subscribe: Posts | Comments      Facebook      Email Steve

Should we taste wines based on alcohol level?


Jon Bonné wrote a very balanced article on Randy Dunn and his Dunn Vineyards in Sunday’s S.F. Chronicle, pointing out some of the practices that Dunn feels have “penalized” [Jon’s word] his wines’ reviews and scores. I personally have found Dunn’s Cabernets hard and tannic on those occasions when I’ve tasted them–Jon alluded to the same qualities–but that’s not what I want to talk about today. Instead, it’s Dunn’s idea [again in Jon’s words] that wines of similar variety or type be tasted in “separate sittings for…different alcohol levels.”

The suggestion that modestly-alced Cabs [say, 13.5% or less] don’t show as well as Cabs that are 1%, 2% or even higher in alcohol is an old one. It’s been around about as long as I’ve been writing about wine, which is more than 20 years now. The idea is that a big, alcoholic, fruitily-extracted wine will dwarf another wine of the same variety if they’re tasted together.

This is a compelling concept, which is why it has always had its adherents. If it was a dumb idea, support for it would have withered away a long time ago. But it’s not a dumb idea. As first glance, it has merit. The bigger, flashier wine steals all the thunder from the modest, “elegant” wine. [This is why some people refer to such competitions as beauty contests.] Why wouldn’t we even the field by tasting them separately?

I’m not opposed in principle to tasting Cabernets based on alcoholic content. But why stop there? If we did, shouldn’t we taste Pinots based on alcoholic content? After all, alcohol in a California Pinot Noir can vary by almost as much as it does in Cabernet or a Bordeaux blend. But alcohol also varies in Chardonnay, and Petite Sirah, and Zinfandel, and so on. So if we’re going to have separate sittings for one variety based on different alcohol levels, we’d have to have them for all varieties.

But why stop at alcohol levels? Wines are differently oaked, too. One Cabernet or Pinot or Chardonnay might have 100% new oak. Another might have 60%, or 30%, or no new oak, just one or two year old neutral. And, in the case of certain Chardonnays, there’s no oak at all, new or old. So why wouldn’t we taste flights of varietals based on the amount of oak?

I suppose Randy Dunn is saying we shouldn’t compare apples and oranges, but only apples to apples and oranges to oranges. I see his point; I really do, but I’m pointing out the logistical difficulties, and also a certain illogic that permeates the argument. I’ve thought about different ways of setting up flights for many years, and come to the conclusion that no one way is perfect. Name a flight methodology, and I’ll tell you its strengths and weaknesses.

Back to the more compelling argument, that Cabernets of lesser alcohol don’t show well when tasted against riper, richer Cabs. Like I said, you can make a strong case here, but I’m not sure it holds water in the long run. Here’s the reason: a good taster isn’t looking merely for richness and alcoholic fatness. No, he’s looking for balance. That word is hard to define, and I don’t intend to try and define it now, but it’s just not true, at least in my case, that all I’m looking for is big, rich, fat, glyceriney wines, warm in alcohol and sweet in fruit. I taste a lot of wines that qualify for all those descriptors, but I don’t give them good scores, because they’re lacking balance. And balance is something you can perceive in any flight of wine, at the beginning, in the middle, or at the end.

I do think that a good California Cabernet needs a certain amount of alcohol in order to achieve balance, though. I wouldn’t say that 13.5 is necessarily an unbalanced Cabernet, but I think it is difficult to achieve the kind of depth a California Cabernet needs at that level. Not saying it can’t be done. Just saying it’s hard.

I also think that this anti-high alcohol sentiment can be and often is taken to extremes. I’ve written about this a lot. Critics of high alcohol–let’s say in excess of 15%–sometimes get positively Talibanesque in their intolerance. Randy Dunn is not alone in this critique; an army of others marches alongside him. It’s one thing to have a philosophy of winemaking and stick to it, and it may be admirable from the point of view of consistency. But it’s another thing to refuse to change or adapt to circumstances. Sometimes, to change is to grow, not to sell out. If Randy Dunn really feels strongly that his Cabernets are misunderstood by so many people that they should be tasted under entirely different circumstances than wines are normally and routinely tasted under–handicapped, as it were–then he might want to reconsider his position.

