FWIW, I'd like to see the gage R and R on this. I doubt the measurement accuracy is much better than +/- 0.5% by volume, the accuracy reported for refractive based estimates. Distillation is on the order of +/- 0.1%. Oh, BTW, they're talking a.b.v. here or alcohol by (per unit) volume. A.b.v. is the standard, unless your talking about 3.2 beer in MN, then it's a.b.weight. 10.0% a.b.w. is about 12.5% a.b.v. As wines gain weight, Chronicle to print alcohol levels Jon BonnéSunday, April 24, 2011 Rajat Parr praised Lee's high-alcohol Pinot Noir. Adam Lee switched alcohol labels during a Pinot tasting. Time for a small industry secret: Most of the nation's top winemakers spend their days making dessert wine. In the eyes of the federal government, that is. As far as the federal Alcohol and Tobacco Trade and Tax Bureau is concerned, any wine with more than 14 percent alcohol must be classified type 88: dessert/port/sherry/(cooking) wine. Marcassin Pinot? Dessert wine. The Cabernets of To Kalon? You got it. Silly, perhaps, but wine's alcohol level is plenty serious. Why else would it be the one piece of technical data required by law to be on every bottle? The easy answer - that taxes are higher on wines with more alcohol - ignores the fact that a lot of wine lovers pay keen attention to the big number in the tiny type. And not just out of curiosity. It's a matter of debate whether high-alcohol wines go with food or show ample nuance. But what's beyond question is that the same-size glass of wine with even a bit more alcohol gets you drunk faster. That's giving pause to more wine drinkers for reasons that have nothing to do with taste. So, starting next week, The Chronicle will print the listed alcohol levels of each wine we recommend in the Food & Wine section. A handful of mostly niche publications have taken similar steps. And Britain's Decanter magazine will publish alcohol levels beginning in May. But few newspapers or magazines have routinely published this information. High-alcohol debate Including us. Occasionally we've noted alcohol levels where germane. But we resisted printing them regularly because the act of bringing alcohol into the discussion of a wine is inherently political. Our decision comes at a time when it is harder than ever to understand the implications of alcohol in wine. Regulations reflect a world of decades past, when "table wine" was meant to hover around 12 percent alcohol and anything that dared surpass 14 percent was assumed to be fortified. Early examples that defied the trend, like the 1968 Mayacamas Late Harvest Zinfandel, left regulators stumped. Yet, the origin of the 14 percent rule is a mystery, even to many in the wine industry. Wine attorney and historian Richard Mendelson suspects it may predate Prohibition. Or it may stem from a hodgepodge of past practices. After all, the quaint remnant of the "table wine" designation - once meant for wines between 11 and 14 percent alcohol - lives on. Most European wines were shipped with labels that indicated them simply as "table wine," enough that Josh Jensen of Calera chose to emulate language from his favorite Burgundies in his inaugural 1975 vintage. Even then, the term was losing its meaning. "Initially we didn't want to put the number on," Jensen recalls, "and then after about 10 years of doing that, people would say, 'Oh, this is just a table wine,' and I'd say, 'This is our best wine!' And so we started putting alcohol numbers on." If 14 percent seems arbitrary as a dividing line, its financial implications are not. Currently the TTB assesses a tax of $1.07 per gallon, or 21 cents per bottle, for wines 14 percent alcohol or less; wines above 14 percent are taxed at $1.57 per gallon, or 31 cents per bottle. For a large winery producing 100,000 cases of wine, nudging above 14 percent costs an extra $120,000. To the largest producers, there are good reasons to keep alcohols low. And for all the kvetching about high alcohols, most wine taxed in the United States remains in that "table wine" bracket: 440.6 million gallons of bottled wine were at 14 percent or less last year, compared with 50.2 million above 14 percent. Hot topic in wine world But for higher-end wines? Life above 14 percent is routine, and for the past two decades it was often a required path to critical success. All this presumes truth in labeling for alcohol levels, which often isn't the case - enough so that we tested 19 wines to check their accuracy. (See story above.) No other topic prompts so many flareups in the wine world. In 2008, Sacramento retailer and wine authority Darrell Corti, a longtime critic of higher-alcohol wines, drew flak when it was revealed that he would no longer