This is the approved final manuscript of “The Science of Sensory Evaluation: An Ethnographic Critique” which appeared in Adam Drazin and Susanne Kuechler, eds., The Social Life of Materials, published by Bloomsbury Academic in 2015 (pp. 81-97), acquired by Routledge in 2020.
The Science of Sensory Evaluation: An Ethnographic Critique
David Howes
There is a class of scientists who specialize in the analysis of the sensory qualities of commodities – the colour, sound, smell, taste and feel of things. Their work has not attracted much scrutiny in material culture studies or social studies in science to date. The original name for this area of research was “organoleptics.” Its origins, at least in the US, can be traced back to the 1930s when the Arthur D. Little industrial consulting firm devised a “Flavor Profile Method” and “Hedonic Index” for use by commercial food and beverage companies and the first panel on “Flavor in Foods” was presented at the 1937 meeting of the American Chemical Society. The field was given a major boost during World War II, when the US Army found that industrially-produced troop rations, which had been designed for their nutritional value, were “not performing their role because the men didn’t like how they tasted and looked” (Shapin 2012: 179). Various studies were commisioned to find out how to make the food more acceptable (Pangborn 1964; Lahne n.d.).
The title of “organoleptician” has since been dropped, replaced by “sensory professonial.” The sensory evaluation of food products remains central to the practice of these professionals (a practice which goes under the name of “sensory evaluation” or simply “sensory science”) but the scope of the products that now fall within their purview has expanded significantly to include everything from personal care to household cleaning products, and home decor to automobiles.1 Sensory professionals have also lobbied hard to expand their role within the companies they work for, seeking to convince management that the application of sensory evaluation techniques is crucial to every stage of product development, from conception to consumption. They like to use the language of driving, as in “sensory properties drive consumer acceptance and emotional benefits” (Kemp et al 2011), and it has had the desired effect. The science of sensory evaluation now forms an integral part of what Steven Shapin has called the “aesthetic-industrial complex.” It is one of the “sciences of subjectivity” which, as he suggests, “are world-making” (Shapin 2012). But what sort of world are these professionals making out of our senses?
The science of sensory evaluation rests on a fundamental paradox. On the one hand: “Most sensory characteristics of food can only be measured well, completely, and meaningfully by human subjects” as opposed to scientific instruments. On the other hand, it is considered important that human subjects behave as much like scientific instruments as possible: “When people are used as a measuring instrument, it is necessary to control all testing methods and conditions rigidly to overcome errors caused by psychological factors” (Poste et al. 1991: 1). In a similar vein, Morten Meilgaard et al (2010: 1) affirm that the key to sensory analysis is
to treat the panellists as measuring instruments. As such, they are highly variable and very prone to bias but 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.
The controls in question include creating a sampling environment that is as sensorially neutral as possible with regard to such factors as temperature, colour, and odour, and ensuring that “irrelevant” sensory factors, such as the size of the samples, do not impinge on the panelists’ judgement. Furthermore, assessors are trained to evaluate products “monadically” – that is, to assess one sensory characteristic at a time: the use of blindfolds, nose clips and “ear defenders” is advised to ensure that panelists maintain the desired focus (Kemp et al. 2011: 2.2.1.5 and 3.2). Focus is also enhanced through isolating one panelist from another by having them perform their tasks in individual booths or cubicles (for illustrations of the design of such cubicles see Meilgaard et al 2010: 24-30). In addition, assessors are commonly instructed not to discuss samples before evaluation since this might create expectations, which are considered one of the most serious potential sources of error; and to work in silence, since “comments or noises made out loud e.g. urgh! or Mmmm! can influence sensory judgments” (Kemp 2011: 2.2.1.2). Panelists are otherwise instructed to disregard their “subjective associations” since the objective is to “provide precise, consistent, and standardized sensory measurements that can be reproduced” (Poste et al. 1991: 15).
There are basically three kinds of tests used in sensory evaluation experiments. “Discriminative tests” are used to determine whether or not a difference exists among samples. “Descriptive tests” are employed to identify sensory characteristics that are important in a product and give information on the degree or intensity of those characteristics. “Affective” or “hedonic tests” are used to measure how much a panelist likes a product sample based on its sensory characteristics. There is at least one kind of test missing from this repertoire, as we shall see presently.
Finally, the variability of responses is controlled for through the use of standardized questionnaires and standard numerical scales (e.g.Stone et al 2012; Meilgaard et al 2010) as well as through statistical analysis of the results of the experiments, and the plotting of such results in the form of graphs and tables. Only those results which are “statistically significant” are considered “meaningful.” In other words, while sensory evaluation experiments are concerned with assessing the qualities of products, it is the quantification of sensation that (really) counts. There are some cautionary voices: “Statistical analysis is not a substitute for thinking”; hence, “Just because one obtains a graphical display or a series of tables with associated statistical significance does not mean it has any meaning or external validity” (Stone et al 2012: 2). Nevertheless, such cautions go largely unheeded, and in the final analysis the interpretation of results boils down to tabulating responses and pinpointing averages so that any trace of the “subjective associations” of individual panelists can be eradicated from the over-all picture of a product’s sensory qualities.
