Subjectivity in wine tasting

Wine tasting: subjective or objective?
Exploring the nature of wine perception

I’m guilty of it. When I’m presenting wine tastings to consumers, I’ll encourage them to trust their own palates, and not to be deterred by the seeming mystical ability of some wine writers to detect a shopping list of flavours, spices and fruits in a simple glass of wine. ‘There is no right and wrong in tasting wine,’ I find myself saying. ‘Your views are as valid as those of the expert.’ Effectively, I’m telling them that wine tasting is totally subjective.

But then in my working life, my colleagues and I act as though wine tasting were anything other than subjective. We give wines a numerical score; we argue about the merits of particular bottles; we take part in competitions that award medals to some and not to others; we think that our expertise in wine assessment is of a level where we can charge others for it.

So which is it? We can’t really have it both ways. To suggest that wine tasting is merely subjective is not intellectually honest in light of the way that we, as professionals, work. This conundrum is highlighted by philosopher Cain Todd in his recent book.[1] Equally, anyone who has been around wine for a while can’t really argue that wine tasting is utterly objective. So the goal of this article is to begin to explore just how subjective or objective wine tasting actually is.

Assumptions of objectivity
There’s a hidden set of assumptions underlying attempts at wine education, all the way from the most basic WSET (Wine and Spirt Education Trust) courses up to the tasting exams of the MW (Master of Wine) qualification. This nested set of assumptions concern the objective nature of wine tasting.

They begin with the assumption that the taste of wine is a property of the wine. They continue along the line that as we taste together, we are all experiencing pretty much the same thing as we taste, as long as we have a properly functioning nose and palate. There is the recognition that good tasters can have subjective preferences, but that these can be set these aside to facilitate objective wine assessments.

Then, when it comes to competitions and the act of wine assessment and criticism, this assumption of objectivity continues. Some notable wine guides and websites are in the business of charging readers for access to tasting notes and scores of wines. Because the wine world is now a large one, these sites involve teams of tasters. The belief underlying this is that this currency of a ‘professional tasting note’, when authored by a competent wine critic, is of some value because competent critics will read the wine correctly. It follows from this that it doesn’t really matter who authors the note from the team, as long as that wine is covered.

I’m increasingly convinced that there is no single correct way to read a wine. There are two reasons I have come to this conclusion. First, from working up from the physiology, psychology and neurobiology of perception, and second from working down from looking at wine assessment in practice.

Wine assessment in practice
I’ll start with the latter. It is plain to see that even the most experienced and acclaimed wine professionals disagree quite significantly in their assessments of wine. This is shown beautifully in the tastings section of the magazine The World of Fine Wine. Here we have teams of three experienced tasters looking at the same wines together. The great thing about these tastings, though, is rather than average out scores to reach some sort of consensus, the individual scores of each taster are given. Often there is a large variance.

If wine tasting were objective and there was one correct way to read a wine, then the only conclusion that could be made is that where tasters disagree significantly, one of them has got it wrong. But this is an uncomfortable conclusion that doesn’t chime well with experience.

Of course, there is such a thing as ‘getting it wrong’ in tasting wine, even if we are to take the view that there is a strong subjective element in wine tasting. Broad-brush categorizations of wine by quality level are appropriate and fairly robust among tasters.

For example, wine pricing varies dramatically, and such price differences would not persist if there was not such a thing as differences in quality levels. The word ‘quality’, however, needs to be used with care here, because we are entering the realm of aesthetics. Who gets to decide that one wine is great and one wine ordinary? Or that a wine is faulty? [After all it is quite common to find wines that are highly acclaimed by some but dismissed as faulty by others: for example, wines affected by Brettanomyces or with some reduction characters.]

Certain characteristics of wine are regarded as positive attributes, and contribute towards a wine’s quality level. We have to tread carefully here, because we are dealing in generalities, not absolutes. Factors such as balance, concentration, elegance, purity and complexity are seen as positive attributes. You might reasonably expect professionals, when faced with an array of wines of rather different categories, to be able to separate these wines out according to their quality level, in a very general sense.

This is a level that is fairly objective, and one which is loosely reflected in the price of the wine. Take for example, a simple Bourgogne rouge available for Ł6 in a supermarket, compared with a Grand Cru red Burgundy from a good grower and decent vintage, costing Ł60 in an independent merchant. Even a vaguely competent taster could attribute quality correctly here. If a taster pronounced the former to be a better wine, then by all accepted standards of quality in wine, they got it wrong.

