“Brett Character” in Wine: How Much Can the Experts REALLY Detect?

Among many in the wine industry, a wine exhibiting “Brett character” is generally thought of as negative and highly undesired in most wines.  This idea has been and still is a source of debate for some wine industry professionals, particularly when it comes to whether or not the presence of Brett is actually all that bad, or if it is just another added level of complexity to the wine.

“Brett character” is caused by the yeast, Brettanomyces bruxellensis, via several chemical transformations that result in increased levels of 4-ethylphenol and 4-ethylguaiacol in the wine.  A wine that has been contaminated with Brettanomyces is often described as having “off” flavors such as band-aid, barnyard, horse stables, and pharmaceutical, to name just a few.

On the other hand, some studies have found Brettanomyces to actually improve the sensory characteristics of a wine, including some leathery notes as well as improving the fruity and varietal aroma balance of the finished wine.

The ability to even detect Brettanomyces contamination is apparently difficult in and of itself, thus adding

 Hector Hanoteau [Public domain], via Wikimedia Commons

Hector Hanoteau [Public domain], via Wikimedia Commons

fuel to the debate regarding whether or not the presence of Brett even matters.  One study found a very weak correlation between the levels of ethylphenols in wine and the detection of “Brett character” in the tested wines.  Another study found that the presence of other compounds, such as isobutyric and isovaleric acids, created a sort of “masking effect” that basically “hid” any Brett character that one would expect to be there based on the wine’s ethylphenol levels.  Yet another study found that even among wine experts, the ability to detect the presence of Brett is not universal—specifically, 1% of 134 wine experts had a diminished ability to detect ethylphenols or Brett character in wine.

One question that arises from this information is whether or not training and specific job type has any influence on the ability to detect Brettanomyces contamination in wine, or if the experts’ ability to identify this quality is completely random and irrespective of the type of training or job experiences they’ve had.  Is the ability to detect Brett characteristics in wine a learned trait? Or is it something some are just “born with”?

The study presented today aimed to address these questions by determining (1) the perception thresholds (i.e. how little of the stinky stuff can they actually smell) for wine experts, as well as (2) the language they use to describe the tainted wines and (3) the decision-making process utilized by the different wine experts (I do not go into this section at all, due to space limitations, but if anyone is really curious, I can look into it for a separate post).


A total of 87 participants were included in the analysis of this study, as out of the 206 that started the study, only 87 of them participated in at least two of the three parts of the study mentioned just previously.

The participants were determined to be wine experts with the following professions/experience:  winemakers, winegrowers, wine merchants, and wine brokers.  All participants were active in the

Photo By JPS68 via photoshop (Scan book) [Public domain], via Wikimedia Commons

Photo By JPS68 via photoshop (Scan book) [Public domain], via Wikimedia Commons

Bordeaux Wine Council and were involved in formal wine tasting on a regular basis.

Wines used in this study were red Bordeaux wines with bag-in-box closure systems, in order to avoid the possibility of cork taint.

In order to determine detection ability levels of each participant, wines were “spiked” with a serial dilution of ethylphenols from very high concentrations to very low concentrations (and everywhere in between).

“Spiked” wines were presented to participants at random along with two untouched “control” wines.  The participant was required to analyze all three wines and to determine which one of the three was different.  Individual detection thresholds were determined by the lowest concentrations of ethylphenols correctly identified three times in a row.

For the language/description portion of the study, the subjects were asked: “What terms would you use to characterize this sample?”. The frequency of use for each term among all the participants was calculated and analyzed.

Brief Results

Detection Abilities

  • Wine expert profession had a significant influence on the ability of individuals to detect Brett character in wine, as well as off-odors in general.
  • There was closer agreement among winemakers in regards to the detection of off-odors compared to all other professions studied.
  • Variations in detection abilities among the individual participants were very large – specifically, the difference in ethylphenol concentrations identified was 2.5×103 between the most sensitive and the least sensitive perceivers.
  • Participants who were winemakers as well as those with academic tasting degrees were found to be significantly better at identifying Brett character in wines compared with other wine experts without academic degrees or in other professions.
  • There were no effects of age on Brett detection ability in the participants in this study.


  • A total of 221 of terms were used to describe the contaminated wines, with 31 unique terms present.
  • Odor descriptors used were always related to animal or pharmaceutical terms.
  • There were no effects of age, profession, or academic qualifications on the terms used among individual participants.