To an outside observer, it might appear difficult to distinguish between the protocol of a sensory evaluation test and the protocol of the sensory deprivation experiments of the 1960s (see Zubek 1969). It is indeed remarkable the degree of sensory restriction to which the sensory professional is subjected in the interests of producing results that are “precise, consistent” and, above all, reproducible (Poste et al. 1991).
A survey of the articles published over the past five years in the Journal of Sensory Studies, one of the leading journals in the field, reveals that many of the papers are concerned with the development of sensory lexicons. The construction of these vocabularies is important both to the standardization of communication among sensory professionals working in different countries, and the communication of sensory product attributes to the consuming public. While coffee and meat are the most studied products, one study concerned the development of a sensory lexicon for the description of the flavour, aroma, texture and appearance characteristics of dry dog food products (Di Donfrancesco et al 2012). No dogs were consulted for this study. The lexicon was entirely based on the perceptions of a five-member highly-trained descriptive sensory panel of Homo Sapiens.
The papers may otherwise be grouped according to whether they use trained panelists or so-called naïve panelists, whether they use forced-choice, projective mapping, or some other scaling method, and whether they are unimodal (e.g. El-Ghezal Jeguirim et al 2010), multimodal or cross-modal (e.g. Piqueras-Fiszman and Spence 2012) in orientation.2 A good example of multimodal product profiling, which is by far the standard, is provided by a paper on “The Perception of Creaminess in Sour Cream” (Jervis et al 2014). This paper starts from the observation that creaminess perception in dairy products is complex, and above all tied to fat content, which in turn determines liking. This poses a challenge for those manufacturers who wish to introduce low-fat and reduced fat products since, despite their health benefits, such products lack the sense appeal of their full-fat counterparts.
The experiment involved the use of an 11-point creaminess rating scale and a 9-point hedonic scale. It unfolded over a series of seven sessions which sought to hold different sensory modalities constant and in this way measure the contribution of each modality to the perception and liking of the array of sour cream products tested. In an initial session, all of the modalities were engaged and the results of this session were used as a baseline. Subsequent sessions involved visual inspection only, visual inspection and physical stirring only, blindfolds and stirring only (to focus attention on the haptic), blindfolds while tasting (to isolate in-mouth texture and flavour), blindfolds and nose-clips while tasting (to control for sight and flavour) and nose-clips only while tasting (to control for flavour). In the result, it was found that olfaction of milk-fat associated flavours has the greatest impact on creaminess perception, followed by visual assessment of flow while stirring. With this information in hand, sour cream manufacturers can know which factors have to “accounted for” in order to maximize consumers’ perception of creaminess and liking.
In another paper (Oberfeld et al 2009), which attracted considerable media interest (Oberfeld et al 2013), researchers at the University of Mainz related how they invited panelists to taste wine under different ambient lighting conditions: red, blue, green and white. The colour of the wine itself was occluded by serving it in black opaque glasses. Among other things, it was found that blue and green ambient lighting made the wine taste spicier than under white light, and that red ambient lighting made the wine taste as much as 50% sweeter than under blue or white light. General liking and willingness to pay a higher price were also found to be augmented when the illumination was set to red or blue rather than green or white. It was the same wine in all cases (a dry white Riesling). This study is of interest for its methodological innovation. It did not just focus on the product, the way most sensory evaluations do. It modelled an environment. And it did not just treat the senses monadically (i.e. by concentrating on a single sensory characteristic) or additively (i.e. toting up the scores to arrive at a “complete” sensory profile of a product). Rather, it allowed that the senses might be interactive. In the result, it was found that the red ambient light, which was not a property of the product (the wine), but rather the environment, decidedly influenced the perception of the product’s flavour, and so on with the other colours.3 Hence, the wine’s taste must be recognized as contingent on its context of consumption, but it is precisely context that the design of the sensory evaluation laboratory (except in the case of this study) is designed to rule out. Ergo the majority of the studies published in the Journal of Sensory Studies are valid to the extent that the products concerned are consumed in the laboratories in which they were tested. But who wants to drink wine alone in a booth in a sensory research laboratory?
Sensory professionals are to be admired for the sensory and social sacrifices they make to test and “perfect” (or bring what is called “quality control” to) the products we consume. However, there are serious questions concerning the validity and applicability of the findings of sensory science outside (and even within) the laboratory which still need to be addressed.
Summing up, it is difficult to imagine a more asocial or, practically speaking, more asensual environment and protocol than the environment and protocol of a sensory research laboratory. This is due to the assumption that, as Meilgaard et al. put it: “we must minimize the variability and control the bias [of panelists] by making full use of the best existing techniques in psychology and psychophysics.”
The Senses in Everyday Life
What if the methodology of some other discipline, besides psychology and psychophysics, such as anthropology, were incorporated into the practice of sensory science? Anthropology has, in fact, begun to make inroads into the field due to the rise of the subdiscipline known as the “anthropology of the senses” or “sensory anthropology” (Howes 2003). It is exemplified by the work of Sarah Pink (2004, 2009), John Sherry (2006), Timothy Malefyt (2014), Jake Lahne and Amy Trubek (2014), and the Concordia Sensoria Research Team (CONSERT), among others. The principles of this mode of inquiry may be summarized as follows:
First, sensory anthropology understands acts of perception to be cultural as well as biological and psychological processes.
