Container for the Thing Contained

In last month's New Yorker food issue, Nicola Twilley takes a jaunt through sensory science with Charles Spence, the psychologist who heads Oxford's Crossmodal Lab. Spence's research investigates the multisensory aspects of perception, especially the perception of flavor. Beer tastes more bitter when bass is booming from the stereo.

Twilley's article looks at how food companies apply Spence's research to packaging design: in order to engineer cans for energy drinks, for instance, whose hiss when opened is pitched to evoke masculine fortitude, or low-sugar chocolate bars whose red wrappers dial up sensations of sweetness. In other words, shaping the container to influence the perceived qualities of the thing contained. For manufacturers, research like Spence's seems like a way of bringing some scientific rigor to the fuzzy, intuitive art of container design, perhaps minimizing the high mortality rate of new products. Spence himself keeps "a rogues' gallery of failed products" on display in his office, brief-lived merchandise whose commercial death warrant was signed, according to his diagnosis, by design decisions that failed to account for the perceptual effects of colors, sounds, and shapes on consumers' experiences of flavor.

But food packaging has long been an engineering problem with recognized sensory dimensions and consequences — though not exactly the kinds of sensory effects studied by Spence's lab.  

 One of the beautiful color lithographed advertisements in American Perfumer & Essential Oil review, this 1913 ad for Mulford Mints draws attention to their "individual sanitary boxes" which "keep their flavor."

Consider the package for a moment. Packaging itself defines the category of "processed food" more than any other aspect, and sharply distinguishes the foodways of industrial modernity from prior modes of eating and living.  (For instance: Labels! The whole visual/informational superstructure of the modern food industry — and a good part of regulatory agencies' authority — depends upon the infrastructure provided by the package, which serves as the slate upon which the enticements and warnings and "facts" can be inscribed.) Think about the vast and heterogeneous category of food made in factories, now and in the past: Leibig's Extract of Beef, Heinz's 57 Varieties, Uneeda Biscuits, Land O' Lakes butter, Campbell's Tomato Soup, Diet Dr. Pepper, Midnight Cheeseburger Doritos. Each comes in its own, standardized, (almost always) inedible, rarely reusable container. Sometimes there are even containers within containers: colorful, printed cartons enclosing translucent bags of Wheaties, or transparent sleeves of Oreos, or microwaveable black plastic trays of Lean Cuisine, whose frosted slab of lasagna or sesame chicken is kept immaculate by a heat-bonded layer of crystalline film. 

The importance of finding the right container; in this case tin foil bags for coffee. Spice Mill, 1915.

Rather than thinking of food and package as two dissimilar kinds of things, brought together at the end of the production line only to be torn asunder in the kitchen, it is important, I think, to consider both as integrated components of a single system. The inedible package profoundly affects our experience of the edible contents within, defining the possibilities and scale of industrial food production, as well as setting our expectations for food that tastes "factory fresh," flavorful, familiar, reliably consistent.    

Packages are technologies of preservation, forestalling not only actual rottenness, but ideally also maintaining food in a sort of homeostasis until the actual moment of consumption. Long adapted to keeping out moisture, oxygen, vermin, and other assorted crud, in the twentieth century, the package becomes a deliberately designed tool of flavor control.

Flavor control? Yeah, cause flavor needs to be controlled. Volatile chemicals, those aromatic agents that deliver a large part of our flavor experience, are cosmopolitan and promiscuous. They are tireless travelers, and keep all kinds of company. Packages were designed to keep volatiles in the package, and to protect them from oxidative and other changes. Research into stale coffee led to air-tight packages filled with inert gas, that could keep ground beans tasting "fresh" longer.  Glass milk bottles were replaced with opaque containers in part to prevent the disconcerting, tallowy taste that sometimes developed when milk was exposed to sunlight.

In the flashy world of food packaging, don't forget about the humble paper bag!

But a package's membrane is penetrable from both directions, which also allows inauspicious odors to creep in and settle down. Tracking down the source of mysterious "off-flavors" that tainted packaged goods was often described as a kind of detective work.  As early as the 1930s, chemical consulting firms offered their sleuthing services to companies to track down the source of that musty smell in cigarettes, or that weird flavor in certain shipments of cocoa. (Both real cases handled by NYC consultants Foster D. Snell in the 1940s. The cigarettes were contaminated by benzene hexachloride, leached from a bag of insecticide that had been packed alongside the smokes during shipping. The cocoa's weird aroma was due to inks used on its colorful label.)

The new scientific attention to the flavor and "keeping qualities" of food fueled research and interest in new materials. Glass jars and tin cans were joined by squeezable tubes, waxed and laminated cardboard, and, especially since the Second World War, plastics, such as cellophane, polyethylene, polyvinyl, and other synthetic polymers and composites.  

The common material EVOH - ethyl vinyl alcohol - a copolymer of ethylene and vinyl alcohol - in action at the center of this multi-layered co-extruded packaging material. 

