Ideophones —vivid sensory words found in many of the world’s languages— are often described as having little or no morphosyntax. That simple statement conceals an interesting puzzle. It is not often that we can define a word class across languages in terms of its syntax (or lack thereof). After all, most major types of word classes show intriguing patterns of cross-linguistic variation. There is no particular reason to expect that the morphosyntactic position or degree of embedding of, say, noun-like or verb-like words will be similar across unrelated languages. Indeed that is why typologists define comparative concepts primarily by reference to semantic rather than grammatical or morphosyntactic properties (Croft 2003; Haspelmath 2007). Continue reading
We have a new paper out. It’s actually been available since February in an online-first version, but for those of us who love page numbers and dead trees, the journal has now printed it in its August issue on pages 1274-1281. Citation:
Lockwood, G., Dingemanse, M., & Hagoort, P. (2016). Sound-symbolism boosts novel word learning (PDF). Journal of Experimental Psychology: Learning, Memory, and Cognition, 42(8), 1274-1281. doi:10.1037/xlm0000235
This is another one for which we’ve made available the stimuli —word lists and sound files— through OSF, contributing to our mission to make research from our lab replicable. Also, we have since replicated the results in a follow-up study where we also took EEG and individual difference measures.
I provide a quick summary in 2×3 points below. For a write-up that’s much more fun and has great illustrations, check out Gwilym Lockwood’s Sound symbolism boosts novel word learning: the MS Paint version. Continue reading
Dingemanse, Mark, Will Schuerman, Eva Reinisch, Sylvia Tufvesson, and Holger Mitterer. 2016. “What Sound Symbolism Can and Cannot Do: Testing the Iconicity of Ideophones from Five Languages.” Language 92 (2): e117–33. doi:10.1353/lan.2016.0034
The basic finding is this: people are sensitive to the meaning of ideophones they’ve never heard, even when they are produced out of context by a computer voice in a difficult forced choice task. Yet there is also reason for caution: the effect is not nearly as strong as what people have found for pseudowords like bouba and kiki.
As we note in the introduction, “there appears to be a tendency to either underplay or exaggerate the significance of iconicity in the study of language and mind”. In this paper we chart a middle way between these extremes. Here’s a quick summary in 3×3 points:
What we did:
- Sound symbolism (iconicity in spoken language) is usually studied using hand-crafted pseudowords in binary forced choice experiments (think bouba and kiki, as reviewed here), but there are three problems with such experimental designs: (i) they run the risk of inflating effect sizes, (ii) it is unclear how they relate to natural languages, and (iii) they usually don’t control for prosody.
- We designed a study to tackle these problems by (i) adjusting the binary choice task to be more realistic and harder, (ii) using real words and meanings from natural languages, and (iii) teasing apart prosody and segmental features. Essentially, we bring linguistic insights to bear on the psychological study of sound symbolism.
- We take 203 ideophones —lexical sound-symbolic words— from 5 languages and 5 semantic domains and present them to 80 participants in 4 versions: (i) full original recording, (ii) full speech synthesized version, (iii) prosody-only condition and (iv) phonemes-only condition. The versions help us control for variation due to different speakers and help us examine the contributions of prosody and segmental features.
What we found:
- People can choose the correct translation of ideophones at a level significantly above chance. So ideophones in Japanese, Korean, Semai, Ewe and Siwu are not fully arbitrary, as is normally assumed of words; they contain iconic cues that even people who don’t speak these languages can pick up.
- Sound ideophones are easiest to guess, but the other semantic domains (movement, texture, color/visual appearance, and shape) come out significantly above chance as well. However, the effect is much more modest than most bouba/kiki studies: in the best versions, people score 57.2% on average (where 50% would be chance level) — quite different from the 95% that has sometimes been claimed for pseudoword studies.
- Performance for the original and resynthesised stimuli is indistinguishable, so our speech synthesis method works. Performance is significantly better for the full versions (i-ii) than for the reduced versions (iii-iv), so both prosody and phonemes contribute to the effect (and neither alone is sufficient).
