, the disagreements among critics to the extent they are not about mere preferences are not evidence that wine tasting is nonsense but evidence that all dimensions of a very complex and subtle object are being explored. Such disagreements are what we should expect if indeed wine is the complex, multidimensional object worthy of the devotion it inspires.

For more ruminations on the aesthetics of food and wine visit Edible Arts.

Like what you're reading? Don't keep it to yourself!
Share on Facebook
Tweet about this on Twitter
Share on Reddit
Share on LinkedIn
Email this to someone