But it’s hard to think of too many examples like this when we get to wines of more-or-less equivalent price. Different professionals apportion value to certain characteristics of wine in different ways. For example, I love the elegance and freshness of some of the ‘natural wines’ that have emerged in recent years, largely from lesser French and Italian wine regions; many others, whose palate I respect, just don’t get them. Some critics swoon over ripe, dense, dark, concentrated red wines with lavish new oak; others criticize these wines for being made in an international style. Some prefer their wines young; others like them very old, at a stage others would refer to as senescent.

These differences of opinion transcend preferences. Most professionals are able to have their own style preferences while still being able to appreciate the quality of wines that are not to their liking. There is, however, a level beyond preferences where professionals score certain wines down as unbalanced, acidic, light or clumsy when others rate these same wines very highly.

So divided are wine professionals about some wines that there are only two ways out. The first is to question the critical competence being displayed; the other is to recognize that wine assessments have a large degree of subjectivity to them. The former is a much more problematic course to take: it means war with your colleagues!

The physiology, psychology and neurobiology of perception
If critical disagreements in the absence of incompetence present problems to the objective view of wine tasting, then our emerging understanding of the biology of sensory perception is the nail in its coffin.

The old-fashioned, simplified view of wine tasting sees us as acting as measuring devices. The taste buds in our tongues and the olfactory receptors in our nasal cavities detect chemicals present in the wine, and our brains read out the information thus gained in terms of smells and tastes: the flavour of wine.

This is quite wrong. Our conscious perception of flavour is formed in the brain. In the interaction between us and the wine we taste, the impression we form is a conscious experience that involves the fusion of inputs from at least four different senses, coupled with some sophisticated brain processing. It’s a unified ‘representation’ that can’t easily be dissected into its component sensory inputs, which we commonly try to do with taste and smell.

Any readers who have ever walked a dog will be aware that for them, the world of smell is as important and dynamic to them as the visual world which we are so familiar with. In contrast, we humans live in a vastly diminished smell world, having traded our olfactory acuity for enhanced colour vision—an evolutionary change that took place during primate evolution. Many of the genes encoding olfactory receptors in our genome are now redundant. Rather than being important for orientation, social organization and mate choice, as it is for many mammals, our use of the chemical senses, taste and smell, is largely restricted to food choice—a reduced but still vital role.

Taste and smell begin with receptors in the mouth and nasal cavity that turn chemical information into electrical signals, which can then be processed by the brain. The nasal cavity contains olfactory receptor neurons which express, between them, around 2000 different receptors, each tuned to recognize different volatile odorants. But it’s wrong to think of our sensory systems as complicated measuring instruments, which give a read-out of the taste and aroma molecules that they encounter. This is a common misconception: that wine perception is a result of a combination of such linear inputs.

Instead, what the brain does is to model the world around us. Our sensory systems are bombarded constantly by a mass of information, which, if attended to uniformly, would swamp our perceptive and decision-making processes. So the brain is able to extract from this sea of data just those features which are most relevant. This is done by a procedure known as higher-order processing.

We experience is an edited version of reality that is based on the information that is most relevant to our survival and functioning. For almost all purposes it does no harm for us to think of the world around as revealed to us to be ‘reality’—indeed, life would become quite complicated if we operated any other way—but for the purposes of this article we should be aware that the version of reality we experience is an edited and partial one.

The higher-order processing in the visual system is the best understood for any of the senses. Scientists have worked out how visual processing extracts features of the environment that are most likely to be relevant. For instance, our peripheral vision is sensitive to motion: moving objects immediately stand out, because neurons are tuned to respond to them. This motion detection ability is much stronger in the periphery than it is in the central visual field. Faces are also likely to be significant cues, so our visual systems have special brain mechanisms for face processing.

Although it is less well studied, this sort of higher-order processing is also important in flavour detection. We are bombarded with chemical stimuli all the time and the brain has to filter this information so that only the important bits get through. It seems that much of the brain is dedicated to producing a suitably edited view of reality, just as the staff in a newsroom work hard all day sifting through the output of their journalists to produce a fifteen minute news bulletin for broadcast that evening.

So, when we taste wine, before we are even aware of it, our brains are doing quite a bit of complex processing of the information we are getting from this wine. In addition, there are inter-individual differences in the ability to detect certain flavour or aroma molecules. For example, one-fifth of people can’t smell the chemical rotundone, recently found to be important for the peppery aroma found in Syrah/Shiraz wines. And around a quarter to a third of the population seem to be quite insensitive to bitter tastes.