What we conclude:
- Findings based on pseudowords like bouba/kiki cannot be automatically translated into claims about natural languages. Ideophones combine iconicity and arbitrariness, and lexical iconicity in ideophones is best characterised as a weak bias, which is supported by multimodal performances in actual use and which may be amplified in cultural evolution (cf our TiCS paper).
- Prosody is just as important as segmental information in supporting iconic interpretations (as predicted here). Prior work, which has rarely controlled for prosody, likely overestimates the role of speech sounds and underestimates the role of intonation, duration and other prosodic cues.
- Speech synthesis offers a viable way to achieve experimental control in the study of sound symbolism. To stimulate its wider use we’re making available all stimulus materials, including the diphone synthesis source files we used to create them. Get them at MUSE or OSF.
Sound symbolism is a phenomenon with broad relevance to the study of language and mind, but there has been a disconnect between its investigations in linguistics and psychology. This study tests the sound-symbolic potential of ideophones—words described as iconic—in an experimental task that improves over prior work in terms of ecological validity and experimental control. We presented 203 ideophones from five languages to eighty-two Dutch listeners in a binary-choice task, in four versions: original recording, full diphone resynthesis, segments-only resynthesis, and prosody-only resynthesis. Listeners guessed the meaning of all four versions above chance, confirming the iconicity of ideophones and showing the viability of speech synthesis as a way of controlling for segmental and suprasegmental properties in experimental studies of sound symbolism. The success rate was more modest than prior studies using pseudowords like bouba/kiki, implying that assumptions based on such words cannot simply be transferred to natural languages. Prosody and segments together drive the effect: neither alone is sufficient, showing that segments and prosody work together as cues supporting iconic interpretations. The findings cast doubt on attempts to ascribe iconic meanings to segments alone and support a view of ideophones as words that combine arbitrariness and iconicity. We discuss the implications for theory and methods in the empirical study of sound symbolism and iconicity.
Just out in Trends in Cognitive Sciences: a review paper by yours truly with Damián Blasi, Gary Lupyan, Morten Christiansen and Padraic Monaghan. It is titled “Arbitrariness, iconicity and systematicity in language”. You can download it here (PDF). Here is a simple summary:
An important principle in linguistics is that words show no predictable relation between their form and their meaning (arbitrariness). Yet this principle does not have exclusive reign. Some words have forms that suggest aspects of their meaning (iconicity). Some groups of words have subtle statistical properties that give away something about their grammatical function (systematicity). To fully explain how words work, we need to recognise that the principle of arbitrariness is not the whole story, and that words can additionally show degrees of iconicity and systematicity.
Here is are six take-away points:
- Often, arbitrariness is thought to be not just necessary but also sufficient to explain how words work. We show this is not the case: non-arbitrary patterns in language are more common than assumed, and they have implications for how we learn, process and use language.
- Often, arbitrariness and iconicity are pitted against each other. We show this is an oversimplification: iconic words have a degree of arbitrariness and the two do not exclude each other.
- Often, the role of iconicity in language is thought to be minimal. We show that it can differ dramatically across languages and also varies as a function of meaning and modality (e.g. signed or spoken).
- Sometimes, iconicity and systematicity have been confused. We show that distinguishing them helps us to better understand vocabulary structure, by showing why we may expect iconicity to show certain universal patterns while systematicity allows more language-specific patterns.
- Sometimes, we may forget that words are not abstract ideas but tools that have their own history. We argue that the way words are learned and used influences their form, and that this may help explain how arbitrariness, iconicity and systematicity pattern the way they do.
- Sometimes, language scientists make far-reaching claims based on studying a small portion of the vocabulary, or a small number of (typically Western) languages. We argue that we can get a better picture of language by looking at a wider range of evidence.
Charles Hockett had interesting views on the relation between iconicity and arbitrariness. Here is a key quote:
The difference of dimensionality means that signages can be iconic to an extent to which languages cannot; and they are, even though, as Frishberg (1975) tells us, the trend in Ameslan for over a century has been towards more and more conventionalization.
Now, while arbitrariness has its points (see, e.g., Hockett 1960a, p. 212), it also has drawbacks (Hewes, ANYAS, p. 495), so that it is perhaps more revealing to put the difference the other way around, as a limitation of spoken languages.
