So, Jason Wilson (still my favorite newspaper cocktail writer) just posted his near-weekly column in the Post, this time on high-proof spirits.  As usual, I enjoyed it, but I’m not sure I agree with his central claim:

[T]here is a logical reason why craft bartenders seek out higher-proof spirits. Alcohol delivers flavor, just as fat does in food. It’s a similar reason why alcohol levels have crept up in wine in recent years: People expect that explosion of fruit in the mouth. In mixing cocktails, bartenders want all the various ingredients to pop with flavor, and the rich mouth feel that high-proof spirits convey.

As many of you know, I just finished my MS, focusing on the aroma compounds of rye whiskey, so this kind of thing is on my mind.  I’m really not trying to start a flame war or interwebs-argument or what-have-you, but, based on what I know of the chemical composition of spirits, the assertion that higher proof equates to better flavor transference is questionable.  I’m going to talk a little bit about my justification for thinking this is incorrect, and then I’d love to hear explanation of why I’m wrong.

The comparison of spirit proof to wine proof may be somewhat misleading – a lot of flavor in wine comes from fermentation by-products, so by allowing more fermentation (by having higher initial fermentable sugar), more flavor and more alcohol are produced.  I’m certainly not a wine expert, but I would suspect this is a case of confusion between correlation and causation.

In terms of production, it is possible that distillation or extraction with a high proof is necessary for proper production.  Ted Breaux (quoted in the article/is the guy behind Lucid/way better chemist than I am) is totally right about high proof herbal spirits maintaining the solubility of their herbal components.  In absinthe, estragole – a flavor compound extracted from anise and other herbs used in making the spirit – is insoluble below a certain proof.  It is one of the main compounds responsible for the louche, which occurs when the addition of water lowers the proof to the point of precipitation for these compounds.  The high proof of absinthe, however, is thus seen to be a production requirement, not a consumption requirement.  In fact, as any of us who’s tried straight absinthe knows, it is traditional to lower the proof of absinthe significantly to induce the louche and make it more – not less – palatable.  I think Chartreuse can also be talked into louching, although I could be wrong (science experiment when I get home!).

Similarly, most whiskeys are diluted upon bottling from a barrel proof that can be anywhere from 110 to 150 proof, depending on the type of whiskey (due to aging conditions, Scotch and Irish whiskeys tend to lose alcohol while aging, while American whiskeys tend to lose water, effectively increasing their proof).  Thus the proof of Rittenhouse 80 and Rittenhouse 100 could be the same (I have no knowledge of their production protocols myself) out of the barrel, and just diluted differently upon bottling.  By US law, “straight” whiskey produced in the US must be barreled at or below 135 proof, and is diluted to the desired strength (or bottled undiluted) upon bottling, so the extraction of flavor chemicals from the barrels during distillation and aging should be similar regardless of bottle strength [1, 2].

I’m really curious, though, about the idea that higher-proof spirits and cocktails taste better upon consumption.  There are several counter-arguments (as well as the practical fact that, in general, dilution with water or ice is necessary to make a strong spirit like absinthe or George T. Stagg whiskey palatable).  First, high alcohol proof can actually conceal the characteristic flavor of a distilled spirit because most aroma compounds are far more soluble in alcohol than in water.  Thus, dilution with water to a lower proof releases them, making them volatile and therefore available to the sensory apparatus in our nose (where most “flavor” is perceived”).  High proof affects both surface tension and micro-structure of the beverage as well, helping keep these aroma compounds contained in the beverage and inaccessible to our senses [3, 4].

Second, alcohol (ethanol) itself is a fairly powerful aroma compound.  A high concentration of ethanol can potentially compete for odor receptors and air space with other, less abundant and less volatile aroma compounds.  This in turn can cause nasal fatigue fairly quickly to the consumer.  I know that in the Scotch industry, the documented nosing procedure involves dilution to 40 proof in order to maximize the release of aroma compounds (as above) and reduce nasal fatigue [5].

All of this is not to say that a spirit can’t benefit from higher proof.  It is entirely possible, for example, that the aroma compounds kept masked or inactive by high proof are unpleasant or uncharacteristic, leading to a more positive impression from the undiluted product [4].  Alcohol itself provides certain aromas and sensations (like mouthfeel, as the article mentions) that are essential for making the spirit taste right.  Having lived in a dry town in Ohio, I’d be the first to agree that diluted spirits (the only type that were sold within the town) are a terrible thing.

Spirits at different proofs will inarguably taste different; all you have to do to experience that is to side-by-side taste the two Rittenhouses I mention above.  But I still am not convinced that “[a]lcohol delivers flavor, just as fat does in food.”  Sometimes that difference will be positive, and sometimes less so (I have several friends who more-or-less touch wood and spit whenever I mention Stagg), but it’s not a clear-cut mo’ alcohol mo’ better situation.

So my idea here is not so much to punch holes in Jason Wilson’s article as to foster some discussion.  I really hope I don’t come off as insanely confrontational – I just defended my thesis yesterday, so, as I mentioned, this subject is very much currently on my mind.  If anyone has a counter-example (especially one based on sensory science, since my analysis relies more heavily on chemistry), I’d love to see it.

And, since I want to end on a good note, let me just say that I love the Greenpoint.  The whole article is totally worth it if I can start walking into bars and ordering them without having to pick up the mixing glass myself.

UPDATE: Jason got in touch with me with the following clarification:

When you paraphrase that “the idea that higher-proof spirits and cocktails taste better upon consumption” — this is not exactly what I’m saying.
I’m suggesting that higher alcohol accounts for bigger flavor, not necessarily better flavor. The higher alcohol makes flavors pop in the mouth in a different, more pronounced way. In the wine world, a lot of people hate this development (and I can’t say I’m too fond of it either). Some blame high alcohol wines on the chasing of Parker scores, etc. In cocktails, it’s a bit different. You have a number of ingredients and flavors competing against one another and you need to bring them into balance. So sometimes, you need the pop of flavor that a higher proof liquor provides.
Here’s another piece I wrote on high-alcohol wines btw:

Jason’s response is interesting, but it still doesn’t answer my central question: given the data I’ve cited, what is the possible synergistic effect between alcohol and other flavor compounds?  How does high proof make flavors “pop”?  It seems to me that, given the data I cite above, as proof increases, the alcohol would have a higher impact while potentially reducing the impact of all other flavors.


[1] Piggott, J.R.; Conner, J.M. Whiskies, In Fermented Beverage Production, 2nd ed.; Lea, A.G.H. and Piggott, J.R., Eds.; Kluwer Academic/Plenum Publishers: New York, 2003

[2] Philp, J.M. Cask quality and warehouse conditions, In The Science and Technology of Whiskies, Piggott, J.R., Sharp, R. and Duncan, R.E.B., Eds.; Longman Scientific and Technical: Harlow, UK, 1989; pp. 264-294.

[3] Nose, A.; Hojo, M.; Suzuki, M.; Ueda, T. Solute effects on the interaction between water and ethanol in aged whiskey. J. Agric. Food Chem. 2004, 52, 5359-5365.

[4] Conner, J.M.; Paterson, A.; Piggott, J.R. Interactions between Ethyl-Esters and Aroma Compounds in Model Spirit Solutions. J. Agric. Food Chem. 1994, 42, 2231-2234

[5] Jack, F. Development of guidelines for the preparation and handling of sensory samples in the Scotch Whisky industry. J. Inst. Brewing 2003, 109, 114-119.