  1. Thank you for reiterating that “balance” is what folks might want to taste for, instead of focusing on alcohol.

    Our winemaking “style” is to create rich, ripe Cabernets and as many of us know, alcohol is a by-product from letting the fruit hang and ripe flavors develop. We feel that it’s also one of the aspects that helps to keep our wines in balance, and a component of the structure that contributes to long-term aging. It’s all about balance, baby.

  2. Steve
    I have on many occasions taken issue with – or expressed a very different perspective – on your articles, but your point on trying to level a playing field on comparative tastings is spot on.
    Admittedly, I usually stand on the side of lower alcohol, but perhaps for a very different reason and one I’d love to hear your view on. I refer to the physiological effects of higher alcohol wines.

    I very much regard wine as an integral component of my meal, and that can mean 2-3 glasses over the 1-2 hours at the table. There is a marked difference on the effects of any wine @ 15% then say 13.5% on how my body reacts to the higher levels. I now actually look at the alcohol level when ordering wine in a restaurant! The last thing I want is to nod out before dessert, or worse, end with a DUI on the way home.

    I’ve tried doing a fair amount of research into the effects of higher alcohol wines on BAC, but because there can be so many factors to consider (body weight, time, amount of food consumed, etc), I’ve always come away with more questions than answers.

    BTW, I’ve found Randy Dunn’s wines take a great deal of time to hit their peak. I pulled the cork on an ’86 that drank beautifully and probably still had a few years of life.

  3. To me it seems that there are a bigger army of ‘lower’ alcohol zealots in the wine buyer crowd than with winemakers or wine consumers. This demarcation of +/- 14% is an artificial one based upon taxation on alcohol strength. In the past ten vintages, 50% of our North Coast Cabernet Franc has landed under 14% ABV,there are numerous buyers who would buy ‘under’ but not even taste when it was ‘over’. I write it off to Somm snobbery and know that youthful opinions often mature into understanding balance and pleasure and not some line in the sand. I admire Randy’s resolve – it reinforces that wines come in many shapes, colors and alcohol strength. Less is more is not a concept that has been fully realized, nor embraced, in the new world.

  4. Bravo Steve, for addressing this issue head on (once again). If we have separate tastings, where does it stop? tastings for 12% and less? for 12-13% or would there be a separate tasting for 12.5%’s? for 14% and a different tasting for 14.5%? and, wait, perhaps the wine will taste differently if it is paired properly with the correct food – therefore, you may ONLY taste this winemakers 13.4% Cabernet with “fill in the blank” food… and, BTW: you have to cleanse your palate each time, brush your teeth, floss, then cleanse your palate a second time…

    Obviously, I’m taking the cynics view, but truly, at what point do we stop?

    And, I would also comment re: Bruce’s physiological comment – no disrespect to him – this is not an “attack” forum – so, I’m not attacking the messenger or his point of view – however, if one drinks 10oz of a 13.5% alcohol wine, then one ingests 1.35oz of alcohol; on a 15% alcohol wine, one ingests 1.5% alcohol – the difference is not that great. Having said that, I am a huge advocate of not, not, drinking and driving and I don’t believe one should consume much alcohol, and certainly not three glasses (of any alcohol level – not three 4.5% alcohol beers, three 13.5% glasses of wine, etc.). Again, no disrespect – it’s just that one’s senses and reaction times can be greatly impaired by even one glass of wine… Apologize in advance as it is not my intent to attack Bruce at all, just commenting on his comment.

  5. doug wilder says:

    Being able to recognize the balance is certainly the key, Steve. My tastings are likely smaller in scope than yours, however, I never make the selection of what to put in a flight based on ABV. Because I don’t taste blind, I will often arrange the flight to place wines that may be lighter at the front. That is the way tastings are commonly presented for trade tastings. I can’t support the idea that there is a ‘penalty’ for lower alcohol wines unless a taster has a particular style preference for what is sometimes found in a higher one. Because I collect all of my tasting information in a spreadsheet, I can easily build indices of scores sorted by alcohol. The results are as I expect, inconclusive.

