Rating beauty

Or, why the ogre was right all along

In the folklore of the mountain villages of Eastern Europe, an ogre roams the high passes holding a skeleton dressed in bridal robes. This is his wife, who died many years ago, but the ogre, maddened by grief, believes she is still alive and beautiful.

The ogre stops unwitting travelers and demands:

“Tell me what you think of my wife.”

The correct answer, and the only one that lets you live, is:

“She is the most beautiful woman I’ve ever seen.”

If the traveler—shocked, horrified, afraid—answers in any other way, the enraged ogre kills them and adds their bones to the heap of other unwise travelers at the back of his cave. The correct answer, though, is met with satisfied approval and safe passage through the mountains.

Still hot. Could use a couple of cheeseburgers, though.

There are a lot of great themes in this short tale—the power of grief, wisdom vs. honesty, watch out for mountain ogres—but what I want to think about is beauty.

People love rating things. The best Beatles songs. Worst heavy metal albums. 100 greatest movies of the 1980s. Worst opening lines in books. Best opening lines in books. How to rate the attractiveness of women on a scale of 1-10. How to rate the attractiveness of men on a scale of 1-10. A bunch of different magazines (e.g. Wine Spectator, Wine Enthusiast) rate wine, assigning numbers on a 100-point scale to different vintages. Untappd is an app that crowd-sources and aggregates beer ratings. I doubt there is any area of aesthetic judgment where nobody has tried to implement rankings.

It’s popular to insist that “beauty is in the eye of the beholder” or “it’s all subjective” or “it’s just a matter of cultural norms and what society says we should like.” I think those kind of claims are mostly false. Still, the degree to which something is beautiful or delicious or enjoyable is somewhat variable. Why? Because aesthetic experience is a function of both context and multimodal sensation. Here’s what I mean.

Context

There’s a microbrewery in Tampa, Florida called Angry Chair Brewing. One of their beers is a double pastry stout called German Chocolate Cupcake. It is described as a bourbon-barrel aged imperial milk stout with coconut, cacao nibs, and Madagascar vanilla, coming in at 11% ABV. It is thick, sweet, and boozy. Now imagine two different drinking situations:

1. You and a friend are sharing a post-prandial bottle next to a hearth fire on a cold winter’s evening.

2. You are sitting sweaty and alone in the hot summer sun on a Jamaican beach.

In which of these cases does a bottle of German Chocolate Cupcake sound more enjoyable? The beer is the same in both cases; only the context has changed. The context, though, makes all the difference.

Here’s another example. Consider Frank Lloyd Wright’s house Fallingwater. A truly great thing about its architecture is how it is integrated into the surrounding forest, hills, and waterfalls. It feels both modernist and like a natural formation that organically emerged out of the Bear Run Nature Preserve. Now imagine Fallingwater transported to the middle of Istanbul.

It would look bizarre, ridiculous, incongruous.

Or think back to times when you experienced something wonderful for the first time, the sublime beyond expectation. No repeat encounter can measure up to that first one; the context of appreciation has changed. The flip side is when the enjoyment of the familiar is increased by a change in circumstances. The normally forgettable experience of take-away pizza is elevated to memorable when shared on the deck with friends on a fine spring evening.

The importance of context is old news in the fine art world: Duchamp’s toilet is one thing in the Tate Modern, another thing altogether in his bathroom. Seeing Mona Lisa hanging in a warehouse full of greasy auto parts is a far different experience from seeing it in the Louvre. Watching a pretty woman walk past on the sidewalk does not measure up to seeing the same woman after she has become the wife you love.

The perception of beauty is the output of a function that includes not only context, but obviously sensation as well. It’s just that sensation is multimodal, which leads to unexpected results.

Multimodal sensation

If you really want to make a pure and objective assessment of this chicken tikka masala, you need to hold your nose while you eat it, so that its taste isn’t improperly influenced by the smell. Wait, what? That’s stupid. Everybody knows that a lot of how something tastes is due to how it smells. Holding your nose while you eat cripples the flavor; it doesn’t isolate it. The nose and mouth work together to produce taste.

It’s not just your nose that contributes to taste. So does sound, as in this tale of two oysters:

The first oyster in the experiment was served on the half shell, the way most restaurants serve them. In the background a sound track of waves crashing on the beach was playing. The second oyster was served in a petri dish, making that quivering, gelatinous, slimy gray mass look like an organ being readied for transplant. In the background they played the discordant sounds of clucking chickens. Not surprisingly, the participants rated the oyster on the half shell with ocean sounds much more pleasant than the petri dish oyster with clucking sounds.

