Hey Jim,
Not too sure that I agree with the fact that Sweet/Dry wine is a proxy for alcohol content. It is certainly possible to get a sweet highly alcoholic wine (i.e. port style, dessert) or a dry lowly alcoholic wine (dry brut, beaujolais).
I don't particularly care for the alcohol content except that it matches the taste profile. I don't want to drink rocket fuel or diluted water.
I also think that focusing on wine alcohol content only is a bit of a snobbery act. I don't care that a wine I am drinking is outside of the alcoholic range that a snooty sommelier is saying the wine should be. As long as it tastes good and achieves balance.
Yes, I am a bit of an heretic....::
Agree or not, sweet/dry must be a proxy for alcohol content in the historical sense, since it was largely not divulged in the 1940s, 1950s or 1960s, the period of rapid growth of the commercial wine industry in the U.S.
I've got a small library now of old commercial winemaking books from those days, and the American commercial processes used then were very different from today and did produce sweet or dry wines strictly according to the length of ferment (i.e., percentage of alcohol). I am talking here of wine, not a fortified product.
All the chemicals in use now to control yeast were not available then, and UC-Davis' fabulous research and industry improvement efforts had not begun or, in the case of the 1960s, were in their infancy. UC-Davis' work started out with basic yeast testing basically just to find out what yeast could do, since the wineries were simply naturally fermenting then with whatever wild strains they had managed to inoculate their cops and equipment with through use. The other leg of UC-Davis' early push was simple basic sanitation, also not a No. 1 priority at early wineries.
As an example of how different the accepted practices were, up until 1982 wines were routinely filtered through a mat of asbestos. So if you are old enough to have drank commercial wine back then, you drank some asbestos particles. Same if you have been saving that 1981 Whatever in your cellar. You will very likely drink some asbestos.
So to assess alcohol content, you pretty much have to infer from sweet or dry if you go back farther than the late 60s. BTW, even though the pendulum does swing back and forth on sweet/dry and high/lower ABV, historically Americans have preferred a sweet, lower alcohol wine of about 10-12% ABV. This makes Europeans and connoisseur types aghast.