06 March 2009

Somewhere, a place for us

[Blair Walter inspects his subsoil]Jamie Goode laid the gauntlet down. Of course, he did it on Facebook, so I doubt most people did more than trip over said gauntlet on their way to superpoking someone. Nonetheless, he got my attention…which might just mean I spend too much time looking at the ground, wary of things over which I might stumble. Anyway, this is what he said:

Terroir matters but what exactly is it? We need a good definition.

Right now, anyone who’s been around the online wine fora block for a few years (or decades, in my case) is rolling their eyes. Terroir is one of the trifecta of grossly overworn subjects, along with the efficacy of scoring wines and their ever-escalating cost, that has been so thoroughly masticated that there’s absolutely nothing new to say, though there’s an ever-revolving crowd of newbies to say it. Long-timers know each others’ arguments and positions by heart.

So why would I – one of those eye-rolling long-timers – want to dip my toes once more into this exceedingly tepid and turgid water?

Blame Jamie Goode.

Usually, the definition offered up is “somewhereness” (I prefer “placeness”). Both are a little un-rigorous. They tell us what terroir means, but not what it is. I think we can do better.

Let’s start with the fundamentals. I’ll begin with one of the most controversial things I could write: terroir exists. It’s true that not everyone believes this. “Terroir is bullshit,” claims one well-known California winemaker of my acquaintance, and he’s hardly alone in saying so. “Terroir is marketing,” claim a number of his brethren elsewhere in the Golden Two-Buck Chuck State. (I’m sorry, was that snarky? Hey, it’s a blog. Snark is what we do.) And so forth. If terroir doesn’t exist, then everything that follows is a waste of time. It still might be. But I think that the definition, or perhaps the argumentation leading up to it, will actually take care of this foundational problem. Judge for yourself, later.

Much later.

Continuing with the fundamentals, wine is a construct made with a defined number of inputs. Only two, in fact. They are:

  1. the grape(s)
  2. the winemaking

You’ll notice the distinct lack of the word “terroir” in that list. Not to worry. Because the grape, too, leaves the vine having accepted its own collection of inputs:

  1. the grape(s) inherent characteristics
  2. the viticultural choices made by the grape grower
  3. all other grape-changing effects not produced by human intervention

That last part includes the weather over that growing season. But it also includes longer-term meta-effects. Anything that changes the chemistry of the vine (and thus the grape) in any measurable way would be one of these meta-effects.

The meta-effects are terroir. Well…more or less.

What meta-effects? The list is well-known, if not necessarily universally agreed-upon. Non-comprehensively, these include overall mesoclimate (this is the proper term, not the usually-deployed “microclimate”), anything that effects water retention, sun exposure, heat retention, soil chemistry (and thus soil type(s)), and so forth. Broadly speaking, these meta-effects can be broken down into two categories: those below the surface of the ground, and those above it.

For some, we’ve gone far enough, and with a little more specificity in that list will have defined terroir in a satisfactory fashion. But if that were all there is to it, Jamie wouldn’t have dropped his gauntlet on Facebook’s scuffed-up floor. Very, very few people argue that what I’ve just iterated doesn’t exist, or doesn’t have an effect on grapes…and those that do argue the point seem unacquainted with basic agriculture. Any farmer will tell you, without necessarily knowing or caring about terroir, that these effects both exist and affect their practices in a fundamental and inexorable way. However, we need to continue, because not everyone is willing to stop their definition in the realm of the sciences, or at least the evidentiary. For this we can probably blame the French.

Part of the definitional problem is that terroir is a borrowed word with no obvious English equivalent (that is to say, there’s a translation, but it’s not what we mean when we talk about terroir in the context of wine). To many French folk, the word is imbued with much more than climate and chemistry. Anyone who has seen a “produits du terroir” sign while driving the French countryside knows this; the word has quasi-historical implications, and certainly is laden with a measure of cultural baggage that is, to the French, inseparable from geography. Some even talk about human inputs as being part of terroir, though a rigorous definition of this type would have to include everything from training methods to complete site (re)constructions of the type practiced by certain mega-wineries, and that’s most certainly not what the human-input proponents mean. Others claim that people themselves are part of terroir, arguing that the majority portion of the word “viticulture” is not “viti,” but “culture.”. Traditions are sometimes mentioned. As are other living creatures. And so on.

