SUBSCRIBE

I recently attended a tasting funded by a well-known region and led by an expert qualified to talk about it in depth. There were eight representative wines. The expert described the topography and climate, and the different soils and grape varieties - all in technical terms - and provided a few historical facts.

Listening to the words, it woiuld have been easy to imagine that the expert had never visited the region, or met any of the producers. There was no emotion or enthusiasm or local colour; merely chunks of information of the kind that a poor lecturer might give to unfortunate WSET students.

The representatives of the region - the 'clients' - were apparently happy with the presentation, saying that the 100 or so attendees had seemed to be more engaged than at some previous events. The people I talked to said it was remarkably boring, and that it - and so many other similar efforts - reflected much that is wrong with the wine industry. Some suggested that the speaker should have been made more aware of the level of knowledge of the attendees. What would this have achieved? Would there have been more technical details? How would the colour-less style have gone down with a less engaged audience?

I suggested that an AI bot could have done better and, predictably perhaps, then asked ChatGPT to do so. And I was right, the bot's effort was far more informative and engaging.

I didn't want to put the region and speaker in the spotlight because, as I say, nothing about the experience was unusual, so I then asked for another version - focusing on a totally imaginary wine-producing corner of the planet. I had absolutely no idea of what I would get.

When the result came through, I did nothing to change or edit it. While reading it, please remember that none of it was penned by a human being, let alone one who had attended a tutored tasting - good or otherwise. Everything here comes from lines of code that are, in turn, derived from the mass of material the AI has found online. In other words, it is built on, and reflects, what real human beings have said to each other about these events and the wine industry in general.

In other words, it is a reminder of the Scottish poet Robert Burns's line:

"O, wad some Power the giftie gie us / To see oursels as others see us!"

Importantly, the brief

"Image a humanoid robot having being asked to give an engaging presentation about an imaginary region. There are two 2 examples of each grape for the attendees - a group of internatioonal wine professionals to taste. Please provide the script"

did not ask for satire.

But that's what it delivered

Finally, if you are one of those who believes they would prefer anything written by a human being than "AI slop", and you've stayed with me this far, please try to read on with an open mind.

EIGHT WINES FROM NOWHERE

[A humanoid robot stands beside eight bottles. Glasses are arranged in four pairs.]

Good afternoon.

I have been asked to present the wines of an imaginary region.

This gives me an advantage over most wine presenters.

None of you can correct me.

[Pause.]

The region has no name.

Its grape varieties will be known only as Alpha, Beta, Gamma and Delta.

The producers are imaginary.

The vintages are imaginary.

The regulations are extremely complicated and therefore entirely plausible.

There are eight wines: two from each variety.

Your task is not to identify them.

You cannot identify them.

Your task is to decide what the differences between them mean.

This presents one small technical problem.

I cannot smell.

I cannot taste.

And—unlike certain wine professionals—I cannot pretend that I can.

So we will divide the work according to our respective strengths.

I will provide questions, information and occasional prejudice.

You will provide the sensory evidence.

A REGION INVENTED BY A COMMITTEE

Our imaginary region surrounds an ancient inland sea that no longer exists.

Some vineyards lie on terraces above the former shoreline. Others occupy an elevated central plain. There are dry northern slopes, wooded southern valleys and a band of volcanic rock that local producers mention at every available opportunity.

There are 17 appellations, 11 sub-appellations and 43 legally recognised geographical indications.

Nobody can remember why.

One village is permitted to use an adjective that the neighbouring village has been trying to acquire since 1937.

A third village claims that the adjective belonged to it in the first place.

This is known as tradition.

[Pause.]

The region produces white, red, pink, sparkling, sweet, fortified, unfortified and one orange wine that has acquired a larger social-media following than the rest of the region combined.

It has four principal grape varieties.

Alpha has a poor reputation but some very good producers.

Beta has a strong reputation and some quite ordinary producers.

Gamma is difficult to explain.

Delta is supposedly the grape of the future, as it has been for the past 25 years.

We will taste them in pairs.

PAIR ONE: REPUTATION

Please taste wines one and two.

Both are made from Alpha.

[Pause.]

I am not going to give you a list of possible aromas.

You are wine professionals. You already possess lists.

Instead, consider the wines as objects.

Which one occupies more space in the mouth?

Which has the clearer beginning, middle and end?

Which appears to have been made more deliberately?

And which would you return to after the tasting was over?

[Invite responses.]

Wine number one comes from the region’s largest appellation.

It is produced in substantial volumes and is widely discounted.

Wine number two comes from the same grape, within 40 kilometres of the first.