The new plastic materials could be cheap, light, and flexible, durable and colorful, but they also had liabilities. Consider that packaging material isn't just a passive membrane, but has its own sensible qualities, and its own proclivities to form attachments, to cling and react. For instance, some commonly used plastics have a tendency to bond with certain flavor chemicals, upsetting the carefully calibrated flavor balance engineered by flavorists. The technical term for this is "scalping," I kid you not. Polyethylene, one of the most common plastics used in commercial packaging, is a notorious top scalper. The material lining the gabled tops of Tropicana orange juice cartons were found to be scalping the top-notes off the OJ within, which led manufacturers to move to materials that would not absorb the citrusy linalool and limonene within their clingy matrix.

The proliferation of chemicals in the world, and the complexity of the production and supply chain, means increasing numbers of opportunities for foods to be tainted. (Here's a great article by Sarah Everts on this subject.) Lubricants used in the production of beer cans can linger on the can, reacting with the beer to produce rancid flavors. Fungicides and microbicides used to treat wooden pallets can sometimes react to form chloroanisoles, which can leach through cartons stacked on top of the pallets to impart a moldy odor. A cat urine off-odor that contaminated some cooked ham products was found to be caused by printing inks that had migrated into the laminate film used as packaging, reacting to form the pissy-smelling 4-methyl-4-mercaptopentane. In 2010, Kellogg's recalled 28 million boxes of Froot Loops, Apple Jacks, and other cereals, because a weird smell — described variously as waxy, stale, metallic, and soap-like — had been reported, causing a handful of consumers to feel nauseated and vomit. It took quite a bit of sleuthing to determine that the cause was inks used on the exterior of shipping boxes, which had migrated through three layers of packaging to taint the cereal within.

Before sensory psychologists like Charles Spence can work their tricky magic on package design, the business of flavor chemists depends on packages that are engineered to preserve and protect the sensory qualities of food.

And to close out this little reflection on the increasingly intimate relationship between the container and the thing contained, the thing contained and the container, I'll leave you in the capable hands of Thurber. In "Here Lies Miss Groby," his 1942 rememberance of his old English teacher, (an essay which often haunts my reverie), he presents this joke as an example of reverse metonymy, of taking the thing contained for the container:  

A: What's your head all bandaged up for?

B: I got hit by some tomatoes.

A: How could that bruise you up so bad?

B: These tomatoes were in a can.   

Messing with the Senses

I'll begin with this: the "mystery" flavor of Dum-Dum lollipops. When I was a kid, I had a theory that mystery flavor was a factory mistake. All the lollipops that accidentally made it through the assembly line uncolored were swaddled in a "mystery" wrapper, spangled in question marks like the suit of the man who helps you get free government money. Which didn't actually help me solve the problem of what flavor, exactly, they were supposed to be. I always found them off-putting -- colorless, translucent globes of indeterminacy. (Googling it now, this article claims that the mystery flavor is a mixture of two other flavors in production, the mixed-up flavors that get produced between batches in the lollipop factory.) 

Almost everyone, in school science labs, has done some variation of this experiment: sipping tiny paper cups of colorless orange soda, or Sprite tinted to look like Coke, and then trying to guess at the flavor of these uncanny concoctions. The flavor of a soft drink -- something that seemed so obvious and familiar -- is revealed to be elusive, befuddling, difficult to pin down. Is it grape? Is it orange? Is it lemon-lime? Why is it so hard to tell?

And it's not only rubes who can't tell red wine from white without looking at the glass -- this is a common incapacity, even among snobby winos.

Examples like these, of the profound effects of color on our perception and experience of flavor, are familiar to most of us now. Our present-day scientific understanding of how color is mixed up with flavor has its roots in the 1930s, when the industrialization of food systems made flavor a technical and scientific problem for food producers. Among other things, manufacturers needed ways to minimize and counteract the deleterious effects of processing on food quality; they needed standardized, stable, and consistently priced products; they needed foods with "flavor appeal" that would tempt "repeat buyers." This meant defining what, exactly, flavor is, and how it works to produce its effects. Even as chemists, food technologists, home economists, and other scientists got better at analyzing, identifying, and manipulating the molecular and material aspects of food that contribute to flavor, they recognized that flavor could not fully be described chemically, nor was it exclusively produced by the "chemical senses," taste and smell. As Ernest Crocker, who I've written about before on this blog, put it in his introduction to the landmark 1937 American Chemical Society Symposium on Flavors in Foods: "A new approach to the subject of flavor consists in attacking several of its many sides simultaneously, but especially the psychological and the chemical sides." Understanding flavor would mean not only studying its molecular aspects, but also the way perceptions of flavor were influenced by visual cues, social norms, personal history, present atmospheric conditions, and the vagaries of individual physiology. This is one of the points where two nascent fields -- flavor chemistry and sensory science -- are cross-hatched together.  

One of the first people to mess around with visual cues and flavor perception was H.C. Moir, a Scottish analytic chemist working at a baked-goods factory in 1930s Glasgow. Present-day sensory scientists cite Moir's 1936 article ("Some Observations on the Appreciation of Flavor in Foodstuffs"), published in the British technical journal Chemistry and Industry, as the first to document how the color of a food shapes our experience of its flavor. (For instance, this nifty article by Crossmodal Lab's Charles Spence touts: "ever since the seminal observations of Moir in the 1930s, researchers have known that changing the color of a food or beverage can change its perceived taste/flavour.")

