Inquiring Winemaker


Varieties, Methods For Making Rosé

September 2009
by Tim Patterson

  • A recent controversy in the European Union about rosé wine production methods raises the question of how much difference techniques and grape varieties make in the quality and taste of pink wines.
  • Adherents and advantages can be found for both the saignée method and for direct pressing--more likely resulting in differences of style than of quality.
  • The common aromatic and flavor elements across rosé bottlings likely reflects the prominence of ethyl esters, acetates and alcohols, which can be extracted with short skin contact and preserved with cool-temperature winemaking and early bottling.
A rosé is a rosé is a rosé--except in the European Union, where the authorities have decided that rosé made by blending surplus red and white wine is a no-no.

The June ruling by bureaucrats in Brussels, Belgium, put an end to a controversy that had flared all spring concerning a proposed rule to allow white-red blends (to be called rosé assemblage) to hit the shelves alongside classic pink wine (re-labeled as rosé traditionelle). Phalanxes of protesting winemakers from France and Italy won the day, defeating a scheme that, no matter what you think of the product in question, would have done nothing to address the underlying problem of too much wine.

But the debate did raise some interesting questions about pink wine and let us ask ourselves, "Aren't most rosés pretty much the same?" Would your average wine drinker really know or care how they're made, as long as they're fresh and juicy and chilled? Do they really taste all that different?

The dry rosé category in the United States, both domestic and imported, is definitely on the rise, registering large annual increases, though on a tiny initial base. (Meanwhile White Zinfandel, widely regarded as the evil twin of serious rosé, shows a slow decline from its huge base of millions of cases.) But the viticulture and enology that go into pink wine have gotten relatively little attention, which makes the EU flap a good occasion to ask: How seriously should we take the making of a wine meant not to be taken all that seriously?

Methods, we got methods
There seem to be four basic methods for making pink wine--two of them prestigious, two of them considered vaguely or entirely disreputable. Most common in the U.S. is probably the saignée method, in which a portion of red wine juice is pulled off the skins a few hours or sometimes a few days after crushing and then fermented with white wine techniques for maximum fruitiness and freshness, making the remaining red must more concentrated--which is often the motivation for the bleeding. John Buechsenstein, who has made rosé over the years for Joseph Phelps, McDowell Valley, Fife and other operations, and has spent a lot of time in the pink wine country of France, thinks this potential for improvement of red wine from lightly-colored, over-cropped grapes was likely the initial impetus for the whole category.

Second, in the south of France--Tavel, Bandol, Provence--and other places where rosé is a big deal, the entire crop of grapes gets made into pink wine, frequently through whole cluster pressing, sometimes by a combination of free-run saignée and pressing, with various combinations of crushing, destemming, whole berries, etc., in the mix. Once again, when the pink juice is secured, white winemaking takes over. This method may result in a bit more phenolic structure than non-pressed saignée pinks.

Method three gets you White Zinfandel--which is, yes, a form of rosé. Here grapes are harvested at much lower Brix to retain acidity (and get the crop processed early), pressed and eventually vinified with some residual sugar or concentrate additions to produce a lower-alcohol, semi-sweet, blush wine. This is, incidentally, a beverage I enjoy from time to time and have served to company in my home with excellent results. When people complain that Americans don't drink rosé, I wonder what sales figures they're reading. But since some of you will cancel your subscriptions if I continue in this vein, let's move on.

The final option is the one the EU got so hot and bothered about: blending finished wines--rosé by the back door. This has the advantage of opening up a lot more varietal options, especially among whites, and allows for precision in color intensity and other attributes.

Aside from the fact that it just seems like cheating (like putting Karo in a dry Riesling and calling it a TBA), we might suspect this method would produce wine with a quite different aromatic profile--if any of us had tasted many of these. Come to think of it, we have: Most Brut Rosé sparkling wines, including a lot of $200 bottles, are made from blends of red and white base wines (more often than from pink base wines). Often, the red wine is added at the dosage stage to prevent browning while en tirage.