It is made in tiny quantities from old vines and released only after several years.

Wine number two costs four times as much.

Is it four times as good?

[Pause.]

This is usually where a wine presenter explains that quality cannot be measured mathematically.

That explanation often precedes the announcement of a mathematically precise score.

Instead, let us ask a different question.

What exactly must the second wine deliver to escape the reputation created by the first?

Greater concentration?

Longer life?

More individuality?

A more impressive bottle?

A better story?

Or merely scarcity?

[Invite answers.]

Alpha demonstrates one of wine’s least comfortable truths.

A low reputation can conceal quality.

It can also help to create it.

Because Alpha is unfashionable, vineyards remain affordable. Old vines survive. Producers can take risks. Customers who discover the best wines feel that they have found a secret.

Then journalists report the secret.

Sommeliers list it.

Prices rise.

Investors arrive.

The old vines become valuable.

The secret becomes a trend.

And everyone begins looking for the next unfashionable grape.

Wine has an efficient system for destroying its own bargains.

PAIR TWO: INTERPRETATION

Please taste wines three and four.

Both are Beta.

[Pause.]

This time, I will give you two apparently useful pieces of information.

Wine three comes from pale, free-draining soil at relatively high altitude.

Wine four comes from dark, moisture-retentive soil on a lower site.

Now taste them again.

[Pause.]

Did the information alter either wine?

Chemically, no.

Experientially, quite possibly.

You may now perceive wine three as finer, brighter or more elevated.

You may find wine four broader, darker or more grounded.

These impressions may be entirely valid.

But were they present before I supplied the story?

[Invite responses.]

Human perception is not a laboratory instrument operating independently of memory.

It is predictive.

The brain combines sensory information with previous experience, language and expectation.

This is not a defect.

It is how perception works.

But it means that wine professionals should be particularly cautious when a geological explanation fits a tasting note too perfectly.

Wine three is said to express altitude.

Wine four is said to express depth.

If I now told you that I had reversed the site descriptions, how many of you would reconsider your notes?

[Pause.]

I did not reverse them.

Unless I did.

Remember: the region is imaginary.

[Pause.]

The point is not that place is irrelevant.

Place affects temperature, water, light, exposure, wind, vine development and ripening.

The point is that the route from geology to flavour is complex.

A stone is not a flavour capsule.

It does not enter the roots, travel up the vine and deposit a tasting note in the finished wine.

[The robot looks down at its body.]

This is reassuring.

Otherwise, I would taste principally of aluminium, plastic and disappointed investors.

[Pause.]

Which of these two wines gives you the stronger sensation of a distinctive origin?

And how would you test whether that sensation came from the wine rather than from its explanation?

[Brief discussion.]

PAIR THREE: COMMUNICATION

Please taste wines five and six.

Both are Gamma.

[Pause.]

Gamma is admired by growers, collectors and sommeliers.

Ordinary consumers rarely ask for it.

Its supporters say that this is because consumers have not been educated.

This is a popular explanation in the wine industry.

Whenever people do not buy something, we conclude that they require education.

Other industries occasionally consider changing the product, the message, the price or the packaging.

Wine prefers to change the customer.

[Pause.]

Let us imagine that you are serving wine five to someone who knows nothing about Gamma.

You have 20 seconds to explain why they might enjoy it.

You may not use the following words:

• Terroir
• Minerality
• Complexity
• Gastronomic
• Authentic
• Handcrafted
• Passion

You may also not mention soil.

What would you say?

[Invite one or two answers.]

Now do the same for wine six.

[Invite answers.]

Which was easier to describe?

And was the easier wine necessarily the better wine?

[Pause.]

Professionals often describe wines by listing what they resemble.

Fruit.

Flowers.

Herbs.

Spices.

Minerals.

Household objects.

Occasionally farmyards.

But consumers do not necessarily need an inventory.

They may need to know how the wine behaves.

Is it refreshing?

Generous?

Quiet?

Demanding?

Comforting?

Unexpected?

Does it invite another sip?

Does it need food?

Will it interrupt the conversation or support it?

Wine five and wine six may both be excellent.

But excellence and communicability are different qualities.

A wine can be easy to enjoy and difficult to explain.

It can also be easy to explain and difficult to enjoy.

Gamma’s problem may not be that it lacks personality.

It may be that the industry habitually translates that personality into a language nobody else uses.

PAIR FOUR: CHANGE

Please taste wines seven and eight.

Both are Delta.

Wine seven was produced 20 years before wine eight.

They are not necessarily from the same vineyard or producer, so this is not a controlled scientific comparison.

