New Paris Cuisine Is A Challenge for Wine

Four days in Paris last week refuted the idea that haute cuisine in France has run out of steam. Every dinner was different and innovative, but a theme that seemed to run through the evenings was the introduction of Asian spices. This leaves me wondering whether the traditional matches of wine and food still stand up in France or we need to rethink.

This question has struck some producers. François Milo of the producers’ association in Provence says that, “The mondialization of cuisine has benefited rosé. In France there has always been a fixed idea of which wines (red or white) accompany certain stages of the meal. But it’s difficult to pair red wines with international foods. I think that for the future, rosé is a vin de liberté.” I did not go so far as to try rosé – for one thing there aren’t that many rosés with enough flavor interest at this level, and for another choices on restaurant lists are very limited – but I did vary my usual thinking on suitable combinations.

Turbot in coconut sauce was a definite challenge the first evening at restaurant Auguste. In fact, I found the coconut influence a bit too strong for the delicacy of turbot. Overall this seemed to offer a similar challenge to dishes of stronger fishes prepared with vanillin a few years back. Then we had gone for a white Côte de Beaune to match: this time we went for Louis Michel’s Montée de Tonnerre 2012 from Chablis to find a bit more contrast. The Chablis didn’t have quite enough minerality to cut through the coconut, but it resisted well. Actually I liked it better than the Valourent of the same vintage, tasted a few days earlier, which seemed to have a surprising amount of forward fruit: the Montée de Tonnerre at least had intimations of minerality, although I’m not sure how far they will develop with time.

There were two fish dishes at l’Arôme: légine australe on asparagus, and turbot on rhubarb. (The first was unknown to me but tuned out on investigation to be the same as Chilean sea bass, a.k.a. the Patagonian toothfish, except that it apparently comes from waters off Africa.) These were quite strongly flavored dishes, too strong I felt to match white Burgundy, but Jonathan Pabiot’s Pouilly-Fumé Predilection from 2012 provided a brilliant contrast. This is very much the New Pouilly-Fumé, all delicacy and elegance: in fact, the Anima Figure (my companion) described it as ethereal. (The antithesis of New World Sauvignon Blanc, demonstrating wonderful range for the variety, if raising the question of its true typicity.) A course of chicken oysters and gambas was less successful.

Friends in Paris had managed to obtain a table at l’Astrance for the next evening. (This had required 35 phone calls on the day booking opened.) Choosing wine is a little tricky since the menu is a surprise, but on the basis of some hints from the sommelier, we decided that a light red would be most appropriate, and went for Domaine Dujac’s Morey St. Denis 2002. (The wine list at l’Astrance is extraordinarily fairly priced, a big contrast with most other restaurants in Paris, although you can usually find some wine where the sommelier has a special connection and price is more reasonable.) The red proved extremely versatile, going well with the famous cake of fois gras, langoustine with Asiatic influences, and légine australis again. (They were a little put out at l’Astrance to discover we had had the same fish the evening before at l’Arôme: apparently there are only five boats fishing for it, one from France, which presumably supplies both restaurants). Curiously, the final lamb dish which should have been the best match for the wine didn’t quite come to life, although the wine showed a wonderful combination of crystalline brilliance reminiscent of Volnay and femininity of Chambolle Musigny. Fighting well above its communal level, you might say.

Finally abandoning restaurants with names starting with “A”, our last evening was at Jean-François Piégé. The main courses were hommard bleu (cooked in blackcurrant leaves) and turbot in a curry sauce. One of the preceding dishes was asparagus in a sauce in which I thought I also detected curry, but which turned out to be saffron pistils. The wine was a no-brainer as there was a strong selection of Raveneau premier cru Chablis at reasonable prices. We had a Vaillons 2005, which turned out to be noticeably richer than usual for Raveneau, but still showing that characteristic anise and minerality on the back palate. Possibly a leaner year would have been an even better match for the food.

I can hear a cry going up: why no Riesling? It’s a wonderfully versatile grape that matches a wide variety of foods, especially good against Asiatic spicing, and is undervalued. I would concede the principle immediately, but my problem with Riesling is that nowhere – Alsace or Germany or anywhere else in Europe – is the principle accepted that there is an international standard for dry wine: less than 4 g/l of residual sugar. So I am almost never certain enough that a wine will be dry. Producers may argue that it tastes dry if acidity is sufficiently high, but that’s a matter of subjective judgment, and I prefer not to take a risk in a restaurant. (And asking the sommelier has resulted in too many wines which were stated to be dry but on which residual sugar could be tasted.)

I believe l’Astrance started the move in Paris to surprise menus. I was struck by the fact that three evenings out of four we had a surprise in at least some courses. At both l’Arôme and Piégé, you choose your main course(s) – you can choose either one or two from a short list – but the starters and desserts are a surprise from the chef. It’s a neat solution to the difficulties of providing many choices at every course which must simplify issues like food wastage and buying-in for the restaurant. Of course, you have to be a top-line chef to pull this off. A consequence is that it does make it more difficult to find an appropriate wine. Wines by the glass chosen to match the food are offered by most of the restaurants, but my past experience is that this can be a bit erratic in providing interest in the wine.