For this reason, it is only correct to say that the taste of wine is not a property of the wine, but a property of our perceptual apparatus, but strongly based on the characteristics of that wine. We bring something of ourselves to the wine tasting experience: our context, our biology, the information we have about the wine, our previous experiences of wine, and even our own internal state at the point of tasting the wine. Because in our sensory perception we are effectively modelling the world around us, our perception itself is largely subjective.

A recent study by New Zealand psychologist and sensory scientist Wendy Parr, working on the perception of Sauvignon blanc, shows that professionals and novices approach wine in quite different ways. ‘We have recently done a study that is quite cognitive,’ says Parr. ‘We gave people a large A1 sheet of paper, and had around 70 descriptors of Sauvignon that had come out of our previous studies. Each one was written on a separate sheet of paper with a sticky back. We asked people to draw a picture of a high typicality Sauvignon in a hierarchical structure, like a tree shape. We had a group of wine consumers and a group of wine professionals. People had to sit and think. They did this experiment twice: they came and did it totally from memory (we want you to think about a good example of a Marlborough Sauvignon), and they drew these pictures. They came back two weeks later and we gave them three Sauvignons that had been rated in a previous experiment as highly typical. They tasted these wines, and did the picture again. So this is a perceptive condition, versus a conceptual condition (doing it from memory). The findings showed some interesting results and differences. Some people had broad pictures of Sauvignon; others had more linear ones.’

The idea behind these two separate experiments is that the first condition was what is known as a ‘semantic’ condition, relying on top-down cognitive processes of memory and linguistic skill. The second is a ‘perceptive’ condition, in which tasting the wine takes place at the same time as a classification task: this is a bottom-up experiential process. The three questions being posed were: (1) would experts and novices employ the same descriptors in the same way?; (2) would expertise influence reproducibility across both semantic and perceptive conditions?; and (3) do both groups exhibit a shared concept of Sauvignon blanc?

The results showed that experts classified the properties of Sauvignon blanc according to their own internal descriptions of typical New Zealand Sauvignon blancs, while the novices were much less consistent in their descriptions. The experts had a stronger underlying cognitive concept of New Zealand Sauvignon blanc, producing stronger hierarchical trees. They were also more consistent among themselves: this higher within-group consistency indicates a shared conceptualisation of Sauvignon Blanc knowledge. The experts and novices used different levels in their hierarchical descriptions, with stronger super-ordinate nodes. That is, the top descriptors for Sauvignon are seen as more important.

From this Parr concludes that the experts use top–down processes in their wine judgements involving prototypes shared by all experts. For the novices, it seems that their descriptions are driven by bottom–up processes, and are based on the taste of the wine.

The notion that the way that professionals describe wine is ‘prototypical’ was also found in some (now famous) studies by Frédéric Brochet (here). Among other observations, his studies showed that in blind tasting, professionals first seem to decide what sort of wine it is, and then select descriptors to match that type of wine. In effect, their assessment is only partially based on the chemical nature (the reality) of the wine.

Conclusions: subjective, but to a degree
There are clearly problems, then, with any view of wine tasting as an objective exercise. There is a good deal of subjectivity involved in the tasting of wine. We bring a lot of ourselves to the tasting of wine. However, it would be a mistake to say that because there is a strong subjective element to wine tasting, that anything goes, and that there is little value in expertise. Or that there is ‘no right and wrong’ in wine.

While our perception may be largely subjective in its nature, there is a strong learned component to wine appreciation. Expertise plays a strong role in being able to appreciate great wine. While there is a hedonic component to the enjoyment of wine (‘this tastes good’), learning about wine, and learning to recognize and appreciate some of wine’s complex flavour characteristics, has the potential to enhance the enjoyment a great deal. And, clearly, despite the fact that our perceptions of wine are unique to us, when we share a glass together, we can usually have a good deal of common ground when we discuss the merits of the wine in question.

So, why discuss subjectivity and objectivity in wine appreciation at all? Because I think it is important that we have a more accurate – if more complicated – understanding of the theoretical basis of wine assessment. People are sitting exams with tasting elements to them, and these exams could have an important bearing on their professional lives. Yet the assumption of those setting the exams is clearly that wine tasting, as carried out by professionals, is an objective practice. As we develop our understanding of the nature of sensory perception, however, this assumption is found wanting. It is time for a more grown up understanding of the wonderfully rich and interesting practice of wine appreciation.

[1] Todd C 2010 The Philosophy of Wine: a case of truth, beauty and intoxication. Acumen Publishing