Indeed, the dimensionality of signing is that of life itself, and it would be stupid not to resort to picturing, pantomiming, or pointing whenever convenient. (Even when speaking we do this: for example, we utter a demonstrative such as there, which indicates relative distance but not direction, and supplement it by a pointing gesture that indicates direction but not distance.)
But when a representation of some fourdimensional hunk of life has to be compressed into the single dimension of speech, most iconicity is necessarily squeezed out. In one-dimensional projection, an elephant is indistinguishable from a woodshed. Speech perforce is largely arbitrary; if we speakers take pride in that, it is because in 50,000 years or so of talking we have learned to make a virtue of necessity (cf. Hill 1972, pp. 313-15).
Linearity means that single devices must serve multiple functions, whereupon structural ambiguity becomes par for the course (see C. R. Peters, Origins, pp. 83-102). We hear Carbon fourteen, Strontium ninety; out of context, we do not know whether this is mention of two radioactive isotopes, or a roadside marker giving the distance to two towns on the road ahead, or the final score in the game between Carbon Free Academy and Strontium Senior High.
It is such ambiguities, forced by limited dimensionality if by nothing else, that have given rise to the- notion of “surface versus deep structure,” which Stokoe evokes for the remark of which the present paragraph is an expanded paraphrase-his trenchant observation (ANYAS, p. 510) that in sign, as over against speech, “surface and depth more nearly coincide.” (pp. 264-5)
Hockett, C. F. 1978. “In Search of Jove’s Brow.” American Speech 53 (4): 243–313. doi:10.2307/455140.
African ideophones and their contribution to linguistics
Africa’s linguistic diversity has impacted the study of language in many ways. The articulatory phonetics of the Khoi and San languages prompted methodological innovations in phonetics, the tonal systems of West-African languages spurred the development of autosegmental phonology, and the ornate morphology of Bantu prompted syntacticians to reconsider the balance between transformational rules and lexical elaboration. In this workshop we consider how the study of ideophones can contribute to theory and methods in linguistics.
Ideophones (also known as mimetics or expressives) are marked words that depict sensory imagery. A major word class in many African languages, they are somewhat of an inconvenient truth for the dogma that spoken languages rarely feature iconicity in the lexicon. Their phonology is marked in a way that bears a clear relation to the broader phonological system of the language, providing for a unique window into phonological structure. Their prosody and morphosyntax set them apart as special words, yet they are more deeply integrated in linguistic subsystems than is often assumed, raising interesting questions about what is in and outside grammar. Their meanings are rich and imagistic, providing unparalleled ways to talk about sensory perceptions. All of these properties represent areas where ideophones can shed light on the design features of language, the iconic affordances of speech, and the nature of human communicative competence.
This workshop gathers international experts to present recent research on ideophones and to put recent developments into theoretical context. Submissions are expected to focus on the connection between ideophone research and foundational issues in linguistics, from phonology to prosody and from syntax to meaning. We encourage papers that show how new approaches can shed light on old questions, and how the systematic study of ideophones can contribute new insights to our understanding of the structure of language and languages. One and a half centuries after the earliest descriptions of ideophones in African languages, the 8th World Congress of African Linguistics in Kyoto offers a unique chance to take stock of what we have learned so far from ideophones, and to explore ways to integrate this knowledge into the broader language sciences.
Deadline for abstract submission: October 31, 2014
Notification of acceptance: December 1, 21014
Conference: August 21-24, Kyoto
Abstracts should follow the general guidelines established for the submission of abstracts for WOCAL 8, which can be found here: is.gd/wocal8abstracts
Where moderation is not utterly overstepped, the wealth of sound in languages can be compared to coloration in painting. The impression of both evokes a similar feeling; and even thought reacts differently if, like a mere outline, it emerges in greater nakedness, or appears, if we may so put it, more coloured by language.
Wilhelm von Humboldt, On language, page 80. Originally published in 1836.
When you hear the word dog, you understand it because you have learned that meaningless individual sounds mean dog when arranged in a specific order into a word – it’s not like d means “fluffy”, o means “four legs”, and g means “enjoys rolling in smelly things”. The sound of the word dog is unrelated to the thing it means – it just so happens that the combination of d, o, and g in English mean dog. The same concept in other languages is expressed with very different sounds; hond in Dutch, inu in Japanese, sobaka in Russian. The idea that the individual sounds which make up words are unrelated to the meaning of the word that those sounds express is called arbitrariness in linguistics.