Each of these methods has adherents. The whole crop/direct press folks say they can grow and harvest and process their grapes with rosé in mind, rather than treating it as a byproduct of grapes grown for red wine. They have taken to calling this rosé by design, or intentional rosé.

Serious saignée method producers insist their wines are made just as much on purpose, and may even be more versatile in the glass because of riper fruit. As for blending, many practitioners of the two higher-status methods have been known to fine-tune their final pinks with a judicious splash of red for extra color--and of course, there are plenty of pinks from the south of France with white grapes in the mix from the start.

All the same?
In the meantime, method aside, I think I'm not alone in perceiving most rosés to have strong common elements--they aren't by any means all the same, but they seem to inhabit a narrower spectrum of aroma and flavor profiles than the bigger world of whites or reds.

Often the variety gets subsumed under the category. Knock back some New Zealand Sauvignon Blanc on your back porch in the summer, and you're likely to say, "I just love this style of Sauvignon Blanc" (or maybe hate it). Knock back a pink wine, and you're likely to say, "Nothing beats a goo d rosé in the summertime. What's this made from, Syrah? Grenache?" And $5 says that when you get down to ferreting out descriptors, strawberry will be on the list.

If these hunches are right, how come? Time to talk to some smart winemakers.

"We do whole cluster press rosé," says Chris Phelps at Swanson in the Napa Valley, "intentional rosé, which gives me a predisposition that this is better than saignée. But I'm doubtful if I really know the answer." After making its Rosato from Sangiovese for several years, Swanson switched to Syrah six years ago, when Phelps arrived, and he now makes about 600 cases per year, which sell for $18 per bottle. The Syrah is harvested at no more than 22º Brix, which leads directly to a wine with relatively low alcohol and high acidity. "If I was picking at 26º," he says, "and did a saignée, I'd have to add water and acid to make it work. A lot of saignée can be flat and alcoholic."

Phelps acknowledges that picking less ripe means a muting of varietal character: "Red grapes at 19º-20º Brix don't show a lot of differentiation in flavor and aroma--you can make a pretty good rosé out of any red grape." Lower maturity means a different chemical soup in the grapes, in this case lots of bright cherry, strawberry, rhubarb and raspberry--plenty of estery stuff. "I'd almost be bothered if someone said, 'This really tastes like Syrah.'"

Jeff Virnig at Robert Sinskey Vineyards, also in Napa Valley, is another direct press advocate, arguing that picking at lower Brix gives a more delicate wine. The fruit picks up color and other goodies in the press until the taste starts to get astringent, resulting in a very pale pink from Pinot Noir. He wonders what you would get if you gave crushed red grapes a weeklong cold soak and then did a gentle press and fermentation. Would you get fuller wine character?

Virnig thinks he can pick up the differences between rosé grape varieties--say, Pinot Noir and Grenache--and thinks he gets more of the fresh, perfumey character with direct press than he would with saignée. Sinskey makes about 1,700 cases of its Vin Gris, listing at a relatively hefty $28 per bottle. Nonetheless, Virnig insists this wine is for refreshment on the patio, not for meditation. "If someone wants to open a second bottle, we've succeeded in our mission."

Saintsbury co-founder Dick Ward says they got into the pink wine business in 1989, using the saignée method as a way to deal with an overly large, rain-diluted crop in Carneros, but soon their delightfully named Vincent Vin Gris became a regular staple--currently 1,000 cases for about $14 per bottle. Winemaker Jerome Chery says Saintsbury is serious about its Vin Gris (OK, serious for a rosé), and thinks of it as much more than a byproduct, utilizing a mix of tank and barrel fermentation, full malolactic, and often a slight addition of red Pinot for tannin and color adjustment. He thinks the difference between saignée and direct press is stylistic, with more phenolic structure in the latter, not a difference in quality.