Second, it takes the study of product perception out of the sensory research laboratory and into the street, the home, the bar or whatever the “natural environment” of the consumer may be. The meaning of a product is in its use, and not its physical structure (abstracted from any context) alone.
Third, its methodology is one of participant sensation, or feeling along with one’s informants, as opposed to subjecting them to some predetermined protocol and list of questions the way a sensory scientist would.
Fourth, the focus of sensory anthropology is on eliciting “the native’s point of view,” or rather, because one doesn’t want to privilege the visual over other senses, the native’s “ways of sensing” (Howes and Classen 2014) – that is, the practices (including technologies) which frame a given group’s perception of the world.
Fifth, sensory anthropology postulates that “the senses interact with each other first,” in culturally-conditioned ways (Howes and Classen 1991: 258). Hence, the focus is on analyzing the relations and transfers between the senses, rather than viewing them as independent channels or separate silos.
Sixth, “As we sense, we also make sense,” in Phillip Vanini’s felicitous expression (Vannini et al 2012). This formulation plays on the polysemy, or double meaning, of the word “sense.” This word includes both sensation and signification, both feeling and meaning in its spectrum of referents, which should be conceived as forming a continuum.
The polysemy of the word sense is lost on sensory scientists. The signifying (or “symbolic”) and also social dimensions of perception are occluded by their research protocols. By limiting the sorts of tests they use to the discriminative, the descriptive and the hedonic, they prevent themselves from ever investigating what could be called the semantics of perception (but see Alcántara-Alcover et al. 2014). A semantic test, such as an anthropologist would be the first to utilize, would seek to “determine the meanings or mental associations stimulated by a given product’s sensory characteristics” (Howes 2003b: 119).
In addition to highlighting the issue of sense-making, as will be discussed further below, research in the sensory anthropology of consumption has shown that consumers may be more or less discriminative in a particular sensory register, depending on how it is weighted or valued in their culture or subculture. By way of example, consider Sarah Pink’s study of “the sensory home,” which was informed by the methodology of “sensory ethnography” (Pink 2009).
The Sensory Home
In this study, which was commissioned by Unilever, Pink (2004) compared attitudes toward household cleanliness and practices of housework in Spain and the UK. Her informants included students and retired people, as well as single and married women (or “housewives”) of middling age. She asked her informants to take her on a tour of their house or flat and recorded their actions and words on video. The tours typically involved poking her head in cupboards, and being invited to “smell this,” or “feel that,” in addition to conversing with her informants (thereby breaking the silence that normally prevails over the assessment of products in the sensory research laboratory).
For the Spanish subjects, “dust” referred to matter that had infiltrated the home from the outside world, and was classified as dirt to be eliminated. For the British subjects, dust referred to the flakings of persons and matter such as paint or plaster inside the home, and people were more tolerant of a certain build-up. It was not dirt as long as it did not smell or appear tacky. One young man stated that when the floor of his apartment started to feel sticky it was time for cleaning.
Pink found that the practice of cleaning house sometimes involved people “dancing uninhibitedly” to their favourite music while wielding a broom or mop. Thus, housework had an audio component, a kinaesthetic component (which involved more than just scrubbing), and it also involved setting out scented products, like incense and essential oils as a finishing touch. In other words, cleaning did not involve eliminating odours so much as enhancing the existing smell of the home. Significantly, Pink found that all of her informants compared themselves (often negatively) to what they suspected a “real housewife” would do, thereby incorporating a social dimension into what might otherwise be seen as a very private practice.
Pink’s study of “the sensory home” brings out how consumers do not necessarily use products “as directed” but rather “negotiate” social meanings through them and in so doing construct identities for themselves. Consumption is a creative process, Pink argues, wherein products do not “drive” or “trigger” responses in a straightforward fashion, but rather are selectively deployed to construct “worlds of sense” within which people can feel “at home.” 4
The Interplay of the Senses
As noted previously, sensory evaluation tests frequently involve the construction of barriers between people, between the senses, and between the “subjective associations” of the assessor and his or her response to the sensory characteristics of the product tested. This is accomplished through training and through the architecture of the sensory research laboratory with its individual booths and “neutral” atmosphere . Assessors are instructed to discriminate, describe and express their preferences, but not associate. Do these firewalls work? They might be made to work in the context of the laboratory, though they would hardly work in everyday life, but even in the laboratory it is dubious that the play of associations between the senses can be forestalled. In one of the studies we reported on in Aroma: the Cultural History of Smell (Classen Howes and Synnott 1994: 194) involving a test of facial tissues, it was discovered that respondents found pine-scented tissues to be “fresher” but also rougher than unscented facial tissues, even though there was no actual difference to the texture of the tissues used in the two samples. The reason should have been obvious: the respondents did not dissociate the scent of pine from the feel of pine needles, which are, of course, prickly. This is because “As we sense, we also make sense” (Vannini et al 2012).