consider wines that exceeded 14.5 percent. In the aftermath, Corti began publishing alcohol levels in his popular newsletter. His business survived what came to be called Zingate. "I have never had a customer who has read any of this stuff say anything like, 'You're a fascist' or 'you're stupid' or 'you have your head up your ass,' " Corti says. "They all say, 'bravo.' " In part, Corti was frustrated because he saw how higher alcohols have come to largely define California's reputation. While most alcohol regulations in Europe mandate a minimum alcohol content - premier cru Volnay must clock at least 11 percent - "in California, it should be the other way around," Corti suggests. Corti's dustup would feel familiar to Michael Mina corporate wine director Rajat Parr, who was caught in a flurry of attention when, during a tasting at the World of Pinot Noir in March, he endorsed what he thought was a lower-alcohol Pinot made by Siduri owner Adam Lee. Lee had switched labels on two wines, and revealed that Parr had praised a 15 percent wine. (See sfg.ly/hyac9X) The intentional switch stemmed, no doubt, from Parr's vocal role in the lower-alcohol movement - including a policy that limits New World Pinot Noir and Chardonnay at his San Francisco restaurant, RN74, to 14.5 percent. (The limit doesn't extend to other wines or other Mina restaurants.) And it prompted a long round of bickering about alcohol, including an online finger-wagging from critic Robert Parker, who suggested that "arbitrary cutoffs make no sense, and are nothing more than a form of wine fascism." The use of alcohol as a barometer leaves many winemakers uneasy - especially since they acknowledge they often take advantage of the built-in leeway provided for wine labels. "The emphasis on alcohol for me is a distraction from the larger picture," Lee says. Parker and others often complain that alcohol rules are unfairly weighted against American producers - that often inaccurate labels on imported wines mask a boozier reality. The trick of balance Sommeliers swap tales of white Burgundies (to say nothing of ripe-vintage Bordeaux or Barolo) marked 13.5 percent but exceeding 15. In winemaking, vintners must consider alcohol along with pH level, acidity and so on. This leads us to the tricky notion of balance in wine. A Parr-organized Pinot event, "In Pursuit of Balance," was further skewered for the word "balance" serving as code for less alcohol. But that's not always the case. Napa winemaker John Kongsgaard, whose powerful (but balanced) Chardonnays routinely exceed 15 percent, hasn't made a wine below 14 percent in a decade. Yet his concentrated flavors retain balance. "We never hear complaints about the alcohol, ever," Kongsgaard says. "We hear about it as an intellectual idea from sommeliers, but I think the sommelier class are the people who are the most interested. I think that's fine. They're serving lots of wine to lots of people and they need to be careful about it." Wine and dine And that's where alcohol - regardless of balance - remains a defining element of a wine's style. For all the knocks against alcohol-watchers as ignoring the fundamental pleasure of wine, a balanced high-alcohol wine still won't please palates looking for a lighter touch. Alcohol takes on a new dimension when you sit down to dine. So sommeliers particularly have become champions of more - and more accurate - alcohol disclosure. Perhaps that's because diners have only their own assumptions to rely on when browsing a wine list. Perhaps it's because wine directors have to pair wines with food - there's a reason one Australian winemaker dubbed high-octane specimens "cocktail wines" - or because they want their customers to walk out the door, not stagger. For his part, Parr - who makes wine in Santa Barbara - finds more customers, especially under 35, asking about alcohol content. "I want to have a glass, two glasses, three glasses, and not feel that I'm going to be intoxicated," he says. "It's the wine I drink, it's the wine I make, it's the wine I serve in my restaurant. I never said the other style is wrong." Increasingly, winemakers agree. "I think every by-the-glass list should have alcohol listed on it," says winemaker Gavin Chanin, who apprenticed with Au Bon Climat's Jim Clendenen, an early scourge of California's higher-alcohol trend, before founding his Chanin label. "Not because it's the be-all end-all, but for those of us who know what kind of style we want, it's such a great indicator." Which might explain why at least one restaurant in Wine Country is serving up hard numbers. At Oenotri in Napa, wine director Sur Lucero lists alcohol levels for every wine he sells on a list that