Tasting the Label

Raymond Oliver, owner of Le Grand Véfour, tasted a 1972 Napa Valley Chardonnay.

"Ah, back to France!" he said.

Another judge tasted a 1973 Bâtard-Montrachet. "That is definitely California. It has no nose."

That wine was French.

At the InterContinental Hotel in Paris, on 24 May 1976, staff served wines one by one. White gloves, white tablecloths, a tasting conducted in silence. The judges could see the glasses, but not the labels.

That kind of certainty is hard to question. If you cannot taste what the expert tastes, doubting them can feel like admitting you lack the skill. Their confidence becomes another reason to trust them.

Oliver was no novice. He belonged to the French food and wine establishment. That made the mistake harder to dismiss.

The judges had the wine in front of them. The country was missing, and their comments kept trying to put it back.

Steven Spurrier organized the tasting with Patricia Gallagher, his American business partner. He was an English wine merchant with a shop and wine school in Paris.

Spurrier had invited nine French judges. Among them were Aubert de Villaine of Domaine de la Romanée-Conti and Odette Kahn, editor of La Revue du Vin de France. Christian Vannequé was head sommelier at La Tour d'Argent.

These were people whose work depended on knowing wine. The tasting asked them to use that knowledge without knowing the producer or country on the bottle.

Spurrier had told them they would taste Californian wines. Only once everyone was seated did he reveal that French wines would be included.

They tasted twenty wines in two groups of ten. Each group had six Californian wines and four French: Chardonnay against white Burgundy, then Cabernet Sauvignon against red Bordeaux.

George Taber of Time was the only journalist there. He held the list showing which wine was which. The judges did not.

That list let him hear something the judges could not hear in themselves. Oliver had been confidently wrong about the country. What sounded like recognition was a guess.

The scorecard

But when the scores came in, California won both groups. The white winner was the 1973 Chateau Montelena Chardonnay. The red winner was the 1973 Stag's Leap Wine Cellars Cabernet Sauvignon.

Six of the nine judges gave Montelena their highest white score.

Kahn herself ranked Stag's Leap first. After the results, she demanded her scorecard back. Her request was unsuccessful.

She did not want the world to know her scores. If they became public, she would have to answer for them under her own name.

Kahn never spoke to Spurrier again. Spurrier was banned from wine tastings for a year. The response reached beyond one judge and one sheet of paper.

Taber's report appeared in Time on June 7, 1976. It ran to four paragraphs.

"It was filler. It didn't even have a photo," Taber later recalled.

For the Californian winemakers, those paragraphs carried further than their labels had. Warren Winiarski, the winemaker behind Stag's Leap, recalled: "The phone started to ring pretty quickly."

But one afternoon, one panel: could California's win have been luck?

A narrow win

The red result leaves room for that question. Stag's Leap beat the 1970 Mouton-Rothschild by just 0.17 points on the judges' average scores.

In a 2026 reanalysis, Richard Ballantyne of Noble Grape argues that Haut-Brion comes first under rank-based methods. That is one analysis of the judges' scores, not the result announced that day.

It does show why the winning order needs care. Different ways of combining judgments can produce different winners from the same scorecards.

One panel on one afternoon proves little about which country makes better wine. Nor does a blind tasting measure exactly how much a visible label changes a score.

I find the guesses more useful than the winning margin. Oliver found France in California. Another judge found California in France, and treated the supposed origin as a reason to dismiss the wine.

Their skill had not disappeared. Their ability to judge wine was precisely what gave the results weight. But the missing labels exposed a gap between what they could taste and what they believed they recognized.

The belief that French wine was superior had been doing work they could not taste in the glass.

Take the names off

That belief has equivalents in how work gets approved.

Imagine your next review with the names taken off the work. You do not know whose idea it is, which agency made it, or which brand stands behind it.

Write down what you think before the names return. Then reveal them and check whether your judgment changes.

A change would not prove the name was irrelevant. A producer's record can carry useful information. But you would have something specific to examine: what did the name tell you that the work did not?

If your judgment held, that would matter too. Removing a label is a test, not a demand that you reverse your answer.

The approval meeting is one place where the Perception Engine works. It cannot tell a bad idea from a good one that is simply strange. It stops both.

Taking the names away gives the strange idea a chance to be judged before its source explains it away.

I understand the appeal of the safer choice. A familiar recommendation comes with reasons other people already accept. The counterintuitive choice has your name on it, and you may have to own the failure without their cover.

Taking the label off is simple. Acting on what you find can take courage. You have to let the record stand after the names come back.

The labels were hidden. The judges weren't.

I write down the labels I catch myself tasting, as I find them.

Part of The Perception Engine