Arbitrariness seems to make sense; if the sounds of language did have specific meanings, if the sounds of words were related to the things they mean, then surely all languages would sound quite similar. Given that they rather obviously do not (for example, Hawaiian has only eight consonants, while Georgian has 28, and !Xóõ, spoken in Botswana, has at least 58), it is safe to say that there is little or no relation between sound and meaning.
However, researchers have recently started to investigate this assumption. Several languages around the world use sound symbolic words called ideophones, which are used to talk about sensory imagery. Interestingly, these words seem to be directly related to their meaning (i.e. the sounds of the words are symbolic of their meaning), and even more interestingly, there seems to be something universal about these words – several experiments have shown that people who don’t speak these languages can still understand (or accurately guess) the meanings of ideophones.
We can try that out now. See if you can guess the meanings of these Japanese ideophones (answers below):
- nurunuru – dry or slimy?
- pikapika – bright or dark?
- wakuwaku – excited or bored?
- iraira – happy or angry?
- guzuguzu – moving quickly or moving slowly?
- kurukuru – spinning around or moving up and down?
- kosokoso – walking quietly or walking loudly?
- gochagocha – tidy or messy?
- garagara – crowded or empty?
- tsurutsuru – smooth or rough?
- nurunuru – slimy
- pikapika – bright
- wakuwaku – excited
- iraira – angry
- guzuguzu – moving slowly
- kurukuru – spinning around
- kosokoso – walking quietly
- gochagocha – messy
- garagara – empty
- tsurutsuru – smooth
Did you guess the meanings of these words better than you would expect? Unlike the word dog, it seems that the individual sounds in these words actually do contribute to the meaning of the words, and this is called sound symbolism. Sound symbolism is the opposite of arbitrariness, but the two can coexist perfectly happily within language.
Speakers of languages with sound symbolic ideophones, such as Japanese, often talk about how the ideophones create a very vivid image or feeling in their minds, whereas normal words don’t. When a Japanese person hears the word kirakira, meaning sparkly, it is like they can actually see the thing that is sparkly. How sound symbolism works, however, is not quite clear, and there have not yet been many neuroscience studies on it, but the research so far suggests that hearing sound symbolic words might involve other forms of sensory perception in a similar way to how people with synaesthesia associate colours to letters. My research at MPI is to investigate why certain sounds appear to be related to certain meanings across languages and how the brain processes these sounds.
Words evolve not as blobs of ink on paper but in face to face interaction. The nature of language as fundamentally interactive and multimodal is shown by the study of ideophones, vivid sensory words that thrive in conversations around the world. The ways in which these “Lautbilder” enable precise communication about sensory knowledge has now for the first time been studied in detail. It turns out that we can paint with language, and that the onomatopoeia we sometimes classify as childish might be a subset of a much richer toolkit for depiction in speech, available to us all and in common use around the globe.
[Written for the Max Planck Jahrbuch 2013. A German version, translated by Gunter Senft, appeared here and in Scinexx Magazine.]
Few professions should be more familiar with the nature of words than academia. Words are the currency of our trade. They record our cumulative progress, and they measure our productivity as we disperse our ideas through articles and books. How easy is it to fall in love with the printed word, black symbols on a white page, tidy spaces separating units of thought like stars dotting the skies of conceptual clarity!
How different are our everyday ways with words. We roll them on our tongues as we speak, whisper, or exclaim. They become pliable as we perform, prolong, and repeat them. We colour them with fine shades of meaning as we exert control over pitch, intensity, and duration. We masterfully integrate them with gestures and facial expressions into what linguists call composite utterances (Enfield 2009). For a long time all of these things were cordoned off as paralanguage: not the real thing, but a side show detracting us from the incisiveness of an idealised formal language. In the philosopher Frege’s musings on language, aesthetic delight and the striving for truth are in direct opposition.
