Thoughts on the Terroir of Distilled Spirits

Clay Risen just wrote a wonderful piece about the terroir of distilled spirits in the New York Times. I cannot criticize much, but I can certainly add to it. It should also be pointed out that the birectifier and its methods can help anyone out in pursuing the terroir concept with more than lip service. This is a tricky new topic and you have to admire anyone brave enough to wade into it in a major publication with a word count.


A starting point for terroir, and for reviving distillation in general, is to note the massive trove of literature on the subject that has recently been unearthed. Distillers have a choice of using this literature and standing on the shoulders of the giants that came before them or sitting in their shadows and poorly reinventing the wheel. We can also study and deconstruct role model spirits to find the patterns of terroir across categories and bring them to new productions. What will mostly encompass terroir within a distillate will be observed in fraction 5 of the birectifier where the fixative terpenes and longer chain esters live (esters, however, are subordinate to these terpenes). [rose ketones]

As this discussion advances, we can start to get really nitty gritty with the science. First up in the article was differentiating three rye varietals by ethyl acetate content in the final spirit. Ethyl acetate is probably the most ordinary and basic congener in a spirit and typically an amount is cut away. This number will start low and increase to an equilibrium as spirits mature in a barrel (and there is tons of historical data). Role model distillation with the birectifier and the 1989 James Guymon lecture from Robert Léauté give us the best ideas of what the level should be. Other ideas, like my sensory sketches for apprentices (where I illustrate how to systematically vary congener categories), give a glimpse of what a spirit is like with elevated ethyl acetate and how when in excess it competes for attention with other features. Birectifier methods teach us how to systematically modify cuts to hit targets for ethyl acetate without resorting to more heavy duty and expensive chemical analysis (which we should still strive for).

Because excess ethyl acetate is cut away, if terroir was to be expressed in a spirit, in the ferment before distillation, nobility would be producing less of the ordinary congener. This would allow less cutting, and help justify distillation at a lower proof which is another marker of nobility. This isn’t always the case in other beverages, and for example, ciders often become distinct and fruitier if a yeast is selected to produce more ethyl acetate.

[To a degree this is proving incorrect, and mainly applies to the world of heavy bodied rum and far less so to other spirit categories like whiskey. As rum pursues longer chain esters, it can accumulate very significant amounts of surplus ethyl acetate while in other categories like whiskey, that is far less the case. This shows we need better comparative looks at spirits and rum cannot be transposed to whiskey so easily.]

A recent role model distillation of Wray & Nephews Over Proof shows how the headiness in spirits can become a unique cultural decision. Ordinary ethyl acetate can overshadow other more extraordinary and harder won features (lower frequency of occurrence), but a culture can prize it enough to do it. In distillation, it may be found, that the cultural aspect of terroir becomes the most salient. [This holds true, but Wray & Nephews is a unique example of a spirit with surplus ethyl acetate in its production. If you put it in a barrel, it would continue increasing.]

Another interesting rum example that was first thought to be site specific terroir and later became cultural and/or an economic limitation is seen in the rums of Cape Verde. The grogues of Cape Verde are distinct and I prize them highly. They are sort of the mezcals of the fresh cane juice genre, but the main thing that sets them far apart from Martinique and the other more famous agricoles is that the cane juice is not centrifuged and defecated. The phenomenon was only revealed recently after recovering more works by Rafael Arroyo. This isn’t the end of the story and we are still left with many questions on how they differentiate from the cachaças which do not all centrifuge and defecate. They may be differentiated by cane varietal as well as yeasting practices. A family member from the Cape Verdean producer Vale D’Paul told me his father has singled out a certain plot of land as producing distinctly extraordinary grogue and he sets the product aside just for the family. [The aroma from these unique rums is called vesouté by the French and is mostly considered a fault.]

Some great tales of grains exist in the literature and I described them in my series of practical distillers versus the science guys: Father, Forgive Them; For They Know Not What They Are Doing. and Who is Dante and Who is Virgil and the Value Proposition of Bourbon.

C.S. Boruff may have led the way in the stripping of terroir from whiskey:

In the selection of the grain the primary consideration is its starch content, since other constituents (proteins, etc.) are always present in ample amounts. In the old days, distillers apparently failed to recognize that differences exist between the starch content of various grades of grain and consequently always bought the cheapest. The fallacy in this has been amply demonstrated and the first and second grades of corn, although selling at higher prices per bushel, have been found actually cheaper sources of starch than the lower-priced inferior grades.

One reason terroir stripping may have been allowed to happen is that at the time they did not have the science to find out what exactly they are compromising. (The science guys were also tricked by the practical guys who knowingly and deliberately built a value proposition into early bourbons hoping to be bought out for big money.) Terroir in whiskey, just like rum, may lie in the divine and hard won carotene derived terpene congeners that you cannot simply titrate to reveal. The birectifier was a big deal because it was the first tool that could isolate and reveal the category. The next leg up came from chromatography which could first-name and then further categorize. Eventually big whiskey had the scientific tools to pursue terroir, but they didn’t have any market demand which has only re-emerged in the  21rst century. Birectifier analysis of the new bonded Overholt rye revealed some extraordinary things and I hope to conduct more surveys in the future.

Fraction 5 showing a terpene emulsion which eventually settles as droplets.

A lot of the cultural terroir in spirits will be found in its relationship with science. Bourbon is a fascinating case study as they transitioned from so called practical distillers to the naive value stripping science guys and back around to what I call guided traditional practices. New American distilling has a really poor relationship with science. There is nearly 1 billion dollars in new American distillery investment and near no new player conducts any analysis unrelated to proofing. Near no one owns equipment as basic as an automatic titrator ($4500) and with possibly 200 million invested in new rum production, to my knowledge, not a single new player is equipped to measure the much talked about ester content of rum without outsourcing. I keep getting mansplained out the door trying to present the birectifier.

