The Science, Culture and Commercialism of Beer Flavor
In front of you are two glasses of India pale ale. They both fit the parameters of the style: stronger in alcohol and containing more hops than a classic pale ale, but less of either than an imperial IPA. Both beers are medium gold in color, clear and bright. Both pour with small but persistent heads of foam that are renewed by fine streams of carbonation.
But you like drinking one more than the other. So, you might simply order the one you prefer and leave it at that―which is how most of us experience beer. But if a friend asked you which beer to order, could you tell your friend why you prefer the beer you do?
“It tastes better. It’s very easy to drink,” you say. That’s a start, but not very persuasive.
You dig deeper. “I like the way each sip is a little bit sweet at first, but then slowly turns bitter.” But is it sweet like maple syrup or sweet like fresh peas? Bitter like broccoli or bitter like quinine?
You try comparing the beers. The one you like tastes more, well, floral than the other one. Floral? Roses or geraniums? asks your friend. “Neither. More like the smell in a florist’s shop,” you answer.
And there you are, describing a beer that tastes the way a flower shop smells. What’s remarkable is that your friend knows exactly what you’re talking about.
“I hate that flavor,” the friend says, and orders the other IPA.
The Puzzle of Flavor
You have just put your impression of that IPA into words, and you’ve done it by referring to an experience that has nothing to do with beer. And even though you and your friend disagree on how you feel about that reference―you like a very floral IPA and your friend doesn’t―you’ve helped him to select a beer he’s more likely to enjoy.
Certainly, we can express whether or not a food or beverage is pleasant without words―we can smile, or we can we push it away, or twist our faces in revulsion. But, unlike other species, we can also convey both how pleasant or unpleasant we find these experiences and why through the use of language. We generally do this by analogy, summoning memories of flavors that we all share. It’s not for nothing that we jokingly say that any bizarre new meat “tastes like chicken.”
We humans actually rely far more on our senses of vision and hearing than most other species. However, the senses of taste and smell are most vividly linked to memories and emotions.
Our other senses may have warned our ancestors “There’s a lion on the next ridge” (sight), or “A leopard just snapped a twig in that thicket” (sound). But nothing else conveys the up-close message “Don’t eat that! It’ll kill you!” with the urgency of smell or taste. You instantly relive bad experiences, you recoil, you gag―and these two most intimate senses may have saved your life.
Taste and smell also steer us towards food and drink that is full of the good stuff: sugar, fat, salt―nutrients that would have been rare in our ancestors’ landscape. They survived by following the cues of their senses, and loading up on rich foods whenever they came upon them. But in our modern environment, surrounded by plenty, our health will suffer if we let our taste buds rule.
The fascinating thing is that our taste preferences are adaptable. We may be hard-wired to eat as much as possible of any food that tastes sweet, but we learn to moderate that urge.
We are also able to reinterpret the negative messages that taste and smell convey. Take our fondness for increasingly bitter beers, for example: in nature, bitterness is usually a cue to avoid a food or drink that might harm us, but legions of beer drinker have cultivated a preference for extreme bitterness, against the bidding of their taste buds.
Our ability to perceive flavor is an essential survival skill―determined by our biology, shaped by our evolutionary history, modified by culture and experience, and expressed through strong individual preference. At the end of this long process, you, as it turns out, prefer a floral, hoppy IPA.
Taste Perception
Remember that map of the tongue that illustrated specialized areas to detect each of the four basic flavors―sweet and salt on the tip of the tongue, sour on the sides, and bitterness on the back? Now, forget the map: although it’s still repeated in lessons and textbooks, it’s wrong. It turns out that all the basic flavors can be sensed anywhere across the tongue (though perhaps in different intensities).
There are thousands of minute, mushroom-shaped taste buds on the human tongue. Molecules and ions in solution in the mouth enter the taste buds. Only in the last decade or so have researchers identified receptor cells, 50 to 100 within each taste bud, that respond specifically to each of the basic flavors, sending messages along sensory nerves to the brain―this is the mechanism of taste (or gustation).
There are now thought to be at least five basic tastes: the old foursome of salt, sweet, bitter, and sour, plus the new addition, umami―the sensation we’d describe as “meaty,” a quality also found in cheese, soy sauce or tomatoes.
