Conceptual Structure, Discourse, and Language Conference

Amy Cook (Indiana)

Saturday, May 19th
Buchanan B213

And as imagination bodies forth
The forms of things unknown, the poet’s pen
Turns them to shapes and gives to airy nothing
A local habitation and a name. (A Midsummer Night’s Dream, 5.1)

The study of linguistic processing – and the embodied and embedded thinking that it reflects and shapes – offers a tremendous tool to those of us wanting to know how and why stories are told by bodies onstage. In this paper I will provide three examples of how conceptual blending theory and gesture research has expanded my readings of Shakespeare’s plays on the page and on the stage. I will stage the connections between an edifying interpretation of Hamlet by the people of a West African village and the casting of Sam Shepard in Michael Almereyda’s Hamlet (2000). By going under the language we think of as literal, we can enrich our understanding of the play, illuminate our conceptual structures, and expand options for staging and embodying the play. Ophelia’s death and Shepard’s Ghost are both conceptual blends that deeply impact our conception of the play as a whole and a cognitive method of analysis allows for a smarter performance analysis.

Of course, research on gesture has shown persuasively that language is profoundly embodied. In the final section of my paper I will look specifically at how Shepard uses his hands to communicate with and against the language. I will also raise some methodological challenges of studying gesture onstage.

The test for me of the application of conceptual blending theory or cognitive science to products of the creative mind is not if it answers with finality some question, but rather whether or not it helps us get to the next question of interest. Conceptual blending theory and cognitive linguistics more broadly put pressure on old readings and generate future research questions. The question, then, is not what does Shakespeare mean when he says that Ophelia drowned or that theatre should be like a mirror held up to nature, but rather what further questions can we ask about Hamlet once we have used blending theory to explain how he makes his meaning?

Download PDF

Robert Carlson (Africa IU & SIL)

Friday, May 18th
Buchanan A210

This paper examines Supyire (Senufo, Gur, Niger-Congo) spells (kulushî) which accompany healing (wempyii) and protecting (sárágá) rituals performed by ordinary people (as opposed to herbalist specialists). As in many African cultures (cf. for example Azande spells, described in Evans-Pritchard 1937: 450ff), Supyire spells use ordinary language. They do not contain archaic or incomprehensible words or out-of-the-ordinary syntax, nor is the exact wording important.

Like Sørensen (2007), this paper analyses the rituals using conceptual blending theory, though the approach differs in some respects from Sørensen’s. For example, Sørensen (p. 74ff) claims that all magic rituals involve the blending of “profane” and “sacred” input spaces, but the latter is not a useful category in the rituals examined here. In fact, the category of “magic” is not a useful category for understanding insiders’ conceptualizations of the activities investigated here.

Supyire medicine, like much folk medicine, is analogical. As one man explained, Wyɛɛ̀re ɲyɛ yasumaga. ‘Medicine is an analogical thing (lit. a thing measured (against something else)).’ The success of the healing or protection depends at least to some extent on the aptness of the analogy between the materia or the manipulation of the materia, and the thing or situation to be altered. The spells accompanying healing and protection rites normally mention the analogy involved. The following (translated literally), uttered while pouring cold water on live coals in the presence of an ill person, is typical:

These coals are hot.
This cold water is cold.
The heat of these coals, the cold water will remove it.
Like the cold water has removed the heat from these coals,
may the heat (=disease) go out of [patient’s name]’s body.

The healing depends on bringing together a domain of directly manipulable items (cold water and live coals) and a domain in which direct manipulation is impossible (removing a disease from a patient). Change in the “effecting” (manipulable) domain is meant to effect a change in the non-manipulable but analogous “affected” domain, and this is referred to explicitly in the spell.

More interesting, as well as more common, are spells which, in addition to prompting for the construction of the effecting and affected mental spaces, explicitly mention in parallel a counterfactual (or extremely unlikely) space and the actual, but unwanted, situation in a fourth space. Following is a typical example, uttered while inserting a small object representing a person who has been losing weight into an incision made in a baobab tree:

If the baobab has become thin,
may [name] become thin.
The baobab has not become thin,
may [name] not become thin.