In another study reported on in Aroma, respondents in a Chicago shopping mall were asked: What odours make you feel nostalgic?
People born in the 1920s, ‘30s and ‘40s said that such odours as rose, burning leaves, hot chocolate, cut grass and ocean air made them feel nostalgic. Persons born during the 1960s and ‘70s, in contrast, grow nostalgic at such scents as Downy fabric softener, hair spray, Play-Doh, suntan oil, Cocoa Puffs, and candy cigarettes (Classen, Howes and Synnott 1994: 202-3)
The trending evidenced by this survey, when the responses are grouped by decade of birth, are significant: there has been a shift away from “natural” odours towards “artificial” ones, and many of the latter come already trademarked. This pattern brings out nicely the extent to which the sensorium is an historical formation:
It is not only in clothing and appearance, in outward form and emotional make-up that men are the product of history. Even the way they see and hear is inseparable from the social life-process …. The facts which our senses present to us are socially preformed in two ways: through the historical character of the object perceived and through the historical character of the perceiving organ (Horkheimer quoted in Levin 1997: 63 n. 1)
A study conducted a number of years back by the Concordia Sensoria Research Team brings out further the sociality of sensation and the indissociability of the senses. We asked: What accounts for the popularity of Corona, the best-selling imported beer in Canada? Preliminary research suggested that part of the answer must have to do with gender. Men are, notoriously, far more avid drinkers of beer than women, and in the case of most brands the ratio is 5 male drinkers to one female drinker. In the case of Corona, however, the ratio is more like 3 to 2. This means that its popularity among women is key to its success. So we set out to investigate why women prefer Corona. Our quest took us to a range of bars and restaurants, many with a Mexican theme, where we talked with both men and women about their preferences. This displacement was essential, since in anthropology one wants to encounter subjects on their own ground and elicit the categories they use to order the world.
We did not go in with a predetermined set of questions. Instead we let the questions emerge in the course of interaction. Some subjects said that they had encountered Corona while on vacation in Mexico. For them, drinking Corona when back in Montreal was a way of injecting some festivity or “vacation spirit” into the drudgery of everyday life. More typically, however, those subjects who drank Corona regularly said they liked it because it is “light.”
Technically, Corona is not a “light beer.” It has the same alcohol content and carb levels as regular, domestic beers. This response, then, is an example of consumer-added meaning (and value). We needed to discover what motivated this categorization. What was it about the sensory characteristics of Corona that could explain this “misperception” (which is not a misperception at all, of course, from the native point of view)?
The design of the Corona bottle struck us as one of the factors contributing to the perception of the beer as “light.’’ Corona comes in tall, slender bottles that are clear and translucent. By contrast, most domestic beers, such as Molson Canadian, come in short, stubby, brown-coloured bottles that even look more weighty, more dense than the former. Furthermore, the colour of Corona is light, like sunshine, compared to the golden colour of Molson Canadian. From our conversations with our research subjects, it appeared that they were condensing – or “associating” – a number of different sensations into one: the bright (or “light”) tint of the beer and the translucency (as well as slenderness) of the bottle was identified in their minds with lightness of taste (or, put another way, absence of heaviness). This impression was borne out by the gestures people used to describe their taste experience. When men talked about what they liked in a beer they would pat their stomachs whereas the women would rub their thumb and fingers together. The latter gesture suggested that what women most appreciate in a beer is a refined or delicate taste, whereas men are more interested in a full (and filling) flavour. Indeed, those men who preferred domestic beers claimed that Corona “has no taste” (by which they meant body) whereas those men who drank Corona with their female friends dissociated themselves from more “heavy drinkers,” as they styled their male counterparts.
The Corona study has implications for the study of the perception of creaminess in dairy products discussed earlier. It suggests that greater attention should have been paid to the interaction of the modalities involved in the trial (instead of holding them constant), and that the meaning or “sense” of fat in everyday life for the test subjects (and the general populace) also needed to be explored. In other words, even though full-fat products can be demonstrated to have higher acceptance on account of their “intrinsically” fuller taste or palatability, the matter does not end there. Also pertinent are the extrinsic associations which fat as a “material symbol” evokes. The “fatness” both of bodies and substances has recently become a topic for anthropological investigation (Forth and Leitch 2014), and this has led to a more nuanced understanding of both its (increasingly ambivalent) cultural construction and its effects.
“Sensory Experience is Social Experience”
A study of consumer perceptions of Vermont arisan cheese conducted by the team of Jake Lahne and Amy Trubek (2014), the former a trained sensory scientist, the latter a sensory anthropologist, can shed light on this question of so-called extrinsic properties and associations. They theorize sensory perception as a learned and active practice (rather than passive reflex). Sensations are held to arise “neither from the food nor from the consumer, but from the encounter between them, that is, it is neither taste nor taster, but tasting” (Lahne and Trubek 2014: 130 citing Hennion 2005). This shifts attention from the search for (putatively) universal, objective sensory qualities of food to the recognition that the sensory qualities of food “emerge” for a particular consumer in a particular context.
