The Annual Red Bordeaux Report

Two thousand four is likely to be yet another underappreciated Bordeaux vintage—the third in four years—that will offer veteran claret drinkers the opportunity to purchase many excellent wines at prices that seem downright reasonable in the context of today's international wine market. It is not a vintage of dense, glossy, flamboyantly ripe and concentrated wines, and it will almost certainly not be a vintage for the investor/speculator. On the contrary, it is a vintage that featured a huge crop load, and a year in which the ripeness came slowly and late. There is a cooler cast to the aromas and flavors of 2004, and the better wines have a classic acid/tannin backbone. On the other hand, early reports of forbiddingly austere wines with underripe flavors and hard tannins were not borne out by my tastings in Bordeaux during the first half of April. I enjoyed the young 2004s very much, and I suspect that this is exactly the kind of vintage that will gain richness during élevage without losing its essentially firm structure.

The 2004 growing season. A cold March delayed budbreak until the end of the month and the first half of April, or a bit later than the average of the past ten years. Following a mild and somewhat rainy April and a close-to-normal May, the flowering started at the beginning of June, just as hot, dry weather arrived. After two years of short production, the vines were prepared to snap back with a vengeance. With weather conditions virtually perfect, the flowering was fast, even and out of control, ensuring a huge crop. Without problems of coulure or millerandage, there were a huge number of bunches, as well as a huge number of berries per bunch. The unusually warm and dry June, which featured sunshine hours far above the norm, set the stage for deep colors and concentrated tannins in the wines.

Then followed a July of average warmth, with the last third of the month the warmest and sunniest, and some rainfall occurring on and off through the month. The veraison, or the point at which the grapes stop growing and begin to change color, began in late July, but the demi-veraison, or the midpoint of this process, was in early August, still a few days later than the norm, and roughly the same as in 2002 and 2001. August was then the worst month of the growing season, with more rain than usual and steadily decreasing temperatures toward the end. Vines had difficulty ripening their heavy crop loads, and an element of dilution was introduced, especially where estates had not ruthlessly thinned bunches by the end of July.

September began warm, but with storms and humidity threatening to trigger grey rot. The vintage looked bleak at that point, as the huge crop was not yet ripe. It then turned dry and sunny for the rest of the month and into early October, with warm afternoons and cool nights. At Château Latour, for example, afternoon temperatures exceeded 77oF during 20 days of September. Overall sunshine hours were slightly above average for the month, and total rainfall was well below average. October was then also mild, with much warmer than average temperatures during the first ten days, normal temperatures during the middle ten days, and then atypically hot weather again at the end. The middle third of the month was difficult, with intermittent rainfall, some of it substantial, and many estates that had not finished harvesting by then stopped picking during part of this period.

As in 2002, the vintage was saved by excellent weather in the weeks leading up to the harvest. Still, lesser properties—those that cannot afford the expense of making additional passes through their vineyards to thin the bunches, pull leaves, and so on—had excessive crop levels and have made wines that lack intensity and ripeness. Many of these wines are distinctly green and meager. At the level of the top estates, however, there are successes in every appellation, and at this early stage it is difficult to choose between the Left and Right Banks.

It was perhaps more a matter of money than of Mother Nature. Christian Moueix explained the stark contrast between 2004 and 2003, in terms of the work that was necessary in the vineyards. "In 2003, we should have done as little as possible, because the season was so hot and dry: no leaf pulling to get more sun on the fruit, no green-harvesting because the crop was already tiny. Petits châteaux did less work because they can't afford to spend much money on labor, and they mostly got better results. In 2004, it was exactly the reverse. It's a disastrous vintage at the level of the generic wines. We spent 19,000 extra man hours in our vineyards in 2004; especially on the Pomerol plateau we worked very hard. But for many properties elsewhere, this extra work was simply too expensive."

Although the harvest was eventually quite late, at the level of the top châteaux the fruit ripened well. (At Chateau Lagrange, to cite just one example, the starting dates for both merlot and cabernet were the latest since 1987, although only marginally later than 2002 and 2001.) While some properties in Pomerol brought in a good portion of their merlot during the last week of September, estates on the cool limestone plateau east of the town of St. Emilion routinely picked their cabernet franc well into the second half of October. Similarly, most merlot in the Médoc was harvested during the last week of September and first week of October, while the cabernet sauvignon, cabernet franc and petit verdot were mostly picked during the first half of October.

The 2004 vinifications and wines. The young 2004s are deeply colored wines, as the dry spring and spectacular June produced high levels of both anthocyanins and tannins. The wines generally possess substantial tannins, but the best wines have enough middle-palate stuffing to support them, and the tannins are sufficiently ripe. Some of the latest-maturing grapes have wonderfully full, sweet tannins; they clearly benefited from more thorough phenolic ripeness. However, with crop levels very high, many estates unintentionally exaggerated the tannins in their wines, either by bleeding off 20% to 30%, or more, of their juice through the process known as saignée, which can throw a wine off balance, or by concentrating the must through high-tech methods such as reverse osmosis or vacuum evaporation. These latter methods are primarily used to raise potential alcohol levels in the must by removing water, but insufficient grape sugars were not normally a problem in 2004 and thus only a handful of the better estates made more than sparing use of these techniques for fear of getting overly alcoholic wines.

While relatively few wines at the level of the better estates are clearly underripe, flavors do run to cool, fresh, fragrant and classic; rarely do the 2004s display any of the exotic or roasted character of 2003. Often even the merlot shows the floral character, fine-grained texture and firm structure of cabernet. Bordeaux-based wine critic Jean-Marc Quarin theorizes that the exceptional June weather explains why there are rarely vegetal smells or tastes in the 2004s—i.e., few traces of methoxypryrazines, the molecules that give a smell of green pepper to wine.

