Postmodern Winemaking


Integrated Brett Management

June 2011
by Clark Smith
It’s high time I post my views concerning this beast and its handling—as usual somewhat at odds with modern enological thinking. For winemakers interested in bypassing sterile filtration when possible (count me in), Brettanomyces management is the central problem facing the making of serious wine.

The reason is simple: The focus of postmodern philosophy is the creation and preservation of beneficial macromolecular structure. This structure manifests in wine as colloidal particles sometimes nearly as large as a bacterial cell. The benefits of good structure (profundity, aromatic integration and graceful longevity) appear to be lost by sterile filtration despite the fact that no tannin material may be retained by the filter. Our hypothesis is that the action of tight filtration somehow disrupts rather than removes structure.

Integrated Brettanomyces Management (henceforth referred to as IBM) advocates another approach that both preserves and takes advantage of the benefits of good wine structure. It utilizes a three-legged approach:

1. Create a nutrient desert
2. Microbial balance
3. Aromatic integration through good structure




  • Modern enological practices cause Brettanomyces and exacerbate its effects.
  • Postmodern techniques both assist and require a new approach to Brett management.
  • To master Brett management is to understand what red wine really is.

Microbial activity when properly husbanded can amplify distinctive terroir characteristics and soulful appeal. On the other hand, wines lacking good structure fail to properly incorporate the influences of microbial activity; the resulting perception of Brettanomyces as a defect has led to draconian control measures that produce wines of less interest.

When these measures become incorporated into a winery’s general protocol, they generally harm the development of its wines. Paradoxically, these wines not only carry Brett less well, they are more susceptible to it, leading to a snowball effect in exactly the wrong direction. Due to its clever survival strategy (see box on page  77), Brett in this environment outlasts organisms whose competition normally controls it. Brett is a hospital disease, fostered by the very sanitation measures designed to suppress it.

In the recent past, such practices have reached the level of academic dogma, with the consequence that Brettanomyces (together with its sexual spore-producing twin Dekkera) is now classified outright in many quarters as a spoilage organism. The intrusive aromas of unmanaged Brett include horse sweat, leather, shoe polish, salami, fois gras, truffles, dog doo and (Ralph Kunkee’s classic descriptor) “wet dog in a telephone booth.”

Neurological studies in guinea pig brains have shown that sensory inputs associated with important events—e.g. an A-flat played to signal feeding time—result in increased neural mapping for the stimulus. In the same way, wine tasters can become sensitized to aromas of oak, bell pepper, VA and other “defects” so that they no longer experience them as untrained consumers do.

Nevertheless, only the most zealous of its critics fail to acknowledge “good Brett” at least occasionally. In his hilarious article “Attack of the Brett Nerds,” wine importer Kermit Lynch issued a call to reason that unfortunately has been largely ignored.1

Why not simply eradicate this organism and make clean, sterile wines of appealing fruit character? Fresh, unstructured wines such as Rieslings are easily stabilized by modern technological tools (sanitation, fermentation with pure yeast strains, temperature control, pH adjustment, maintenance with inert gas and sulfites, and sterile bottling). But winemakers often find these tools are best placed aside in the development of fully-evolved, structured wines for which profundity rather than varietal purity is the goal. This discovery is the beginning of the path to postmodernism. A fully evolved Brett strategy completes the journey.

The emergence of ‘conventional’ wine
The modern technological system developed in Germany shortly after World War II brought into being a kind of wine previously unseen in history. Without sterile filtration, the off-dry German wines we regard as “traditional” were impossible to make because they were unstable in the bottle.

Such wines, more properly called “conventional” wines, did not exist in commerce in the 6,000 years of truly traditional winemaking. Reductive winemaking methods were extremely effective for the production of aromatic white wines whose appeal is based on freshness, varietal purity, crisp acidity and, sometimes, mineral depth.

Fine. No problem. I love a good Mosel. But take note: This style has in the past half-century almost entirely eradicated the market presence of traditional white wines, which once shared the mature evolution, oxidative development and microbial complexity that today is the exclusive province of red wines.