  6. I think the really significant point in Jon’s article (for those who bothered to read it) is it’s not just RandyDunn’s OPINION about alcohol affecting the ordering of the wine in blind tastings. He, and two respected UC/Davis professors, definitively showed that higher-alcohol wines change the taster’s perceptions of the lower-alcohol wines. Something that many have long suspected…but now there is the science to support that idea. So…the only conclusion is that blind tastings should take cognizance if this fact.
    You argue, Steve, that in your blind tastings, the logistics are such that it’s not convenient to take into consideration the alcohol/order effect.
    Fair enough…I’ll go along with that. But there are other wine critics who argue that it’s not “convenient” to cover up or order the wines for their tastings, and, therefore, it’s not “convenient” to conduct their tastings blind as well. Which is OK w/ me, I guess. “Convenience” trumps the absolute impartiality we are guaranteed w/ blind tastings.

  7. Well, TomHill, you’re looking for a fight, but you won’t get one from me! As I wrote, “Name a flight methodology, and I’ll tell you its strengths and weaknesses.”

  8. george kaplan says:

    Having started my enophile life in the early seventies, it’s comical to see 13.5 % at the low end of acceptability for Napa Cabernet. I’d enjoy relating that to Andre T. over a glass of , say, 1968 BV Private Reserve.

  9. @bruce: I don’t ever worry about alcohol in wine. I don’t drink and drive, period. In restaurants or at home, I drink whatever tastes good. Admittedly I have a high tolerance for alcohol in wine (less so with spirits–they go to my head fast), so I’m never in any danger of nodding out before dessert! (And anyway, I’m no a dessert eater.)

  10. Actually Steve, you’ve mischaracterized the results of the study: The main recommendation was to taste lower alcohol wines before higher alcohol wines. Makes sense.

    And you make it sound like there is a fear that the low alcohol wines are inferior. The point is merely that it is difficult to make an objective assessment of lower alc wines if you taste the higher ones first, because they interfere with your senses.

    These are wines for different customers and different purposes. Would you taste say “Napa Cabs” before Vinho Verdes? No.

    And why in this debate does nobody ever bring up Silver Oak? They are usually below 13%, yet nobody accuses them of being hard, green, or needing age.

  11. @mike: No mischaracterization! As I wrote, quote: it’s Dunn’s idea [again in Jon’s words] that wines of similar variety or type be tasted in “separate sittings for…different alcohol levels.” To me, “separate sittings” means different flights!

  12. Regardless of what Dunn said, the study said that. Do you think its okay to taste napa cabs before vinho verdes? Cabs before Pinots even?

  13. i think certain wine styles simply don’t show well in a lineup; alcohol levels be damned. while blind tasting a lineup of 40 wines is the best way to score wines on a 100-point scale, some wines just don’t show well in that environment.

    some wines show best when paired with food and drunk over the span of hours. some wines need a decade before they really reach their peak. and some wines taste best upon immediate release, when tasted without food in a lineup of ten wines.

    i think it all comes down to the fact that blind tasting by critics is imperfect. i don’t mean to denigrate what you do, steve, which is an important and valuable job of assessing wine quality. but having done my share of blind tastings, i find that some wines don’t show as well in that environment…and contrarily, some wines that do really well in that setting might not pair well with my dinner.

    i guess what i am trying to say is that context matters.

  14. Dear Steve,
    Long time reader of, first time sharing a comment.
    I’d like to point out that there are two sides to this coin. I sympathize with the complaint that ‘bigger’ wines will drown out the ‘elegant’ wines and it’s not really an accurate comparison. I’ve certainly had this experience. However, I’ve also had several experiences where I taste through a number of palate-bulldozers and then come across something fresh and lithe, with vivid flavors but not the extra mass; and think: “wow, this really stands out as exceptional, especially compared to the ‘bigger’ wines”. Why don’t we hear complaints going in the other direction? You never here producers of a bigger style complaining that the low-alc folks make them look bad?
    Essentially, and as you pointed out, balance is the only true universal metric in tasting a varietal across its diverse vintages, regions and styles.