The emphasis on context is carried further by Lahne and Trubek’s insistence that “sensory experience is social experience.” This is reflected in the way their preferred methodology involves holding focus-group discussions around a plate of cheeses instead of relegating participants to individual cubicles, and inviting participants to recall past experiences instead of simply check off boxes. The sorts of questions Lahne and Trubek ask may be characterized as conversation-starters. They include questions about the participants’ own consumption practices, about what makes a cheese “artisan” in their estimation, and challenges like: How would you convince people who’ve never tried that cheese to give it a try? Is there anything besides taste that is very important about this cheese? In this way, participants are encouraged to explore the subjectivity of their own responses, instead of screening out any trace of subjectivity so as not to appear biased. Interestingly, none of the participants felt that their responses were any the less valid for being subjective, and their subjective judgments about subjectivity became part of the findings. This is in stark contrast to the objectification of taste (through the privatization, bureacrarization and pacification of the senses) in the conventional sensory science research laborartory. The context Lahne and Trubek created activated the senses instead of restricting and objectifying them.
Lahne and Trubek found that the sensory experience of Vermont artisan cheeses, for those who customarily consume them, “stems from a mix of intrinsic, organoleptic properties and extrinsic socially embedded properties,” but the two are “mutually constitutive,” and so it is not fruitful (or entirely possible) to disentangle them (Lahne and Trubek 2014: 132). The extrinsic properties include such things as the social context in which a cheese was first encountered by a participant (e.g. a wedding, a family meal) and the memories which attach to the cheese in consequence, and information about the conditions of production of the cheeses sampled (e.g. the cheesemakers’ animal husbandry practices, ethos of workmanship, scale of operation, sustainability, etc.). For example, one respondent remarked on the “grassy,” “earthy” flavours in a particular cheese possibly due to knowing that the animals were permitted to graze instead of being exclusively fed grain; another sensed the “care” a cheesemaker put into his cheese. Lahne and Trubek also recorded instances of participants modifying their perception and judgment of a particular cheese in response to what other participants had to say.
Alice: I would say it’s tangy, and it has a nice – the flavors change from when you first bite it to … the aftertaste.
Ben: A little citrusy, maybe?
Alice: Yeah … I would maybe even say nutty, like it’s … I don’t know … like it rolls around in your mouth and the flavors change (Lahne and Trubek 2014: 135)
Summing up, Lahne and Trubek argue that “through an active, iterative, and social practice of sensory perception, consumers integrate their past personal experiences, socially transmitted and valued information about producer practices, and the material properties of the cheese into a single instance of sensory experience..” (Lahne and Trubek 2014: 130)
The Lahne and Trubek study is unusual in the annals of sensory science for what could be called the free-range character of the focus-group discussions (though the use of focus-groups is not uncommon) and for introducing a “social theory” of sense experience. Its publication in the journal Appetite is a reflection of how the field of sensory science is changing, opening up to new methodololgies and theories. One can nevertheless imagine the questions and objections that a conventional sensory professional might put to the authors, such as: What would a blind taste test involving artisan and generic cheeses reveal about which “tastes better”? Why no attempt to plot liking in relation to fat content? Aren’t the participants all pre-selected?5 Aren’t the questions somewhat leading questions? How can any generalizations be made on the basis of such contingent results? However, these objections can be turned around: Lahne and Trubek freely admit that their results are contingent on context, but would point out that sensory evaluation experiements are no less contextual, despite their appearance of objectivity, on account of being staged in a lab.6 Furthermore, they could point to the all too frequent practice of using the employees of the company that is conducting the study as stand-ins for “the ordinary consumer” (see Resurreccion 2008), and ask: How representative is that? More seriously, as Lahne (2015 n.d.) has argued, the research protocols of sensory science are fitted, and even “overfitted,” to industrial production where products are standardized and theefore “portable across contexts” (e.g. a batch of Coca-Cola) in contrast to artisan production, where products are “unfinished,” often quite variable and tied to locale. There is risk involved in artisanal production, whereas variation is virtually eliminated in industrial production (Lahne 2015; see further Paxson 2013). Hence, the protocol doesn’t fit the product, or vice versa .
In Lahne and Trubek’s study, social context does not simply refer to the exchanges between the participants in the focus group, but also extends to the geographic region or terroir of Vermont Part of what makes the artisan cheeses “taste better” is that they embody “the taste of place” (Trubek 2007). But the taste of place is not a function of geography alone, Lahne and Trubek insist. It also has to do with what they call “cultural saturation” — that is, the ubiquity of Vermont artisan cheese in Vermont, such that it is impossible for a Vermonter not to be aware of artisan cheese, and many consumers have in fact developed “personal connections” to such products, and their producers.