While merlot tended to be harvested under dry, sunny conditions during the last part of September and first third of October, the cabernets may have benefited from longer hang time, even if they were often picked during and after periods of rain. Although the grape clusters were huge in 2004, and the number of grapes great, the size of each individual grape was closer to normal. In general, the cabernet grapes were smaller than the merlot, and thus the late-ripening varieties may be more likely to have creamier middles and nobler tannins. But relatively few 2004s possess outstanding flavor authority and concentration, or the grip and palate-staining length of truly great wines. This should generally be a vintage with at least medium-term ageability, with the better wines likely to be at their best between 6 and 20 years after the vintage—and more like 8 to 25 or 30 for the first growths. Most of the rest should be consumed within their first 15 years, particularly those wines made from huge crop levels.

Clearly there are similarities in style between the 2004s and the 2002s. The earlier vintage had the advantage of much lower crop levels, while 2004 generally had the edge in weather. Where top estates could control yields in 2004, they have generally made wines with greater density, lower levels of acidity and riper tannins than the 2002s. More than one Médoc proprietor compared the cabernet sauvignon in 2004 to that of 1996, while a couple of proprietors on the Right Bank felt that they have made their best wines since 1998, a superb vintage, especially in Pomerol. In many cases this may be wishful thinking. Several proprietors on both sides of the river described their young 2004s as finer and more floral than the 2000s, a crop of wines that are mostly in a dumb stage today.

The wines presented at the Union des Grands Crus tastings during the first week of April were extremely young. The harvest had been late, the malolactic fermentations were late, and the wines were blended late. Wines that did not finish their secondary fermentations by early winter were further delayed by a freakishly cold period in Bordeaux in late February and early March. A few wines even had some unconverted malic acidity at the time of my visit. Most tasters I spoke to in early April told me that the wines had been extremely difficult to taste during the last week of March. They had become more accessible by the first week of April, and showed even more texture and personality the following week.

To buy or not to buy. Vintage 2004 is not likely to be an investment-grade vintage, and thus it would appear to make little sense to purchase 2004s as futures. With relatively few out-and-out exceptional wines, with worldwide demand generally restrained, with a lot of available cash having already been spent on the 2003s, and with the Euro currently strong, there is little reason to believe that any but the very scarcest items will quickly disappear from the marketplace and escalate in value. Obviously, demand for the vintage will depend to a great degree on opening prices, and it is entirely possible that some wines will open at prices lower than the 2002s. Even so, many U.S. importers are reluctant to commit to purchases now with a weak U.S. dollar. They are hopeful that by the time the wines are ready to be shipped, the dollar will have improved.

In my extensive tastings in early April, I focused on the 2004s but also sampled as many 2003s and 2002s as time, and the châteaux, would allow. (At a few large group tastings, for example, I limited my attention to the young 2004s and the bottled 2002s, simply for logistical reasons.) Vintage 2002, like 2001, is underpriced in the retail market (2000 and 2003, in contrast, are distinctly overpriced for their quality), and claret drinkers who are not afraid of wines with firm acidity can find many bargains among the better 2002s. Many cabernet-based wines of the Médoc are very good, and numerous borderline-outstanding bottles can be found at retail today in the $30 range, making them exceptional value against today's cabernets from California. (In my Bordeaux coverage in this issue, price ranges shown for 2002s come from a dozen or so major retailers across the country who have these wines on their shelves or have published prices for wines about to arrive.) I will offer a report on Sauternes, with notes on 2003s and 2002s, in the next issue.

A second look at 2003. Vintage 2003 is a fascinating though extreme vintage on which I advise extreme caution. Although some wine critics are describing 2003 as a great vintage, it must be pointed out that the Bordelais themselves are generally lukewarm on these wines. They are less likely to view 2003 as a vintage with wonderful ripeness than they are simply to consider the wines to be extreme, if not freakish. The exotic roasted character of the vintage often dominates; many wines show distinctly liqueur-like aromas—coffee and chocolate liqueur as often as fruit liqueur—that seem far too advanced even when the wines are reasonably fresh on the palate. I have a hard time considering vintages so far outside the norm to be outstanding, even if at the level of individual properties numerous flamboyantly ripe and plush wines were made.

Many 2003s also struck me as extremely oaky in early April, perhaps more oaky at this stage of their evolution than any recent vintage. With their high pHs, high alcohol and low acidity, these wines have soaked up the oak and are wearing the wood today like a winter coat. Négociant Bill Blatch refers to this as the Jack Daniels syndrome: high alcohol and low acidity. He compares some of these wines to the 1989s, a number of which he says are still too dominated by their oak today. It must be noted that many châteaux purchased more new barrels than they ultimately needed in 2003. When the crop shrunk as the grapes dehydrated in late summer, some properties probably used a higher percentage of new oak than they should have. (A high percentage of new oak may also have been a bad idea with the 2004s, as most of these wines already have enough skin tannins.)

This is not to say that there aren’t some great 2003s. The best of these wines (and here I’m talking about wines like Montrose and Latour) are monumental; they may well turn out to resemble the ‘47s. While I was not around to taste the ‘47s in their youth (I did not taste a ’47 until 1980 but have tried since then to make up for lost time), I would have to say that I’ve gotten more pleasure from this vintage than virtually any other one of the 20th century.