Looking back over my previous posts on minerality, it seems as though this wine-tasting phenomenon isn’t related to the uptake of minerals (chemical or geological) from the soil, as described here and here. Nor, it seems, is it likely that it is due to petrichor, as described here. It seems increasingly likely, re-reading these posts, that what we call ‘minerality’ no more reflects the presence of minerals in the wine, than describing a wine as ‘floral’ implies there are flowers in it, or describing it as ‘rich’ implies it has somehow accrued some wealth.
‘Minerality’ seems to be a somewhat metaphorical term; what we sense in the wine is reminscent of minerals, but it is not really mineral. If we can accept this then we can call this hopeless search for minerals to a halt, and look for other more original explanations for what we call ‘minerality’ in wine. And on doing this up pops a rather strong candidate: reduction. To get to grips with this possible explanation, though, I think we first need to reconsider what we mean by reduction. As with minerality, ‘reduction’ is not what it at first seems to be.
When it comes to winemaking and oxygen, there are two extreme schools of thought. First, at the oxidative end, the wine is allowed increased contact with oxygen during the winemaking process; this might involve fermentation in an open environment, using more porous materials (i.e. wood), racking more frequently to allow contact with the air, and not adding an antioxidant such as sulphur dioxide. At the other end of the spectrum is reductive winemaking, avoiding contact between the must and the air, working in closed vessels, using carbon dioxide or other gases to prevent such contact, not racking and so on. [NB. As an aside, the processes of oxidation and reduction are more correctly described as transfer of electrons; oxidation is the loss of an electron often to oxygen (but also other molecules, often oxygen-derived, can take the electron) and reduction is the gain of an electron. In addition, the actions of sulphur dioxide are not accurately described when we refer to it simply as an ‘anti-oxidant’. But I don’t want to get sidetracked; for the moment let’s just stick with the man-in-the-street definitions of oxidation, reduction and sulphur dioxide.]
During fermentation, yeasts produce many substances other than alcohol and carbon dioxide. In particular, when fermentations are carried out under nutrient-poor conditions, yeasts become stressed. Yeasts require nutrients – such as nitrogen – to thrive and multiply, and carry out the fermentation, and this nitrogen is usually obtained from ammonia or common nitrogen-containing amino acids (the building blocks from which proteins are made). In the absence of these yeasts look around for more esoteric nitrogen sources, including less common amino acids such as cysteine, which contains nitrogen but also sulphur. As a consequence, the yeast now starts pumping out sulphur-containing waste products, starting with hydrogen sulphide. This gives the wine a rotten egg smell, and it is fortunate that hydrogen sulphide doesn’t hang around for long. Other sulphur-containing waste products, including mercaptans (also called thiols) and other more complex sulphides, do hang around though. And the array of different aromas these complex molecules produce can be very broad; there’s everything from the less pleasant scents of cabbage, garlic and rubber, through to more (or less!) appealing aromas of passion fruit and cat’s urine (mercaptans/thiols are important determinants of the characteristic aromas of Sauvignon Blanc), and then there are also the very suggestive aromas of flint, struck match, smoke and gunflint. This latter group look like very good candidates for ‘minerality’, don’t they?
So what we call ‘reduction’ is really the presence of mercaptans and other aromas. Once present their existence can be assured by nursing the wine using a ‘reductive’ philosophy like that described above, but they originate from stressed nitrogen-poor fermentations. So from this comes a theory, and several questions.
1. Could it be that the most nutrient/nitrogen-poor fermentations originate from the most nutrient/nitrogen-poor soils? If the vines have struggled to find what they need from the soil, and produce nitrogen-poor fruit, is it not logical that the must will be nitrogen-poor?
2. If so, very stony, nutrient-poor sites might produce the wines most prone to mercaptan and complex sulphide production. Is it not possible that we could interpret some of these substances as minerality, and then in a typical wine-writing leap of faith ascribe this sensation to the stony vineyard soils? Could minerality be mercaptan-related and dependent upon a paucity of vineyard nutrients not the stones themselves?
3. If this is the case, how ‘minerally’ a wine is could depend on work in the vineyard, and how easy or difficult a life the vines have. This seems to fit with the view of some – including Olivier Humbrecht of Zind Humbrecht – that ‘minerality’ doesn’t come without careful viticulture.
4. Likewise, it seems that work in the cellars could enhance or inhibit the generation of ‘minerally’ characteristics, and there is some evidence that this is also true, for example the new wave of more minerally white wines coming out of Australia (I’m on shaky ground talking about Australian wines, but I hope you get the point).
5. Personally, this would explain why I find the characteristics of ‘minerality’ in some wines – especially the flinty-smoky-gunflinty character of Pouilly-Fumé – to be remarkably similar to ‘reduction’ (in other words mercaptans and other sulphides) that I find in other wines.
There are problems with this theory, admittedly, not least the fact that although there are correlations between minerality and reductive winemaking, and the lack of minerality and oxidative winemaking, it isn’t a pure or strong correlation. At the extreme end it seems clear (i.e. I don’t recall finding a minerally but oxidised wine) but in the middle ground wines can be made in a manner more oxidative (oak fermentation vats and racking for example) and still be minerally, whereas more reductive winemaking can still produce wines stubbornly free of minerally character. Of course, it might well depend on the soils (as indicated above) or other factors (such as variety, or climate) and maybe the reductive character is merely part of the story. Nevertheless, I feel it is probably a large part of the story. I would be fascinated to hear comments on this, especially from readers with more knowledge of winemaking than I have.