Generic cheeses, and processed cheeses, such as Cheez Whiz, lack such personal connections, and indeed are designed to be “portable across contexts.” The same is true of most of the standardized commercial products, hatched in laboratories, which saturate our existence as consumers. These products have the effect of standardizing our perceptions, shaping our tastes to conform to those of their designers and manufacturers. But this does not prevent us from personalizing, or, as it were “domesticating” them through incorporating such products into our everyday lives. Cheez Whiz is a case in point: it has been discovered by consumers to have many uses never imagined by its manufacturers (Green 2000). Consumption is always a matter of context in the final analysis. The meaning of goods is in their use – that is, in the sense we make of them, and not simply the design characteristics which the producer put there (Howes 1996)
The implication of these observattions is that rather than dismiss the Lahne and Trubek study for being too parochial, sensory science needs to develop new, more historically-, and culturally-grounded methods for understanding (not just assessing) the sensory qualities of the full range of commodities, materials, substances that pervade our everyday lives, both those which are artisanally produced and those which are mass-produced. The anthropoogy and history of the senses has a vital role to play in generating such an understanding by attuning us to the “social preformation” of the senses and the extent to which “As we sense we also make sense” (Vannini et al 2012; Howes and Classen 2014). Perception is not a passive process, a physiological reflex, it is an active, “world-making” activity (Shapin 2012; Classen 1993), which is nevertheless contingent on the materials at hand. A number of highly stimulating studies that illustrate this point have emerged in recent years, having to do with such materials as aluminium, the “material of mobility” (Fallan 2013; Sheller 2014) and lycra, the fiber that “shaped America” (O’Connor 2011). By way of closing, I would like to offer a sensory history of perhaps the most ubiquitous material of the twentieth century: plastic. Plastic was at once the substance that characterized the physical world of the twentieth century and provided a material base for much of its cultural expression.
The Plasticity of the Material World
The term plastic, in fact, covers a variety of synthetic or semi-synthetic subtsances, from the celluloid used in film and cheap jewellery to the vinyls employed in records, raincoats and exterior siding. However, plastic became the umbrella term for all these creations of the chemical industry. Technical advances resulted in plastics that were amazingly durable, as well as low cost. “Plastic is forever” touted one industry pioneer, “and a lot cheaper than diamonds” (cited in Miekle 1993: 9).
Over the course of the century the material became a familiar component of ordinary life. A family celebrating Christmas in the United States in the 1970s, for example might have a plastic Christmas tree decorated with plastic ornaments and featuring plastic Barbie dolls and Lego blocks as presents. In a famous line from the 1967 movie The Graduate, a recent college graduate who is uncertain about his future is told by his businessman father: “I want to say one word to you. Just one word … There’s a great future in plastics. Think about it” (cited in Miekle 1993: 3).
Like its material uses, the sensory properties of plastic were multiple. Easy to shape, colour, and texture, plastic might approximate anything. It could be made to look like wood or it could be made to look like glass, it could resemble flowers or it could mimic gemstones. Nonetheless, on close inspection, it always retained something “plasticky” in its look and feel. Plastic itself had no imitators, for who would imitate such a cheap and indeterminate substance?
Due to its mutability, plastic engendered a notion of the malleability of the material world. The French philosopher Roland Barthes wrote of plastic in the 1950s that it embodied “the very idea of … infinite transformation” (1972: 79). Plastic’s mutability coincided with twentieth-century desires to reshape not only the physical environment, but also society, and even the human body through cosmetic and surgical procedures. Limits set by nature or by custom no longer seemed to hold in a plastic world. Anything could take on a new form.
While plastic was embraced by the twentieth century for its malleability and low cost, however, it was despised (at least by the educated classes) for its “inauthenticity”. Over the course of the century, in fact, the word plastic came to be a synonym for fake. Social critics saw plastic as sign and symptom of a society in which simulations had a greater appeal than reality. When the businessman in The Graduate affirmed that there was a great future in plastics, the line was not intended to serve as an indicator of commercial acumen, but as an indictment of the superficiality and materialism of Western culture.
It was not only the look and feel of the twentieth-century industrialized world that breathed artificiality, however, but also the taste. Convenience foods – from the quick meals served up by fast food restaurants to the prepared foods stocked at the supermarket (such as the frozen “tv dinners” made to be warmed and eaten while watching television) – became increasingly popular during the century. The new processed foods also had new artificial flavours and colours, many of them derived from petrochemicals just like most plastics (see Classen, Howes, and Synnott 1994: 187-200). In their song of 1972, “Plastic Man,” The Kinks sung disparagingly of a “plastic man” who “eats plastic food with a plastic knife and fork.” Many processed foods, of course, were packaged in plastic, if not canned or boxed. The contents of the supermarket thus seemed, from one perspective, to represent one more triumph of modern technology, and, from another, one more of the shams of contemporary life.
Acknowledgments
An earlier draft of the first half of this chapter was published in the Proceedings of the Ethnographic Practices in Industry (EPIC) 2013 conference (Howes 2013). I wish to thank the copyright holder of this material, American Anthropological Association, for allowing me to reproduce it here. In being transposed, however, it has been substantially modified and expanded. The cultural history of plastic presented in the concluding section of this essay derives in large part from my introduction to A Cultural History of the Senses in the Modern Age, 1920-2000 (Howes 2014)
Notes
1. For a survey of these parallel developments in the management of sensation in other fields of mass production-consumption besides food see the discussion of “giving products sense appeal” in Howes and Classen 2014: 139-41; Sheldon and Arens 1932).
