Postmodern Winemaking


Oak Reconsidered: Its Seven Functions

April 2010
by Clark Smith
My postmodern definition of winemaking is: “the practical art of touching the human soul with the soul of a place by rendering its grapes into liquid music.” If grape flavors of origin are primary, what is the role of oak?


  • Consider what deficiencies a wine may possess in order to choose the right oak to supplement them.
  • Oak has five primary functions that must be addressed before the two aromatic functions are relevant.
  • Using old barrels and high-quality oak chips are good ways to achieve consistency and complexity.

Even U.S. federal regulations, which are no treasure trove of winemaking wisdom and guidance, distinguish wine from beer and other beverages through the basic principle that wine has no ingredients—yet oak flavors are an exception. Barrels got grandfathered in as a container long before anybody paid attention to their roles as a flavor source, and oak alternatives rode those coattails.

“Oak is like cosmetics for wine,” Terry Lemaire of Oenodev taught me. “In the best case, you can’t tell she’s wearing any.” Jim Concannon likens oak to garlic in cooking: At best, it’s invisible, lifting out flavors. 

A purist would say a self-respecting woman should wear no makeup. Yeah, right. In practice, most like to look their best. Winemaking is a commercial enterprise, and few brands willingly sacrifice fiscal viability for philosophical purity—nor is unbalanced wine forgiven for its purity.

Randall Grahm likes to talk about vins d’effort and vins de terroir, essentially wines that show their processing vs. those that highlight natural aspects. There is a place for oak in the latter, but as a supporting actor or even as an extra with no lines. Oak isn’t the enemy of terroir any more than yeast inoculation or any other tool. But inept, clumsy winemaking certainly is. Over-oaked wine is just bad cooking.

Complete wines do not need oak. They often do benefit from time in old barrels, to out-gas off odors and facilitate slow oxidative development. It pays to consider what deficiencies a particular wine may possess in order to choose the right oak to supplement them.

Subliminal intentions
Novice winemakers look at wood first and foremost as a flavoring agent. This is a mistake. Oak has five primary functions that must be addressed before its own aromatics can have any relevance.

1. Co-extraction: Red wine is a lot like chocolate syrup, containing tiny phenolic beads that in turn contain tannin, color and flavor. When we ferment crushed grapes on their skins, it is challenging to extract the anthocyanin pigments as well as flavor components such as the spicy derivatives of cinnamic acid, which are not soluble.

Untoasted oak is a rich source of hydrolysable tannins called ellagitannins, which break down in must to yield prodigious quantities of gallic acid, a powerful cofactor. The toasting process turns these useful small molecules into large polymers that won’t form colloids. Toasted wood doesn’t work to enhance color extraction. Worse, the barbecue aromas of toasted wood are amplified by yeast action to produce a strong Worcestershire aroma.

Green, untoasted wood contains trans-2-nonenol, a nasty, planky sawdust aroma that persists in wine for years. To prevent this, oak needs curing outside, in weather that will leach tannins and foster subtle microbial transformations. Curing wood is an art, and skill is required to avoid the formation of TCA. Thus, only highly reputable coopers should be entrusted with the production of untoasted chips or barrelheads.

2. Anti-oxidative power: A common deficiency in musts is reductive strength, without which structural integrity, good texture and graceful longevity are not possible. Besides its co-extraction properties, gallic acid is also a wonderful antioxidant. As a vicinyl triphenol (see “Building Structure—The Postmodern Tool Kit” in the March 2010 issue) it imparts supplemental reductive vigor to weak wines, which are poorly concentrated or which have lost their energy to field oxidation during excessive hang time.

In oxygen’s absence, ellagitannins will lend a harsh crudeness, but proper exposure to oxygen will convert this to a fat, round texture. Oak plays a particularly important supporting role in Pinot Noirs, which often cannot support the extensive aging required to elaborate their flavors without supplemental anti-oxidative power.