Fresh wines target quite distinct aesthetic goals from mature wines. Freshness, varietal simplicity and crisp acidity are undesirable traits in even today’s conventional reds. The point of most red wine (with exceptions such as nouveau Beaujolais) remains to evolve its initial simple aromas of berries and herbs into something richer and warmer and that more profoundly touches the soul. Tannins must themselves evolve from a coarse astringency into a rich underlying structure that supports and integrates developing flavors into a coherent single voice.

Reductive winemaking has several disadvantages for vins de garde (wines intended for cellaring). These wines depend on properly formed phenolic colloids, where almost all red pigment and tannin reside, to provide a refined structure that does not interfere with fruit expression through harsh aggressiveness. Aromatic products of microbial metabolism, oak influences and vegetal characteristics also are integrated by good structure into a coherent background that is positive rather than interruptiv e. Modern oxygen-free practices suppress the development of good structure, and sterile filtration disrupts it. Low pH and high titratable acidity also present challenges for mature reds.

Ecology of the organism

1. In an aerobic environment, Brettanomyces can utilize proline as a sole source of both carbon and nitrogen. This amino acid is available in high quantities in both wine and beer, and it is not metabolized by Saccharomyces spp.

2. Brettanomyces employs a parasitic, secondary contaminant strategy--rarely taking over. It lacks the genetic capability to synthesize biotin, folic acid and other essential micronutrients that it depends on other organisms to exude during their autolysis. Brett is well controlled to less than 500 cells per mL by other organisms normally present in wines. Sterilized wines can reach 10,000,000 cells per milliliter.

3. It principally enters wineries from bees and is spread by barrels and bulk wines from other wineries. In wineries, Brett establishes ubiquitous sequestered low populations that are extremely difficult to eradicate.

4. Brettanomyces can ferment cellobiose, a synthetic sugar produced by toasting of new wood.

5. It cannot grow at temperatures below 59°F.

The fallacy of acidity as a virtue
High titratable acidity exacerbates tannic aggressiveness by drawing excessive salivary protein into the mouth and coarsening the impression of tannin, thus counteracting the winemaker’s efforts at textural refinement. Low pH inhibits wine development and microbial stabilization, as do sulfites and cool cellars. Palate liveliness is to be prized, but its best source is minerality, which confers liveliness to the finish whose sources and mechanisms are still poorly understood.2

The mastery of winemaking at high pH levels is essential to postmodern work. The reaction speed necessary to useful phenolic development requires elevated pH. If pH is the gas pedal of aging, then pH 3.70-3.85 is analogous to the 55-70 mph of freeway driving, allowing us to cover some developmental distance in the cellar.3

Creating a nutrient desert in the vineyard
The first steps toward microbial balance should be taken in the vineyard. The goal of these procedures is that a vigorous fermentation can occur that will consume not only all traces of sugar but also essential micronutrients, the absence of which can hold Brettanomyces secondary growth in check. Paradoxically, the production of a wine poor in nutrients requires a must rich in nutrients to promote healthy yeast action.

Brett’s strategy is insidious. It is designed not to compete with Saccharomyces (normal wine yeast) during primary fermentation, but emerges later during aging. The nutrient status of wine in support of secondary microbial growth can be considered to have four aspects: fermentable sugars, nitrogen sources, micronutrients (vitamins and other cofactors) and oxygen. Brettanomyces growth must be suppressed in both of its modes—fermentative (requiring sugar) and respiratory (requiring oxygen.)

We can assess the risk of Brett fermentation by measuring enzymatic glucose + fructose. Levels above 1,000 mg/L are unsafe, and we’d prefer to see less than 500 mg/L. The winemaker must determine empirically the best method to achieve sugar dryness. Considerations include the choice of commercial yeast inoculum vs. wild yeast, temperature of fermentation, avoidance of highly elevated sugar in the must and a host of other factors.

Since oxygen is not necessary for later fermentative growth, we must depend on primary fermentation to reduce fermentable sugars. Even given good consumption of sugar, toasted wood contributes to wine the fermentable sugar cellobiose, which Brett can also utilize. Thus a secondary inhibitory strategy is advisable.

