Is wine flavour an objective property?

Is wine flavour an objective property?
A response to an article by Professor Barry Smith

I have been thinking a bit about Barry Smith’s Nature article on flavour perception, and whether or not flavour is an objective property of a wine. Smith asks the question in his standfirst: ‘Is flavour an intrinsic objective property, or a subjective experience that varies from person to person?’ This is an important question, because it touches on the theoretical underpinnings of what many of us in the wine business do for a living.

If wine tasting is entirely subjective, then every opinion about wine has equal validity, and expertise is of little value. Everyone is their own expert. Critic recommendations are rendered so personal as to be redundant.

Smith’s argument presents (what I interpret to be) a three-part model of flavour perception. Here I will explore this arguments using wine as an example, although this could equally well apply to other foods and drinks.

State 1. First of all we have the chemical properties of the wine. This is an objective property of the wine: we can measure the wine’s chemical composition, and when we share a bottle together, the liquid in each of our glasses will have the same chemical composition. Some of these chemicals have tastes and smells.

State 2. Second we have the flavour of the wine as an objective property of the wine. The sum of the flavour active chemicals, working together, produces the wine’s flavour.

State 3. Finally, we have our own perception of this flavour, which is subjective. This is because we each have different biology, and different experience.

The neat trick here is the creation of this second state – the flavour of the wine as distinct from its chemical composition. This allows us to regard wine flavour as an objective property of a wine aside from its perception by individuals.

This chimes with the way all of us in the wine trade work. While it might seem democratic to insist that wine tasting is subjective (so people are free to like what they like), the way we behave indicates that we consider wine assessment to be largely objective.

We take part in wine competitions, where scores or medals are awarded. We share our tasting notes, often with scores attached. We take wine exams where tasting is part of the examination. We sell our expertise. We recommend wines to others. We discuss wines we taste together.

And what about how we approach tasting wine itself? I pour myself a glass and take a sniff, followed by a mouthful, and I contemplate it. I am trying to ‘get’ the wine. I am interrogating it. As I come back to it on repeated occasions, it reveals itself to me. Some days I seem to taste with more clarity than other days; the wine shows more of itself to me. My behaviour indicates that I believe the wine possesses flavour that I then try to assess.

All of these activities indicate that we consider the flavour of wine to be an objective property of the wine. According to Smith’s view, the taste is in the wine, and we try to ‘get’ this when we taste.

The three part model is also quite good at dealing with inter-individual differences in flavour perception. We all differ in our olfactory receptor repertoire, and also our sensitivity to different smells. A great example here is rotundone, the chemical responsible for the peppery smell found in some red wines. One-fifth of people can’t smell it, so their experience of a peppery cool climate Syrah will be quite different to mine. The three part model deals tidily with this by saying that these individuals are not experiencing the whole flavour of the wine.

But do we need this three part model in order to salvage some objectivity for wine tasting? And, more importantly, can it be defended as a concept?

The answer to both questions is no. Let’s deal with the second first.

The taste is not in the wine. Flavour is not a property of the wine, but rather of the interaction between the wine and the taster.

Consider the earth as it would have been gazillions of years ago, before animals existed. Chemicals existed then which we would now call flavour chemicals. But before animal life evolved, did any of these chemicals have flavour?

So is the flavour of, say, salt, a property of salt? Salt cannot have had a flavour before animals evolved. The sea would not have been salty because saltiness is a property our perception of the taste of salt conveys on the sea.

Chemical senses evolved because there was utility for organisms to respond in some way to the chemicals in their environment. Initially, bacteria with the ability to move (through possessing a flagellum) could move towards or away from chemicals in their environment; fast forward a long, long time and we see the development of sophisticated chemical senses such as those found in humans.

Our perception of flavour has been shaped through evolution. Chemical entities smell and taste because there is selective advantage for us to be able to smell and taste these chemicals. Our perception has been shaped by evolution. The flavour of chemicals is a property conferred on them by us.

