Back in the day, Michael Silverstein began his course “Language and Culture” by questioning the very juxtaposition of the two words, which had been cast sometimes as “language in culture” (with a covert nod to an entity called “language” being encompassed by an entity called “culture”), sometimes as “language and culture” (with a tacit nod to the uneasy coexistence of distinct disciplinary traditions that made holistic claims to their subject matters). Silverstein warned us not to commit the nominalist error of assuming that the existence of disciplines by the names of (social or cultural) anthropology and linguistics meant that there truly were objects in the world that we could bracket as “culture” and “language,” respectively, though if my memory is correct, he alluded to his own position as being closer to “culture in language” than to the other encompassment. For graduate students who had just committed themselves to disciplinary training in a social anthropology and in a linguistics that took for granted the easy correspondences between their objects of study and the ways in which their disciplines had carved them up, Silverstein’s remarks challenged the very ground of our intellectual existence yet I, like—I believe—most of my classmates, let them pass without further consideration, much as readers of Foucault’s similar admonition (Reference Foucault1969, 33) often assume that it applies to other disciplines. It was only years later that I came to appreciate that the nominalist error left us with a truncated view of language and a truncated view of culture, both reduced to systems of representation, with the intricate causal threads binding them severed.
Let me turn to a classic, influential statement of the problem, from the 1940s, by the philosopher Susanne Langer (Reference Langer[1941] 1948), who distinguished two types of cultural forms, discursive and presentational. As a philosopher of art, her task was to challenge the received wisdom that the limits of conscious, denotational discursiveness marked the limits of rational thought; she did so by defining a domain of presentational forms, such as visual representations, gesture, and music. These, she argued, were equally bound by reason, equally articulate, and equally amenable to analysis. The crux of her distinction was that language (notice the slippage here, from conscious denotational language to language as a whole) “requires us to string out our ideas” like links on a chain, even though it might express objects or events that are simultaneous (cf. Saussure Reference Saussure1915, 103). “Any idea that does not lend itself [to a sequential arrangement] … is ineffable” (Langer Reference Langer[1941] 1948, 66). In contrast, presentational forms, which are not bound by the restrictions of sequentiality, can be apprehended simultaneously. Since they are not constrained by the limitations that short term memory places on the length of a discursive sequence, they have a potential for greater complexity than discursive forms do. I quote Langer at length on this point:
Language in the strict sense is essentially discursive; it has permanent units of meaning that are combinable into larger units; it has fixed equivalences that make definition and translation possible; its connotations are general, so that it requires non-verbal acts, like pointing, looking, or emphatic voice inflections, to assign specific denotations to its terms. In all these salient characters it differs from wordless symbolism, which is non-discursive and untranslatable, does not allow of definitions within its own system, and cannot convey generalities. The meanings given through language are successively understood, and gathered into a whole by a process called discourse; the meanings of all other symbolic elements that compose a larger, articulate symbol are understood only through the meaning of the whole, through their relations with the total structure. Their very functioning as symbols depends on the fact that they are involved in a simultaneous, integral presentation. This kind of semantic may be called “presentational symbolism,” to characterize its essential distinction from discursive symbolism, or “language” proper. (Reference Langer[1941] 1948, 79)
Langer initially distinguished language (reduced to denotational “discursive meaning”) on the basis of a single, relatively uncontested criterion—that talk unfolds in time. But in the course of distinguishing presentational from discursive meaning, Langer folded in a set of other, distinctly different criteria (a polythetic distinction, “a complex network of similarities overlapping and crisscrossing” [Needham Reference Needham1975, 350]) that bled into each other, concealing a tacit metaphysic:
Langer’s metaphysic of language was absorbed as dogma into mid-twentieth-century interpretative anthropology, especial in the anthropology of ritual, even when it was not cited directly. And these criteria in turn accrued others (all midcentury attributions, usually made quite innocently):
The reemergence of linguistic anthropology over the past fifty years (a movement in which Silverstein has been a key player) has challenged Langer’s metaphysic of language theoretically and ethnographically.Footnote 1 At the same time, however, linguistic anthropology has positioned itself institutionally astride a social-cultural anthropology that (for the most part) has not shifted significantly in treating language as a scalar supplement to social forms and to wordless behavior.