See if you can predict the results of these experiments:

(1) Pringles. The control group ate Pringles with no background noise, but the experimental groups ate them with amplified crunching noises in the background. Guess which group judged their Pringles as being fresher.

(2) Carbonated water. The control group got a glass with no background sounds, but the experimental group had a glass with loud fizzy bubble sounds in the background. Guess which group thought their water tasted extra-carbonated.

Smell affects taste, sound affects taste, how about sight? Yep, that one too. It turns out that coffee served in a white mug is perceived as more intensely flavored/bitter than one served in a clear glass mug. The same coffee in a blue mug or a clear one is perceived as sweeter. Hot chocolate served in an orange cup with a white interior, or a dark-cream colored cup was judged as sweeter with a more intense chocolate aroma. Strawberry mousse on a white plate was sweeter and had a stronger strawberry flavor than the same served on a black plate.

Maybe you’re inclined to say that crunchy sounds or white mugs are just faking out your brain, and the Pringles don’t really taste crispier and the coffee doesn’t really taste more intense. The problem is that it’s inconsistent to insist that sight and sound are just confusing your brain about taste while at the same time conceding that scent actually contributes to taste. All in all it’s best to conclude that taste isn’t just made up of one sensory modality; it is a function of several.

Here’s another example on the taste front, which I mentioned in my earlier piece on the sublime. In S1E7 of Penn & Teller: Bullshit, the hosts conspired with a high-end restaurant to serve “artisanal” bottled waters to their diners, all with fancy names like Agua de Culo, Amazone, Mt. Fuji, and Le d’Eau Robinet. Then the water steward, a.k.a. an actor working for the show, described the supposed flavor profiles of the different bottles to unsuspecting diners. True to form, when asked for their impressions of the taste differences, the diners all sincerely reported that “Oh, yes, I can taste the glacial qualities in this one! So much better than tap water!” All bottles had been filled with water from a garden hose on the restaurant’s back patio. The diners just tasted what they expected to taste.

Penn & Teller argued that bottled water is just, well, bullshit. I want to push for a slightly different conclusion: it is entirely possible that the diners really, genuinely, tasted differences between the different bottles. The packaging, presentation, circumstances, and patter of the water steward were all inputs that, along with the taste buds, produced flavor. This hypothesis shouldn’t be too shocking; we already know from the McGurk Effect that what we see changes what we hear, why shouldn’t what we see change what we taste?

Relatedly, there have been many claims that wine lovers can’t taste the difference between actual red wines and white wines with red food coloring added. This has been followed by angry denunciations from oenophiles insisting that they couldn’t be fooled. (There are loads of examples of this dynamic online. I’m too lazy to link them all.) I’m splitting the baby on this one: the oenophiles may taste colored whites as being true reds, but they aren’t fooled. The appearance affects the taste, or, better, taste is a function of multiple inputs, including appearance. Let’s call this effect “rosé-colored glasses.”

The brain is relentlessly trying to predict future experiences, and works mighty hard to fit whatever input data it gets into that mental model. If the data more or less fits expectations, the brain just kind of says, “Welp, close enough. Prediction satisfied.” You expect that red-colored wine will taste like a merlot or a cabernet franc or something similar. Given what the waiter said, you think the Mt. Fuji water will be slightly different from Agua de Culo. The coffee sure looks dark in that white mug, so you bet it is strong. Then, lo and behold, those predictions all come true.

How does all this affect ratings? The idea that there is a neutral scenario in which one can objectively rate beers (or anything else) is nonsense. Yes, of course there are better and worse exemplars of a style, but there is no objective, transparent context for fine-grained rating. What would that be? The artificiality of a competition? With food? Without food? Alone? With friends? In the sunshine? Should we drink blindfolded? What kind of glass?

Aesthetic experiences are the output of a function that takes context and multimodal sensations as input. These inputs vary along several dimensions, and it may well be that when suitably amplified they yield surprising judgments about beauty, deliciousness, elegance, disappointment, pleasure, and ugliness. In a sufficiently powerful context of love, even the beloved’s corpse may be beautiful.

The ogre’s judgment of his dead wife’s beauty is not distorted by his grief, and he is not deluding himself, any more than drinking coffee from a clear mug distorts its taste. The mug authentically affects the taste, or rather, the taste is a function of many inputs, including the mug. In the same way, the ogre’s boundless love affects her appearance. She truly is beautiful to the ogre; his love is so strong he continues to perceive her beauty even in death. He cannot help it, any more than you can help hearing the McGurk effect. For a thoughtless traveler to insult such genuine loveliness, well, they deserve the ogre’s punishment.