On the last evening at Piégé, I said to the maitre d’ that a series of interesting dinners seemed to put paid to the idea bruited a few years ago in the Anglo-Saxon press that haute cuisine in France had died a lingering death. “I would have agreed with the idea five years ago,” he said, explaining that the rush of innovation is a revival of the past few years. Granted that there are similar influences, each interpretation is different: I wonder where it will go next.

Can Rosé Be Interesting?

I admit that I have always had doubts about rosé, where the compromises you have to make to get the color right sometimes seem to be at the expense of flavor. I could understand the position of the sommelier at a Michelin-starred London restaurant, who when I said I was going to visit producers in Provence, commented “Rosé is not wine.” The tricky thing is that you make rosé more or less like a white wine, but from black grapes. Can you really make an interesting wine from black grapes if you’re constrained by needing to harvest earlier (to maintain freshness), so the grapes don’t reach the ripeness levels they would have for making red wine?

For my palate, rosé should avoid showing overt red berry notes and should be savory, perhaps even slightly saline. That is probably why in general I don’t much favor rosés made from Grenache, which basically accounts for most of French production, from Provence, Languedoc, and the Rhône. Nor do I like the slightly sickly quality of most rosé from the Loire, which is required to have 10 g/l residual sugar. (I am inclined to think it’s a bit of a waste to use Cabernet to make sweet rosé.) My preference in rosé tends to be for Sancerre, coming from Pinot Noir, and at its best showing an elegant minerality. (One of the best rosés I have had was declassified from the Clos des Lambrays in Burgundy.)

The movement towards producing “premium” rosés in Provence seems to be in the right direction as I find they usually tend to show less red berry fruits and more savory minerality. (I’m not convinced, however, by the move towards defining sub-appellations for the Côtes de Provence on the basis of terroir, as so far differences in the wines seem to relate more to vinification than to terroir. I will be very interested if anyone demonstrates terroir in rosé.)

Château d’Esclans is at the extreme end of the move towards premium rosé. I tasted all four of its cuvées this week. Whispering Angel, of course, has taken the world of rosé by storm, although actually it comes mostly from purchased grapes. Even though it’s the entry level, I think it has a very good balance of fruity to savory elements, giving it some flavor interest. Moving to the chateau wine, which goes further into the premium class, the balance shift a little more in the savory direction. With the two super-premium cuvées, Les Clans and Garrus, there’s a real sense of underlying structure. Tasting the 2011, Les Clans is now just right, but Garrus has a slightly phenolic or even tannic impression suggesting that it actually needs a little more time.

I asked owner Sasha Lichine what makes the difference between the cuvées: terroir? yields? grape varieties? vinification? “All the wines start out in the same way. Temperature is brought down to about 8 °C with dry ice, the wine goes into closed presses under nitrogen, and as soon as juice comes out it is tasted and assigned to one of the four cuvées. The wines come 90% from free run juice, with the rest from pressing, so there is essentially no maceration. The cold system is the key,” Sasha explains. “What used to happen in Provence was that people were picking too early and then macerating to get the color out. We manage to pick a bit later and riper.”

All the wines are dominated by Grenache, but there’s a small change in other varieties going up the ladder. There’s also a change in vinification: Whispering Angel spends a short period in cuve, Château d’Esclans spends six months in cuve and wood, and Les Clans and Garrus spend ten months maturing half in cuves and half in old demi-muids (600 liter casks) with frequent battonage. You can see the extra richness that comes from the battonage.

There are some differences in sources–Garrus tends to come from the oldest vines–but the basic difference between cuvées is selection. “For the first vintage of Garrus we just creamed off the best three barrels, and then for Les Clans we creamed off again,” Sasha recollects. What’s the intended difference in style? “As you go up the ladder, the wine fills your mouth more, you should not necessarily taste the wood, the wine should keep its freshness, and by Garrus you should get tiny tannins.” The change along the range is a striking demonstration that rosé is not necessarily just rosé, but that it can have distinctive character and (to some extent) ageworthiness. As this is unprecedented for rosé, it’s hard to judge how softening with time will play out against development of flavor, but at a rough guess I would be inclined to drink Whispering Angel and Château d’Esclans when they are released, Les Clans about three years after the vintage, and Garrus about five years after. The relative prices of the cuvées more or less indicate their relationship. When Sasha purchased Château d’Esclans in 2006, he was quoted as saying his aim was “to make the best rosé in the world.” Whether Garrus is the best rosé in the world is debatable, of course, but it is probably the most expensive.