In an anthropological reading, nondiscursive cultural forms were fluid, shifted with (and organized) social situations, and functioned through a logic of their own. Presentational forms were untranslatable into other cultural media: talking about the jazz improvisation you just played is not the same thing as playing it, after all. It was necessary, then, to regard presentational forms as analytically independent of other cultural forms, especially of discursive forms. Insofar as presentational forms were specific to particular social situations (this is the sense of Langer’s assertion that presentational forms cannot convey generalities), they were especially useful as analytical entryways into the nature of the social situation. Insofar as they were implicit and outside the awareness of the social actors, presentational forms were especially powerful as entryways into the cognitive orientations of the actors.
The polythetic distinction between discursive and presentational form thus gathered additional associations, more as a pervasive attitude than as an explicit program; discursive and presentational became ciphers for “language” and “culture,” respectively. I would characterize this attitude by a core set of associations:
and a more fluid, secondary set of contrasts that overlap with the first set. The core set is pervasive; the second is composed of attributes that are less frequent but still common enough. It is typical for polythetic categories to combine a core set of characteristics with a more peripheral set, which is not as widely shared. The view of the relationship between language and culture concealed in this polythetic set is roughly that language is “about” culture; culture is constituted independently of language, below the threshold of consciousness, through experience; language is interpretive of culture, largely above the threshold of awareness, whereas cultural meaning is implicit; language is under the subjective control of individuals, whereas culture is shared; and, finally, language is a domain of thought, whereas culture is a domain of social action.
The commonplaces about language and culture (or language and society, language and social interaction, and so forth), then, are several: (1) that language is reducible to a linear, segmentable, denotational system (Silverstein Reference Silverstein and Basso1976, 49); (2) that the typified time of segmentable units can encompass the real time of social interaction analytically (Auer et al. Reference Auer, Couper-Kuhlen and Müller1999, 3–34); (3) that language is a superstructure constructed on a social base, sometimes as abstract structure, sometimes as local and situated, but always scaled small relative to “larger” social phenomena (see Carr and Lempert Reference Carr, Lempert and Summerson Carr2016, 8–9); (4) that talk reduces to individual practices and representations (and indeed provides warrants for a separating individual and social forms, under various guises). These, together with the initial separation of language and culture, along the lines of disciplinary linguistics and disciplinary social anthropology, are factitious objects.
Simplest Social Interaction
A cumulative finding among scholars of social interaction since the 1960s (Goffman, Sacks, Schegloff, Ericksen, and McDermott principal among them) is that social interaction requires mutual entrainment among the interactants, an entrainment that coordinates their interpretative frameworks (though not their interpretations) and their expectations of the each other’s contributions to the interaction (Goffman Reference Goffman1979; Collins Reference Collins2004, 47–101; Lempert Reference Lempert2014).Footnote 2 Their expectations are looped, which is to say that the actions of each interactant are predicated on other interactants interpreting their contributions in particular ways (Bennett Reference Bennett1976; Lewis Reference Lewis1976). Social interaction, then, is more like a dance than a series of soliloquies. When it is successfully established, coordination among interactants is indexed by a common rhythm, a rhythm that synchronizes all contributions to the interaction, be they verbal, gestural, gaze, or physical positioning. Beyond the rhythmic and gestural affordances for mutual entrainment, interactants draw on the each other’s speech in constructing syntactic parallelisms and mimesis (Ochs Reference Ochs, Ervin-Tripp and Mitchell-Kernan1977; Silverstein Reference Silverstein and Schiffrin1984; Lempert Reference Lempert2014, 383), creating an interactional poetics. A description of verbal interaction as the holistic phenomenon that it is, is impossible within the limits of linguistics, linguistic anthropology, sociology, or social anthropology as segmented disciplines—indeed it must also draw on the psychology of perception, cognitive psychology, and biology.