Susan Schiffman, formerly of Duke University, and now a consultant specializing in taste and smell, is convinced there are even more basic tastes yet to be discovered.
“Metallic, for example,” she says. “You know certain beers have a metallic taste to them? That’s a distinct taste.” She also has experimental data suggesting that calcium salts are a separate taste: “They have a very strange taste to them that’s not sweet, sour, salty or bitter.”
And while astringency is definitely a feeling―the dry, wooly texture in the mouth―it may also qualify as a separate basic taste, and the same may be true of carbonation. In her research, Schiffman has determined that human patients who have lost the ability to feel astringency or carbonation on the tongue can still distinguish astringent or carbonated compounds through the taste buds alone.
As to whether fat is a basic sense, as some writers posit, Schiffman is doubtful. “I don’t believe it exists,” she says.
When you take a sip of beer, how does your brain process the information it gets from your taste buds? Reading just the basic tastes, does the brain register the beer as scoring, say, 15 for bitterness, zero for salt, eight for sweetness, one for sour and two for umami? Is the brain like a painter working with only a few basic colors, but employing them in a near-infinite number of combinations and strengths?
“That’s one of the theories,” says Schiffman, “that you get a continuum. I would say that’s one of the big fights in the taste field, but I would be on the continuum side.
Three different cranial nerves receive stimulation from taste buds located on different parts of the tongue: the first two-thirds, the back of the tongue, and in the throat. So your experience of a beer really does have a beginning, a middle and an end―yes, there is such a sensation as “finish” in a beer’s flavor, and beer reviewers are smart to swallow, not spit, like their wine colleagues.
What’s more, taste buds at the back of the throat can be stimulated by compounds that enter the nose and reach the back of the tongue where the nasal passage joins the throat. This means that we know something about the taste of everything we smell. When beer reviews suggest that a beer tastes of things you’d never dream of putting in your mouth―wet rocks, horse blanket, and (heaven forbid) cat box―don’t laugh: we’ve all “tasted” these compounds, because we’ve smelled them.
Of course, the tongue is also the source of valuable tactile information about a food or beverage: it detects viscosity, texture, pain and temperature―as well as giving additional cues about the feelings of astringency and carbonation.
How Do You Smell?
The sense of smell (olfaction) is far more complex than the sense of taste.
In contrast to the palette of five (or more) basic tastes, the human nose contains specific receptors for detecting hundreds of different odor molecules. “Perfumers will say they can distinguish among 10,000 different odors,” says Schiffman. “Again, I believe you have the same thing you have with taste, a continuum―it’s the relative amount of firing across different receptor types that gives you a certain experience. The pattern across these different types of receptors gives us the aroma.”
When you drink a beer, the sense of smell plays two roles: on inhalation through the nose, we detect the beer’s aroma. But when we exhale, airborne odor molecules from our mouth pass back through the nose, and contribute to our perception of taste. Your experience of flavor depends a lot on your nose: if you doubt that, think of how flat everything tastes when you have a bad cold.
Like taste, smell is modified by experience and culture. The odor of cheese is mouth-watering to many Westerners, but repulsive to many Asians, who associate the sharp, acidic smell with spoilage and sickness.
How Different Are We?
Not all animals share the repertoire of five basic tastes: cats, for example, can’t detect sugar, and this is a genetic mutation of great antiquity. So, a cat’s flavor world lacks sweetness, just as a dog’s visual world lacks full color. But, before we get smug about our own abilities, remember that our own sensory world is also incomplete compared with other animals: many species detect things we do not. Doubtless, these differences in perceptive abilities reflect the different evolutionary histories of different species.
Even among our close relatives in the primates, other species include things in their diets that we find unpalatable―presumably, these foods taste “good” to the animals that consume them.
Richard Wrangham, professor of anthropology at Harvard University, has studied dietary differences in the primates. In research on primates at the San Diego Zoo, he looked at animals’ responses to tannins, plant compounds that are both astringent and bitter.
“I found that the willingness to eat tannins, which are quite an important feature of alcoholic drinks in general, is strongly associated with how much leaf you eat in the wild,” says Wrangham. “Species that don’t eat any leaves in the wild completely reject tannins in captivity. Chimps and gorillas were tolerably tolerant.”