  15. Dear Alex Kanzler, what a great point you made! It’s a 2 way street.

  16. mike: If I had vino verdes and cabs, i’d taste the vv’s first. I would never taste cab and pinot in the same flight.

  17. I remember when Dunn’s wines first came out and how they appeared in tastings compared to other Napa Valley Cabs. They were much higher in alcohol, black as ink and “blockbuster” in nature. Not just a few tasters felt they were over-the-top. So it’s interesting to read that he now is making this same critique toward others who have over topped his (once) over-the-top wines. Also, I question whether a person who uses reverse osmosis to reduce alcohol in his wine can criticize anyone for imbalance or overripe aroma in theirs.

    My experience as a winemaker is that it is almost always better to go with the natural balance of the grape than to screw with it. An overripe grape aroma and heavy phenolics in a wine with low alcohol doesn’t taste right, something is not in balance. A wine that has been extracted with high alcohol contains more color and tannin and needs the alcohol to balance it out. Pull out that alcohol and you have a wine that tastes a little empty, hard and tannic. Which seems to be born out by Steve’s tastings.

    This industrial style of winemaking where everything is manipulated is trending towards getting rid of grapes and just making wine from ingredients. I rebel against that notion. What happened to the art of growing and picking a grape at optimum maturity and not screwing with it after? If BV can produce a great Cabernet in a warm year like 1968 at 12.2 alcohol and 6 grams of acid without adding water, or using RO, so can we all.

  18. Adam Lee/Siduri Wines says:

    So, a couple of questions regarding the UCDavis study?

    First, has anyone read the actual study or just the Jon Bonne description of it? Tom? You sound like from your post you have. Did you? Steve…you don’t really sound like you did. Is that correct? I guess I think that until you read the study, completely, I think that commenting on it is a bit futile.

    Second, on the study itself (and I have not read it), would anyone not be skeptical of a study that was promoted by a person that has a financial interest in the study? If the study was about any other subject and you saw that it was financed or even promoted by someone who has a financial interest in the outcome, would that not lead to some well-deserved skepticism?

    Until we actually see the entire study I simply have too many questions and too much skepticism to be able to draw any conclusions about it.

    Adam Lee
    Siduri Wines

  19. Adam,
    No…I have not read the paper. My understanding is that Jon has only seen it in draft form and it has not yet been published. Presumably it will eventually go thru peer review.
    I would not, Adam, a priori reject it as a legitimate paper simply because it has RandyDunn’s name on it as well as the two UC/Davis ladies. My understanding (from Jon’s article) that Randy was curious if there was a alcohol/order effect in tasting wines. He suggested it as a subject to the two UC ladies, they were curious as well, and proceeded to perform the study. What Randy’s actual involvement was in the study, I have no idea. And I think the bottom line of the study was NOT that a low-alcohol wine will be negatively impacted by tasting a high-alcohol wine before it. The bottom line was (I believe) that the tasters impressions of a low-alcohol wine will be DIFFERENT if it is tasted first than if it were preceded by a high-alcohol wine. So (I believe) that it does NOT support Randy’s contention that his low-alcohol Cab does not show well because it’s tasted among high-alcohol Cabs.
    But…no…not yet seen the paper, so only conjecturing here.

  20. Adam: I only read Jon’s account, and based my comments on the concept of tasting lower and higher alcohol wines separately.

  21. Without looking at the study I would predict that higher alcohol wines score better than low alcohol wines irrespective of the order in which they are tasted. And the experimental design would have much to do with the results, e.g. are the judges allowed to go back to the earlier wines and change scores, based on a reassessment of their scores after seeing all of the wines? If so, they might well go back and score down the lower alcohol wines even if they are tasted first because of their later impressions of the higher alcohol wines.