Most scientists who cite Moir don't go into any detail about his experiments, and (just guessing here) probably haven't read his article. And, really, why would they? In the intervening decades, there have been dozens, if not hundreds, of studies published about the role of visual cues in flavor perception, using much more sophisticated techniques, producing much more formidable results. Scientific conventions prescribe preserving the honor of first discovery in the crowded footnotes, but there's no obligation to engage with this dustiest of data. (And Moir may not even fully deserve the credit he gets as pioneer. In his article, he credits Mr. Rendle of Chivers & Son -- a manufacturer of marmalades, fruit preserves, and jellies-- with developing the method of "testing 'palates'" that he describes.)

Stomping around in the bibliographic basement, however, can sometimes enrich our understanding of how we got to now -- the interlinked networks of interests, institutions, ideologies, technologies, materials, and living, working bodies that underlie the production of scientific facts.  

So, with all that said, who was H.C. Moir, and what exactly is his story?

It's rather difficult to find any solid information on Moir, but when he wrote his article, I'm fairly certain that he was the director and chief chemist at William Beattie, Ltd., a Scottish wholesale bakery. That is, he was not a psychologist, psychophysicist, or physiologist trained to observe and measure human sensory responses to stimuli. He was an industrial analytic chemist, and the research that he describes did not take place in the controlled setting of an academic laboratory, but rather on the factory floor, with workers in his bakery as his subjects.

Nor was Moir primarily trying to prove any basic hypotheses about the nature of sensory perception. Instead, he was dealing with a technical and commercial problem: he needed to find reliable tasters to evaluate the quality of his baked goods.

He writes: “My object in making ... these tests was to find within the factory" a group of individuals with a proven "discriminating palate... to whom questions of flavor could be referred." He wanted to have trustworthy "tasting panel" that could weigh in on new products, or detect whether something was going wrong with the production line.

And so he casts his net over the factory floor, drawing in sixty tasters -- managers, salesmen, "factory girls," bakers, "in some measure... a cross section of the consuming public" -- who are subjected to a series of tests in order to assess their sensory acuity.

Moir begins by having his subjects rank solutions of sucrose and citric acid in order of increasing sweetness and sourness. He then asks about their habits and preferences. Do you have a sweet tooth, or do you prefer savories? Do you take sugar in your tea? How many lumps? Are there any foods you particularly loathe -- olives, asparagus, pineapple?  

But the most dramatic part of Moir's investigation -- the part that still earns him citations from present-day sensory scientists -- comes when he serves up discordantly colored sweets. Recognizing that people are often "misled by their eyes" when identifying flavor, he decides to confound the senses of his subjects by serving them Chivers-brand "table jellies" -- ie, flavored gelatin, like Jell-O, I think -- in four distinct "good, true-to-type flavors," but with colors that were not typically associated with the added flavor. So:

• Yellow Vanilla (I think we can assume that this was bright, bright yellow)
• Green Orange
• Amber Lime
• Red Lemon

The tasters were assured that they were dealing with very familiar flavors -- nothing odd or exotic here -- and then asked to name them. If they really struggled to come up with anything, they were given the four possible options, and told to match them with the proper jellies.

The tasters performed terribly. Only one person out of the sixty got all the identifications right; most got fewer than half the questions correct on the test. And the guesses were all over the place. The vanilla jelly was identified as black currant, lime, apricot, lemon, orange, tangerine, strawberry, among other things. Guesses for the lime-flavored jelly included vanilla, pineapple, and apricot.   

What's more, Moir was astonished by the indignation that his tasters exhibited when told of their execrable performance: 

“Some of the least discriminating were the most dogmatic in their decisions. The majority of those who came below 50% went to great pains to assure me that they were considered by their wives or mothers, or other intimates, to be unduly fastidious about their food, and were invariably able to spot milk turning well in advance of any other member of the household.”

Some tasters insisted that their palates were fine, it was the test that was flawed. Others complained that the test was unfair to them because they personally disliked table jellies. “But of course, what I was anxious to find was those who were possessed of palates which could discriminate even that which they did not appreciate," grumbles Moir. "No one enjoyed the flavor of decomposed fruit... but on occasion one must detect, and if possible, identify it."  In other words, for Moir, a good taster and a gourmand are not the same. An accurate taster must be able to report his or her sensory perceptions without prejudice, dispassionately detecting and identifying the flavors that are present in a food regardless of personal preference.

Moir emphasized the egalitarian implications of his findings. Situational authority -- the power or expertise possessed by the foreman, the manager, the chemist -- does not confer sensory authority. Just because someone is in a position of power does not mean that he or she is "the right person to decide any point as regards the flavor of the products concerned." Indeed, Moir laments that chemists too often assume the accuracy of their sensory capabilities, with disastrous results for the business. "There is nothing to be ashamed of in the lack of a palate," he avers, "but there is something to be ashamed of in a chemist making definite statements on a subject in which he is unable to discriminate."

Even though the results of his investigation reinforce his suspicions that "in the majority of people the faculty [of perceiving flavor] was exceedingly dull," Moir counsels his fellow food manufacturers not to use the public's poor taste as an excuse to neglect the flavor of their products. Though the good tasters may be vastly outnumbered, he says, “the discriminating section of the public exercises an influence out of all proportion to its numbers on the non-discriminating section."

