1997, 1996 and 1995 Bordeaux

At the beginning of April of last year, I enjoyed a perfect week of tasting weather in Bordeaux: dry, sunny days followed by cold, clear nights. But the chateau owners and winemakers I was tasting with were in a state of constant fear-and it wasn't my ratings they were worried about. Following a warm, dry February and March, the vines were far ahead of their normal schedule, and as a result were extremely vulnerable to a sharp spring frost. Indeed, each night during my tour of the Bordeaux region temperatures reached the freezing mark, but almost miraculously never went lower. Chateau proprietors, keeping their fingers crossed, were well aware that the '97 vintage could be the earliest since 1893, and thus were quietly hoping for a great year. Somehow, the region made it through the frost season, except for some pockets of damage in the Graves in late April. But then came the growing season from hell.

The growing season. The key characteristic of the '97 growing season, the harvest, and the resulting wines is wildly inconsistent ripening of the grapes, not just from site to site, but from vine to vine, bunch to bunch-and even, amazingly, within bunches. The problem started with the early but prolonged bud-break. Then a bone-dry April delayed the flowering. May featured wild weather: a period of cold nights early in the month, rainy spells during the middle third of the month, and then a return to cold nights followed by a blast of unseasonable heat toward the end of the month. Under such difficult conditions, coulure and millerandage were widespread, especially in right-bank merlot. Ultimately the flowering was spread out over a full month, bringing about an extremely uneven start to the ripening cycle (with the merlots affected more than the cabernets and the sauvignons more than the semillons), but the official "mid-flowering" date, May 23, was still the earliest on record.

Although the first half of June was warm, and there was moderate rainfall, it soon became clear that the bunches of large grapes were not evolving well, possibly due to the cold snap at the end of the flowering. Then came wet and cold weather during the second half of the month, raising the specter of grey rot and requiring treatments to prevent oidium. Most of July was cool and humid; the vines grew quickly while the development of the grapes remained slow. The most conscientious growers repeatedly trimmed their foliage.

The mid-veraison, on July 31, was also the earliest on record, but the grapes were considerably larger than average, due in part to the June rains. Still, as sugars were high and acidity adequate, there was still a chance for a superb vintage-especially if the grapes could lose some of their excess weight during a period of dry, warm weather. Most estates did green harvests to eliminate the less-ripe berries and bunches.

From late July until late August, under hot and humid conditions verging on tropical, the vines once again spent much of their energy on growing their foliage rather than on ripening and concentrating their grapes. As one negociant put it, "In 1997 the vines just didn't want to have sex." Quality-obsessed growers took desperate measures in July, August and even early September to eliminate hopelessly underripe fruit, and to reduce crop loads further so that the vines might have a chance to ripen what remained.

The harvest. The white wine harvest began on August 18, the earliest start since 1893. But relatively few of the best white grapes were ready to be picked before the beginning of September, and in the meantime there was fairly heavy rain during the last week of August and on the first of September. The quality and ripeness of the white grapes, especially sauvignon, were variable, and the harvest was protracted.

For red wines, too, the late-August rains dashed any hopes for greatness. Much early-maturing merlot, especially on the right bank, had to be picked soon following the rain, and was distinctly weak. But then the weather turned warm and dry, the early panic picking stopped, and growers began to push back their harvest dates in hopes of getting better ripeness. As it turned out, there was virtually no rain between September 2 and October 5, an almost unheard-of streak of luck. The harvest for merlot generally began on September 8 in the Medoc and continued for up to two weeks. Most right-bank merlot was brought in by the 22nd, but some was harvested into early October.

The cabernets, which had resisted the rains more successfully than the merlot, were harvested over an even wider time span. In general, those who picked early tended to get weaker material, while those who waited benefitted from longer hang time (in some extreme instances, a record 140 days), which brought greater development of flavors even if grape sugars increased only grudgingly. Showers on October 6 helped the remaining cabernets reach ideal maturity.

The wines. Owing to the vastly irregular ripening of the grapes, problems with the sauvignon, and a drawn-out harvest, the dry whites are uneven, though the handful of top examples I tasted seemed quite good. Sauternes benefitted from a slow onset of botrytis, and the wines have retained a vivacity that has many early tasters comparing them to the '88s. 1997 appears to be a worthy follow-up to 1996, which itself was the first truly interesting Sauternes vintage since 1990. Due to space constraints, I will cover Sauternes in the next issue.

In theory, the early-ripening right-bank merlots were most affected by the rainstorm on September 1, because these grapes had to be picked shortly after the rain. But the Medoc, which received less rain on September 1, enjoyed more days of dry weather before the harvest began. And, of course, the Medoc features a higher percentage of later-harvested cabernet, whose skins weathered the rains more effectively. Very early reports were that the Medoc outperformed St. Emilion and Pomerol, but my own extensive tastings suggest that there will be some surprisingly good right-ank wines, in most instances from vines harvested late.

As a rule, the '97s possess rather gentle structure, with polyphenol levels lower than those of the two previous years. The early start to the season and the fine September weather ensured that most chateaux picked a reasonably high percentage of ripe fruit. Still, many wines I tasted on my tour of Bordeaux at the end of March and the beginning of April of this year betray some signs of less-than-ideally-ripe fruit. Many winemakers went for gentler extraction, pumping over less frequently or for shorter intervals, or shortening the duration of additional maceration after the end of the alcoholic fermentation. They feared that heavier extraction would not give them what the fruit did not possess in the first place, and would risk throwing the wines off balance with astringent tannins. But some properties employed more extractive vinification techniques: some of these wines are impressive for the vintage, while others lack balance and charm.

Prices. Yields varied widely in '97, depending not just on the success of the flowering but on the amount of crop eliminated through green harvesting and subsequent passes through the vines to eliminate less-ripe fruit. Many proprietors declassified a higher percentage of their production than ever before in order to maximize the quality of their grand vin. They point to their considerable financial sacrifices when justifying high pricing for '97s, which in most cases are above the opening prices set for '96s last spring. The problem, in a nutshell, is that 1997 has produced many attractive wines best suited for drinking over the medium term, but they are being priced like cellar treasures. Relatively few '97s possess the solid structure, acidity or grip for long-term aging.

1996 and 1995 revisited. My coverage on the following pages includes follow-up tastings of most important '96s-in most cases sampled within two months prior to their bottling-and notes on the finished '95s. As I reported in my earlier coverage of these vintages, each produced many outstanding wines. In '95, many of the high points came from the right bank, where silky, very concentrated and thoroughly ripe but not overripe wines were made in both Pomerol and St. Emilion. But there were also a number of outstanding '95s made in the Medoc and the Graves. Nineteen ninety-six, on the other hand, was a year that clearly favored the cabernets. In many instances, Medoc cabernet sauvignon is extraordinarily expressive: ripe without being roasted, and supported by substantial ripe, sweet tannins. The best of the cabernet sauvignon-based '96s are uncommonly classy, suave wines-classic versions of cabernet that avoid the stewed fruit character of so many '82s and '89s, the harder tannins of some of the '86s, and the green notes found in so many '88s, '93s and even '94s.

As always, I have provided ranges for unfinished wines: my notes and projected scores should be regarded as preliminary in nature for the '97s. Prices listed for '95s come from a handful of major retailers who are offering these wines; but please note that prices can vary widely depending on when, and how, retailers purchased these wines. Many '95s are already sold out at the wholesale level.