Bottle Battle.

The latest fight over wine labels is part of the ongoing struggle between wets and drys.

Since 1989, every bottle of beer, wine, and liquor sold in the United States has carried a two-part government warning that is by now almost as familiar as the bar code: "(1) According to the surgeon general, women should not drink alcoholic beverages during pregnancy because of the risk of birth defects. (2) Consumption of alcoholic beverages impairs your ability to drive a car or operate machinery, and may cause health problems."

Soon you may start to notice a new sort of label on wine bottles. "To learn the health effects of wine consumption," the proposed label says, "send for the Federal Government's Dietary Guidelines for Americans." If you like to be lectured, or you simply can't get enough of the thrilling prose generated by government-appointed committees, you can get a look at the guidelines by writing to the U.S. Department of Agriculture's Center for Nutrition Policy and Promotion or by visiting its Web site; addresses for both are helpfully provided.

But Strom Thurmond would prefer that you didn't. In fact, the Republican senator from South Carolina, who sponsored the legislation that brought us the surgeon general's warnings about drinking and pregnancy, drinking and driving, and drinking and health, was said to be "absolutely furious" when he heard that the Bureau of Alcohol, Tobacco, and Firearms (BATF) planned to let wineries mention the Dietary Guidelines on their labels.

What is so subversive about the federal government's own nutritional advice? Thurmond objects to two sentences in a discussion of alcohol that is otherwise unrelentingly negative: 1) "Alcoholic beverages have been used to enhance the enjoyment of meals by many societies throughout human history." 2) "Current evidence suggests that moderate drinking is associated with a lower risk for coronary heart disease in some individuals."

Although both of those statements are verifiably true, Thurmond worries that acknowledging any positive aspect to drinking will contribute to alcoholism. It's the sort of attitude you might expect from a Southern teetotaler who is old enough to be a paleoprohibitionist. What's surprising is that Thurmond's position is echoed by so many people who claim to speak in the name of science and public health.

Public health may be blind to the pleasure that people get from drinking, but a discipline aimed at minimizing morbidity and mortality has to take into account the large body of evidence that moderate alcohol consumption reduces the risk of heart disease and prolongs life. Beginning in the 1980s with the famous Framingham study, epidemiologists discovered that alcohol is good for the cardiovascular system. It combats atherosclerotic buildup in the blood vessels, which eventually results in the blockage characteristic of coronary artery disease - by far the leading killer of both men and women in this country. Because they are less prone to coronary artery disease, moderate drinkers live longer than abstainers.

Despite this discovery, public health information about alcohol in the United States continues to be almost uniformly negative. As a result, having learned about alcohol from grade school on, American students still don't appreciate the difference between hazardous and beneficial drinking. Seven in 10 high school seniors disapprove of adults having "one or two drinks nearly every day." Yet this is just the sort of drinking that is associated with greater longevity in the epidemiological studies. (Such a pattern also avoids the dangerous aspects of the drinking binges that are typical among high school and college students.)

The Dietary Guidelines for Americans, which are produced jointly by the Department of Agriculture and the Department of Health and Human Services (HHS), originally reflected the federal government's general tendency to portray drinking as something to be avoided. The 1990 edition, for example, said drinking alcoholic beverages "has no net health benefit, is linked with many health problems, is the cause of many accidents, and can lead to addiction. Their consumption is not recommended." But by 1995 (the guidelines are revised every five years), the evidence of alcohol's health benefits had become so strong that it could no longer be ignored. After considerable debate, the committee of scientists appointed to revise the guidelines decided to include the two sentences that so upset Thurmond.

Committee member Marion Nestle, chairwoman of the Department of Nutrition and Food Studies at New York University, seemed amazed that she and her allies had managed to add some balance to the discussion of alcohol. "It's a miracle, a miracle," she told The New York Times. "It is a triumph of science and reason over politics. The committee process was very contentious, but the outcome makes the fuss seem worthwhile."

Philip Lee, assistant secretary for health at HHS and a wine drinker whose father owned a vineyard, was also pleased. "In my personal view," he said, "wine with meals in moderation is beneficial. There was a significant bias in the past against drinking. To move from anti-alcohol to health benefits is a big change." Elisabeth Holmgren, director of research and education at the Wine Institute, the vintners' trade group, pronounced herself "full of joy."