Concluding Thoughts

Most studies that I read focus on comparing the ability of experts to detect a given characteristic compared with novices.  For me, this is the first time I’ve read a study focusing on within the expert group themselves, in order to tease apart some of the variation in tasting and detection abilities.  The results of this particular study, which when I thought about it make a lot of sense, indicated that wine makers as well as experts with academic degrees in tasting were better able to determine if a wine was “spiked” with Brett character, and were able to detect Brett characteristics at a much lower concentration than other wine experts of different professions.

If you think about it, this makes a lot of sense.  Winemakers taste many different kinds of wine ALL the time.  Specifically, they are tasting them earlier on in the process and can therefore attempt to make adjustments to allow the finished wine to be as “Brett-character free” as possible (I realize that’s not a real term, but you know what I mean).  Other wine experts, such as wine brokers and wine merchants, primarily taste the finished wine that’s had a chance to be manipulated already and thus by the sheer nature of their jobs, they don’t taste as many Brett-laden wines as do winemakers.  Additionally, a wine contaminated with Brett might not even make it to the market, giving the edge to winemakers again in terms of Brett tasting experience compared with other wine professionals.

Overall, this study provides some information regarding how wine experts perceive Brett character  in

Photo By Craig Camp [CC-BY-2.0 (http://creativecommons.org/licenses/by/2.0)], via Wikimedia Commons

Photo By Craig Camp [CC-BY-2.0 (http://creativecommons.org/licenses/by/2.0)], via Wikimedia Commons

wines, and helps illustrate why there might still be some debate in the wine community as to whether or not Brett is actually “good” or “bad” for a wine (not counting when it’s utilized on purpose).  Maybe those on the “good” team haven’t had enough experience with Brett to know what they are tasting?  Maybe it doesn’t even matter. If some wine experts are not able to detect a Brett-like fault in a particular wine, how in the world would the average consumer be able to tell?

In general, results like these in this study are interesting from a psychological standpoint and would be useful in wine competitions and judging, however, for the average consumer?  Just drink what you like, Brett-laden or not!

Feel free to share any of your thoughts/ideas/questions related to this short study.

Source: Tempère, S., Cuzange, E., Schaaper, M.H., de Lescar, R., de Revel, G., and Sicard, G. 2014. “Brett character” in wine: Is there a consensus among professional assessors? A perceptual and conceptual approach.  Food Quality and Preference 34: 29-36.

7 comments for ““Brett Character” in Wine: How Much Can the Experts REALLY Detect?

  1. DB
    May 28, 2014 at 1:29 pm

    I believe the bigger problem, is that even at low levels of brett contamination the fruit flavors of the wine are masked, so that the consumer who may not recognize the brett characteristics, just knows he doesn’t like the wine.

    • Becca
      May 29, 2014 at 10:05 am

      That’s a great point, DB! Thank you for your insight on this topic! Cheers!

  2. Bob Henry
    May 30, 2014 at 2:42 am


    One aspect of accepting or rejecting Brett in a wine is your “internal reference standard.”

    California winemakers trained by UC Davis and Fresno State have been inculcated in a philosophy that having Brett spoilage yeast in your wine is universally “bad” thing — and should be expunged.

    By contrast, European winemakers (perhaps self-taught who embrace techniques handed down from one family generation to the next) may be inured to Brett . . . thus “internalizing” it as a “typicity” signature style of the region’s wines. (Example: red Rhones.)

    A personal anecdote.

    Some years ago, I judged a California county fair wine competition in Paso Robles covering “Central Coast” wines: Chards in the morning and Syrahs (Syrah blends) in the afternoon.

    At my table was a friendly acquaintance UCLA English “classics” professor-cum-wine enthusiast, and two winemakers from large commercial wineries located in the Central Coast.

    The wine flights were presented on a tray — each glass “randomly” numbered.

    We tasted in silence, recorded our “rankings” in silence, and at the end of the session submitted our individual written and signed ranking sheets to the competition administrators.

    Only then did we discuss at our table the wines for assigning medal rankings (Gold, Silver, Bronze, or no medal).

    During the session assessing Syrahs and Syrah-blends, the two commercial winemakers at my table insisted on disqualifying any wine that exhibited even the hint of “Brett.”

    (Aside: A 2011 Los Angeles Times article on Mouvedre stated that grape may intrinsically have a “Bretty” character. So a wine with a game-y/barnyard-y aroma and matching flavor may be “true to type.” And that by definition is not a “defect.”