Champagne Diary part 2: Will the Real Rosé Come Out Please. Visits to Bruno Paillard, Jacquesson, and Laurent-Perrier

I am going to start with a confession. I do not usually like rosé Champagne. It tends for me to have a softer, slightly muddier flavor profile that loses the essential liveliness I look for in Champagne. Sometimes there are faint red berry sensations that seem disconnected from the rest of the flavor profile, at extremes giving a slightly synthetic impression. It can be difficult to resolve what seems to be an intrinsic conflict between being rosé and being Champagne.

Naively you might think that a rosé champagne should come from Pinot Noir and so should be a sort of über Blanc de Noirs. But not a bit of it: rosé can come from any of the grape varieties. Elsewhere in France, a still rosé wine can be made only by allowing skin contact between juice and the skins of black grapes. My problem with rosé Champagne may relate to the fact that it is an artificial construction, the only wine in Europe that is allowed to be made by blending red wine and white wine. So it can in fact be Chardonnay, with a little red wine (usually from Pinot Noir) added to give color. But for me, a rosé should have some Pinot Noir character.

But why should it make a difference whether a rosé is made by allowing 100% of the juice to have a small amount of skin contact with the skins or by adding a small amount of red wine that has had much more extensive skin contact? Perhaps the character of what is extracted from the skins is different. Actually, the general view is that it’s not so easy to tell the difference between a rosé Champagne made by assemblage (blending) and one made by saignée (allowing limited skin contact for all the juice). I was given pause for thought about this, however, when during one day in my week in Champagne I tasted three rosés made by saignée. I spent the morning and lunch with Bruno Paillard, and then—running late as always—the afternoon at Jacquesson and the early evening at Laurent-Perrier.

The standard bearer for the saignée style is Laurent Perrier. “The rosé was launched in 1968 at a time when rosé was considered a froufrou,” says Ann-Laure Domenichini at Laurent-Perrier. It comes exclusively from Pinot Noir, mostly from Grand Cru, and maceration lasts up to 72 hours. It gives a sense of structure you don’t get with a rosé d’assemblage. You get a certain sense of the earthy quality of Pinot Noir here, almost a faint tannic character to the structure, perhaps more a sense of texture. I would say this has more structure than the Grand Siècle prestige cuvée but less evident refinement. I suppose it all depends what you want from your Champagne, but this is undoubtedly a rosé with character.

At Bruno Paillard, the exact details of production aren’t revealed, but Bruno says that the Pinot comes from a mix of short and long maceration. “At first I tried to make rosé only by short maceration, but I think that makes sense only for vintage.” About 15% Chardonnay is included also to help with freshness. I suspect this means that the nonvintage rosé is a blend between Pinot Noir exposed to short saignée (a genuine rosé), some Blanc de Noirs pressed straight off the Pinot Noir and also white wine from Chardonnay, and a little red also directly added (made by long maceration). All of Bruno Paillard’s wines are precise and pure, and the rosé is no exception. A vinous nose leads into a light palate with the citrus edge just taken off. It doesn’t exactly have red fruits on the palate, but makes a softer impression than the Brut. It’s just a little more textured than you usually find with rosé.

A rosé with real attitude comes from Jacquesson (well, actually, I would be inclined to say that all the wines at Jacquesson have attitude). I tasted the 2008 Terres Rouge, which comes from Dizy. It has a strikingly dark color, close to a light red wine. It is 100% Pinot Noir made by skin contact using two macerations: one for 20-30 hours, the other more briefly using free-run from pressing. The palate shows a rare case where red fruits are evident but completely integrated, with an aromatic, almost perfumed, quality of Pinot Noir. Softer on the palate than the other wines from Jacquesson, this is a rosé of unusual character.

Perhaps the question is whether a rosé Champagne should be more like a rosé or more like a Champagne? At Billecart-Salmon, which I visited last year, and which is the standard bearer for rosé d’assemblage, they are in no doubt. “If you served the rosé in black glasses, it would be important not to be able to tell, it should have the aroma and taste of Champagne not red wine,” says cellarmaster François Domi. Billecart-Salmon’s nonvintage and vintage rosés alike have the trademark elegance of the house, and are a major exception to my usual complaint about sparkling rosé. “In the rosé we balance the power of Pinot Noir with the Chardonnay. We look for elegance and delicacy, it should be discrete and not too heavy,” says François. “It’s a secret,” he says when asked what’s special about Billecart’s rosés, but later he relents and explains that it’s the quality and character of the red wine. “There shouldn’t be too much tannin, enough to stabilize the color but not more. If the red isn’t good, you cannot make rosé.”

Whether made by assemblage or saignée, the key to my mind is that any red fruit elements have to be extremely discrete and integrated into the flavor spectrum. Perhaps this is more easily accomplished when the wine comes exclusively from Pinot Noir, which should be true when it is made by saignée. But I haven’t done the acid test yet of seeing whether differences between the two production methods are evident in blind tasting.