[Geisberg & Osterberg over Ribeauvillé]While these diversions appeal to the romantic in me, they pose a definitional problem. I’m going to solve that by insisting on another fundamental precept: terroir must have clear, scientifically-measurable, and scientifically-repeatable boundaries. If it does not, then a definition is going to be impossible, because it can mean different things to different people. Since we’re here to define terroir, that’s not of much use. Moreover, imposing a structure on the concept of terroir doesn’t preclude the consideration of any of these other categories of influence. Far from it. I certainly think that human inputs exist (I’ve mentioned some of the forms already), and I think that traditions absolutely matter for certain wines, though not for others. What remains to be determined is whether or not they have anything to do with terroir.

Tradition has a more specific name in wine lingo: typicity. Some think that typicity should fall under the definition of terroir, and most who think that live in the Old World, in places where the traditions of wine run deep into the centuries. However, following from the determination that terroir must have scientifically-measurable and scientifically-repeatable boundaries, typicity cannot be part of terroir. Why not? Because while there are components of typicity that can, under certain circumstances, be derived from the immutable qualities of grape varieties and terroir, those components are not the entirety of the concept. Winemaking – practices and style – must also be considered. Thus, taken as a whole, typicity isn’t inherent, it’s artificially imposed. And remember that there are wines whose typicity is a matter of great debate among their advocates. Does a Cotat-produced wine from the Mont-Damnés really taste of Mont-Damnés despite being so different from other producers’ interpretations? Does Brun’s “l’Ancien” taste like Beaujolais or not? How much sangiovese is too little in a Chianti, or tempranillo in a Rioja? There are also wines whose typicity is a matter of arbitrarily choosing date ranges; is Bordeaux “typically” blended with Hermitage or not, and what sites and/or grapes does a given label actually comprise? Is a Mâcon botrytized or not? Montrachet: sweet or dry? What is a Rasteau “supposed” to be? Are varietal bottlings or blends the true Alsatian tradition?

I could go on for a long while, but these examples are all well-known among controversy-loving wine geeks. And they all serve to illustrate the basic impermanence and subjectivity of typicity. Thus, it cannot be part of our definition of terroir.

What about culture? Again, which culture? Greek? Etruscan? Roman? Roman Catholic monastic? Germanic? French? Many cultures may have contributed to the viticulture of a long-planted site. But the cultures were different. And even within narrower groups, culture is not steady-state, it’s a shared environmental construct in constant motion. To say that terroir includes culture is to introduce a permanent variable into the equation. That may satisfy a romantic urge, but it’s of little use when trying to construct a definition with any utility or rigor.

How about creatures other than man? It depends: are said creatures a permanent influence on a given set of vines, and – this is important – in a way that affects grape chemistry, or are they transitory? Most pests, like glassy-winged sharpshooters, or the phylloxera louse, are transitory; they (or the parasites they carry) may affect grape chemistry, but it’s hard to call something a permanent influence when it eventually kills the vine, and when it’s fully mobile under its own power. Grape-noshing birds might seem to be a permanent fixture, but they don’t influence grape chemistry, just quantity (if they ate underripe grapes, performing a sort of avian green harvest, then maybe we could include them…though maybe not, since unless they’re trained their influence is hardly predictable, and thus not scientifically measurable).

One biological entity might be part of terroir, though, and that’s botrytis cinerea, the fungus known in certain forms as noble rot. That botrytis affects grape chemistry is unquestionable, its effects are predictable, and thus the effect is scientifically-measurable. And if a site has a clear and permanent predilection to be affected by botrytis (or the opposite), then it can be said to be part of the site’s inherent characteristics…thus, more or less scientifically-repeatable. So it should be part of terroir, right?

[St-Jean-de-Minervois]One objection is obvious: if terroir is to be scientifically-repeatable, its form should be a constant, or close to it. Thus, if a vineyard isn’t botrytis-affected every single year, botrytis cannot be part of terroir. Right?

This seems an easy escape clause from what appears to be a thorny definitional issue. In fact, it’s too easy, because it misapplies the concept of terroir. In any case, there’s a better reason to eliminate botrytis from the list of terroir-influencing sources. Bear with me here…what follows will seem to be a bit of a diversion, but it will eventually come back to and explain this point.

One more thing we can probably blame the French for goes right back to one of those frequently-heard Californian objections to the concept: terroir is marketing. Because the fact is that, for many, it is very much a point of differential and qualitative marketing. And it has been used in both positive and negative ways. The latter is what gets other winemakers’ backs up, because some will insist that only certain wines “show terroir”…those wines usually being those with a long tradition of site, varietal, and winemaking continuity, and thus (obviously) few of which are placed anywhere in the New World.

