But as consumers, we ultimately care about terroir only insofar as it manifests itself in the wine, not in the plant or fruit.
Actually, this is precisely where I took issue before, and will again. I believe terroir exists (because it must...a plant must be responsive to its growing conditions, it can't help but be) whether or not we can identify the manifestation.
Now, you specify that identification is the part we care about. That may be, and it's undoubtedly true for some people, but here's a hypothetical for you: does a non-wine drinking grape farmer care about terroir? I'd suggest that she does. She has to, because the way the site affects her crop determines her livelihood. But clearly she does not care about the manifestation of the terroir in the wine, because she doesn't drink it. She cares about its manifestation in the plant, full stop. Since that's the only point at which terroir actually affects the weather/soil/vine/grape/wine/consumer cycle, I'd also suggest that it's the one that matters more than any other.
That there's a whole separate category of interest in terroir, whether the INAO's puzzlings or the parlor game of "guess the site" we all play to generalized failure and occasional successes, is fun and part of what makes terroir-expressive wines enjoyable. But it's only an ancillary effect. Terroir is manifested in the plant, whether you or I care about it or not.
The DOC/AOC authorities are hardly enlightened, but if they defined terroir as the effect on plant, grapes picked in a Barolo vineyard could be vinified in a New Zealand cellar with Kiwi yeasts and still be called Barolo because the terroir is preserved.
No, that's not correct. They could define the
appellation as being that, or anything else, but they don't define the terroir, they just (in some cases, not in others) attempt to identify it. The borders, the composition laws, the min/max alcohol laws, the modification laws...they're all arbitrary.
But really, this is done anyway. You've talked about native-to-the-vineyard yeasts being an important terroir element, but few wineries exist entirely within the only vineyard that supplies their grapes, and even fewer vinify from just one site. No one, to my knowledge, has an entirely separate facility for each site's wine, existing within that site and only vinifying wines from that site. (At this point, I bet someone is going to tell me about one.) In every defined appellation of which you can think, the majority of grapes are vinified outside the terroir, and thus influenced -- at least in part -- by yeasts from other terroirs, yeasts native to the winery rather than the vineyard, or yeasts non-native to either.
But how is your case all that different from what Duboeuf used to do, or what any number of ngociants/producers do now? His wines all carried their respective appellations, but only one of the many could actually have come from vines even close to his winemaking facility, and he used an aggressive yeast that is now well-known as the cherished aroma-enhancer of industrial Marlborough sauvignon blanc. There are some pretty well-known Rhne producers, both inoculate-users and spontaneous-advocates, who farm and/or buy grapes from a pretty vast geographical area and bring the wines to a single facility. Is the difference between two hours and twelve important to the yeasts? Certainly not if you're inoculating, but I can't see how it would be even if you're not. Do the yeasts die? Are the winemaking facilities free of yeast populations of their own?