He also looked at responses to bitterness, using quinine. “There are so many different kinds of poisons that taste bitter. No one to my knowledge has understood the fundamental thing that makes something bitter.”
And yet, bitterness and astringency, tastes that are generally rejected, are sought out in beer and wine. Wrangham reasons “Basically, what happens is the receptors that taste bitter are able to attune themselves to things that―from that particular species’ point of view―would be bad.” Under the right circumstances, we learn to accept new flavors.
Looking only at humans, there are important differences in our ability to taste and smell. A baby is born with taste buds on the sides and roof of the mouth, which presumably is useful at a time when new flavor memories are being formed. Over time, the distribution and number of taste buds are reduced; older people are not as sensitive to flavor as youngsters.
Our capacities don’t just change through life, they also vary from one individual to another, and some of this variation is genetic. The best-documented example involves sensitivity to PTC (phenylthiocarbamide), a compound that ranges from unpalatably bitter to undetectable, depending on the subject’s genetic make-up.
Sensitivity to PTC has been used to identify so-called “super tasters,” yet there is little evidence that the ability to taste PTC generalizes to greater flavor sensitivity overall. In popular literature, “super tasters” are said to be more sensitive to bitterness and to alcohol in general―but the scientific evidence is yet to weigh in.
So, are hop-heads more likely to be “non-tasters,” unable to perceive the bitterness in PTC and similar compounds? Are “tasters”―those of us able to sense PTC―more likely to prefer malt-accented beer?
Susan Schiffman can’t establish a connection. She notes “In Japan, there are many more PTC tasters than in the American population, and yet Kirin beer is pretty bitter, right? In consumer tests, strong PTC tasting doesn’t seem to affect beer preferences.”
We also seem to differ in our sensitivity to some compounds that are relevant to brewing: diacetyl, dimethyl sulfide (DMS) and methyl mercaptan, in particular. Diacetyl, a product of fermentation, can be perceived at extremely low concentrations. At low levels, it is reminiscent of buttery theater popcorn, then butterscotchﾑbut not every drinker is equally sensitive. The same is true of DMS, which imparts cooked vegetable flavors, and methyl mercaptan, which contributes light sulphur notes appropriate to some beer styles, but which at higher levels smells to some drinkers like skunk.
Schiffman makes the good point that we don’t experience beer one taste or odor element at a time, but as part of an integrated whole that blends taste, feel and smell. Flavors that we dislike in isolation can be pleasant in the finished beer―when they occur in the appropriate amounts and combinations.
From Flavor to Sensory Analysis
Since the highest attribute of a beer is that it tastes good, as beer production scaled up to industrial levels, it makes sense that the emerging science of flavor analysis would be enlisted to make sure that beer was consistent and free of flaws.
Beginning in the 1960s until his death last year, the Danish chemist Morten Meilgaard defined the field of sensory analysis for the brewing industry. He saw this formal practice of analysis as one with a long history, closely linked to the rise of trade and the need of merchants for an objective assessment of the quality of the goods they were buying. For his part, Meilgaard’s goal was to codify beer flavor for the industry, establishing a common vocabulary of taste.
In his book Sensory Evaluation Techniques, Meilgaard explained that his approach was to “treat the panelists as measuring instruments.” He recognized that trained human tasters were far more sensitive than instruments in evaluating beer flavor, but that they were also prone to bias. But, he acknowledged “they are the only instruments that will measure what we want to measure, so we must minimize the variability and control the bias by making full use of the best existing techniques in psychology and psychophysics.”
Meilgaard’s Beer Flavor Wheel, adopted by the major professional brewers’ associations, is still a standard tool today. The wheel displays 97 attributes of beer, taste and aroma, both good and bad. Many attributes refer to familiar foods such as apple, coconut or cooked sweet corn. Others refer to familiar sensory experiences, though not foods: Bakelite, leather, or can-liner. Still others name specific chemical compounds or qualities that the educated analyst could trace to stages (or errors) in the brewing process: acetaldehyde, caprylic, or autolysed.
An analyst who is trained to recognize these attributes and to accurately weigh their presence in a beer is invaluable to the beer industry. In the field of quality control, panels of trained tasters can monitor the consistency of a beer brand, advise on recipe modification or ingredient substitution, and help identify the source of flaws in a brewery’s beer.