Others will point to a generalized inability to, in controlled double-blind conditions, inerrantly identify specific terroirs as proof that terroir does not exist. Two things are worth mentioning here. First, anyone with enough experience has met tasters – many, but certainly not all, of them grape-growing winemakers – who seem to be able to identify sites with surprising regularity. Sometimes, they can even do this by tasting the grapes themselves. A remarkable talent? Maybe. More likely, it’s long familiarity. The wine generalist may not be able to reliably differentiate Schlossberg from Sommerberg, but a specialist in Alsatian wine will be better, and someone who grows grapes on those sites will often prove to be quite adept. Second, however, is the more fundamental objection: the “proof” thus demonstrated by such tastings is not that terroir does or doesn’t exist, but that it is not always useful for the consumer of wine. Again, hold onto that thought for a moment or two.

The (indeed highly marketable) idea of a “terroir wine” is an entirely different concept from the basic definition of terroir. Note that nowhere in my proposed definition have I mentioned a requirement for terroir to be organoleptically identifiable. A wine may or may not show its site-derived characters , just as it may or may not show its varietally-derived characters, and just as its winemaker-derived characters may or may not be obvious. In no case would a taster, having failed to discern certain qualities in a wine, deny the existence of the grape(s) or the winemaker. Yet for some reason, when terroir is not discernable, they’re perfectly willing to deny its existence. This is remarkably insensible; remember our farmers, who would weep at the notion that one plot of ground is pretty much the same as any other. If farmers acted on such an absurd belief, a lot (more) of them would be out of work.

This brings us to the key point: it’s not that terroir is useless or irrelevant in terms of wine appreciation, but that its actual point of application in the process that leads from vine to wine is wholly within the vineyard. In other words, in the purview of the farmer, not the taster. There may be terroir for the taster to discern, or there may not, but there is always terroir for the farmer to discern and deal with.

So to return to our moldy conundrum, is botrytis part of terroir? From the farmer’s perspective, the question is not about botrytis, but about an affinity for botrytis. That predilection is something with which the farmer must deal, compensating (or not) according to the demands of the wine, just as a mesoclimate-derived predilection for extreme August heat is a factor with which the farmer must deal. Botrytis can be prevented or encouraged, yes, but the predilection itself cannot be eradicated without significant changes to other aspects of the terroir (proximity to water, altitude, water retention, diurnal temperature effects, etc.). In other words, the terroir is not the fungus itself, but the predilection for the fungus…a property of the site, not of the mold. This keeps us safely within the boundaries of our earlier assumptions; the yearly presence of botrytis may or may not be fully predictable and thus not scientifically-repeatable, but the chance of a site’s embrace/rejection of botrytis most certainly is predictable and scientifically-repeatable. What follows from this seems to be a firm, clear standard: no biological entities aside from the vine itself have terroir effects.

[Craig Camp at Anne Amie]Right?

Well, what about the wee beasties in the soil itself? Worms, bugs, bacteria, and so forth…are they or are they not part of the terroir of a site? And what about grasses and other things planted in and around the vineyard? How about eucalyptus trees in the neighborhood, from which oils adhere to grapes, affecting the taste of the resulting wine? And how about pollen from nearby lavender fields, often cited as a “natural” flavorant in wines from Provence?

The fashion in which we dealt with fungus shows the way forward. It’s certain that the physical and chemical makeup of the soil is indeed affected the creatures living in it (and, it might be added, by cover crops and other in-vineyard plants of that nature). Since we previously asserted that soil chemistry is part of terroir, surely these chemistry-modifying biological entities are also part of terroir.

Again, no. Botrytis affects the chemistry of the grape in predictable and measurable ways only given its presence…which is not assured, only predicted. The same holds true for things living in the soil; they can move away, or be killed by means physical and chemical, or experience a growth spurt one year and a decline the next. In other words, they’re a variable influence, like the weather. The only aspect of their existence that may be part of a site’s terroir is, as with botrytis, a predilection of that site to encourage or inhibit such biological entities.

In both cases, the key point is that our fungi, annelids, bacteria, and so forth are an effect of terroir, not a cause thereof. They exist, or not, as a response to the site…just as the vines themselves do. And while it’s true that they also may affect the site, the same is true of vines, whose questing roots may change the physical nature (and thus the water retention, and as this proceeds over a very long time the geological composition) of a site. Yet the vines themselves are a response to the terroir, not the terroir itself.