3. Sweetness: As a general rule, the desired palate architecture for red wines is a sweet core of fruit contained by an angular frame of tannin. Some wines start off with excellent fruit core but lack definition—typically Barbera, Grenache and some Merlots. Others have the frame but not the fruit, a common problem in Mourvédre, Carignane and Cabernet Franc.

Oak can be a source of a wide variety of sweet elements. Untoasted wood supplies a coconut element (whisky lactone) that lifts out varietal fruit aromas, rich in forests where sessile oak species predominate, such as Alliers, Vosges and Argonne. Toast can enrich vanilla, toffee and sometimes sweet coffee elements. The flame converts cellulose to cellobiose, an exotic sugar that can feed Brettanomyces. It is important to match the wood you are using to the deficiency you are trying to address.

4. Framing: The oak selections to supply framing are entirely opposite those that provide sweetness. Certain forests, particularly those of Eastern Europe and Limousin, which favor the pedunculated species, supply the most tannin. Moderate toast can bring out spice elements, which accent mouthfeel. Very heavy toast can produce deep espresso notes that frame fruit well and enrich flavor persistence in the finish.

5. Structure for Aromatic Integration: Wines containing vegetal or microbial notes often can benefit from enhanced structure, which can serve to integrate these aromas into the background and allow them to merge with and support fruit character. Oak can be used to assist this process in several ways already discussed, such as anthocyanin extraction and struc tural supplementation. Oak introduced during primary fermentation can also provide sacrificial tannins that remove protein and deactivate yeast enzymes destructive to color. Structural enhancements call for extra wood, usually in the form of high-quality, untoasted oak chips.

6. Curing Aromatics and 7. Toasting Aromatics: Oak aromatics receive so much emphasis that little needs to be rehashed here. A year of air curing is essential for degrading plankiness (trans-2-nonenol) and to enhance whisky lactone. The longer curing occurs, the more ellagitannins are leached. High temperature is necessary to create clove spice, vanilla, caramelization and espresso aromas.

These functions are listed last because, all too often, the first five functions are ignored when choosing oak. Decisions are based solely on aromatic embellishment, resulting in unbalanced wines—cloyingly sweet or overly framed, with poor structure and problematic integration of aromas, tiring easily in the cellar

Ships into chips
The magical changes that occur during the maturation process in barrel are not easily replaced. Barrels breathe. They inhale a small, steady dose of oxygen. More uniquely, they exhale, cleaning the wine of funky off odors. They facilitate settling, good lees contact and interaction between reductive and oxidative zones within a small space, with intriguing flavor benefits.

Old barrels do all these things as well. New barrels as a source of barrel extractives are fiscally foolish and environmentally reprehensible. High-quality chips, when prepared with skill and care, provide these extractives much more reliably, responsibly and economically.

A French oak barrel is made from a 200-year-old tree planted by Napoleon to build future navies. When bark is stripped away and heartwood removed, the premium wood remaining yields about 25% staves for barrels. The rest is perfectly good wood, and it is discarded. Why? Because winemakers want to look cool.

Scoring the sublime
To paraphrase Forrest Gump, a barrel is like a box of chocolates: You never know what you’re gonna get. The innate variability in oak forests, even within a single tree (south vs. north side, high vs. low), is absolutely staggering. An exhaustive French government-funded study by INRA in the mid 1990s documented vast inconsistency in wood composition everywhere that was examined.

The point was hammered home at a 2001 California State University, Fresno, seminar about oak. First we were generously treated to a book of splendid prose from the hand of Jeff Cohn, then at Rosenblum Cellars, with one-page sensory descriptions of 27 cooperage house styles: an extremely well articulated and perceptive guide.