The phrase “mutual entrainment,” like the words “linguistics” and “anthropology,” suggests a unitary referent, one that I can refer to plainly in further academic writing. But it conceals great complexity drawing on a assemblage of behaviors and capacities, “‘complex and mediated’ absorption of indexically linked values and presuppositions” (Silverstein Reference Silverstein2003, 195; McNeill 2016). In the absence of external perturbation, humans among other animals synchronize the rhythms of their organs and the rhythms of their behaviors, a phenomenon known as “internal synchrony” (Moore-Ede and Salzman Reference Moore-Ede, Salzman and Aschoff1981; McNeill Reference McNeill2016, 20ff). Social interaction itself is synchronous, in the broadest sense of the word, grounded in establishing a joint rhythm, as the critical axis of mutual entrainment. Joint rhythm is readable from speech sound, but also from syntactic parallelism, repetition, gesture, and bodily repositionings, as a single gestalt, though the regularity of an entrained rhythm varies. Along with typological differences among languages in the role of linguistic stress in timing, the degree of mutual entrainment can vary locally (arrhythmia often signaling “trouble” in the interaction—in the form of recognition that another participant is not taking up an utterance in the way the speaker expects) and external perturbations. Even fully entrained rhythmic synchrony can depart from absolute synchrony by approximately 40 milliseconds (Auer et al. Reference Auer, Couper-Kuhlen and Müller1999), and speech/gesture synchrony by about 80 milliseconds (McNeill Reference McNeill2016, 20). Interactional synchrony overlays and subsumes internal synchrony, so that the individual bodily inhabits speech (McNeill Reference McNeill2016, 12) and consequent participation roles. Social interaction does not consist of one individual interacting with another individual interacting with another individual. Rather the social interaction is an emergent, inhabited bodily by individuals, through both rhythmic entrainment and recruitment of individuals to ever-shifting participant roles in real time (Goffman Reference Goffman1979).
Rhythmic entrainment draws on three key features of human verbal behavior: (1) The double-resonator design of the human vocal tract, mediated by the larynx, affords a stable acoustic fundamental frequency (Lieberman Reference Lieberman1973) and the perception of rhythmic regularity. (2) A tacit theory of mind, an “everyday construal of each other in terms of underlying mental states … including understanding how human action is shaped by such states” (Wellman Reference Wellman, Banaji and Gelman2013, 69). We entrain to other social beings (or at least what we assume to be other social beings); rhythmic entrainment assumes that we can read other interactants as social beings like ourselves. (3) The metrical system of a mutual, particular language, which provides a matrix for the timing of the interaction, the absence of which is a source of interactional trouble. While the first feature can safely be assumed to be anatomically grounded and universal, there is variability in theory of mind (though the relevance of that variability to face-to-face interaction has not been established) within a broadly established core of universal principles, and metrical systems—including the relationship of timing to units of speech production and perception—are language-specific. Metrical systems vary across languages, but do not vary randomly; they are a nomic (lawlike) domain, in which a typology of metrical systems opens into a universal hierarchy of units (not all of which are relevant to all systems)Footnote 3 and a universal set of organizing principles, some of which are cognitive (e.g., the incorporation of smaller units such as morae and syllables into feet) and others of which are cognitive-perceptual (e.g., that the weak-strong sequence of syllables in what traditional handbooks of poetics call “iambs” is perceived as a difference of intensity, but that the strong-weak sequence of “trochees” as a difference of length).
In accounting for social interaction, there are no clear distinctions to be drawn between the linguistic and the nonlinguistic aspects of interaction, between discursive and presentational (to return to Langer’s distinction), between the individual and the mutual relationality of the interaction as an emergent whole, between biological and linguistic, between universal and language-specific. To understand even the simplest social interaction, one must make reference to a myriad of disciplines: linguistics, acoustics, linguistic anthropology, social anthropology, sociology, cognitive psychology, perception, and human anatomy, with no discipline having special purchase over the others, and with no scalar organization by discipline. One can imagine the social sciences organized very differently, by subject matter rather than by the accidents of nineteenth-century institutionalities: Social interaction, Event analysis, Institutions, Social cognition, and so forth, rather than the familiar disciplines in which we lecture, publish, and train students, and whose traditions we honor, mainly in the breach.
And this brings me back to Silverstein’s comment that the words “language” and “culture” should not mislead us to imagine that they are the cleanly bounded subject matters of cleanly bounded scholarly disciplines. At the time he made these remarks, Silverstein was working in the same office that Edward Sapir had occupied decades earlier, in the Social Science Building at the University of Chicago. And one can’t help but imagine that he was channeling the disciplinary skepticism of Sapir when the latter referred to anthropology, sociology, and psychology as “preliminary disciplines” ([1934] Reference Sapir, Darnell and Irvine1999, 306) whose “obscure opposition of spirit must be transcended.”
