Ideophones at the intersection of theory and methods in African linguistics

Plenary address at the 43rd Annual Conference on African Linguistics, New Orleans, March 2012.

Ideophones, marked words that depict sensory imagery, are found in many of the world’s languages, but they are especially prevalent in sub-Saharan Africa. The first recognition of their prominence in African languages dates back to some of the earliest descriptive grammatical work in West-Africa (e.g. Schlegel, Vidal, Koelle, Junod). They were defined as a word class by the Bantuist Doke, and their study has since been “part of the Africanist subtradition in linguistics” (Tedlock 1999). What can we learn from the prominence and ubiquity of ideophones in African languages? In this lecture I focus on some matters that are crucial to understanding ideophones, and, I argue, language at large. The data come from Siwu, a Kwa (GTM) language of Ghana; methods range from traditional pencil and paper fieldwork to video recordings of natural conversation and from stimulus-based elicitation to the collection of folk definitions. An underlying concern is to register some of the challenges that ideophones have posed to scientists of language, and to explore ways in which they can contribute to the ongoing development of theory and methods in (African) linguistics.

The starting point is the relation between language and perception. Perception is our point of contact with the outside world; language is our way of making social sense of it. It is significant then that languages may differ quite radically in the resources they provide for talking about perception; and striking that so many African languages feature large word classes dedicated to evoking sensory imagery. Although the sensory meanings of ideophones have often been regarded as elusive, methods from cognitive anthropology and psychology help chart the lexical domain of ideophones and bring to light the existence of fine-grained semantic-perceptual categories such as in-mouth sensation, surface appearance, firmness, granularity, texture, and spatial extent. Gesture is another important source of information. I show how the iconic gestures that often accompany ideophones help elucidate their meanings; and I argue that the tight coupling of ideophones and gesture is further evidence for the fact that their meanings lie in the domain of sensory imagery. I explore the implications of these points for the language-perception interface.

Ideophones have often been eschewed as exotic words with elusive meanings. I hope to show that once we rise to the challenge by bringing in new methods and adopting a holistic perspective on language, we find that ideophones shed light on many themes that are crucial to our understanding of language. This is what the study of ideophones can contribute to African linguistics in the 21st century: fostering a resolutely interdisciplinary and multimethodological approach to linguistic science, and refreshing our views of what is possible and probable in human language.