However, this view is fast becoming outdated as linguists increasingly realise that the written word is a poor model for our true communicative competence. Language evolved in a much richer environment, and it has always done more for us than just deliver disembodied information. We use language to build social relations, to relate our experiences and to express our stances. We do not just inform, we also perform. This calls for a renewed study of how words work.
Often the best way to develop a fresh view of one’s field of study is to radically change one’s starting point. Linguists at the Max Planck Institute for Psycholinguistics do this by collecting new data on languages far afield. In the Language & Cognition Department, researchers carry out sustained fieldwork in over twenty locations across the globe. Over the last few years, there have been several in-depth studies of a type of words known as ideophones. Characterised as “vocal images”, these words are probably the best approximations of painting in speech. They were long seen as exotic, out-of-the-way words, but new research shows that they are ubiquitous in conversations across the globe, and used in unforeseen ways.
Ideophones are words whose form is suggestive of their meaning. Familiar examples include English kerplop and boom or German holterdipolter and tick-tack. But whereas in European languages, these words tend to be small in number and limited mainly to imitating sound, many of the world’s languages have hundreds or even thousands of ideophones, covering a much broader range of sensory meanings. Take tuŋjil-tuŋjil ‘bobbing, floating’, ulakpulak ‘unbalanced scary appearance’ and c’onc’on ‘woven tightly’ from Korean; or dhdŋɔh ‘appearance of nodding constantly’, praduk pradɛk ‘noises of scattered small drops of rain’ and grɛ:p ‘crispy sound’ from Semai, a language of peninsular Malaysia; or mukumuku ‘mumbling mouth movements’, fũɛ̃fũɛ̃ ‘elastic, flexible’ and kpɔtɔrɔ-kpɔtɔrɔ ‘walking like a tortoise’ from Siwu, a language of eastern Ghana (Dingemanse 2012).
When studying video recordings of conversations in such languages, one sees that these words, with their peculiar forms and colourful meanings, are not spoken like ordinary words. They are delivered as performances. They bring to life the events in ways that ordinary words never do. Observing them in use, we get a sense of what the linguist and psychologist Karl Bühler meant when he wrote, “If there were to be a vote on who is more richly equipped with resources, the painter with colours or the painter with the voice, I would not hesitate to give the second my vote” (Bühler 1934).
How can we paint with speech? In the ideophone systems of the world’s languages, there are three basic ways in which speech is used to depict sensory imagery — three types of iconicity (Dingemanse 2012). The first is to imitate sound with sound, as in English boom ‘sound of explosion’. This is called direct iconicity. It is the most simple way, but also the most limited. After all, not all events involve sound. What all events do have is internal temporal structure. This is where the second method comes in: the structure of words may resemble the structure of events. Bühler recognised this when he noted that words may be “Gestalt faithful” to the events they represent. Therefore we call this type Gestalt iconicity. For instance, words can be prolonged to evoke duration, closed syllables can evoke end points, and repeated syllables can evoke repetition — as in several of the examples given above. Third, and finally, sometimes similar words are used for similar events. Take the three Semai words grɛ:p ‘of chewing fruit’, gra:p ‘of chewing crisps’, grɨ:p ‘of chewing cassava’. They share a common template gr_p which we can characterise as ‘crispy sound’ (Tufvesson 2011). Because related words map onto related meanings, we call this type relative iconicity. Together, these three ways of suggesting meaning are part of the word painter’s toolkit. They allow depictive words like ideophones to form perceptual analogies of events.
But how does one know that some stretch of speech is intended as an ideophone —a vocal image— rather than an ordinary word? Comparative research shows that languages converge on remarkably similar ways to do this. For instance, across languages, ideophones sound special because they exhibit certain liberties relative to other words. They have a wider range of possible syllable structures and word forms, and they are remarkably susceptible to playful word formation processes like reduplication and lengthening. In spoken utterances, they stand out because they show a great measure of syntactic independence and they tend to be delivered as an intonation unit of their own. Last but not least, in many languages, ideophones are introduced by quotative markers or “say” or “do” verbs, emphasizing their performative nature. All of these features work together to mark ideophones as depictions, much as the frame around a painting tells us to interpret it as a painting and not as the wallpaper.
