It’s like one of those old riddles: when is a dry wine not dry? Answer: when it’s German Riesling. Or for that matter Riesling from Alsace and (less often) from Austria. The old question of perception of sweetness was brought back for me by dinner at the Setai Restaurant on Miami’s South Beach, where I felt that the Asian-dominated cuisine called for a Riesling. But I wanted a dry Riesling. Although there was range of choices, largely from Germany and Alsace, it was not immediately obvious whether the putatively dry Rieslings were really dry.

So I had a discussion with the sommelier. I liked the look of the Zind Humbrecht Clos Hauserer, which comes from the foot of the Hengst Grand Cru and should have that quintessential mix of richness and minerality. The sommelier assured me that it would be completely dry and went off to find the bottle. After a  few minutes he returned to say that in fact the wine is not dry. I wondered how he had discovered this without opening the bottle, but decided to move on to avoid any possible problems.

Most of the German Rieslings on the list were clearly sweet, but there was one Grosses Gewächs, so I felt safe ordering that without much further discussion, as the whole idea behind Grosses Gewächs is that first the wines should come only from designated top vineyard sites, and second, they should taste dry (more of this in a moment). The wine was Georg Mosbacher’s Forster Freundstuck (Freundstuck is a relatively little known vineyard close to the famous walled Kirchenstuck vineyard in the town of Forst).

This turned out to be a nice wine, inclined to richness rather than minerality, but palpably sweet. If it had been labeled as a halbtrocken (where up to 18 g/l residual sugar is allowed) I would not have blinked, but I found it hard to accept that any reasonable person could describe it as dry. Perhaps it has residual sugar right at the limit for the trocken classification, with low enough acidity to let the sweetness dominate the palate. The sweetness was far too evident for the wine to be a good food match, although it did make a nice aperitif. I suggested to the waiter that the sommelier might like to come back and taste the wine so that he could avoid telling future clients it was dry, but for whatever reason he did not reappear.

When it was time for another bottle, I asked for the wine list, and the waiter, perhaps reacting to my earlier comments, said that he would fetch another sommelier. So I went through the list with head sommelier Dwayne Savoie, and we decided that there really wasn’t another completely dry Riesling. But an excellent alternative was found in the form of F. X. Pichler’s Loibner Berg Grüner Veltliner (Smaragd). I’m usually rather cautious about Grüner Veltliner outside of Austria, because I’ve had so few that have been truly expressive (as opposed to some interesting wines, especially older ones, in the Wachau region. Also, the Smaragd classification is based on ripeness/alcohol and so does not guarantee dryness). But I trusted Dwayne’s judgment on this, and wine turned out well, showing much the same quality as Riesling in ability to match Asian cuisine, although I would say it showed less refinement than Pichler’s Rieslings. My only complaint was that this wine is really too young now – but as I agreed ruefully with Dwayne, restaurants can no longer afford to hold the wines for ripe old age.

But I was left with the feeling that perhaps I should go back to my old rule: never order Riesling in a restaurant because you can’t tell whether it’s dry or not. With the exception of a small number of producers whose wines are always absolutely dry, it’s a pig in a poke. My experience in trusting sommeliers to know whether the wine is dry has not been great either. For years I never ordered Alsace Riesling because of this uncertainty, and I only abandoned the rule for German Riesling when the Grosses Gewächs classification was introduced. But it turns out this has the same flaw as the old system: trocken can allow up to 9 g/l residual sugar, on the assumption that acidity will be high enough to hide the sweetness. But it takes only one case, such as my recent experience, to throw the whole system into doubt; I’m not sure I will take the risk again. Uncertainty is completely lethal in a restaurant, especially given the prices on wine lists these days.

I will certainly concede that the playoff between sugar and acidity can give quite misleading results. Take a series of Rieslings that are supposedly trocken and try to place them in order of sweetness or in order of acidity. Often enough, the order conflicts with the numbers, because of the way in which high acidity disguises sweetness. But there is really no guarantee that a wine will taste dry unless its sugar level is below the level of perception. As Armin Diehl, of Schlossgut Diehl, former editor of the Gault-Millau guide to German wines, said when I asked him about the 9 g/l limit for trocken wines, “This is a nonsense: internationally dry is less than 4 g/l.” If they are really serious in Germany about persuading consumers that they make dry Rieslings that are suitable for matching food, trocken needs to be limited to 4 g/l residual sugar. That’s the level at which almost no one can detect sweetness. As I’ve suggested before, there should be another classification for wines that have more than 4 g/l but less than 9 g/l, and which are intended to taste dry. I am sure the German language is up to producing some very complicated compound adjective, which means, “has some residual sugar but actually tastes dry.”

Tasting Notes

Forster Freundstuck, Grosses Gewächs Riesling, Georg Mosbacher, 2007

The first impression is richness rather than finesse: not surprising for the Pfalz. The nose is fairly muted but gives more of an impression of stone fruits than citrus. The palate follows the same spectrum, but with more of an edge of citrus on the finish, and just a touch of petrol to identify the variety. But the wine is palpably sweet: it does not taste as though residual sugar is below the 9 g/l for the trocken designation. The soft richness of the wine attests to relatively low acidity, which presumably lets the sweetness show through so clearly. Alcohol at 13% attests to the ripeness of the grapes. At all events the balance of sweetness to acidity does not justify the trocken label. This is not what Grosses Gewächs is about, and betrays the whole principle.  87 Drink now-2016.

Loibner Berg, Franz Xaver Pichler, Grüner Veltliner Smaragd, 2007

Very dry is the first impression on nose and palate, almost spicy, although not as finely textured as Pichler’s Rieslings. The fruits show as citrus, inclined towards grapefruit, but with a spicy edge (some might call this white pepper). The citrus fruits give a strong impression of minerality, which again brings an almost Riesling-like note to the finish. The overall impression is that the wine is very young, it really needs some time to integrate and rub off the rough edges: it’s a shame to drink it so soon, it seems to be made for aging.  89 Drink now-2017