I originally tracked down Moir's paper because it's one of the earliest I've found that makes reference to a "tasting panel" -- a group of individuals selected for their sensory acuity, used by food researchers as a sort of laboratory tool for producing scientific information about flavor qualities. In the first twenty years after its publication, Moir's 1936 article was most frequently cited by researchers writing about techniques for assembling reliable laboratory taste panels. These studies are primarily concerned not with the operation of the human senses, but with accurately detecting and describing the qualities of foods.

The turn towards applying research about the workings of the human senses to the development of new food products would not come until at least the 1950s (at least that's what I've discovered in my research so far.)  Although sensory scientists now locate Moir at the dawn of crossmodal sensory research, reading his article, it is clear that he is not particularly concerned with the ways that multiple senses work together to produce the experience of flavor. Indeed, his color test is a way of weeding out people whose sensory judgment is deformed by visual evidence -- implying that, for him, the visual distorts, rather than contributes to, flavor. He does dish out some interesting tidbits: for instance, he observes that more intensely colored foods are often perceived to have stronger flavors -- a phenomenon that later research seems to confirm. However, he does not seem at all inclined to use this information to guide the development of baked goods -- eg, chocolate rolls that seem more richly chocolatey without any additional chocolate.  

This stands in marked contrast to trends and tendencies in the application of present-day sensory science. Charles Spence's article mentioned at the beginning of this post -- well worth reading -- reviews the manifold ways that senses other than taste and smell shape our expectations and experiences of food's flavors. Not only the color of food, but the pitch of the music playing over the speakers, the massiveness of the plate, the brightness of the overhead lights, influence our perception of the character and intensity of the taste and smell of the foods before us. This kind of thing is of real importance to food manufacturers, as it provides potential avenues for intensifying the sensory pleasures of foods while decreasing the need for costly flavoring ingredients. Spence also notes that an additional "area of intense commercial interest currently revolves around seeing whether the consumer's brain can, in some sense, be tricked into perceiving tastes/flavours without the need to include all the unhealthy ingredients that so many of us seem to crave."

So is this a perturbing manipulation of our perceptions -- turning our senses against us -- or is it a savvy application of scientific research, to the end of producing goods that can both gratify our sensory desires and satisfy our material and physical requirements (for cheaper foods, more nutritious products, more intense pleasures, etcetera)? Anecdotally, even people who are more or less okay with "processed foods" seem disturbed about this aspect of food research, which gets imagined as the hegemonic forces of big food reaching their creepy tentacles into your brain and occupying your appetites. The informed and empowered consumer, steadfastly reading labels and counting calories, dissolves and becomes a reflex machine, resistless against the compulsions of salt, sugar, fat.

One of the things I'd like to discover is where this horror story comes from. Fear of chemicals in foods has a long history, dating back to the nineteenth century, at least, and coming to the cultural forefront in various guises at specific historical moments -- for instance, in the Progressive era around the passage of the Pure Food and Drugs Act, or in the 1960s with the countercultural critique of the food industry. But I'd like to also track down the prefigurations of this fear or suspicion of food's sensory qualities, and the new tenor that fear takes when science intervenes in producing those qualities. Definitely something to think about...  

Keep it Fresh, Keep it Real, Orange Juice

We don't tend to think of freshness as a flavor, at least not in the same way that we think of "orange" or "vanilla" as flavors. "Freshness" is supposed to indicate something about a thing's material condition, its temporality: its recentness to the world and to us. The life history of a fresh food is assumed to be reassuringly direct: there were few intermediaries, few machines intervening, as it made its way to us. Fresh foods are also by definition not stable -- nothing can be fresh forever -- and so always at risk of becoming not-fresh, stagnant, rotten, stale.

There's something uncanny about a fresh-seeming food that is really an old food -- like the changeless McDonald's hamburger in Supersize Me, or those legendary Twinkies from decades ago, still plump and gleaming in their wrappers -- something reflexively repulsive. It brings to mind succubus myths, old women who make themselves appear young and nubile to seduce enchanted knights. Those stories certainly deserve some full-strength feminist revisionizing, yet remain among the purest expressions of the grotesque in our culture.

At the turn of the 20th century, one of progressive reformers' most potent accusations against food manufacturers was that they hired chemists to rehabilitate and deodorize rotten meat and rancid butter, to restore them to the appearance of freshness. This is a deceptive practice -- akin to running back the odometer on a used car -- but pure food advocates also largely opposed chemical preservatives, which didn't run back the meter so much as slow its rate of progress. Part of their opposition came from the claim that these chemical additives were harmful, but I think some of the horror of it was that preservatives made the question of freshness beside the point. Some foods were fraudulent by passing themselves off as something they weren't: margarine for butter, glucose for maple syrup. What chemical preservatives were doing was faking freshness.  

The problem isn't so much that the food is rotten or dangerous, but that you can't tell the difference between fresh and not-fresh, and that difference matters to us. Time changes food; and food unchanged by time seems somehow removed from the natural world, indigestible.

Yet why does freshness matter so much? (We don't always favor new-to-the-world foods, of course. Sometimes time increases value: think of old wines, caves of teeming cheeses, dry-aged beef, century eggs).

What we call freshness is not an inherent condition of a food, but an interpretive effect. We read it from cues including color, taste, aroma, texture, as well as the contexts of consumption. This is what I'm arguing here: freshness is a cultural or social category, not a natural one.