2. Cross-modal invest igations are new to sensory science, and many of the studies in this vein in the Journal of Sensory Studies have as one of their co-authors the maverick experimental psychologist Charles Spence. Spence directs the Cross-Modal Research Lab at Oxford University and is a frequent collaborator with Heston Blumenthal, the proprietor of The Fat Duck restaurant. Spence’s focus on cross-modal relations, or what we call “intersensoriality” (Howes 2011: 177-9), is at the forefront of the critique and transformation of the compartmentalized understanding of the sensorium that traditionally prevailed in the brain sciences into a more integrated vision. While Spence’s research is still lacking a comparative or cross-cultural dimension (see Howes and Classen 2014: ch. 6), it nevertheless comes close to the interactive understanding of the sensorium that is fundamental to research in the anthropology of the senses.
3. In the on-line summary of their conclusions, the authors of the Mainz study write:
Ambient lighting influences how wine tastes, even when it has no effect on the color of the wine in the glass. Our results show that the context has a stronger influence on the taste perception than formerly believed. These findings can be relevant for the architectural designing of restaurants and wine shops.
How can the effects of ambient color be explained?
The simple hypothesis that whenever a certain light color makes a person feel comfortable he or she likes the wine better could not be affirmed. The emotions elicited by a certain light color do not seem to be the cause of the effects.
An alternative explanation could be an influence of color on cognition, for example by making us more accessible and responsive for a certain taste. Likewise, associations could play a role (Oberfeld-Twistel 2013).
As examples of the role played by associations, Oberfeld et al propose that green may connote “immature” and red may connote “sweet”. The Mainz study departs from the vast majority of research in sensory evaluation by acknowledging the significance of context, recognizing the senses as interactive, and refusing to reduce the explanation of the observed effects to the mobilization of the emotions alone: cognition (or what we would qualify as sensuous cognition) also plays a role.
4. Pink does not discuss how this information was operationalized by the study’s sponsor, Unilever.
5. The participants were “pre-selected” in the sense that they were recuited by means of advertisements that solicited “consumers of Vermont artisan chese who were interested in participating in a research study on their opinions” (Lahne and Trubek 2014: 131; see further Lahne, Trubek and Pelchat 2014)
6. The subjectivity of perception is made to appear objective through what Bruno Latour (1987) calls the process of “inscription” – that is, all of the grapohs and tables which “represent” the object of study. But this objectivity is a product of the process of insciption itself. It depends ultimately on a visualization of taste.
References
Alcántara-Alcover, Enrique et al. 2014. « Exploratory Study of the Influence of the Sensory Channel in Perception of Environments.” Journal of Sensory Studies 29(4): 258-71
Barthes, Roland,. 1972. Mythologies. London: Paladin.
Classen, Constance. 1993. Worlds of Sense: Exploring the Senses in History and Across Cultures. London: Routledge.
Classen, C., Howes, D. and Synnott, A. 1994. Aroma: The Cultural History of Smell. London and New York: Routledge.
Di Donfrancesco, B., Koppel, B. and Chambers IV, E. 2012. “An Initial Lexicon for Sensory Properties of Dry Dog Food.” Journal of Sensory Studies 27(6): 498-510
El-Ghezal Jeguirim, S., Dhouib, A.B., Sahnoun, M., Cheikhrouhou, M, Njeugna, N., Schacher, L. and Adolphe, D. 2010. “The Tactile Sensory Evaluation of Knitted Fabrics: Effect of Some Finishing Treatments.” Journal of Sensory Studies 25(2): 201-15
Fallan, Kjetil. 2013. Culture by Design: Co-Constructing Material and Meaning. In Kjerstin Aukrust (ed.), Assigning Cultural Values . Frankfurt: Peter Lang Publishing, 135-163.
Forth, Christopher E. and Leitch, Alison (eds.) 2014. Fat: Culture and Materiality. London: Bloomsbury.
Green, Jay. 2000. Clean Your Clothes with Cheez Whiz: And Hundreds of Off-Beat Uses for Dozens More Brand-Name Products, Los Angeles: Renaissance Books
Hennion, A. 2005. “Pragmatics of Taste.” In M. Jacoibs and N. Hanrahan (eds.), The Blackwell Companion to the Sociology of Culture. Malden, MA: Blackwell.
Howes, David (ed.). 1006. Cross-Cultural Consumption. London: Routledge.
Howes, David. 2003a. Sensual Relations: Engaging the Senses in Culture and Social Theory. Ann Arbor: University of Michigan Press.
Howes, David. 2003b. “Evaluation sensorielle et diversité culturelle,” Psychologie française 48(4): 117-125
Howes, David. 2005 “Hyperaesthesia: the sensual logic of late capitalism”, in D. Howes (ed.) Empire of the Senses: The Sensual Culture Reader, Oxford: Berg
Howes, David. 2011.”Hearing Scents, Tasting Sights: Toward a Cross-Cultural Multimodal Theory of Aesthetics.” In Francesca Bacci and David Mellon, eds., Art and the Senses. Oxford: Oxford University Press.
