Two: the poetry of French wine is expressed through the verse of the vintage. Everywhere, that is, except one region… where most of the wines carry none. Three: we all know that great wine is in part a consequence of low yields. To suggest that the highest average prices obtained by any French wine region go to the zone with the highest yields would, of course, be ludicrous – were it not true.

Champagne lies on the other side of the looking glass: it’s French wine, contrariwise. What can we learn from it? The first Champagne lesson is one I deplore: terroir simply doesn’t matter to most consumers. The elephantine simplicity of its single AOC is one of Champagne’s greatest assets.

The second Champagne lesson is that sparkling is a different animal to still wine. It is less agricultural, and more industrial; it’s a wine of process as much as of place. There are two consequences, both with flavour implications. The first is that many Champagnes taste of process (those creamy, yeasty, biscuit flavours) more than place. The second is the sometimes sublime, but more often bland tastes which result from blending. As in the world of grain-based drinks like beer and whisky, an emphasis on process rather than place gives larger producers an intrinsic advantage over smaller ones. That advantage is reinforced by the global nature of Champagne sales. Whenever larger producers become the dominant force in a wine region, blending will become the production norm, regardless of whether or not that is the best way of maximising quality. Quantity quickly requires it.

The third Champagne lesson is that vineyards on the viticultural margins work via different mechanisms to those which lie beneath a generous sun. Vines in the far north (or, in the southern hemisphere, the far south) face a nervous, jumpy growing season beset with disappointments, and they respond nervously in their fruit bearing, too. The best you could hope for in Champagne in 2003 was 7,500 kg per hectare. In 2004, the equivalent figure was 23,000 kg. It may seem counter-intuitive to winegrowers in Bandol or Châteauneuf, but high quality in Champagne often coincides not with low, but with high yields – as in 1982, 1990 or 1995. Low yields usually mean lousy weather. Moreover, if you do deliberately restrict the vines by hard pruning, you can
end up not with Champagne but with sparkling Burgundy. The stamp of variety and place is over-strong; the defining finesse and restraint has been pruned away.

The fourth Champagne lesson is linked to the second and third. It is that in producing sparkling wine from (as tends to happen) vines on the viticultural margins,pressing assumes undue importance. It becomes a key act of vinification. Exactly how this is so is a matter of controversy, but everyone takes the matter seriously and throws away juice percentages which would be incorporated further south. Champagne is not as good as it could be; no wine ever is. There are improvements to be made, most notably in the overall quality of grapes and in the mental readiness of producers to take new risks. Its lessons are of more interest to sparkling wine producers worldwide than to any of its French confrères – though as yet no region has managed to convince as a Champagne rival. Could England be the place where strong Champagne challenging wine brands eventually emerge and where the lessons are most profitably drawn? In 50 years, with a 1°C or 2°C temperature rise, just possibly.