For all the celebrating, the section on alcohol in the Dietary Guidelines remained daunting. It said up front that alcohol "has effects that are harmful when consumed in excess. These effects of alcohol may alter judgment and can lead to dependency and a great many other serious health problems," including increases in "high blood pressure, stroke, heart disease, certain cancers, accidents, violence, suicides, birth defects, and overall mortality." Hence the guidelines urged moderation for those who choose to drink, meaning "no more than one drink per day for women and no more than two drinks per day for men." Finally, considerable space was devoted to specifying who should not drink, including alcoholics, children and adolescents, women who are pregnant or trying to conceive, and people using prescription drugs or operating machinery.

Still, from the perspective of vintners who were eager to counter the surgeon general's warnings on their products with something positive, the new guidelines presented an opportunity. In June 1996, the Wine Institute proposed a label similar to the one that was ultimately approved. The only difference was that the Wine Institute originally referred to "the health benefits of moderate wine consumption," which the BATF thought should be changed to "the health effects of wine consumption," lest it sound like a reason to drink.

Even though the label, especially as edited by the BATF, amounted to little more than a plug for the Dietary Guidelines, a coalition of 20 health organizations attacked the Wine Institute's request. The coalition included HHS, co-author of the very document cited on the label. In July 1997, John M. Eisenberg, acting secretary for health (Philip Lee had left HHS by then), told the BATF he was "deeply concerned" that the label "would be construed by the public as encouraging the consumption of alcoholic beverages." Surgeon General David Satcher warned that it would send "mixed messages."

Seeing the labeling request as part of a general campaign by vintners to make hay over reports of health benefits from alcohol, the Center for Science in the Public Interest issued a report in October 1997 entitled "Vintage Deception: The Wine Institute's Manipulation of Scientific Research to Promote Wine Consumption." The CSPI warned that publicizing alcohol's health benefits would encourage people to drink, resulting in more alcohol-related disease and social problems.

Given its longstanding resistance to anything that could be construed as a health claim for alcoholic beverages, the BATF was probably sympathetic to this argument. But as then-Treasury Secretary Robert Rubin noted in a March 1998 letter to Thurmond, the bureau simply does not have the legal or constitutional authority to reject a label statement that is neither false nor misleading. So last February, almost two years after the initial proposal, the BATF finally approved the Wine Institute's language, along with a label from California's Laurel Glen Winery that urged consumers to "consult your family doctor about the health effects of wine consumption."

The National Council on Alcoholism called the decision "potentially disastrous." Thurmond immediately proposed legislation that would overturn the BATF's decision and transfer authority for alcohol labels to HHS. The senator considers the bureau too cozy with the industry, and he seems to believe that HHS could use stricter criteria in reviewing labels, taking into account broader health issues associated with alcohol.

But the fate of the Wine Institute's label may not matter if anti-alcohol forces are successful in deleting the positive statements about moderate drinking from the 2000 edition of the Dietary Guidelines. Many of the submissions received by the committee that is working on the guidelines offer the sort of one-sided treatment that typifies a public health establishment preoccupied with alcoholism.

Consider the recommendations of Enoch Gordis, director of the National Institute on Alcohol Abuse and Alcoholism (NIAAA), a subdivision of HHS. After the standard opening sentence, "Alcoholic beverages supply calories but few or no nutrients and can be addictive," Gordis proposes the line, "Individuals who do not now consume alcohol, for religious or any other reasons, should not begin to drink." Is it the government's role to advise people about religious matters? If a nondrinker changes his religious beliefs, and decides he wants to consume alcohol, why is that the government's business?

Tellingly, Gordis does not offer similar words of support for individuals whose religions condone the moderate consumption of alcoholic beverages. Indeed, he would delete from the 2000 edition of the Dietary Guidelines the seemingly noncontroversial statement that "alcoholic beverages have been used to enhance the enjoyment of meals by many societies throughout human history." (Though this sentence has been interpreted as a kind of endorsement, one of the scientists who worked on the 1995 pamphlet told The New York Times "it was written into the guidelines merely to justify having any language about alcohol in the first place.")

