Minerality and the relation between Intuition and Correlation

Rick VanSickle (@RickWine) of winesinniagra.com recently posted an article by Mike Risk about some work he did with Alex Brunton on minerality in the Niagra region.  The article sparked some reflections on this, one of my favorite (see my earlier post http://goo.gl/nfq1Tp ) often hotly-debated subjects and how it is (or most often, isn’t) studied scientifically.


(Rocks that potentially contribute various minerals to a vine near Saumur)

This is a subject that I worked a lot with while I was in New Zealand, looking into anything and everything written about it to try to understand the knowledge base that exists on the subject.  My conclusion?  There is no base.  There are lots of contradictory arguments thrown around, but no side seems to present a satisfactory case.

Here’s where I found this article intriguing.

Right of the bat, Risk makes what is often considered to be a big concession by scientists, but what I, and many, many winemakers and wine drinkers see as completely undeniable – he says “There’s no doubt about the importance of terroir – it is intuitively obvious [my emphasis] that wines from different locations will taste differently.”

So this intuition thing is what really intrigues me here – as a question to throw out there, what is the weight that we give to our intutitons when we’re defining what is or isn’t true? It depends what we’re defining, right?  In everyday life, sure, we go with our gut (in any case, if we don’t, we regret it).  But you don’t see articles in Nature about how quantum physics is right because its obvious.

Intuition ≠ rigor.

Okay, obviously, but hold that thought for a second.

Risk goes on to say that even though the concept of terroir is logical, minerality is not so simple.  And he’s absolutely right that science disproves the thinking that an intact mineral could enter a vine root and travel up to the grape bunch and end up in our glass. (and I’m not suggesting that this occurs in some magical – that’s the opposite of science, right? hah – way either. that’s not the point. the point is simply to question how we think about the science that says all of this).

But despite the complexity of the question, his team collects som data on trace elements and mineralogy in a few vineyards, and sits down with the excel file open and a flight of wines from different plots.  And he tells us that the wines are distinctly different from eachother, despite quasi-identical winemaking practices (except in one case, which he throws out – my thoughts on this decision, which I wholeheartedly agree with, will have to wait for another time).

But the data don’t correlate with the differences in the glass. (Okay correlation is my word – he uses ‘explain’, but, correct me if I’m wrong – and I’d love to be, this is a scientist reflex – because it’s the tool we have – to start looking for explanations based on correlation).

But why, anyways, do we rely so much on correlation? We know that correlation does not equal causation, we learn that early on.  We’ve seen many examples where the cause and effect relationship has been inversed because of assumptions based on correlations.

Here we have a case where the intuition, that wines from different places will taste different, matches with the reality that the wines taste different.  But our tool, our scientific implement of explanation, correlation, tells us next to nothing.

What happens when intuition corresponds more to reality than correlation does?


(Lets dig in!)

Link to the article is here if you missed the hyperlink above:


*Thank you Rick VanSickle, Mike Risk and Alex Brunton for your article, I am a big proponent of any and all research in this field because it is something that absolutely needs more serious attention such as this.  Enough with the bickering like school children about whether minerality exists or doesn’t or whether rocks have an odor etc etc.  We need people out in the field just like you have done.  I hope that you continue your work and I’d love to hear how it pans out! And as for all scientists, I hope that you always keep a little meta-perspective in your work, asking yourself questions about how you pose your questions!

**PS I am very curious about your passing remark about Brandi MacDonald and the fact that she runs an analytical lab at the same time that she is doing a PhD in anthropology. I’m very curious if there is a link between the two and I’d love to hear more!