Howes, David. 2013. “‘The Race to Embrace the Senses’ in Marketing: An Ethnographic Perspective.” Pages 5-30 in Ethnographic Practices in Industry Conference [EPIC] 2013 Conference Proceedings. Washinton, DC: American Anthropological Association.
Howes, David. 2014. “Introduction: ‘Make It New’ – Reforming the Sensory World” in David Howes (ed.), A Cultural History of the Senses in the Modern Age, 1920-2000, London and New York: Bloomsbury.
Howes, David and Classen, Constance. 1991. “Epilogue: Sounding Sensory Profiles.” In David Howes (ed.) The Varieties of Sensory Experience: A Sourcebook in trhe Anthropology of the Senses. Toronto: University of Toronto Press.
Howes, David and Classen, Constance. 2014. Ways of Sensing: Understanding the Senses in Society. London and New York: Routledge/Taylor & Francis.
Kemp, S., Hollowood, T. and Hort, J. 2011. Sensory Evaluation: A Practical Handbook. Oxford: Blackwell
Jervis, S.M., Gerard, P., Drake, S., Lopetcharat, K. and Drake, M.A. 2014. “The Perception of Creaminess in Sour Cream.” Journal of Sensory Studies 29(4): 248-57.
Lahne, Jake. 2015 n.d. Sensory Science, the Food Industry and the Objectification of Taste.
Lahne, Jake and Trubek, Amy. 2014. “ ‘A Little Information Excites Us’: Consumer Sensory Experience of Vermont Artisan Cheese as Active Practice.” Appetite 78: 129-38.
Lahne, Jake, Trubek, Amy and Pelchat, M.L. 2014. “Consumer Sensory Perception of Cheese Depends on Context: A Study Using Comment Analysis and Linear Mixed Models.” Food Quality and Preference 32: 184-97.
Latour, Bruno. 1987. Science in Action: How to Follow Scientists and Engineers Through Society. Cambridge, MA: Harvard University Press.
Levin, C. 1997. Modernity and the Hegemony of Vision. Berkeley: University of California Press.
Lindstrom, Martin. 2005. Brand Sense: How To Build Powerful Brands through Touch, Taste, Smell, Sight, and Sound. New York: Free Press
Mack, Adam. 2014. “The senses in the marketplace”, in D. Howes (ed.) A Cultural History of the Senses in the Modern Age, 1920-2000, London Bloomsbury.
Malefyt, Timothy de Waal. 2014. “An Anthropology of the Senses: Tracing the Future of Sensory Marketing in Brand Rituals.” In Rita Denny and Patricia Sunderland, eds. Handbook of Anthropology in Business. Walnut Creek, CA: Left Coast Press
Miekle, J.L. 1995. American Plastic: A Cultural History. New Brunswick, NJ: Rutgers University Press.
Meilgaard, M., Carr, B.T. and Civille, G.V. 2010. Sensory Evaluation Techniques, 3rd edition. Boca Raton, FL: CRC Press
Oberfeld-Twistel, D. 2013. Wine and Color: Effects of Ambient Light on Taste and Aroma. http://www.staff.uni-mainz.de/oberfeld/wine2.html [accessed 15 August 2014]
Oberfeld, D., Hecht, H., Allendorf, U. and Wickelmaier, F. 2009. Ambient lighting modifies the flavor of wine. Journal of Sensory Studies, 24(6): 797-832.
O’Connor, Kaori. 2011. Lycra: How a Fiber Shaped America. New York: Routledge.
Pangborn, R.M. 1964. Sensory Evaluation of Food: A Look Forward and Back.” Food Technology 18: 1309-24
Paxson, Heather. 2013. The Life of Cheese: Crafting Food and Value in America. Berkeley: University of California Press.
Pink, Sarah. 2004. Home Truths. Oxford: Berg
Pink, Sarah. 2009. Sensory Ethnography. London: Sage
Piqueras-Fiszman, Betina and Spence, Charles. 2012. “The Influence of the Color of the Cup on Consumers’ Perception of a Hot Beverage,” Journal of Sensory Studies 27(5): 324-31.
Poste, Linda et al. 1991. Laboratory Methods for Sensory Analysis of Food. Ottawa: Agriculture Canada.
Shapin, Steven. 2012. “The Sciences of Subjectivity.” Social Studies of Science 42: 170-84
Sheldon. R. and Arens, E. 1932. Consumer Engineering: A New Technique for Prosperity, New York: Harper.
Sheller, Mimi. 2014. Aluminum Dreams: The Making of Light Modernity. Cambridge, MA: The MIT Press.
Sherry, J. 2006. “Sporting Sensation.” The Senses and Society, 1(2): 245-8
Stone, H., Bleibaum, R. and Thomas, H. 2012. Sensory Evaluation Practices, 4th Edition. San Diego, CA: Academic Press.
Trubek, Amy. 2008. The Taste of Place: A Cultural Journey into Terroir. Berkeley, CA: University of California Press.
Vannini, Phillip, et al 2012. The Senses in Self, Society and Culture: A Sociology of the Senses. London: Routledge
Zubek, J. (ed.) 1969. Sensory Deprivation: Fifteen Years of Research. New York: Appleton-Century-Crofts.