If we don’t ask hard questions of terroir and how it relates to the works of past industry giants like Rafael Arroyo and D. Kervégant, Herman Willkie and Paul Kolachov, and Louis Fahrasmane’s INRA, it is in danger of becoming a smoke screen for an industry not properly investing in analysis or figuring out its relationship with science.

From Clay Risen: “Conventional wisdom, and most distillers, contend that the rigors of the distillation process strip out whatever nuances a grain might carry with it.”

Birectifier analysis of brewery beers (I have more unpublished case studies) have revealed that they are dramatically different than distillery beers. Grain character in brewery beers are almost expressed like gin botanicals (very apparent and spread out in other fractions besides #5). The bulk grain character selected for distillery beers may be far more complicated and best compare to molasses in rum manufacture. The character of grains for distillery beers may not be so readily apparent and may be in the form of hard to unlock precursors like glycocides and carotenoids. When unlocked, they provide divine fixative aromas of incredible persistence best compared to civetone, musk, ambergris, saffron, or true vanilla [rose ketones]. When not unlocked, they get flushed down the drain with the effluent. Unlocking hard won aroma was the quest of a chain of scientists over the course of the 20th century. Distillery beers may be allowed to have percentages of apparently aromatic grains like the ryes of mash bills and no doubt a lot could be debated here.

Introducing new grains capable of contributing this divine fixative aroma is going to be complicated. It is not well known how to unlock this aroma from commodity grains so how is anyone not in on the secret going to do it with something new? How are they going to learn to avoid mixing up grains better suited for a brewery with readily apparent character from distillery grains if the chemical targets are not well known? [We now know the big targets are the rose ketones and that Suntory was working on this congener category in the early 1980’s.]

Grains also likely deal with issues of congruence that grapes have. How do we know character is site specific and not a defect of poor congruence or poor stewarding? Many salient aromas in wine varietals like Carmenere have been revealed to be defects of pruning and not something inherent to the varietal or terroir. Some corns have rank-vegetal character not suited for distillates and though new, and possibly terroir, it is just lame terroir, nothing noble. There are plenty of trash grapes we know to avoid. Some grains may just be the equivalent of foxy Concord grapes. The grain literature is extensive, but I do not think much deals with beauty and aesthetic value.

There is however enormous potential and I described some of it with fruit eau-de-vies. We can likely, with the help of reference orchards, find fruit for distillates of unique organoleptic potential. An automated birectifier may affordably help us locate the divine and most extraordinary far better than chromatography. No doubt this multi trial work will be conducted on strained budgets. Fine fruits are often on the cusp of viability or need amelioration. Could we make a raspberry eau-de-vie of terroir if we had to ameliorate our mash with a percentage of apple or pear juice? The short answer is yes, but a lot more could be said.

As far as yeasts go, the potential is startling, but we have to learn to differentiate yeast practices for distilling from yeast practices for brewing and wine making. A surprising newly recovered primer comes to use from S.F. Ashby in Jamaica at the beginning of the 20th century. The early Jamaicans systematically tried a lot more than we thought. Spontaneous fermentation in wine has become quite common and integrates into the terroir idea very readily. These fine wine ferments form at their own distinct pace and grow from zero. Distillery ferments however, are quite abstracted and nth degree. Yeasts are pitched at a quantity to effect the rate of distillation which in turn dictates a ton of other variables such as aroma-beneficial bacteria growth which is a definite voice for terroir we can look to. Other practical things have to be noted to make comparisons. Grapes are not sterilized by cooking like a whiskey mash and they also have tremendous surface area to harbour yeasts. Spontaneous fermentation of grapes may start with a far higher yeast count than say apples or pears because of surface area. What this means is that yeast counts, diversity, and vectors all relate to terroir which is a lot to sort out.

One of the most exciting yeast stories that hasn’t hit popular drinking is that of schizosaccharomyces Pombe. The fullest bodied rums use a yeast all their own which is a fission yeast unlike the budding yeasts that produced 99% of all other fermented beverages. These yeasts have a lower frequency of occurrence and are typically crowded out by typical saccharomyces yeast. Rum ferments have to be abstracted and ameliorated to promote this obscure voice which only dominates in skewed situations of high osmotic pressure or extremely sterile environments where it can be pitched as a pure culture. Their voice can be one of terroir even if they do not start fermenting spontaneously. Champion yeasts also have to be hunted for and vetted in a sort of Olympiad. When we find a champion capable of unlocking the divine, we need to celebrate it in new ways. Terroir is a romantic concept and past scientists were not known for romance. Yeasts previously were just given arbitrary numbers related to a trial, but we have the opportunity to better connect them to the land.

Before I depart, because I’m at the beach, another vector for terroir in rum historically was the source of the lime plaster used in sugar refining. The lime could naturally vary in composition and that could possibly change its ability to unlock rum oil. Did the Arracks of Batavia have the best pombe yeast or did they have the best plaster? Variance in lime composition for sugar refining is well documented if anyone wants to explore the idea. [We’ve since learned more about this.]

Like wine, the depths of and potential to investigate the terroir of spirits is staggering and will engage us for decades to come. Past spirits researchers have already been encircling the concept without explicitly naming it for quite some time.

Oh wow, I almost forgot this incredible gem, Excise Anecdotes From Arrack Country. At the end, there is the most beautiful realization of terroir after the author is shipped to Cognac to study production practices.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.

search previous next tag category expand menu location phone mail time cart zoom edit close