Turning to the benefits of alcohol, Gordis tries hard to minimize them. He says the coronary effects of moderate drinking are "primarily seen in men over 45 and post-menopausal women." These two groups are at the highest risk of death from heart disease, so it is not surprising that they are the two groups in which epidemiological studies find the biggest decreases in mortality. But since alcohol is believed to reduce the risk of heart disease by increasing production of HDL (the "good cholesterol"), it stands to reason that moderate drinking earlier in life would help prevent the accumulation of fatty deposits, although the impact of this effect would not be notable until later on, when people are prone to heart attacks.

Gordis also fails to note that all women with risk factors for heart disease - overweight, smoking, diabetes, high blood pressure, or high cholesterol - can reduce their chance of death by consuming alcohol. The Harvard Nurses' Health Study, the major study of the connection between drinking and mortality among women, found that about three-quarters of all subjects had at least one such risk factor and were likely to live longer if they drank moderately.

Gordis emphasizes what he calls ambiguities and contradictions in the research on the coronary benefits of alcohol. He notes that some research shows benefits from drinking can be gained from having as little as one drink every other day. And he goes on to cite dangers not only from heavier drinking but also from moderate drinking (breast cancer, for example).

Neither Gordis nor the Dietary Guidelines ever explain that a large body of persuasive evidence shows that people who drink moderately live longer. The main exception is premenopausal women with a family history of breast cancer, who should abstain or drink only very lightly. That is, when all of alcohol's potential effects are combined, alcohol is beneficial for most individuals. This would seem to be a relevant piece of information for people weighing the costs and benefits of drinking.

At a June meeting of epidemiologists in Montreal, several prestigious research organizations confirmed the link between moderate drinking and increased longevity. One of them was the Alcohol Research Group, the epidemiology center funded by Gordis's NIAAA. The group's study found that men who drank two to four drinks daily had the lowest mortality rate. If it is relevant for Gordis to tell Americans the minimum level of drinking that has been associated with prolonged life, surely it is also relevant to cite the maximum level at which benefits have been measured.

Another study finding a significant reduction in mortality, for both men and women, came from a source even closer to Gordis - the NIAAA's own epidemiology division. NIAAA researcher Deborah Dawson reported that drinkers who were not alcoholics had a significantly lower mortality rate than abstainers.

Epidemiological research is complex and evolving, but it is by now nearly indisputable that alcohol protects against death from heart disease. A large number of Americans do not follow the drinking practices most likely to help them live longer, but this more often involves abstaining than drinking too much. According to surveys, half of Americans have not had a single drink in the last month.

Yet Gordis and like-minded public health officials wish to present what we know about alcohol in the most negative light possible, rather than offering straightforward scientific information. This is consistent with a longstanding anti-alcohol tradition in America, one whose advocates once spoke in the name of religion and morality but now wear the mantle of science. The battle over how to portray alcohol in the Dietary Guidelines and whether to allow wineries to publicize, even obliquely, the benefits of moderate drinking is a battle between those who portray alcohol as an unmitigated evil and those who have a more nuanced view.

Recall that the main objection to the Wine Institute label was based on the expectation that it would foster excessive drinking. In response to that concern, HHS commissioned a survey aimed at assessing how wine drinkers would react to the proposed label language. Only 3 percent of respondents said they would be inclined to drink more. The researchers concluded that "the risk of alcohol abuse resulting from the directional labels is negligible because they will not encourage a change in consumption patterns."

If so, you may wonder, what is the point of the label? Whether or not it actually sells more wine, the industry sees it as an important public relations victory. As Wine Institute President John De Luca puts it, "the label is an essential educational component of our public policy mission to counter efforts in some quarters to mischaracterize wine as a 'gateway drug' and a 'sin' that warrants higher taxes, trade limitations, and advertising restrictions." It is, in other words, part of the ongoing war over alcohol's place in our society, a war that did not end with Prohibition or its repeal.

Stanton Peele ([e-mail]), a psychologist specializing in drug use and addiction, consults with the Wine Institute and other industry groups on social and cultural aspects of drinking.