Such spells may be analysed as prompting a network of four spaces, arranged in two parallel pairs: a counterfactual situation in the effecting domain (the baobab has become thin) is linked conditionally to an actual but undesirable situation in the affected domain (the person being treated has in fact become thin). This parallels a factual statement in the effecting domain (the baobab has not become thin) which is linked to an unreal but desired situation in the affected domain (may the patient not grow thin). Although not explicitly encoded, an interpretation of contrast emerges from this network.

References

Evans-Pritchard, E. E. 1937. Witchcraft, oracles and magic among the Azande. Oxford: Clarendon Press.

Sørensen, Jesper. 2007. A cognitive theory of magic. Lanham, MD: AltaMira Press.

Download PDF

Natalia Blackwell, Marcus Perlman, & Jean E. Fox Tree (UCSC)

Sunday, May 20th
Buchanan A201

Introduction
People commonly produce quotations with the devices like, said, and goes when reporting another person’s speech, and to varying degrees, when describing non-speech or non-human sounds

‘He was like “…..”’

‘The cows were like “mooo”’

‘The babies were going “ga ga ga”’

These quotations are often coordinated with imitative bodily gestures (Streek & Knapp, 1992) and might also involve some amount of vocal imitation of the speech or other sound (Clark & Gerrig, 1990). Past research has shown that the particular device that is used can be selected based on a variety of factors, such as whether the quoted information displayed attitudes or emotions (suggesting like, Buchstaller, 2001), contained sounds (suggesting goes, Butters, 1980), and whether the person quoted or addressed was of a higher status (suggesting said, Blackwell & Fox Tree, manuscript).

However, given how imitation is often incorporated into quotation, we wondered whether the choice of quotative device might also be associated with the amount of vocal and gestural imitation enacted by the speaker. Such an association would link the planning of words and grammar (i.e. relating to how the quotation is to be constructed) to imitative processes like those involved in iconic gesture.

Methods
Fifty-two undergraduates at the University of California, Santa Cruz, participated in the experiment in exchange for course credit. Participants watched a series of videos, and were asked to retell the videos in an entertaining way were recorded as they offered spontaneous, quotation-rich descriptions of thirteen video clips to a confederate addressee. The clips contained a variety of sounds including human speech, animals producing human-like speech, animal vocalizations, and inanimate noises. Analysis focused on video clips in which participants used goes, likes and saids to report the same content, allowing us to evaluate imitativeness in association with device independently of the content being quoted. This included two clips: a dog howling what sounded like the words ‘I love you,’ and a talking parrot that produced imitations of human speech and the roar of a tiger.

Vocal and gestural ratings demonstrate that when like and goes were used, the quoted content was associated with significantly more vocal and gestural imitation than with said. The mean rating for vocal imitation for goes was 4.6 (SD =.72), for like was 4.7 (SD = .70), and for said was 2.00 (SD = . 38), F(2, 36) = 68.2, p < .001. The mean rating for gesturing for likes was 2.7 (SD =.79), for goes was 2.74 (SD =1.17), and for saids was 1.84 (SD = .79). These results suggest that devices like and goes are used to highlight the imitative or demonstrative aspects of the subsequent quotation, a point that is further supported by the positive association between gesture and vocal imitativeness, r(93) = .363, p < .001. Thus, quotation may be aptly treated as a multimodal construction that serves to combine linguistic structures with the iconic processes of gesture. In this light, quotation might be compared to certain constructions in sign languages, like classifier constructions, that similarly integrate linguistic structures with iconicity (Liddell, 2003). Both cases point to the broader consideration of language as a multimodal system of communication that makes use of both arbitrary (i.e., traditionally “linguistic”) and iconic elements (Permiss, Thompson, & Vigliocco, 2010).

References

Blackwell, N., & Fox Tree, J. E. (2011). Quoting with Said. Unpublished manuscript.