At Genessee Brewing Co. in Rochester, NY, a tasting panel of between eight and a dozen people meets every work day to taste each of the brewery’s beers before it is released to the next stage of production. Every batch of beer is tasted half a dozen times by the “release panel.”
“We line up eight glasses, and pour samples from lighter to more intense flavor,” says brewing supervisor Jim McDermott. “We taste silently, make our own notes, then go from person to person and compare. If there’s an issue, we go into it a little more.”
McDermott stresses the importance of human taste perception: “Everything that comes to us is in spec―it analyzes fine. If you just looked at the numbers, you could release the stuff. But the final arbiter is human taste.”
The panel generally uses the flavor wheel to guide their comments, but sometimes the feedback isn’t specific to particular Meilgaard descriptors. “Someone might say ‘I think this beer is just brighter than that one.’ It’s a very relaxed process until something comes up that requires attention, then―whoosh―everybody narrows in.”
The brewery also convenes panels when a new beer is under development. Here, the process may start with the identification of a style the company wants to explorer. With a new Irish red lager, for example, McDermott already had experience and understood the accepted parameters of the style.
“After a pilot batch, that’ll be our first pass,” says McDermott, “then we’ll subject that to sensory panels: ‘we need more of this less of that.’”
“Every brewery has something like this. It might be more formal than what we do―if you read the sensory analysis literature, they want each person in a little cubicle away from other people―but the purpose is the same,” he continues. “It’s probably the most important thing we do. You can have differences that emerge during the tasting, and ultimately you may fail a beer that looks fine on paper.”
Tasting for the Consumer
Formal sensory analysis, the Beer Flavor Wheel, and blind panels are vital tools for industry quality control. Mastering their details can also be critical for trained beer judges who are ranking beers on their technical merits or advising brewers on how to improve their craft.
But the breakdown of beer flavor into diagnostic elements doesn’t express to the drinker the experience of consuming the beer. As rated by the Beer Flavor Wheel, a beer may score high on the spokes labeled ‘malty,’ ‘banana,’ and ‘walnut’―qualities that have different origins in the brewing process and which fall some distance from one another on the diagram. But if a beer reviewer writes that the beer evokes memories of “warm banana bread with nuts,” that phrase begins to describe the flavor from the beer drinker’s point of view.
At the Beverage Testing Institute in Chicago, panels of tasters try to capture beer flavor in such a way that restaurants, servers and consumers will know what kind of experience a beer will deliver.
“We don’t try to break it down into components, then rate those components and add them all up,” says Jerald O’Kennard, director of the BTI. “We’re looking at the totality of the beer and answering the fundamental question, ‘OK, is this a good commercial example of this product, and how does it compare to the very best examples of this style?’”
The BTI panel tastes hundreds of beers every year., the panel drawn largely from the hospitality industry. Members of the panel all have an idea of how a world-class example of a given beer style would taste. The also have extensive vocabularies of flavor on the tips of their tongues.
Some rather florid tasting notes can come out of the panel’s efforts. Instead of the technical descriptors from the Flavor Wheel―where beers can be caprylic or husky, or smelling of phenylethanol―from the BTI, the beers are redolent of “chocolate-covered orange peel” or “multigrain toast and orange marmalade.”
For all the lush descriptions, O’Kennard is still cautious about the terms he uses. “I use descriptors very carefully, because if I start talking about roasted corn and cabbages, those words can suggest infection to brewers.”
Some sensations are hard to pin down. “‘Earthy’ is a default term,” he continues. “A lot of times, you don’t have anything more exact to say. It is an earthy experience: it’s minerally, it’s of the earth, it’s soil, it’s old grass―it’s a term you use for hop character. There are earthy hops and citrus hops as default terms. You put down ‘earthy’ and hope that translates.”
So we return to the challenge of communicating about one modality by resorting to another. There may be more precise ways to understand the biological underpinnings of flavor perception, the evolutionary pressures that have shaped our diets and food preferences, or the flexibility that culture builds upon in shaping our choices. But, face to face, when we want to express the pleasure of drinking a beer, we use words to talk about taste―and we struggle to describe the India pale ale that tastes like the aromas in florist’s shop.