ranges from steely Italian whites to California Cabernet (and the occasional heavy-lifting Chardonnay) that top 15 percent. In part, this was a practical consideration - "there's no way I can keep alcohol levels for 650 wines in my head," he says - but Lucero also considers it a useful guide both for his patrons and himself in pairing wines to food with similar weight. While those numbers might seem to target the muscle of California, Lucero notes that they can be equally helpful with Europe's high-alcohol offerings, including Barolo, southern Rhone wines or even Alsatian whites that can push 15 percent. "Here in Napa we have a lot of educated wine drinkers," Lucero says, "enough that it doesn't take anything away from those who think that number is inconsequential, but it adds for those who find something in that number." The end of 14 percent? Whether you believe alcohol levels should come down or that the law should acknowledge a higher-alcohol reality, what's evident is that the arbitrary 14 percent barrier needs reform. "Everybody starts to draw the line at the same place," says Wendell Lee, general counsel for the Wine Institute, the industry's main trade group. "But does it make any sense? Today it probably doesn't." That could mean raising table-wine limits to 15 percent or enacting a sliding tax scale. Clearly the alcohols printed on labels could be far more precise. For those who don't care? It's easy to tune out this particular bit of information. But for a lot of wine lovers, alcohol levels will remain a flash point. So we're adding this one data point to our coverage. We believe that helps everyone make more informed decisions. And rare is the wine lover who doesn't want to be informed. Inside Label leeway: Only one wine label proves completely accurate in a Chronicle test. H6 Labels accurate? Often not How accurate are those numbers in tiny print? Since we intend to begin printing alcohol levels, we wanted to check how trustworthy these numbers are. We sent samples of 19 wines - all of them received for review by The Chronicle - to an independent wine lab in St. Helena for ethanol testing. Each test costs around $20. Only one wine was exactly accurate (see chart below). Three were virtually a percentage point off - or more - flirting with the leeway the federal government allows from what's printed on the label (1 percent for wines above 14 percent alcohol; 1.5 percent below 14 percent alcohol). The biggest gap came on the 2008 Pepper Bridge Walla Walla Valley Cabernet Sauvignon, Washington, which we recommended last week. Its tested alcohol of 15.17 percent was far above the 14.1 percent on the label. "I'm going to send it back" to the lab, said winemaker Jean-Francois Pellet when told of the results. Pellet said his own test landed the wine at 14.8 percent, but he had used an older technology called an ebulliometer. He said he would send another sample to his outside lab. Another surprising wine atop the list: the 2009 Calera Central Coast Viognier, made by The Chronicle's 2008 Winemaker of the Year Josh Jensen, who prints fastidiously informative wine labels. It tested at 15.54 percent, but the label lists it as 14.5 percent. "The discrepancy ... just blows my mind," Jensen said. He ruled out one frequent cause of inaccuracy - alcohol levels are often tested months before bottling so that labels can be ordered - because Calera tests directly from the bottling tank. Also pushing the 1 percent tolerance was the 2007 Chasseur Lorenzo Russian River Valley Chardonnay made by veteran winemaker Bill Hunter. A listed 14.8 percent alcohol was dwarfed by the actual number in our test, 15.79 percent - making for a nearly 16 percent Chardonnay. Another significant discrepancy came from Siduri, whose owner Adam Lee vocally opposes focusing on alcohol levels. We tested two Siduri wines - the 2007 Keefer Ranch Pinot Noir and the 2007 Amber Ridge Pinot Noir. The Amber Ridge landed at 15.02 percent, nearly three-fourths of a percent above its stated 14.3; the Keefer tested at 14.66, well above its stated 14.1. Yet our numbers were within 0.1 percent of levels Lee tested in 2008. His explanation? He rarely modifies the alcohol numbers on his labels, in part because of the cost of filing for a new label approval from federal and state agencies. On the flip side, some major names were remarkably precise. The 2008 Migration Russian River Valley Chardonnay from Duckhorn was spot-on: 14.1 percent tested and on the label. And the 2006 Robert Mondavi Reserve Napa Valley Cabernet Sauvignon lists an alcohol of 15.5 percent, essentially the same as its tested result of 15.46. That seems painfully high for a wine that once landed in the mid-13s, but