WASHINGTON — Washington. -- Although I have never been pregnant, I know what it's like to want a drink at the end of the day. And I can't imagine that lugging around the extra weight would make me want that drink any less. Yet it has been official U.S. Government policy since 1981 that women should not drink alcoholic beverages during pregnancy because of ''the risk of birth
Total abstinence. As of November 1989, this warning has been required on every bottle of beer, wine and hard liquor. In a celebrated recent incident, a pregnant woman ordered a strawberry daiquiri in a Seattle restaurant and was handed the warning label from a beer bottle instead.
A government pamphlet for pregnant women says no alcohol ''even on infrequent occasions,'' and recommends instead meditation, pounding a pillow ''or writing out your feelings.'' I have chosen the latter course. My feeling is that this campaign is totally out of hand.
Fetal Alcohol Syndrome (FAS), first identified in 1973, is an undisputed risk of heavy drinking while pregnant. But it does not follow that light drinking -- say, one drink a day -- does any harm. According to ''Alcohol and the Fetus,'' by Henry Rosett and Lyn Weiner (Oxford University Press, 1984), the recommendation that all women should abstain from drinking during pregnancy is not based on scientific evidence, since ''no risks have been observed from consumption of small quantities.''
Dr. Weiner, director of the Fetal Alcohol Education Program at Boston University, says that nothing has happened since 1984 to temper that conclusion. The emphasis on getting the vast majority of light drinkers to go cold turkey, she says, has undermined efforts to reach heavy drinkers and alcoholics.
Dr. Weiner directed me to the preliminary report of a study by Dr. Robert Sokol, director of the federally funded National Fetal Alcohol Research Center in Detroit, suggesting that there is a threshold of about 1.6 ounces of alcohol a day -- that's more than three drinks -- before any of 16 FAS traits (low birthweight, smaller head circumference, etc.) begin to show up.
It may seem logical that if six drinks a day are very bad for baby, one drink a day might be sort-of bad, but there's no evidence that it works that way. And why should it? It doesn't work that way for adults. Six drinks a day are bad for anybody, whereas light drinking may even be good for you.
Warring studies are impossible for the lay person to evaluate, but there doesn't seem to be much ammo on the other side. A fat volume called ''Alcohol and Health,'' a special report to Congress from the secretary of Health and Human Services, has section labelled ''Effects of Lower Levels of Alcohol Drinking During Pregnancy.'' Yet most of it just repeats the findings about heavy drinking. It cites one study reporting negative effects from ''moderate'' drinking -- defined as two to six drinks a day. Not a word about harmful effects from truly moderate drinking.
The boldest government claim I could find of actual harm from light drinking was in a 1990 HHS booklet for health care professionals: Even babies whose mothers consume only one drink a day during the first three months of pregnancy have ''more erratic sleeping patterns.'' And for this, every bottle of booze orders all pregnant women not to drink at all?
Dr. Weiner says that studies showing even such mild effects from light drinking are suspect because they take mothers' words about how much they drank, and because an average of one or two drinks a day can include getting plastered every Saturday night.
The case for abstinence is not based on evidence but on the logic of better safe than sorry. It is tempting, especially for an expectant mother, to say that any risk, however small or theoretical, is too great. But that is absurd.
As Rosett and Weiner put it, ''The pregnant woman is confronted with many ordinary activities in her life which pose a small but identifiable risk, such as driving an automobile and breathing city air.'' These risks, true, are harder to avoid, but complete avoidance of one risk usually creates others. Rosett and Weiner note that stress from an overly restricted lifestyle and from worrying about theoretical harms can itself cause neuro-endocrine changes that are damaging to the fetus.
Everything about light drinking during pregnancy makes it the kind of theoretical risk that is unlikely to be evaluated sensibly. Doctors are innately cautious and made more so by lawyers hovering overhead with malpractice suits. Both these ''syndromes'' (to use the fashionable term) are aggravated in the field of obstetrics, where emotions run high and where there is a built-in failure rate (birth defects) of about 7 percent. Women who suffer these unavoidable tragedies are needlessly wracked with guilt -- or persecuted by their husbands -- over a few drinks they might have had over the course of nine months.
Above all, judgment is clouded by the presence of alcohol. This is a double disadvantage to clear social-policy thinking. First, any use of alcohol is associated, however irrationally, with its abuse: alcoholism and drunk driving. Second, the benefit side of the cost-benefit calculus is primarily mere pleasure.
Despite our alleged national commitment to ''the pursuit of happiness,'' we are far more suspicious than other nations of any shortcuts.
TRB is a column in The New Republic written by Michael Kinsley.