A while back I wrote a lengthy piece examining the thorny issue of 'terroir', which provoked some strong responses and robust discussion (link here). One of my main points was that it's quite difficult discussing a concept that means different things to different people. Consequently, I thought it might be worth trying to nail down some sort of working definition for 'terroir'. So here's my stab at this -- and as usual I'd welcome your input and criticism.

Terroir as a philosophy
First off, though, I'm aware that some readers will find my approach puzzling and even frustrating. Why is it that I'm so anxious to try to define things, pin down mechanisms, and get all scientific? For many, terroir is an ethos or philosophy. It's a unifying theory encapsulating a certain approach to wine that encompasses the almost metaphysical circle of soil, nature, appellation and human activity. As a philosophy it clashes with the reductionist new world grape-variety and 'winemaker'-dominated approach. For these people, the mechanisms and scientific underpinnings of terroir are somewhat irrelevant. I'm not unsympathetic with this approach, but I'm coming at this from a different angle. I have a fascination for how things work, and I firmly believe that in the right hands, understanding can bring progress, but dogma rarely does -- unless, of course, you are happy in the belief that the sun revolves around the earth.

Soils matter
Consider the following theoretical scenario. Producer X has a vineyard on a south-facing slope, which she has spilt into ten different plots according to differences in soil type, but all of which planted with the same grape variety/clone/rootstock. She then vinifies each of those plots separately before blending, but in an identical fashion. Would you expect the wines all to taste different? I would. Of course, in the real world it's likely that different plots would not only differ in soil composition, but also other microenvironmental factors such as sun exposure, average temperature and rainfall. In reality, many producers do vinify grapes from different plots separately, and these soil and microenvironmental differences are commonly reflected in the character of the wines from each. These differences extend to the regional level: I've tasted wines made by the same winemaker from Syrah grapes grown in different regions: as you might expect, the wines tasted completely different. To my mind, these sorts of site-specific or regional differences are at the heart of terroir.

Just how do soils affect the flavour of the final wine?
Firstly, there is the purely physical factor of drainage. This is likely to have a significant impact on the fruit quality: vines are known to prefer well-drained soils. Another physical property of the soil has to do with how it reflects the sunlight back onto the vine. Darker soils will absorb and radiate heat better, and there's the oft-cited example of the galets (large, slab-like pebbles) in the vineyards of Châteaneuf du Pape that act as storage heaters, soaking up the heat of the day and then slowly radiating it back to the vines. Finally, the chemical composition of the soil is likely to be important. Mineral ions in the rhizosphere have been shown to alter plant metabolism, and although to my knowledge this has not been specifically studied with grapevines, it is probably also true for them. But I'd exclude from my definition of terroir the concept that soils can directly influence the character of a wine, for instance, by flavour compounds in the soil being directly translocated to the grapes. This sort of mechanism is not impossible, but it does seem to be hugely implausible. I'm not a root physiologist, but I do have a PhD in plant biology, and I've yet to hear a convincing explanation for how soil components can directly alter the flavour of a grapes and hence the finished wine. This is a subject I've discussed in (rather too much) depth in my former article, so I will point readers to this for a more detailed coverage. Suffice to say, stony, earthy or mineral flavours in a wine are not necessarily 'terroir' notes. If a wine grown on chalky soil tastes chalky, it's an unjustified leap of faith to say that this is the 'terroir' speaking. The minerals in the soil may be fortuitously imparting chalky notes to the wine indirectly by altering the vine's metabolism, but you'd only be able to tell this by comparing this wine with one made from the same grapes grown on different soils and vinified in a similar way.

Other exclusions
I'd also exclude winemaking differences from my definition of terroir, otherwise the subject just gets far to complex and woolly. I'd add here that certain winemaking practices can completely obscure any differences in terroir, which tend to be best expressed in more traditional-styled wines. Full flavoured, highly extracted and overoaked new world reds tend to taste quite similar even from continent to continent!

Implicit in the above is the idea that terroir expresses itself in the differences between wines. It isn't possibly to taste the 'terroir' in a wine, unless you are comparing wines produced from different vineyard sites, or you have prior knowledge of the wine. For example, you may know that Chardonnay grown on one soil type in Burgundy has particular characteristics that set it apart from Chardonnay grown on a nearby vineyards with different soils. It would be fair to say that these characteristics are from 'terroir'. Likewise, if you are faced with 10 wines produced by the same producer from different vineyard sites, the terroir would be apparent in the different flavours of each.

My definition
So here it is: my working definition. I'd maintain simply that terroir consists of the site- or region-specific characteristics of a wine. I think this is broad enough to encompass all the useful potential meanings of the term, yet narrow enough to exclude the controversial or plain misleading ones. What do you think?

Jamie Goode
wineanorak@i.am