Researchers of the Language & Cognition Department are studying ideophones in a number of fieldsites. They use specially designed stimulus materials to study how these words encode sensory perceptions (Majid and Levinson 2011). They make high quality video and audio recordings of everyday conversations to understand how people use these words in face to face interaction — the kind of setting in which language evolved and keeps evolving. In the course of this research it has become clear that ideophones are far from the stylistic flourishes that people once took them to be. They are dedicated sensory words, on a par with industry-designed sensory vocabularies in terms of their structure and coverage of sensory spaces (Dingemanse and Majid 2012). They are used to communicate expert knowledge during joint work and to share and interpret experiences in storytelling. In languages which have thousands of ideophones, they are seen as the ultimate sign of eloquence.
Every challenge represents an opportunity. Ideophones challenge us to innovate theory and methods, encouraging us to move away from a view of language limited by ideologies or traditions to a perspective that is informed by as wide a range of data as possible. In the end, it should come as no great surprise that depiction may be just as important as description in linguistic life. Even our own findings are not communicated just in abstract words. We illustrate them with gestures and visualise them in figures and diagrams. There is truth, then, to the saying that an image says more than a thousand words. But what we may have overlooked was that our words have been peppered with images all along.
- Bühler, Karl. 1934. Sprachtheorie: Die Darstellungsfunktion der Sprache. Jena: G. Fischer.
- Dingemanse, Mark. 2012. “Advances in the cross-linguistic study of ideophones.” Language and Linguistics Compass 6 (10): 654–672. doi:10.1002/lnc3.361.
- Dingemanse, Mark, and Asifa Majid. 2012. The semantic structure of sensory vocabulary in an African language. In Proceedings of the 34th Annual Conference of the Cognitive Science Society, ed. N. Miyake, D. Peebles, and R. P. Cooper, 300–305. Austin, TX: Cognitive Science Society.
- Enfield, N. J. 2009. The Anatomy of Meaning: Speech, Gesture, and Composite Utterances. Cambridge: Cambridge University Press.
Majid, Asifa, and Stephen C. Levinson. 2011. “The Senses in Language and Culture.” The Senses and Society 6 (1): 5–18.
- Tufvesson, Sylvia. 2011. “Analogy-making in the Semai Sensory World.” The Senses and Society 6 (1): 86–95. doi:10.2752/174589311X12893982233876.
Just a heads-up to let interested readers know of a newish article on the morphosyntactic typology of ideophones by yours truly: Expressiveness and system integration. On the typology of ideophones, with special reference to Siwu (PDF). Completed in May 2012, it has been peer reviewed and accepted, and is due to appear in a special issue of Language Typology and Universals, though the special issue editor tells me that it may, regrettably, take another while before it actually comes out.
Anyway because this is now being referred to in numerous places I have decided to make the pre-preprint available here. The basic approach is to exploit corpus data on the morphosyntactic variation of ideophones within a language to shed light on some larger questions in the morphosyntactic typology of ideophones. Some of the proposals of possible interest to typologists include the following:
- an inverse correlation between morphosyntactic integration and various expressive features (the more syntactically independent an ideophone is, the more susceptible it is to the typical processes of expressive morphology and prosodic foregrounding)
- a functional explanation of the inverse correlation (ideophones are prototypically syntactically independent to help signal their status as depictions of sensory imagery; this also explains their common occurrence at utterance edge)
- a generalisation about ideophone morphosyntax in relation to frequency (higher frequency ideophones tend to be easier to integrate into morphosyntax, a Zipfian effect that may have to do with the erosive role of frequency)
- a prediction with regard to the areal diffusion of ideophones (to the extent that ideophones are typically characterised by a low degree of morphosyntactic integration, this should increase their borrowability)
- a scalar conception of the differences between ideophone systems across languages (looking at morphosyntax and expressive morphology allows us to state more explicitly what makes the ideophone system of Somali different from that of Siwu and these two different again from Semai)
Enjoy. Here’s to hoping the article won’t take too long to appear in print. I’ve already been working with Kimi Akita on an exciting follow-up project testing some of these proposals quantitatively on Japanese corpus data.
- Dingemanse, Mark. accepted. “Expressiveness and system integration. On the typology of ideophones, with special reference to Siwu.” STUF – Language Typology and Universals (special issue).