Transcript

1. Ideophones at the intersection of theory and methods in African linguistics Mark Dingemanse 43rd Annual Conference on African Linguistics Max Planck Institute for Psycholinguistics, Nijmegen March 2012, Tulane University, New Orleans
2. Ideophones
3. sinisinisinisini it holds itself together ‘closely woven’
4. kɛlɛnkɛlɛn it shines ‘glittery’
5. marked words that depict sensory imagery Dingemanse 2011 “perform” “enact”
6. “the lunatic fringe of language” Frankis 1991 “the lexicographer’s worst nightmare” De Schryver 2009 “a step-child of modern linguistic science” Voeltz & Kilian-Hatz 2001
7. “Ideophones can be highly variable and difficult to elicit away from the field … They do not regularly show up in written texts … In traditional elicitation sessions, native speakers will have great difficulty in explaining the meaning of an ideophone…” Tucker Childs, 1994
8. traditional “pencil and paper” work
9. specially designed field tasks
10. recordings of everyday social life
11. Language and perception
12. fieldwork in the ideophonic heartland of West-Africa on Siwu, spoken by 12,000 people in Kawu, Ghana na-Togo < GTM < Kwa (Heine 1968, Blench 2006)
13. kàmɛ̃rɛ̃mɛ̃rɛ̃í ìpɔ̀fa Monellin and thaumatin are two intensely sweet proteins extracted from African berries. These proteins are ~ 100 000 times sweeter than sugar on a molar basis and several thousand times sweeter on a weight basis. (Kim et al., Protein Engineering, 1989) Thaumatococcus Danielii, Katemfe or “miracle fruit” (Daniell 1855) Dioscoreophyllum cumminsii or “serendipity berry” (Inglett & May 1968)
14. nyɛ̃kɛ̃nyɛ̃kɛ̃ “intensely sweet” language & the environment are closely intertwined ideophones target very specific sensory perceptions
15. three problems 1. comparability 2. capturing elusive meanings 3. analysts’ vs. speakers’ intuitions
16. problem 1 comparability “We have hardly begun to study the classification of human experience revealed by ideophones” William Samarin, 1965
17. 6 tasks, 130+ stimulus items Majid & Levinson 2007, see fieldmanuals.mpi.nl Stimulus-based elicitation tasks
18. example touch task item #1, piece of felt 1. wùrùfùù ‘fluffy’ 5x 2. wòsòròò ‘rough’ 4x 3. fũɛ̃fũɛ̃ ‘malleable’ 3x ì-se wòsòròò gàke wùrùfùù anaà it-be IDPH.rough but IDPH.malleable too ‘it is wòsòròò [smooth] but also wùrùfùù [fluffy]’
19. Results for 14 participants for all six tasks Dingemanse 2011
20. 1. ideophones can be elicited 2. stimulus-based elicitation gives us comparability
21. problem 2 capturing elusive meanings “Ideophones are not like normal words to which meanings are readily assigned” Isidore Okpewho, 1992
22. folk definitions video-recorded informal explanations in the language under study Dingemanse 2010 60 ideophones explained in Siwu data from 4 different speakers (m/f)
23. gestures example sentences antonyms and synonyms cultural background information
24. example gìlìgìlì vs. minimini ‘roundness’ Video clips at http://thesis.ideophone.org/part-3/folk-definitions
25. gìlìgìlì ‘circular’ (round, 2D) minimini ‘spherical (round, 3D)
26. 1. gesture sheds light on the imagistic meanings of ideophones 2. offers crucial evidence in many areas information structure motion & manner verb semantics serial verb constructions & event representation semantics & pragmatics of demonstratives
27. . problem 3 analysts’ vs. speakers’ intuitions “Our knowledge of ideophones ... is in part defective because the choice to record them reflects the worldview of the compiler” Roger Blench, 2010
28. classifying the meanings of ideophones 1. Doke 1935: “manner, colour, sound, smell, action, state or intensity” ? why these 7 domains? (cf. Westermann 1927, “all sensory perceptions”) 2. Alexandre 1966: “auditive, visual, tactile, gustative, and olfactive” (i.e. the five senses) ? is this Western folk model of perception applicable? (sensory science finds a much larger taxonomy, e.g. Moller 2002) 3. Samarin 1966: “appearance, arrangement, emotion, measure, motion, odour, quality, shape, sound, state, taste, temperature, time, touch, and weight” ? are these English distinctions universal? (cf. Malt et al. 1999, 2011 on words vs. concepts)
29. capturing speakers’ intuitions with a sorting task Miller 1969, Weller & Romney 1988, Bernard 2006
31. surface appearance taste sound firmness granularity smoothness
32. ideophones & sensory science: the case of touch Dingemanse & Majid 2012 salient dimensions found in psychophysics research rough-smooth, hard-soft, springiness, sticky-slipperiness, and firmness (Bhushan et al. 1997) salient dimensions identified through the sorting task rough-smooth, hard-soft, malleability, slipperiness, firmness features of an industry-developed touch lexicon abstract (not source-based) items, uniformity in linguistic resources (Guest et al. 2010) design features of ideophones abstract (not source-based) items, usually a fairly uniform lexical class
33. 1. sorting tasks empirically capture native speaker intuitions 2. shed light on the organisation of the mental lexicon 3. reveal that ideophones may rival industry-designed sensory lexica
34. summary so far 0. random forest outings for serendipity 1. elicitation tasks for systematic sampling 2. folk defs & gesture for insight into semantics 3. sorting tasks for capturing native intuitions
35. Using ideophones
36. “Ideophones are part of an informal language register, and their function is to dramatize a narration.” Kilian-Hatz 2001 (and Burbridge 1938, Doke 1948, Jungraithmayr 1983, etc.)
37. problem 4 our dataset is skewed
38. A corpus of naturally occurring everyday speech 3000+ utterances across 30+ episodes 1 in 12 utterances contains an ideophone
39. assessment sequences Pomerantz 1984, Heritage & Raymond 2005 A T's tsuh beautiful day out isn't it? (JS:II:28) B Yeh it's jus' gorgeous... Assessment sequences involve the management of knowledge and information: who knows what, who has access to the thing assessed, who agrees with whom, etc.
40. Tetteh: a neighbour Komla: the gunpowder expert T When you sieve the gunpowder, it’ll be kɛlɛnkɛlɛnkɛlɛnkɛlɛnkɛlɛnkɛlɛnkɛlɛn! You’ll see it for yourself. K It’ll go gelegelegele! T Indeed. It’ll shine just like that.
41. ideophones are appeals to direct experience 1. good for emphasizing shared experience (e.g. communal stories, co-constructive) 2. good for stressing epistemic authority (e.g. expert knowledge, competitive) Dingemanse 2011b
42. 1. some functions of ideophones can only be seen if you look beyond narrative 2. this invites comparison to other devices for epistemics in interaction • evidentials (Chafe & Nichols 1986, Aikhenvald 2004) • reported speech (Holt 1996; Clift 2006)
43. Conclusions
44. ideophones challenge us to understand African sensory worlds to never lose sight of visual language to study everyday language use
45. Ideophones “illuminate things so that people will see and learn.” (àkparara ara sɔ máanyà nɛ màɖi ite) Ruben Owiafe
46. I thank the Mawu people the ACAL organizers European Research Council Max PIanck Society for the Advancement of Science email mark.dingemanse@mpi.nl visit thesis.ideophone.org
47. references and other supplementary materials at thesis.ideophone.org