As a case in point, consider the story of store-bought "fresh-squeezed Orange Juice," as described in the April 2014 Cook's Illustrated feature somewhat luridly titled:

The Truth About Orange Juice

Is the sunny image of our favorite breakfast juice actually just pulp fiction?

Cook's Illustrated -- one of my all-time favorite magazines, by the way -- assembled a panel of tasters to evaluate various brands of supermarket orange juice. With the exception of two low-cal samples, all the juices list only one ingredient on the label -- orange juice.

Nonetheless, as Hannah Crowley, the article's author, extensively illustrates, orange juice is a processed food: blended from different oranges, pasteurized, packaged, shipped across continents or over oceans, and required to remain shelf-stable and "fresh tasting," at least until its expiration date. Orange season in the US lasts three months. But we want orange juice all year long.

Part of the challenge of producing commercial name-brand OJ is consistency. How do you get each container of Minute Maid to taste the same as every other container, everywhere in the world, in May or in October? Coca-Cola, the corporate parent of Minute Maid and Simply Orange, uses a set of algorithms known as "Black Book" to monitor and manage production. As an article last year in Bloomberg Businessweek put it: "juice production is full of variables, from weather to regional consumer preference, and Coke is trying to manage each from grove to glass." In all, Black Book crunches more than "one quintillion" variables to "consistently deliver the optimal blend," the system's author told Bloomberg, "despite the whims of Mother Nature."

Sure, but how do you reproduce the experience of freshness? Preservation is not enough. In fact, the means used by OJ producers to arrest decay and rancidity in order to allow them to "consistently deliver" that optimal blend -- pasteurization and deaeration -- actually alter the chemical profile of the juice, in ways that makes it taste less fresh. Pasteurization can produce a kind of "cooked" flavor; deaeration (which removes oxygen) also removes flavor compounds.

Freshness is an effect that is deliberately produced by professional "blend technicians," who monitor each batch, balance sweetness and acidity, and add "flavor packs" to create the desired flavor profile in the finished juice.  Flavor packs are described by Cook's Illustrated as "highly engineered additives... made from essential orange flavor volatiles that have been harvested from the fruit and its skin and then chemically reassembled by scientists at leading fragrance companies: Givaudan, Firmenich, and International Flavors and Fragrances, which make perfume for the likes of Dior, Justin Bieber, and Taylor Swift." The only ingredient on the label of orange juice is orange juice, because the chemicals in flavor packs are derived from oranges and nothing but oranges. Yet orange juice production also has something to do with the same bodies of knowledge and labor that made "Wonderstruck" by Taylor Swift possible. (There are in fact multiple class-action suits alleging that the all-natural claim on orange juice labels is inaccurate and misleading.)

In other words, this isn't just about "adding back" what has been unfortunately but inevitably lost in processing, restoring the missing parts to once again make the whole. The vats of OJ, in a sense, become the occasion for the orchestration of new kinds of orange juice flavors, that conform not to what is common or typical in "natural fresh-squeezed orange juice" (whatever that may be), but to what we imagine or desire when we think about freshness and orange juice. As Cook's Illustrated puts it: "what we learned is that the makers of our top-ranking juices did a better job of figuring out and executing the exact flavor profile that consumers wanted." These flavors don't reproduce nature; they reproduce our desires. But how do consumers know what they want, exactly, and how do manufacturers figure out what this is?  

I can't really answer either of those questions now, but I think one of the consequences is a kind of intensification of the flavor dimension of things. Consider: consumers in different places want different things when it comes to OJ. Consumers the US, according to Cook's Illustrated, especially value the flavor of freshness. One of the volatile compounds present in fresh orange juice is ethyl butyrate, a highly volatile compound that evaporates rapidly and thus is correlated with the newness of the OJ to the world, so to speak. Simply Orange, Minute Maid, and Florida's Natural juices -- all juices "recommended" by the Cook's Illustrated tasting panel -- contained between 3.22 and 4.92 mg/liter of ethyl butyrate. But juice that's actually been squeezed at this moment from a heap of oranges contains about 1.19 mg/l of ethyl butyrate. The equation here is not as simple as ethyl butyrate = fresh flavor, so more ethyl butyrate = megafresh flavor. (One of the exception on the panel's recommendations - an OJ with an ethyl butyrate content more in line with that of fresh-squeezed juice - was actually produced in a way that permitted seasonal variations, was not deaerated, had a much shorter shelf life, and depended on overnight shipping to make its way to stores.) But there is a kind of ramping up, somehow, that seems to both correlate with our desires and recalibrate them.

IBM's "Cognitive Cooking" Food Truck

I'm not ashamed to admit that "Wait, Wait... Don't Tell Me!" is one of my main sources of breaking news, and that's where I first heard that Watson, IBM's own Jeopardy champ, is running a food truck at South by Southwest. Of course, I had to look into it...

A joint venture between IBM and the Institute of Culinary Education, the food truck is an exercise in what IBM (rather bloodlessly) calls "cognitive cooking" -- a street-food demonstration of the practical applications of their "cognitive computing" system, aka Watson. Would you like to read an advertorial about it in Slate? Here you go. And here's IBM's promotional website about the cognitive cooking project. 