Buchstaller, I. (2001). An alternative view of LIKE: its grammaticalisation in conversational American English and beyond. Edinburgh Working Papers in Applied Linguistics, 11, 21– 41. Butters, R. (1980). Narrative go “say”. American Speech, 55, 304–307.

Lidell, S. (2003). Grammar, Gestures, and Meaning in American Sign Language. Cambridge: Cambridge University Press.

Permiss, P., Thompson, R. L., &∓ Vigliocco, G. (2010). Iconicity as a general property of language: Evidence from spoken and signed languages. Frontiers in Language Sciences, 1, 227-241.

Streeck, J., & Knapp, M. L. (1992). The interaction of visual and verbal features in human Communication. In: F. Poyatos (Ed.). Non-verbal Communication. Amsterdam: John Benjamins.

Download PDF

∓

Melissa Zheng & Robert F. Williams (Lawrence)

Sunday, May 20
Buchanan B213

Download PDF

Kasper Kok (U Amsterdam) & Alan Cienki (VU, Amsterdam)

Sunday, May 20
Buchanan B213

Recent, embodied theories of meaning known as ‘simulation semantics’ posit that language comprehension is accomplished via mental simulation (Barsalou, 1999; Kaschak & Glenberg, 2000). However the meaning of ‘comprehension’ varies according to the researcher’s perspective on interpersonal communication. Influential models in social psychology have emphasized that in realistic communication, comprehension not only involves a process of decoding information, but also involves exchange of intentions and perspective-taking during collaborative interaction (Krauss & Fussel, 1996).

In this paper, we argue that the way comprehension is characterized in experimental simulation semantics is rather narrow, as it largely ignores these social aspects of communication. Despite providing an innovative approach to the study of meaning, simulation-based theories of comprehension draw upon traditional encoding-decoding paradigms in the sense that they assume that the meaning of a sentence is taken to be inherently carried by the sentence itself, i.e. meaning is separated from the communicator’s intentions. We propose four hypotheses that need to be tested in order to assess the extent to which simulation semantics is a fruitful framework for studying comprehension in realistic face-to-face communicative scenarios.

First we consider the role of mental simulation in language production. Based on McNeill’s (1992) and others’ suggestion that mental representation underlying language production has a partly imagistic character, we hypothesize that meaning representation in language production and in language comprehension have a similar embodied nature. In light of this, we argue that comprehension in interpersonal communication can be viewed in terms of the degree to which interlocutors’ simulations of some content are sufficiently matched for the current purpose. The literature from simulation semantics helps make explicit predictions as to what it means for interlocutors’ mental representations to be homologous.

Second, we propose two conditions for integrating simulation semantics with a ‘dialogic perspective’ on communication, a paradigm in social psychology that takes comprehension to be the product of a collaborative process of establishing shared conceptualizations (Rommetveit, 1983). In order to define dialogical comprehension in terms of mental simulations, it must be supposed that simulations are (1) dynamically co-constructed during conversation, and (2) endowed with a ‘confidence dimension’: interlocutors retain a level of certainty of having synchronous representations of the expressed information. These suppositions make it feasible that interactive alignment and feedback mechanisms help to establish mutual confidence of having coordinated simulations.

Third, we consider how simulation semantics relates to the ‘intentionalist’ view of comprehension, taking communication as the exchange of intentions (vid. Relevance Theory [Sperber & Wilson, 1986]). Recent research (Gallese, 2007) suggests neural overlap between systems responsible for attributing intentions to others and those responsible for mental imagery. Although the precise relation between understanding linguistically presented information and understanding the underlying intentions remains underspecified, converging evidence suggests that these capacities rely on strongly intertwined embodied mechanisms.

Finally, taking the multimodal character of spoken language use into account, we argue that simulation semantics calls for an account of the differential ways in which verbal, nonverbal and co-verbal signals contribute to simulation of content and intent.

By taking these four issues into account, simulation-based theories may more closely approach a complete account of language comprehension.

Download PDF