the number is accurate. For our tests, the lab used an NIR Alcolyzer, which measures the absorption of a near-infrared beam in liquid to determine the amount of ethyl alcohol in a liquid - a new technology with precise results. The cost of an Alcolyzer is $16,000, according to Margit Svenningsen, a technical representative for Austria's Anton Paar, which makes the device, and an assistant winemaker at Calera. Svenningsen echoed a frequent assertion that the new method supersedes more common means of testing: the gas chromatograph, which can test a wide range of winery data but has potential for inaccuracy, and the ebulliometer, a 50-year-old technology based on boiling points, which can be significantly hobbled by elevation and weather. Sometimes even technology can't explain everything. The 2007 Hall Napa Valley Cabernet Sauvignon - 14.7 on the label, but 15.21 in our test - was tested by winemaker Steve Leveque at 15.0 percent with an Alcolyzer on its bottling date of May 29, 2009. Such discrepancies have become a way of life in the cellar. "It's not perfect," said Leveque, "but I'm not terribly surprised." - Jon BonnéWhy it's tough to trust the label All labeling has some leeway. The Food and Drug Administration allows rounding on food nutrition labels by 5 to 10 calories, for instance. But in an environment where even a half-percent difference in alcohol can chase away a customer, wine labels should be more accurate - especially because the numbers are self-reported and rarely checked. Wines 14 percent or less in alcohol are "table wines" under federal law. They have a latitude of 1.5 percent in listed alcohol - a flexibility largely born out of the prospect that a 12.5 percent table wine could range from 11 to 14 percent. That range, frequently seen on imported wines from the 1960s and '70s, can still be used. Above 14 percent, the leeway shrinks to 1 percent. But that still allows a 15.4 percent wine to be labeled as 14.5. (See the full regulations at sfg.ly/hherzG) U.S. wineries that sell overseas also must comply with EU regulations that round alcohol to the nearest half-percent, which also explains why some European wines have less accurate alcohol listings. And some states and Canadian provinces also have their own labeling requirements - and penalties. The federal Alcohol and Tobacco Tax and Trade Bureau, which oversees wine labels, will occasionally test alcohol content using a time-consuming but accurate distillation method. But it rarely tests, and has little enforcement for those who exceed the stated tolerances. Because any change to a label requires a winery to pay for a new label approval, many wineries leave labels unchanged from one vintage to the next - similar to a European practice to change only the separate vintage label. Differences in testing technology make it hard to reach an exact number. But the leeway could be reduced to a half-percent without causing undue costs to wineries. That certainly would come closer to truth in labeling. - Jon BonnéThe test results Wine Printed % alcohol Actual % alcohol Difference 2008 Pepper Bridge Estate Walla Walla Valley Cabernet Sauvignon 14.1 15.17 1.07 2009 Calera Central Coast Viognier 14.5 15.54 1.04 2007 Chasseur Lorenzo Chardonnay 14.8 15.79 0.99 2008 Siduri Amber Ridge Vineyard Pinot Noir 14.3 15.02 0.72 2007 Peachy Canyon The Vortex Paso Robles Zinfandel 14.8 15.47 0.67 2007 Siduri Keefer Ranch Russian River Valley Pinot Noir 14.1 14.66 0.56 2009 Pascal Jolivet Pouilly-Fume 13 13.51 0.51 2007 Hall Napa Valley Cabernet Sauvignon 14.7 15.21 0.51 2009 Fritz Haag Brauneberger Kabinett Riesling 8 8.47 0.47 2009 Liberty School Paso Robles Cabernet Sauvignon 13.5 13.93 0.43 2008 Vietti Tre Vigne Barbera D'Alba 13.5 13.88 0.38 2008 Williams Selyem Central Coast Pinot Noir 14 14.36 0.36 2008 Bouchard Bourgogne Rouge 12.5 12.71 0.21 2008 Frog's Leap Napa Valley Zinfandel 13.7 13.89 0.19 2008 Duckhorn Migration Russian River Valley Chardonnay 14.1 14.1 0 2006 Robert Mondavi Reserve Napa Valley Cabernet Sauvignon 15.5 15.46 -0.04 2010 Coppola Yellow Label Sauvignon Blanc 13.5 13.35 -0.15 2008 Chehalem Reserve Ribbon Ridge Pinot Noir 13.2 13 -0.2 2009 Rodney Strong Estate Vineyards Pinot Noir 14.5 14 -0.5 Jon Bonnés The Chronicle's wine editor. Find him at jbonne(a)sfchronicle.com or @jbonne on Twitter. http://sfgate.com/cgi-bin/article.cgi?f=/c/a/2011/04/24/FD311J4I7H.DTL This article appeared on page H - 1 of the San Francisco Chronicle -- ------------------------------ * Dr. James Ellingson, jellings(a)me.umn.edu * * mobile : 651/645-0753 * * Great Lakes Brewing News, 1569 Laurel Ave., St. Paul, MN 55104 * * james(a)brewingnews.com James.Ellingson(a)StThomas.edu *