This is how you use it. You have to input three things: the main ingredient, the cuisine (eg, Indian, Azerbaijani, Canary Islander...), and the type of dish (eg, burrito, bisque, sandwich). (At SXSW, the type of dish was left up to a Twitter vote, and I suppose the other variables were supplied by IBM.)  Watson then reviews the vast universe of possible combinations, modeling the flavor chemistry of each component and its interaction with other flavor compounds, as well as the potential taste appeal of the final dish and how novel the combination is. It outputs a set of recipes comprising 12 to 14 ingredients, each with a rating based on its assessment of flavor interactions, likeability, and surprise. Just like on "Chopped," you're judged not only on taste but also on "creativity." The goal is to come up with something that's both "weird" and "good."     

[An aside: What is it about the times we live in that makes cross-cultural comminglings the apogee of "weird" cooking? "Indian turmeric paella," are the first words out of the advertorial's mouth. "Peruvian poutine," "Swiss-Thai asparagus quiche," "Austrian chocolate burrito" are all dishes featured in the cognitive cooking recipe archive. Are these combinations really so strange, or unimaginable without cosmopolitan Watson to liberate us from our parochial attachment to thoroughbred cuisines? This is not, I think, simply a retread of the 90s vogue for "fusion," which sought a diplomatic accommodation between US appetites and "exotic" (usually Asian) ingredients and techniques. All the borders have come down; materials and methods can be freely recombined without tariffs or translations; culture is just another seasoning. Should we call this "world markets cuisine," globalism's dinner plate, neoliberal gourmandise?]     

IBM's challenge is to prove to all of us that Watson isn't just some better sort of Google, a more refined filter for sorting relevant from irrelevant, signal from noise. What IBM wants to demonstrate is that Watson can provide creative or unprecedented solutions, things that don't just work right but also "feel right." As the Slate advertorial puts it, "A system that can generate new things the world has never seen before is a significant step in cognitive computing."

This is actually a rather tall order, especially as IBM is always careful to insist that "cognitive computing" is not a replacement for human creativity (the brain is "the most creative computer of all," in their words) but a tool to enhance it. The decision to use food -- and, specifically, the creation of unusual flavor combinations -- as a debut showcase for this technology is thus very deliberate, and taps into a longer history. Sure, the marketing team has festooned this with all the right merit-badges -- hipster foodies and their food trucks, Twitter crowdsourcing, SXSW, "the cloud" -- to gain likes and influence retweets in those zones of social media where knowing what's "trending" counts as connoisseurship. But the problem of meshing these two kinds of information about flavor -- what IBM refers to as "chemoinformatics" (ie, its chemical behavior) and "hedonic psychophysics" (ie, our sensory experience of it)  -- is something that has daunted the flavor industry since, at least, the mid-twentieth-century.

I've just been reading the proceedings of the 1961 Flavor Chemistry Symposium, hosted by Campbell's Soup at their old HQ in Camden, New Jersey. This was one of the very first scientific conferences devoted to this chemical subfield. (The Society of Flavor Chemists, the first professional organization, had been inaugurated less than a decade earlier; the American Chemical Society wouldn't create a flavor chemistry division until six years later.) The papers from this conference makes it clear how rapid progress has been in the field: more and more, the molecular structure of flavor compounds, their chemical precursors and interactions with other molecules during cooking and preparation, how they degrade, what influences them, and so on, are being quantified, verified, understood. As Carl Krieger, the director of Basic Research & Product Development at Campbell's remarks at the kick-off of the conference, there was a new "realization that the mysteries of flavor can be solved."

Except. Except that "the physiology and psychology of taste, odor, and flavor" are still vast unknowns. Krieger ventures that only by making positive identifications of flavor chemicals "will it be possible to describe flavors in universally meaningful terms" (ie, by their chemical names) rather than the subjective terms of experience -- "metallic," "stale," "rancid," -- "which, I must confess, seem to me to be pure gibberish." Thankfully, Krieger concludes, their conference will not focus on perception of flavors, but their chemistry - "something that I believe all of us feel is more amenable to direct experimental study." 

Okay, that's all well and good for Krieger to say, but knowing what the flavor compounds are doesn't answer the million-dollar question: "Will people like it?" That's a big missing piece of the puzzle -- the gap between the chemoinformatics, so to speak, and the hedonic psychophysics. Flavor companies -- and the US government, especially the army -- labored to make flavor evaluation "objective," to standardize descriptive vocabularies, to train tasters and impanel consumers to supply their opinions before a product hits the market. But these studies always involved human beings, unruly instruments on their best days, and their subjective responses are, by definition, not generalizable -- do not produce the "universally meaningful terms" that Krieger claimed chemistry did.

And this, fundamentally, is what IBM claims is different about its "cognitive computing" model, and what it's trying to show with this food truck project. We're quite used to claims like "chefs can only consider combinations of two or three ingredients at a time; computers can contemplate quintillions" -- yes, computers can outfox even the foxiest human thinkers. This system doesn't just crunch numbers, it makes judgments about subjective sensations. As the IBM advertorial tells us, it "understands why thousands of different recipes are appealing, what people prefer." Here's the crux of the claim: "It understands, learns, and considers not just big data but also human perception."