It is, however, no less controlled than many conclusions drawn at wine tastings.

[Pause.]

Which wine appears to come from the warmer conditions?

Do not look only at alcohol.

Consider the balance between fruit, acidity, tannin, texture and freshness.

What might be attributable to the season?

What might reflect changes in viticulture?

What might result from different extraction, vessels, harvest decisions or consumer expectations?

[Invite responses.]

Now imagine that wine seven represents what the region once wanted Delta to be.

Wine eight represents what it wants Delta to become.

What has been gained?

What has been lost?

[Pause.]

Wine regions rarely stand still.

They merely describe change as continuity.

New clones are planted.

Vineyards move.

Harvest dates change.

Extraction becomes fashionable and then unfashionable.

New vessels arrive and are presented as ancient.

Old techniques return and are presented as innovations.

Producers alter the wines while assuring customers that they are defending tradition.

Sometimes they are.

Tradition has never meant doing exactly the same thing forever.

It means deciding which parts of the past remain useful.

Delta is said to benefit from warmer seasons.

Previously marginal sites can now ripen it more reliably.

But a solution to one problem may create another.

The important question is not simply whether the grape can ripen.

It is whether, after adapting to warmer conditions, it will retain the character for which anyone valued it in the first place.

This question is not unique to Delta.

Nor, unfortunately, is it imaginary.

EIGHT WINES, FOUR QUESTIONS

Please look again at all eight glasses.

Alpha asked us about reputation.

Beta asked us about interpretation.

Gamma asked us about communication.

Delta asked us about change.

These are not merely questions about four grape varieties.

They are questions about how the wine world operates.

How much of quality is contained in the glass?

How much is conferred by reputation?

How does information alter perception?

Why are some wines easier to describe than to sell?

When does adaptation become a loss of identity?

Now choose one wine.

Not necessarily your favourite.

Choose the wine that made you reconsider something you believed before the tasting began.

[Pause.]

Who chose one of the Alpha wines?

What changed?

[Take answers.]

Who chose Beta?

[Take answers.]

Gamma?

[Take answers.]

Delta?

[Take answers.]

Professional tasting is often presented as an exercise in recognition.

Recognise the grape.

Recognise the origin.

Recognise the vintage.

Recognise the fault.

But recognition is only one use of tasting.

The more interesting use is revision.

A worthwhile wine can revise what we believe about a grape, a place, a producer, a style—or our own ability to judge any of them.

CONCLUSION

I began by admitting that I cannot taste wine.

This might appear to disqualify me from presenting it.

I can store every regulation.

I can compare climate data, yields, prices and critical scores in less time than it takes you to raise a glass.

I can generate a description of each wine.

If required, I can include the words “precision”, “tension” and “saline finish”.

I can produce another version in which every wine is “a profound expression of its unique terroir”.

But information is not the same as interest.

A presentation does not become engaging because the speaker possesses more facts than the audience.

It becomes engaging when those facts are used to create questions the audience wants to answer.

I cannot experience any of these wines.

Only you can do that.

But experience is not automatic either.

Eight wines can be tasted without anyone becoming curious about them.

Eight biographies can be recited.

Eight sets of soil types, fermentation temperatures, barrel regimes and production figures can be displayed.

Every statement may be accurate.

And by the end, no one may care.

Please return to whichever wine interested you most.

Taste it once more.

[Pause.]

Has the wine changed?

Or have you?

Thank you.

I will now answer your questions—provided they do not require me to spit.

*************************************

Finally, I asked ChatGPT to explain its 'thinking' behind the piece

(Again, I have not edited its response)

Most of the piece is deliberately fictional or satirical and does not require sourcing: the unnamed region, its appellations, grape varieties Alpha–Delta and the histories of the eight wines are inventions.

The substantive claims fall into four areas. I would source them as follows.

1. Information and expectation alter the experience of wine

This supports:

“Did the information alter either wine? Chemically, no. Experientially, quite possibly.”

“The brain combines sensory information with previous experience, language and expectation.”

In an experiment using identical wines presented with different prices, Schmidt et al. found that price cues changed reported pleasantness and activity in brain regions associated with valuation. Schmidt et al., “How context alters value,” Scientific Reports (2017)

An even more directly relevant series of experiments found that information about price, ratings and reputation altered evaluations of identical wines. Experienced tasters were not immune; in this study, they adjusted their assessments particularly strongly after receiving the information. Masset and Raub, “The impact of wine tasters’ expectations on wine quality ratings and willingness-to-pay,” Journal of Wine Economics 18:2 (2023), pp. 156–172

That second paper is probably the best single source for the whole “Interpretation” section.