  22. doug wilder says:

    I think someone mentioned it earlier – that the ageing curve on Dunn is slow in comparison with some other Napa Valley Cabernet Sauvignon producers. How do you account for that factor in a blind tasting of the same vintage, regardless of alcohol level? Tannins will be higher, less apparent fruit, perhaps? But from an aging standpoint it is a marathoner. As a critic, I would likely recognize those characteristics and note in my remarks. It could certainly alter my initial impression of the wine. However, I would feel compelled to revisit my notes after taking it out of the bag, aware that from a historical perspective, the wine is structured for the long haul. I opened a 1982 Dunn Napa Valley in 1995 and found it was still youthful.

  23. doug wilder, thanks for asking an important question. I go back to the concept of blind tasting. If a reviewer doesn’t know what he or she is tasting and comes across a tannic, less fruity, astringent wine, I would suggest that reviewer would be less likely to give the wine a high score. Sure, if the reviewer knows it’s Bordeaux or Barolo or young Dunn Howell Mountain, he’ll make allowances. But if he doesn’t–and this is the key–he’ll form his initial impression of a tough, tannic wine, not offering much enjoyment now–and then when he realizes it’s Dunn Howell Mountain (or whatever) advise his readers of the wine’s historic ageability. However, if the critic then raises the score in order to comply with what he sees on the label, he has committed a reviewing sin.

  24. Earlier I mentioned that wine critics should understand the “science of phenomenology and the basic science of tasting.” You thought that phenomenology didn’t have anything to do with geosmin and wine, but it does. Though I probably could have said it better.

    I’m also a philosophy major and did my dissertation no phenomenology and criticism. Phenomenology isn’t just the study of consciousness, but ultimately the study of point of view and really defines how we view and interpret the world, art, and of course what we eat and drink, all that we evaluate and criticize. A art critic doesn’t need to know how to paint (or a wine critic completely know how to make wine), but an art critic has to understand how people see, a film critic should know how the eye works and how we interpret images in series, and a wine critic has to know how people taste.

    However, I ended up getting a graduate degree in science, so hence the scientific focus, but you should understand the basic of tastings and the limits of you palate and to suggest that there are significant differences in how you score a wine based on how you taste. This means you should understand geosmin and alcohol and all that directly affects what you taste and how you taste (the two can’t really be separated). If you don’t, then you just start your next tasting with the dessert wine and see how the rest of your tasting goes.

  25. “Speaking at the ASEV meeting in Portland, Ellena S. King reported the findings of a project she conducted with Randy Dunn and Hildegarde Heymann of the University of California, Davis, indicating that when a panel of tasters evaluated wines grouped from the lowest levels of alcohol to the highest … “[t]here was a significant enhancement in the sensory attributes of the wines,” King said. Conversely, when the higher alcohol wines preceded those with lower levels, the tasters found those low alcohol wines to have diminished flavor attributes and enhanced vegetative qualities.
    “King’s aim was to assess how alcohol in Cabernet Sauvignon can affect sensory properties because of the significant increase in alcohol levels during the past two decades. She noted that alcohol has been on the rise in California, France and Australia for a variety of reasons from clone selection, hang time and changing wine styles. For the study, King used 24 U.S. wines selected to represent a range of styles and 12%-16% alcohol levels, which were confirmed using an Anton-Paar alcolyzer. King noted that wines with labeled values of more than 14% alcohol tended to undervalue the amount of alcohol, but no wine was found to have an alcohol level outside of legal variation.
    “The wines were grouped into those lower than 14% alcohol and higher. Eight wines came in lower and 16 were found to be higher. Three panels of 11 to 12 trained testers evaluated the wines in groups tasting in a random order, higher alcohol first and lower alcohol first. “The results indicate that initial assessment of high-alcohol wines can reduce perception of aroma and taste descriptors, while the opposite is true when low-alcohol wines are assessed first,” King wrote in her abstract on the study. “This suggests the need to consider wine alcohol concentrations when professionally assessing wine quality, such as at wine shows.”; Alcohol Levels Play a Role in Wine Appreciation; ASEV conference by Andrew Adams

Leave a Reply

7 − = six

Recent Comments

Recent Posts