Or if there is, the Seminar on Minerality organized by the Institute of Masters of Wine failed to find it. The seminar had a great format: first three speakers presented views of minerality from geological and sensory perspectives; then there was a tasting to assess minerality.
I thought it had long been established absolutely beyond contradiction that, whatever minerality might be in wine, it is not due to uptake of minerals from the soil, but geologist Alex Maltman presented several amazing examples from supposedly respectable sources, such as textbooks, where minerality was attributed to soil elements. So it’s maybe worth repeating that this cannot be: measured quantities of trace elements in wine are far below the threshold for taste. Any effects they have on taste must be indirect.
Debunking another myth, Alex pointed out that insofar as soil might influence any uptake by the plant, it’s the surface that is important: deep roots basically take up water, but it’s the roots towards the surface (or at least in the top meter) that take up nutrients. So all those efforts to drive roots deeper, all that pride in the deep roots of old vines, if not misplaced is at least misunderstood. Deep roots may be important in ensuring water supply, obviously this may have a big effect on ripening and therefore quality, but if nutrient uptake were to have any effect on character, it would come from the surface. (And if there were any such thing as microbial terroir, which I take leave to doubt, it would be superficial.)
You might even question whether minerality relates to the actual character of wine (that is, some chemical or physical property) or is due to some form of association (think of Proust’s madeleine). Wendy Parr’s experiments show that it’s associated with people’s descriptions of other properties in the wine, and so does at least appear to result from what they actually smell and taste.
But there is the most extraordinary range of characteristics associated with “minerality.” Jordi Ballester finds that people who call Chardonnay mineral fall into three groups, loosely characterized as: flint/seashore, oaky/smoky/wet dog, and floral/apple/banana. Personally I’m pretty much in the first group, I can understand the second group (sort of), but the third leaves me totally mystified as to what people mean by minerality. However, Jordi points out that whereas producers in Chablis have a relatively clear idea of what they mean by minerality, consumers show little agreement, to the point at which he is not working with consumers any more.
We blind tasted 5 Chablis and 10 Sauvignon Blancs and were asked to assess minerality for each wine on a scale from 1 to 10. I got a completely different view from participating in the survey from reading papers on the subject, and at the least a much better idea of the limitations. Here was a group of around a hundred professionals, but the assessment of which wine was the most mineral was totally dispersed in each set, not quite equally, but certainly showing no consensus.
I wondered whether this was because none of the wines (to my palate) actually showed strong minerality. I use minerality as a descriptor quite often, but I wouldn’t actually have applied it to any of these wines. I also wonder whether Sauvignon Blanc is a good variety to test, because its varietal typicity can be so strong. I was surprised that there wasn’t an internal control, that is, the same wine included twice: at a minimum, in a research setting I would not accept the validity of any study that didn’t show that individual tasters rated the same wine reliably—otherwise all we’re looking at is scatter in the data.
So is minerality at all useful as a descriptor? I know what I mean by it, but evidently this is not necessarily the same as anyone else means by it. There is an amazing panoply of components that have absolutely no taste but that are used to describe the flavors or smells of wine: graphite, flint, rocks, iodine just to start with. Where a smell is ascribed to an odorless compound, it may come from association—the solvent used in tincture of iodine, or the aromatics released by sparking flint, for example. I don’t think it would matter particularly if iodine was used as a descriptor, even though the smell is not actually of iodine, if it was a reliable descriptor.
The problem is that minerality is anything but reliable. There is a cynical view (people were too polite to express it directly) that minerality is nothing more than a marketing ploy. I don’t accept that, because I do find it useful in my tasting notes, although maybe what I really mean is gunflint or smoky. I guess we end up with the old philosophical question of how we know whether any two people smell and taste the same thing, which of course implies that tasting notes are useful only for the person who wrote them.