    Link: http://articles.latimes.com/print/2011/feb/10/food/la-fo-paso-mourvedre-20110310)

    I asked the two commercial winemakers if they had much drinking experience with Rhone Valley wines.

    “No,” was their reply.

    Asked if they had they had ever traveled and tasted through the Rhone Valley.

    Once again, “no” was their reply.

    So the UCLA professor and I as a united front “agreed to disagree” with our winemaker co-judges across the table, and voted — enthusiastically — for the “hint of Brett” wines. (All two of them.)

    As it turned out, other judges in the room likewise found favor with the wines, and they were awarded Gold medals.

    At the conclusion of the tasting, a master list of all wines by number code was presented to each judge.

    Now knowing the identity of my favorite contentious wine, on my drive home to Los Angeles I made a detour and visited the winery’s tasting room.

    I offered them an “off the record” report of their Gold medal award — and bought half a case of the wine for my own personal consumption.

    Yes, I voted for the wine at the event, and voted with my wallet in the marketplace.

    ~~ Bob

    Postscript. I haven’t been invited back to judge a subsequent annual competition. (I leave it to others to infer any nefarious motives by the provincial-minded commercial judges.)

  3. Patrice Issartel
    July 2, 2014 at 6:48 am

    After all whether the final customer is able to identify Brett as such is not so important, what counts at this stage is whether he enjoys the wine or not. If the ethyl phenols have not covered completely the fruit aromas the wine may, at this point in time, be OK. On the other hand it is essential that winemakers are able to identify these contaminations at an early stage. As the growth of the Brett population can very quickly get off-hands, if it is not identified at an early stage and if the required corrective actions are not taken. The “terroir” or “leather” notes may eventually and incurably evolve towards less desirable “stable” or “barnyard” notes.

  4. July 16, 2014 at 5:04 pm

    In paragraph 3, you say some studies show brett to improve sensory eval. Can you name those studies or link so I can read them?

    My own opine: 1. nobody brags about having brett or saying I have good brett in my wine. The only mention comes as a apologetic, after the fact, uh, yeas, we do have a little bit but we don’t think it is too bad, we think it adds a little enhancement, uh, sort of . . .

    2. Nobody adds or introduces brett on purpose. They either spend money to get rid of it, or tolerate with the line of , uh, yea, we have a little but . . .

    3. A head to head tasting of the same wine or as close as possible the same wine, in my humble, is the way to learn to detect brettt, and the follow on is that very very few consumers will pick a brett over a clean version of any given wine.

    • Becca
      July 19, 2014 at 11:14 am

      Hi Donn,

      Thank you so much for your insightful comments and for adding great points to the discussion!

      RE: studies showing brett improving sensory characteristics….the paper that I reviewed pointed out one of these studies, and I will be completely honest, I did not attempt to seek out any more (thus why I just mentioned “a few” as was mentioned in the original paper and left it at that for this short piece. I probably should have changed this word to “one” since I did not pursue that part further, and I apologize for that).

      This is one study they cited: http://www.scopus.com/record/display.url?eid=2-s2.0-0001923779&origin=inward&txGid=589F0F3A07DBAC12B46095ACC738CCF3.ZmAySxCHIBxxTXbnsoe5w%3a2 This one is tough to find, but in the paper they mentioned that they found that at very low concentrations, the compounds associated with Brett improved the leathery aromas in the wine.

      I agree with all three of your points you made. Definitely adding Brett would be an awful idea and nobody (that I know of) does it, however in very low quantities, they can play up the “enhanced leathery aromas” to the best of their abilities if the wine is still passable as decent and leave it at that. That’s really the extent of the “improving sensory quality” that this study refers to—just playing up a bad thing to make it sound like a good thing so people will still buy it? It kind of becomes all semantics at a certain point, and I think that’s what we’re seeing here.

      Thanks again for your excellent comments/points!

  5. WP
    April 24, 2015 at 10:12 am

    Donn’s point #3 hits the nail on the head. I’ve seen many an instance where brett doesn’t ruin the wine, and it still presents as interesting. However when presented with 2 samples of the same wine, one clean and one bretty, almost all will prefer the clean sample. Take home message: it may not take away from the wine, but it will never improve it.

    Matrix also plays a big role in perception from what I’ve seen, and I’ve looked at some AWRI work with thresholds in different wines: young and fresh vs green vs oaky. If I recall correctly the threshold in the oaky wine was double the threshold in the young, fruity number.

    Love your blog Becca, keep up the good work.

Comments are closed.