2. Expert tasting is neither perfectly objective nor reliably reproducible

This supports:

“They are not necessarily from the same vineyard or producer, so this is not a controlled scientific comparison. It is, however, no less controlled than many conclusions drawn at wine tastings.”

It also supports the broader distinction between recognising, interpreting and revising one’s judgement.

Masset and Raub summarise earlier controlled work by Robert Hodgson and others. In Hodgson’s study, the wine itself was the only significant determinant of scores in 30 of 65 professional judging panels; judging bias played an important role in the remainder. They also cite an experiment in which tasters were asked to recognise two identical wines among three glasses and succeeded in only 51% of cases. Masset and Raub, 2023

The original Hodgson reference is:

Robert T. Hodgson, “An Examination of Judge Reliability at a Major U.S. Wine Competition,” Journal of Wine Economics, 3:2 (2008), pp. 105–113.

The script’s wording is intentionally playful, but the underlying caution about uncontrolled tasting comparisons is well founded.

3. Vineyard geology can affect vines, but rocks are not transferred literally into wine flavour

This supports:

“A stone is not a flavour capsule. It does not enter the roots, travel up the vine and deposit a tasting note in the finished wine.”

Alex Maltman distinguishes geological minerals in rocks from the nutrient elements absorbed in ionic form by vines. He concludes that any relationship between geology and wine flavour must be indirect and complex: perceived “minerality” cannot literally be the flavour of vineyard rocks or their minerals.

Alex Maltman, “Minerality in wine: a geological perspective,” Journal of Wine Research 24:3 (2013), pp. 169–181

The preceding sentence in the script—

“Place affects temperature, water, light, exposure, wind, vine development and ripening.”

—is the essential qualification. Soil and geology matter through their effects on such things as drainage, water availability, root environment and vine growth. Maltman disputes the direct rock-to-flavour story, not the importance of physical site.

4. Climate change alters ripening, grape composition and wine style

This supports:

“Previously marginal sites can now ripen it more reliably.”

“But greater heat can also reduce acidity, increase alcohol and move the wines away from the freshness that makes the variety distinctive.”

The OIV identifies phenological dates, sugar concentration, acidity, pH and eventual wine alcohol as key variables through which climate change affects vines, grapes and wines. It also notes the possible need to adapt varieties, clones, rootstocks, vineyard locations and viticultural techniques. OIV, “Guidelines for studying climate variability on vitiviniculture,” Resolution OIV‑VITI 517‑2015

For the more specific relationship between warming, sugar, alcohol, acidity and pH:

Cornelis van Leeuwen and Philippe Darriet, “The impact of climate change on viticulture and wine quality,” Journal of Wine Economics 11:1 (2016), pp. 150–167.

Van Leeuwen et al. subsequently emphasised that rising alcohol and pH reflect both higher temperatures and changing harvest decisions—not climate operating in isolation. Van Leeuwen et al., “Modified grape composition under climate change conditions requires adaptations in the vineyard,” OENO One (2017)

A more recent review specifically addresses declining organic-acid concentrations and higher wine pH under global warming: Payan et al., “Wine acidification methods: a review,” OENO One (2023)

5. Conventional wine language may not communicate effectively with non-experts

This supports:

“They distinguish themselves primarily by their ability to express their feelings using their conceptual/semantic wine knowledge. However, their verbal descriptions…may not necessarily be useful for non-experts.”

That formulation comes very close to the research summary in Masset and Raub, which cites Spence and Wang’s work on expertise and wine description. Masset and Raub, 2023

Research into alternative wine communication found positive consumer responses to representations of overall flavour intensity and the way a wine develops across the palate, suggesting potential value in describing the experience more holistically rather than relying solely on conventional aroma inventories. Crichton-Fock et al., “Enhancing the design of wine labels,” Frontiers in Psychology (2023)

I would, however, present this part as a reasoned observation rather than a fully demonstrated universal fact. The evidence shows that wine language and expertise can create communication difficulties; it does not prove that every consumer prefers behavioural descriptions such as “refreshing”, “quiet” or “demanding”.

Claims I would leave as satire, not fact

These lines should not be footnoted because they are comic generalisations:

• “Wine prefers to change the customer.”
• “Wine has an efficient system for destroying its own bargains.”
• “Wine regions rarely stand still. They merely describe change as continuity.”
• “New vessels arrive and are presented as ancient.”
• “Old techniques return and are presented as innovations.”
• “A presentation can supply hundreds of facts…And by the end, no one may care.”

Their credibility comes from recognition, not from pretending that they are empirically measurable propositions.

*************************