As part of its strategy as a secondary infectant, Brett is a nutritionally fastidious organism lacking the ability to synthesize for itself many micronutrients. In order to inhibit Brett in both of its growth modes, it is beneficial to encourage consumption of micronutrients during primary fermentation. To do so, the vineyard conditions to deliver a healthy, nutrient-rich wine should be mastered. Petiolar nitrogen measurements at bloom can be used to evaluate where deficiencies require fertilization, often in “hot spots” rather than throughout the field. Over-fertilization, which results in excessive vigor and poor ripening, still must be avoided through careful topological nutrient management.

In nutrient-deficient musts, the addition of simple refined chemicals such as diammonium phosphate (DAP) should be minimized. This chemical is a favorite of modern enologists because it relieves yeast stress and brings about a vigorous, smooth fermentation without sulfide production. However, this mode of fermentation is not useful for consuming micronutrients. When the yeast is fat and happy, it does not need to make enzymes to digest micronutrients as a food source. (If you feed them Twinkies, they won’t eat their oatmeal.)

Our primary ally in suppressing respiratory growth in the barrel is the wine’s reductive strength. It has been shown that Brettanomyces is able, in the presence of oxygen and ample micronutrients, to feed on ethanol as a carbon source. Unlike Saccharomyces spp., Brett can oxidatively metabolize the amino acid proline, ubiquitous in wine, as a sole source of both carbon and nitrogen. Primary fermentation inevitably leaves behind plentiful amounts.

If the wine is to protect itself, it must maintain its ability to consume oxygen. In the vineyard, our goal is to maximize reductive vigor through good concentration of tannins and to avoid excessive maturity, which can damage reductive vigor.

Microbial equilibrium in the cellar
Exactly like Integrated Pest Management4 in the vineyard, Integrated Brett Management seeks in the wine cellar to utilize the natural competitiveness of a complete ecology to maintain the activity of each type of microbe at an acceptable level. It seeks to play out in the cellar any metabolic conversions to which the wine is prone, so that sterile filtration is unnecessary. Any influence inhibitory to this goal should be dialed back to the point where microbial processes achieve completion. Among inhibitors to be considered are alcohol, temperature, sulfites, volatile acidity and pH. Storage temperature must, for example, be held above 60°F for a sufficient period to permit activity to proceed (a couple of summers, for example.)

Just as the most flavorful and distinctive grapes derive from vineyards employing Integrated Pest Management (IPM) rather than draconian pesticides, so a microbial equilibrium results in more interesting flavor development in the cellar. Like the great unpasteurized cheeses of Europe, wines permitted a natural microbial balance can achieve richness and profundity beyond comparison. Since many dangers await a wine or cheese so exposed, the transformation must be handled with great skill and attention through a carefully thought-out program that considers application of postmodern principles to every facet of the wine’s development.

Except in new cellars, Brettanomyces is a ubiquitous organism, a fact of life. Like athlete’s foot, one usually cannot hope to eradicate it, and like keeping ones feet dry, control of this organism is based on suppressing growth by denying it facile growth conditions. Keep in mind that the goal is to facilitate a truce with Brett so that a stable condition exists at bottling. Through nutrient depletion and good reductive strength, we create the best environment to allow the wine time to play out its inevitable development. Any reductions in alcohol and volatile acidity as well as exposure to new toasted wood should occur early, prior to the period when the wine is held above 60°F to resolve its development.

Final blends should be assembled well in advance to avoid the possibility that a blend may promote additional activity of which its parts were not capable. Different wines can hold back activity for different reasons. The blend may be less stable than its parts.

An appreciation of the wine’s reductive strength is critical to good cellar management—and to the timing of bottling. Reductive strength is a function of phenolic reactivity, mineral composition and lees contact. Late-harvest reds may appear heavy in tannin and color, yet be very low in reductive strength. Sensory properties of reductive wines are vibrant purplish hues, a closed aromatic profile and the presence of sulfides. (Direct measurement of reductive strength is the subject of an upcoming article.)