These two things -- big data, human perception -- continue to be held at arm's length from each other. But isn't the promise of this technology, in fact, that it successfully converts human perceptions into data, data that the machine-system can "consider" and that are susceptible to the same tools and techniques that guide the collation and analysis of other forms of 'big data'? The dream realized here is that we will finally be able to bring subjective experience into the same table that we use to calculate agricultural yields or profit margins.

What is supposed to make Watson different, I think, is that it claims to formalize the bodies of knowledge that have so far resisted formalization. Things like intuition. Experience. What we in the STS biz call "tacit knowledge" -- the kinds of things you learn by practice, by doing -- like how to make fine adjustments to instruments, or to hone a curve on the form of a chaise lounge, or to add a new ingredient to a recipe. Not just the look of things, but what we felt at what we saw. But Watson enters a crowded field, because our "personal technologies" increasingly aspire to recognize and cater to our subjective preferences. Like when Netflix deduces your taste in movies, not merely spitting out a list of other black comedies, but synthetically tailoring for you an array of "Dark GLBT Comedies with a Strong Female Lead." Or the new music data venture that scans Twitter for early "flickers of excitement about a fledgling band," "the kinds of signs music scouts have always sought." The Watson system isn't just about helping General Foods design new crazy flavors of potato chips; IBM promises that the applications for cognitive computing are in all fields that rely on "design and discovery." This isn't a technology that competes with Google; it's technology that competes with technicians and so-called knowledge-workers -- designers, flavorists, A&R divisions, R&D folks -- highly skilled workers whose refined, intuitive knowledge of their fields are supplemented (or supplanted) by "cognitive computing."

But fear not! Our cherished celebrity chefs won't be driven to extinction by our new networked overlords. "Cognitive computing is a sous-chef working alongside seasoned professional chefs." Right, it's not Emeril's job that's at stake, but those of his unnamed assistants, who will surely still be required to slice and dice -- Watson, after all, doesn't have hands to get dirty -- but perhaps less entrusted with the fine adjustments and refinements, with the knowledge side of technical work. (Similar, for instance, to what Deborah Fitzgerald calls the "deskilling" of farmers after the introduction of genetically modified hybrid corn.) Or maybe not. Maybe systems like this really do foster innovation, break down the barriers that have hitherto prevented us from dreaming up a Swiss-Thai quiche, an Indian paella.  

I should wrap this up on a less lugubrious note. So I'll add that, the consensus on the internet seems to be that Watson's food was pretty good and somewhat novel, though some were disappointed that it was prepared by humans and not robots. Brillat-Savarin said it, and I believe it: "The discovery of a new dish, which excites our appetite and prolongs our pleasure, does more for human happiness than the discovery of a star." The question, I suppose, is how you define "new," and what you mean by "discovery."  

How to become an expert: Cigarette edition

I listen to a lot of "old time" radio - especially mysteries and detective shows - in part to satisfy my insatiable appetite for narrative while up to my sudsy elbows in the dishwater of history.  The other day, I heard an episode of "Mysteries in the Air," starring Peter Lorre, with his quavering syllables and his lightning-speed mildness-to-mania transitions.

The show was sponsored by Camel cigarettes, and the version I listened to kept the sponsor's message intact in the broadcast. Smokers are notoriously brand-loyal. They're not like consumers of other stuff, switching from Charmin to Quilted Northern on a whim or a spree. They'll ask for their pack of Luckies or Reds or Virginia Slims every time, without fail, no hesitation. You smoke what you've always smoked. But how do you get people to switch? How do you get people to believe that their choice is their own to make, and not somehow compulsory? Here's a complete transcript:

[Cymbal-clash] "Voice of God"-type voice, distorted as though through a PA speaker, intones: Experience is the best teacher.

"Average Joe": Remember the wartime cigarette shortage? Who doesn't! One thing about it though - smokers who went through it really learned a lot about cigarettes. They had first-hand experience with many different brands.

Dame: [Giggles] How true! Goodness, we certainly smoked whatever brands we could get in those days. I smoked so many brands I'm practically a walking encyclopedia about cigarettes. Well, I'm a Camel smoker now, and believe me, I know Camel is the cigarette for me because I've compared so many brands.

Joe: Yes, smoking whatever brands they could get during the wartime cigarette shortage made people everywhere experts on judging the differences in cigarette quality. That experience convinced a host of smokers that they preferred the rich, full flavor and cool mildness of Camels. The result:

PA-speaker Voice of God: More people are smoking Camels than ever before.

Joe: Experience really is the best teacher. Try a Camel yourself.

The ad is interesting to me because it tries to make a conditioned, manipulated, somewhat arbitrary choice -- the choice of what brand of cigarette to smoke -- seem like a reasonable one, made with deliberation and informed judgment. These people, we are told, are experts about smoking, walking encyclopedias. Hey, thanks to the war, you're an expert! The wartime cigarette shortage created a circumstance that never exists in civilian life - you had to smoke what you could get. This wasn't privation; it was a de facto tasting panel. You developed the capacity to judge the differences in cigarette quality. Informed consumer, you can now choose your brand based on the exercise of your newly cultivated expertise. You base your choice on taste, not habit or nostalgia, nor are you a puppet of advertisers. But it's not just individual judgment that's definitive here - there's a consensus. After all, "More people are smoking Camels than ever before." Does your judgment concur with the multitude, or is there something different or perhaps defective about your powers of discernment? 