The next talk featured Steve Pessagno, who stated disarmingly that he would be more comfortable with barrel alternatives if he only knew what he was doing. Meanwhile, he could get all the complexity he desired from the variability in Seguin Moreau’s medium toast. If he filled 50 barrels with Cabernet, he could expect after a year to select 10 for reserve, 10 to dump on the bulk market, and the rest would become his regular bottling, imbued with far more nuance and complexity than he could ever intentionally bring about.

This is not very good news for the guy who only has six barrels. And for the vast barrel warehouses at Mondavi Woodbridge, Bronco and any number of other behemoths, averaging has taken the place of the human attention that might be given a tank.

Glorious stave, humble chip
Full disclosure: I used to sell Oenodev’s chips, and I still love them. But I cannot tell you how I longed to sell staves. So sexy, the oak alternatives’ halfway house. I begged my French colleagues to make them. They would love to, they said, if not for the unfortunate fact that (as they explained in that patient, kind, diplomatic way all Frenchmen have) staves are very stupid.

The biggest challenge confronting winemakers and oak vendors is the problem of reliable consistency. You can’t just try one sample, because products vary from lot to lot. Cohn’s treatise is precious because it reflects decades of experience from which he extracted an average profile for each cooper, an ideal to which no single barrel actually conforms.

A piece of wood as big as a stave, my French colleagues instructed me, can never be produced consistently. Sure enough, I remember unloading a truckload of staves, stacking them into two identical tanks, and filling each with the same Merlot for micro-oxygenation. One tank took 35 mls/liter/mo. for three weeks; the other took 75 for six weeks. Subtracting for the oxygen uptake from the Merlot’s native tannin, that’s a LOT of difference. I saw their point. Not a good business to be in.

Why do we use staves? Americans cling to several myths.
Myth No. 1: Staves are split along the grain and have no exposed endgrain, causing chips to impart a planky harshness. But if you look at oak under a microscope, you can see that it has lots of rays (tubes that run perpendicular to the grain). That planky dryness arises from poor wood selection, inadequate air drying or inconsistent toasting—not exposed endgrain.

Myth No. 2:
Staves, like barrels, have a toast gradient that imparts complexity. The reality is that the untoasted interior imparts a green pithiness at just the wrong moment, just as it does in barrels. It would be so much better if the toast were on the inside and the green wood exposed to the young wine, which can better handle it.

Myth No. 3: Toast fixes color during fermentation. Ain’t so. Anthocyanins are not fixed by aldehydes unless oxygen is present. If they were, the high concentration of aldehyde in fermentation would fix everything. Doesn’t happen. What fermentation does do is to amplify the barbeque furfurals to the level of Worcestershire sauce. Ugh. Toasted staves provide no copigmentation or antioxidative ellagitannins to fermentations.

Myth No. 4:
Staves replicate the complexity of cooperage at lower cost. In reality, barrel complexity arises from varying distance from the fire, and also from the traditional use of untoasted heads, which add sweetness and tannin. Stave manufacture in no way resembles coopering.

Three cheers for the chip
The lowly chip is the perfect format for blending to achieve consistency. It allows rotary toastin g (similar to coffee beans), which produces even, predictable results. Unlike dust, chips retain aromatics and do not surface-adsorb wine flavors. Any degree of complexity can be achieved by blending different chip products, preceded by single-bottle trials. Staves share the inconsistency and inconvenience of barrels and offer no technical advantage over chips.

To be sure, there are plenty of crappy chips on the market. My advice for starting out down this path: Buy only the most expensive chips on the market, and get your flavor from them instead of new barrels.

There will always be enough lunkheads around who will sell you used barrels at a small fraction of the cost of new. Barrel sterilization with ultrasound or microwave can alleviate the most fastidious winemaker’s microbial fears. Once they no longer leach out pithy green tannins, you will come to regard your oldest barrels as your greatest asset.

Clark Smith is winemaker for WineSmith, founder of the wine technology firm Vinovation.  He lectures widely on an ancient yet innovative view of American winemaking. To comment on this column, e-mail


Currently no comments posted for this article.