Let’s consider colour perception. This is clearly subjective. While a red object has certain physical properties that make it look red to most of us in most conditions, we can change the context or lighting so that we don’t perceive red as red. I have a pair of sunglasses that have a very strong orange tint to the lens. When I first put them on, the world appears very odd indeed. But after a while, my vision adjusts and while there’s a slight orange warmth to what I see, it’s all a lot more normal.

I take my glasses off as I head indoors, yet rather than return to normal, all the colours are just as distorted as when I first put my glasses on. A spectral shift has occurred in my vision: I know that my shoes are brown, the grass is green and my pen is blue, and so the higher-order processing in my visual system makes an adjustment when faced with a change in the quality of the light. Somewhere between our eyes and our conscious experience of colour, some interpretation is going on. There’s an automatic white balance at work in our brains that is surprisingly effective.

It is almost as if we learn what red is, rather than red being a fixed property of red objects.

Wine flavour is a property of the interaction between the taster and the wine. What we experience as we taste wine is a perceptive event, created somewhere in our brains, prompted by the chemical properties of the wine, but interpreted by our subconscious and conscious processing of this information. It is impossible for you and I to have identical experiences as we taste wine together, because each of us is genetically different and has different past experiences of wine, which help shape our current experience. There is inevitably, therefore, some subjectivity in the flavour of wine.

If this is true, how can we rescue some objectivity for wine tasting? There are two ways. Knowledge influences flavour perception, and our sense of taste is plastic, designed to be able to learn to appreciate particular novel flavours.

Knowledge influences perception, and we have knowledge in common

I argue that a large component of our understanding of the flavour of wine comes from our knowledge of the wine. This shared body of knowledge helps to remove some of the difficulties caused by the fact that we can never have exactly the same experience of flavour.

Consider your first attempts to write tasting notes. Mine were very bad, and incredibly short. I learned to write notes by reading those of others, and discussing wine with others more experienced. I began to develop a language for wine; a cohort of descriptors that I could marshal in my attempts to describe my sensations, and which undoubtedly acted as hooks that I could hang my perceptions on as I attempted to get to grip with the wines I was tasting.

I tasted wines considered to be fine by others, and thus developed sensibilities for what constituted fineness in wine. This shared aesthetic system of wine is hugely important, and isn’t subject to the problems of subjectivity that bedevil the actual perception of flavour.

Knowledge itself modifies perception of flavour, and something we can share in common is the body of knowledge concerning wine. This helps compensate for individual differences in perception.

Taste is malleable: we learn to like

Now to the issue of the malleable nature of our sense of taste. Taste and smell, in large part, function to steer us towards foods that are good for us and away from those that are bad. There are reward systems, also: we gain pleasure from eating. It is not just a question of eating to alleviate hunger, but also eating because it is a nice thing to do. Certain smells and tastes are aversive, because we need to keep bad things out of our bodies, and avoid hanging around unhealthy places.

There is a selective advantage gained if we can make use of novel food sources that aren’t bad for us. Therefore we are able to develop a taste for something that is aversive the first time we try it. If we eat something with a distinctive flavour that makes us ill, then the smell or taste of that gets logged as something we should avoid. But if we taste something with an unusual flavour that doesn’t make it ill, we can develop a liking for that odd flavour.

This means that we have the capacity to change our tastes with time. Wine is a complicated flavour and interesting wines often taste quite challenging. As wine drinkers we develop a taste for wines. It is often these acquired tastes that are most enduring and appealing. This development of a shared taste can offset some of the problems of different taste worlds.

So I would argue that we can’t defend the three step model of flavour perception, where the flavour of wine is in some way separate from its chemical composition, and is then sensed differently by different tasters. But to reject this model doesn’t mean we have to take flight to utter subjectivity. We can salvage a good deal of objectivity. We acknowledge that each person’s biology, knowledge and prior experience will be important factors in shaping their perception of flavour. But we understand that where there is shared knowledge (an aesthetic system of fine wine appreciation which can be shared among many individuals) and shared experience (our malleable ‘taste’ changes with experience; we learn to like the wines we are presented with that are highly regarded by experts), these help offset the differences in perception that would otherwise render the practice of wine tasting and assessment a very personal, private act.