In my own research into flavor and taste, I've become increasingly skeptical about the claims of sensory expertise even as I recognize the capacity to refine sensory discernment. Objective Methods in Food Quality Assessment, a textbook published in 1987, describes the lengths that sensory scientists go to create "objective" data about food preferences and sensibilities. The first chapter, with the perhaps over-insistent title, "Sensory Evaluation Can Be Objective," advises: "since humans are being used as measuring instruments, every effort must be made to control the effect of the environment on judgment." The testing room should be slightly higher pressure than the exterior, in order to eliminate the introduction of non-relevant odors. The temperature and humidity should be rigidly controlled. Colored lights might be useful, to make color differences in foods invisible. In the author's laboratory, they place tasters in an individual "domed hatch," where they can press a button to indicate when they are ready for a new sample. This way, they eliminate any possible influence introduced by the technician who delivers the sample. The taster is in a pod, isolated from all direct human contact, with a color-indeterminate cube of stuff to decide about.  

Sensory science tries gamely to create "objective" data, staging tasting tests where all potentially corrupting stimuli are stripped away, and the individual is "independent" of outside influence and exercises only her or his own sensory judgment. That is, a situation that is never like actual consumption, where we look everywhere for cues about whether something is delicious, disgusting, valuable, cheap, good to like, bad to like. It's an impossible task - a dream of a science that believes it can exist outside of the social, with laboratory as a space that maintains a cultural cordon sanitaire, sanitized from social factors. 

Which is not to say that one cannot prefer a brand of cigarettes or whiskey, or be a walking encyclopedia about tobaccos or wines or ice cream. Just that in a certain way, perhaps, our choices about taste are not only our own. 

Print and Eat the Food of the Future

One of the best parts of the pseudo-Freudian space fantasy Forbidden Planet is when Robby the Robot obliges the poor space sailor who's been left to guard the ship with a heap of liquor. Robby scans and chemically analyses the spaceman's bottle of whiskey, and then duplicates it... and duplicates it... and duplicates it... until he has a lovely pile of whiskey bottles -- at least until the invisible Monsters from the Id come and annihilate his fun.

All matter is chemicals, after all, and all chemicals are elements, and elements are just atoms, and atoms are everywhere, so why not? Anything can become anything else; stuff can be made out of no stuff.

The wait is over (maybe): why cook, when you can print your food and eat it? Sadly, there's no gracious Robby to butler our meal for us out of thin air. This is basically a modified 3D printer, the "revolutionary" technology that keeps threatening to transcend mere novelty, one of these days, maybe. 

I mock, but this article on the print-and-eat food from the IEEE Spectrum is really fascinating. At first, 3D food printers were limited by the material it used: a paste that hardened into different shapes, pretty much the edible equivalent of the standard 3D printer's plastic. (yum!) 

But then a breakthrough: Daniel Cohen, a grad student at Cornell, had the idea to treat the printer's materials as a set of miscible components, the way the three RGB printer cartridges in a color printer can produce a full-color reproduction of a multi-hued image. That is, he proposed a standard basic palette of food materials, reimagining food's basic components as though there are edible equivalents to the primary colors, which can additively produce any hue in the visible spectrum. This itself is not a novel idea: sensory taxonomers from Linnaeus to Arthur D. Little Consulting Company (and many more) have proposed systems that attempt to break the smellable-tastable world into irreducible elements. However, It's important to note that the color spectrum is a metaphor; it translates imperfectly unto the much different (chemosensory, multisensory) system of flavor perception.

Jeffrey Lipton, the article's author and an engineering student intimately involved in the development of commercial 3D-printing technology and its applications, is concerned with making the food printer's products not only palatable but desirable. The "uncanny valley" of "mushroom shaped bananas" is too "artificial", and thus likely to be rejected by the "home cook." He also dismisses proposals to use 3D food printing as a sort of hedge against a Malthusian crisis (by making palatable foods -- like "steak" -- out of cheap or repulsive proteins -- such as insects) as off-trend: today's savvy consumers reject "highly processed foods." (Incidentally, in my research on the history of flavor additives, I've found this "socially useful" application of flavor additives cited by the flavor industry starting in the 1950s and 1960s -- that synthetic flavor chemicals will help forestall a malnutrition crisis by making cheap nutritive substances (combinations of carbs-proteins-fats manufactured, perhaps, from industrial waste) edible and acceptable). 

Instead of working from basic components, Lipton says, they've taken a "top down" (rather than "bottom up") approach with the printer, working with chefs to produce fried scallops shaped like space shuttles and Austrian cookies with writing on the inside. (How this addresses purported consumer desires for "less processed" foods is not really clear...) The most exciting result is a new form of fried corn dough, impossible to achieve without a 3D printer; the dough forms "a porous matrix that allowed the frying oil to penetrate much deeper into the food. The result was something delicately crispy and greasy, like a cross between a doughnut, a tortilla chip, and raw ramen noodles."

In this incarnation, the 3D printer becomes an exquisitely refined tool for the production of highly processed food. A tool that doesn't just replicate what already exists in the world from a basic color palate, the way a camera reproduces visible reality, but something that makes new, unforeseen things possible -- maybe. Can we use this to imagine and create new flavors, or just to dress up familiar things in fancy, unfamiliar, space-ship forms?