Ward and Chery agree that the short, low-temperature, ethanol-free skin contact extracts less of the varietal character expected from a full-tilt red wine fermentation. "It's not a simple wine," Chery says, "but we don't expect people to still be drinking it the following year."

Ward says, "After six or eight months, I'm ready to move on to the next vintage." And Chery also notes, "There is probably less terroir expression" with rosé--which surely qualifies as classic Gallic understatement.

Also on the saignée track is Tablas Creek on the Central Coast, making 800-1,000 cases of rosé blended from Mourvèdre, Grenache and Counoise and selling it for $27 per bottle. Winemaker Neil Collins says they started making rosé in order to make rosé, not to improve their reds--and in fact, he has to be careful not to bleed too much juice and end up over-extracting the reds. He agrees that pale rosés are hard to tell apart, but says theirs get enough skin time to "push the boundary toward a light red wine." Still, what they get is mainly what he calls "the easy stuff." And despite the price, it's still designed to be a fun, summertime, patio lunch wine.

Strawberries and beyond
My round robin of calls started with Jeff Morgan, who, along with Daniel Moore, makes pink wine and only pink wine at SoloRosa. Over several vintages, they've worked with Syrah, Pinot Noir and Sangiovese. In the process, Morgan says he's changed his tune a bit. "I used to think that varietal differences among rosés were minimal, but each of these wines has a certain unmistakable distinctiveness," he says.

SoloRosa's methods have varied as well: sometimes direct press, sometimes saignée, formerly a lot of barrel fermentation and malolactic, now nearly exclusively tank fermentation and no malolactic.

Morgan, who wrote a book on the subject, Rosé, a Guide to the World's Most Versatile Wine (Chronicle Books, 2005), thinks the quality potential of rosé is yet to be tapped, and the fact that styles are all over the map in the U.S. probably doesn't help. While agreeing that short skin contact and low temperature mean less varietal extraction, he also thinks the generic, quaffer quality of many rosés reflects the (low) quality of the fruit and the (lack of) care in the winemaking. With premium fruit and good cellar work, he thinks rosés are not only better but easier to tell apart.

John Buechsenstein isn't making rosé at the moment, concentrating instead on Sauvignon Blanc at the mono-varietal Sauvignon Republic. His preference is for fully ripe grapes, accepting the higher alcohol that comes with them in order to get a "full-service" wine that can go anywhere and pair with anything. In his role as wine sensory analyst and educator, he says the methods and grapes are often hard to tell apart: There can be some clues, but "mostly you can't tell, and who cares? It's easy to think too hard."

Trying to explain the common threads in the category, Buechsenstein zeroed in on the prominence of esters and higher alcohols, which get pulled out early from the short, aqueous skin contact, preserved by low-temperature fermentation, and retained by early bottling--most producers I talked to bottle in February. By the time these same red grapes go through a full red ferment, these components either get lost in the larger organoleptic mix or hydrolyzed.  

Sure enough, in a recent American Journal of Enology and Viticulture Research Note (V.60. no.1, 2009), Buechsenstein pointed me to a team of researchers who profiled Provence rosé and found supra-threshold levels of several compounds--all either ethyl esters or acetates or alcohols, the stuff of fruitiness. High on the list are our old friends, the strawberry volatiles--estery ethyl hexanoate, the alcohol furaneol--along with other primary fruits. Buechsenstein surmises that the frequency of strawberry on descriptor lists refracts not only the fact that it is probably there, but that it's familiar and easy to articulate at any level of tasting skill.

Enough for now; there's no reason to over-geek this one. All these folks make smashing good pink wine, and they should all just stay at it. All we really need is the Virnig test: If somebody opens a second bottle, mission accomplished.

Tim Patterson writes about wine and makes his own in Berkeley, Calif. Years of experience as a journalist, combined with a contrarian streak, make him interested in getting to the bottom of wine stories, casting a critical eye on conventional wisdom in the process. Contact him through
Currently no comments posted for this article.