Which brings up another question. Doesn’t the preceding demonstrate feedback effects that suggest terroir is an evolving system? Yes. Without external management (which is in direct opposition to the concept of inherent terroir; management is man-made), it is impossible to think that a site does not change over time. Soils change. Vines are uprooted, and their younger replacements’ roots access different soil realms, leading to different vine chemistry. Mesoclimates change, not least in response to anthropogenic climate effects. Farming methods change, causing chemical and biological discontinuities in the soil as viticulturalists adopt, then abandon, various treatments and theories. Weather “events” and regular old erosion change entire vineyards, permanently. And as the previous paragraph demonstrates, the biome created by a given terroir has its own inexorable effect on the terroir.

So how can there be terroir if there’s no continuity? Well, remember what I wrote earlier: terroir is not tradition. Continuity is not a foundational requirement for the existence of terroir. Identity is…but even then, the identity that matters is an agricultural one, not the kind required by a taster for the purposes of identification. Again, terroir is about farming, not tasting.

[Schoenenbourg]This is, I think, a hard mindset for people to accept. Of what use is terroir if 1) it’s not about identifiable qualities in wine, and 2) isn’t even a consistent factor?

The response to this will begin to sound familiar: this isn’t a significant question, because terroir isn’t about tasting. That there’s enough identity and continuity for some experienced tasters to identify some terroirs is both a marvelous thing and a demonstrable truism, and in fact without identity and continuity as expressed in finished wine we wouldn’t be having this conversation in the first place, because no one would care whether or not there was such a thing as terroir. But the entity to which terroir actually matters is the grape hanging on the vine, not the person putting his nose and lips to a glass. Everything else is a mere ancillary benefit, not the effect itself. Terroir is of tremendous utility to the grape, even if it lacks the consciousness to know it.

Oh…and as for eucalyptus, lavender, and the like? An apparent vineyard signature, perhaps (until the offending plants are cut down, or the wind shifts), but an external influence no more intrinsic to the site than a “pool” of humidity birthing dormant fungal spores, or farmer spraying fungicide. So no…not terroir.

Moving on…

Here’s another fundamental concept that follows directly from the above discussion of marketing: all sites have terroir. All sites. Despite what those who which to use it as a wedge marketing term would like you to believe.

“But what about wine X, or Y? There’s no terroir there!”

What did I just say? “All sites.” Terroir is in the vineyard. Whether or not it is in the wine is irrelevant to its definition (though a given taster may care about this very much; I, myself, have a general preference for wines that reveal terroir). The most industrial multi-site blend comes from sites that have their own measurable terroir, even if the only thing discernable in the finished wine is the chemical stew used to bludgeon unpalatable grapes into commercial submission. (Sorry. Again with the snark.) Which is another way of stating yet another fundamental concept: quality has nothing to do with terroir.

Again, we run counter to traditional usage. Terroir, for many commentators, is all too often a synonym for the qualitative phrase “good terroir.” Certainly that’s as misguided as employing “wine” as a synonym for “good wine.” It may stand to reason and the law of averages that not all terroirs are “good,” whatever standards one may wish to apply to that qualitative assessment, and we could delve into the reasons a terroir may or may not be “good,” but they’re all subordinate to one of our guiding principles: scientifically-measurable properties. All qualitative assessments of the “good/bad” type are subjective, and thus not scientifically-measurable.

So are we any closer to an actual definition of terroir? I think so. And here – heaven knows you’ve waited long enough for it – is my proposed definition of terroir:

Terroir is a biological outcome derived from the interaction of mesoclimate, geography, and geology (including soil chemistry), expressed by the entity sustained by that interaction and possessing a chemically identifiable identity, but excluding the influence of external biological entities.

Now comes the important part: let’s pick at it. I’m quite sure it can be improved – certainly it can be made shorter – with a little external biological input.

(And Jamie…are you happy now?)


vinosseur said...


This is one of the best-written definitions of Terroir that I have seen. I was so impressed that I provided a link to "Somewhere, a place for us" on my blog at www.vinosseur.com

Thank you

thor iverson said...

Thanks very much, Joseph.

You have an interesting blog yourself, and of course with a name like mine, the fact that you're in Norway was an added bonus.

I'm adding a link to it now.

Arvid Rosengren said...

I finally worked my way through it all.

And it was worth the time. Keep up the good work.

thor iverson said...

Thanks, Arvid.