Aromatic integration
The third leg of the IBM system is the aromatic integration that takes place in wines of refined, stable structure. The basic idea is that in wines of good structure, microbial aromas that otherwise would appear as spoilage elements can be integrated as elements of positive aromatic complexity. Just as in a good béarnaise sauce we cannot perceive distinct aromas of tarragon, shallot, vinegar and peppercorn but instead a rich “single voice,” so the aromas of varietal veggies, oak and microbial activity can be integrated into a good phenolic structure.

I have discussed the essentials for creation of good structure in previous articles.5 The finer the colloids in such a structure, the more surface area will be available for aromatic integration. Monomeric color is essential for building fine structure. Oxygen, properly applied to a balanced tannin/pigment phenolic blend, acts like a wire whisk in the creation of a rich, light tannin soufflé, and lees stirring after this is accomplished can add fatness and refinement. These steps are also similar to chocolate making, in which the conching process uses oxygen to convert cocoa powder into dark chocolate, and milk protein softens this to milk chocolate.

Wines of proper ripeness and good extraction afforded early structural refinement can carry many times the supposed “threshold” of 400 ppb of the Brett metabolic marker 4-ethyl-phenol without apparent aromatic expression. Indeed, the nuances added to the flavor impression in the nose and by mouth imparted by a Brett manifestation in these conditions are likely to be absent of objectionable intrusion of spoilage characteristics, resulting in wines of greatly enhanced profundity and soulfulness.

Although presented here as a third leg, the creation of good structure should be addressed quite early in the wine’s life, creating the conditions for integrating subsequent microbial activity. Wines benefit most from oxygen immediately after completion of alcoholic fermentation. It is best to delay malolactic fermentation through the use of slightly elevated SO2 addition (I use 45 ppm) at the crusher and also to warm or cool the new wine to 65°F, immediately commencing treatment.

Since many aspects of Phase I oxygenative structuring are counterintuitive, it is best to work with someone skilled in the art.6 Properly administered, oxygenation is homeopathic, increasing rather than degrading the wine’s reductive strength, bizarre as that sounds.

Make up your mind


  • 1. "Attack of the Brett Nerds," Inspiring Thirst: Vintage Selections from the Kermit Lynch Wine Brochure, Ten Speed Press (2004).
  • 2. Speculations on Minerality, Wines & Vines, November 2010.
  • 3.
  • 4. and
  • 5. Ageing Gracefully, Wines & Vines, February 2010.
  • 6. Tools for Building Red Wine Structure, Wines & Vines, March 2010

Most problems with Brett are the results of indecision. The organism is easily controlled in fresh, unevolved white wines by a combination of sanitation, low temperature, low pH, SO2 maintenance and sterile filtration. However, since these con ditions preclude the development necessary for the production of top reds, a strategy of moderate suppression in the cellar can lead to completion in the bottle. The balanced ecological strategy advocated here cannot be implemented halfway. If you’re making classic red wines, you’re probably already a little bit pregnant, so maybe it’s time to embrace the baby and take up responsible parenting.

Clark Smith is winemaker for WineSmith and founder of the wine technology firm Vinovation. He lectures widely on an ancient yet innovative view of American winemaking.

Posted on 11.09.2011 - 00:13:06 PST
A very good question, answered to the best of my ability in "Speculations On Minerality," Wines and Vines, November 2010. I would not say minerality is preferable to acidity, but that it is often confused with it, and has, I believe, much to do with wine longevity that pH is given credit for.
Clark Smith
Santa Rosa, CA USA

Posted on 08.29.2011 - 22:32:50 PST
Clark, hi, I have met you via S.Krebs et al @ NVC. I have more re-reading of this piece on Brett to do but for now, a question. You are a science person, you use precise terms to describe specific things. But, in the paragraph about fallacy of acidity as a virtue, you say that minerality is preferred over acidity, or something of that nature. What, pray tell, is minerality? A common term, but what in science terms is minerality in wine? This is an excellent piece and I wish more writers and consumers would read stuff like this. Once I learned what brett tasted like, I developed a decent ability to detect it and to question it whenever I find it in a tasting. Even of Cain.
Sam Diego, CA USA