None of us has a problem dealing with a wine with "cherry notes" or "silky tannins." No one thinks that actual cherries or silk are involved, just flavors or perceptions that resemble them in some way. But ever since "minerality" came into vogue, the notion that, say, a vineyard set in slate bedrock yields wine with a "slaty" taste has persisted in some circles. In a session at the 2009 GSA Annual Meeting dedicated to terroir, geologist Alex Maltman told us that "mineral" wines, before the 1990s, used to be described as "lean" or "austere." That basically means "not especially sweet, fruity or aromatic." So I think that "mineral" is an advance because it allows us to note the presence of flavor elements from the mineral kingdom like soil, chalk, wet pavement, dust and other evocative scents, not just note the absence of the usual wine flavors.
The mistake is in thinking that wine corresponds directly in flavor to the ground it's grown in. Chalky soils don't yield wines that taste of chalk, thank goodness. It shouldn't take much effort to separate "minerality" from minerals the same way we separate "fruitiness" from actual fruit. But as I listened to Maltman demolishing naïve "minerality," I found myself wondering about the real smells of real rocks. Limestones and cherts often have interesting smells, but I don't think they're trustworthy guides for rockhounds. Maybe they're just smells of soil fungi or something else unrelated to mineralogy.
Do any of you use smell in the field?
UPDATE: One news story about the session was misleadingly headlined "Geologists debunk soil impact on wine." Blogger Dr. Vino followed up with a more substantive entry on the GSA session on terroir.