Three misconceptions about ideophones

In a previous post I have outlined the history of the term ideophone. This post takes on three common misunderstandings about the nature of ideophones. As an added bonus, if you read all three, you get one for free. The working definition I adopt for ‘ideophones’ is the following: Marked words that depict sensory imagery. In lay terms, ideophones are words that stand out (are ‘marked’) and whose form betrays something of (is depictive of) their meaning.

  1. ‘Ideophone’ is just jargon for onomatopoeia. Not quite. Onomatopoeia are generally understood to be limited to words imitating sounds. Ideophones however evoke all sorts of sensory events — not just sounds, but also taste, gait, visual effects, texture, smell, and so on.1
    Consider the following Siwu ideophones: vɛlɛvɛlɛ ‘a dizzy, giddy feeling in the body’; yuayua ‘a sensation of burning (the visual impression, the feeling, or both)’; kpotoro-kpotoro ‘moving jerkily like a tortoise’; ɣɛkpɛtɛɛ ‘delicately fragile, for example of autumn leaves’. These words do not imitate sounds, yet to a Siwu speaker they vividly depict sensory events in a way that is reminiscent of onomatopoeia.2 The German linguists had an excellent term for this: Lautmalerei ‘painting with sound’, the result of which was a Lautbild ‘sound picture’ (Westermann 1907, 1927, cf. also Bühler 1934).
  2. English and other Standard Average European (SAE) languages lack ideophones. Not quite. Given the definition of ideophone above, ideophony is probably a universal phenomenon. English, for example, has ideophonic words like glimmer, twiddle, tinkle which are depictive of sensory imagery: their form betrays something of their meaning in ways that words “chair” and “dog” do not.
    All the same it is true (and interesting) that languages differ in the extent to which they systematize and elaborate their ideophonic (expressive) resources. In that sense English is definitely a much less ideophonic language than, say, Semai (Central Aslian, Austroasiatic, Malaysia), where ideophones are a word class as big as the two other major word classes, nouns and verbs3, or Gbeya (Adama-Eastern, Niger-Congo, Central African Republic), where over 5000 ideophones have been collected4. In the latter type of language, ideophones make up a large and clearly recognizable class of words, whereas in English, ideophonic vocabulary is sprinkled all over the lexicon (and probably less common overall).
  3. Ideophones are a feature of primitive languages. Not quite. This idea is at least as old as the first descriptions of ideophones in ‘exotic’ languages. It was made popular by anthropologist Lévy-Bruhl’s musings about ‘primitive mentality’,5 in which ideophones were adduced as evidence for the ‘irresistible tendency’ of the native to ‘imitate all one perceives’ (1926:142). One thing we have learned since then is that the notion of ‘primitive language’ makes no sense outside the highly problematic model of cultural evolutionism in which it was coined.
    I don’t even want to give counterexamples in the form of supposedly non-primitive languages which nonetheless are ideophonic; anyone interested can look up some relevant literature (start with Voeltz & Kilian-Hatz 2001). For linguists, languages differ in interesting ways and along all sorts of dimensions; but the supposed dimension of primitivity is not one of them.

And because it’s ‘read three, get one for free’ today, here is your bonus misunderstanding:

  1. An ideophone is a type of musical instrument. Not quite. Xylophones, marimbas and the like are called idiophones, lit. ‘self-sounding’. This is distinct from ideophone, which derives from ‘idea-sounding’. Don’t worry, it’s a common error.


  1. Bühler, Karl. 1934. Sprachtheorie: Die Darstellungsfunktion der Sprache. G. Fischer.
  2. Diffloth, Gérard. 1976. Expressives in Semai. Austroasiatic Studies 1, Oceanic Linguistics. Honolulu: University of Hawaii, Special Publication 13: 249-264.
  3. Lévy-Bruhl, Lucien. 1910. Les Fonctions Mentales dans les Sociétés Inférieures. Paris.
  4. Lévy-Bruhl, Lucien. 1926. How Natives Think. Trans. Lilian A Clare. New York: Washington Square Press.How Natives Think. Trans. Lilian A Clare. New York: Washington Square Press.
  5. Samarin, William J. 1970. Field procedures in ideophone research. Journal of African Languages 9, no. 1: 27-30.
  6. Voeltz, F. K. Erhard, and Christa Kilian-Hatz, eds. 2001. Ideophones. Ed. F. K. Erhard Voeltz and Christa Kilian-Hatz. Typological Studies in Language, 44. Amsterdam: John Benjamins.
  7. Webster, Anthony K. 2008. ‘To Give an Imagination to the Listener’: The neglected poetics of Navajo ideophony. Semiotica 171: 343-365.Semiotica 171: 343-365.
  8. Westermann, Diedrich Hermann. 1907. Grammatik der Ewe-Sprache. Berlin: Dietrich Reimer.Grammatik der Ewe-Sprache. Berlin: Dietrich Reimer.
  9. Westermann, Diedrich Hermann. 1927. Laut, Ton und Sinn in Westafrikanischen Sudansprachen. In Festschrift Meinhof, 315-328. Hamburg: J.J. Augustin.


  1. Some consider onomatopoeia a subcategory of ideophony, in which case onomatopes would be those ideophones that evoke auditory events. However, this is a choice that can only be made on a language by language basis.
    A related issue is the fact that languages differ in interesting ways with regard to the relative elaboration of the different sensory modalities in their ideophone inventories. Certain indigenous languages of North America for example seem to have ideophones that are predominantly auditory (cf. Webster 2008 on Navajo), whereas auditory ideophones make up only a small fraction of the inventories of many African languages.
  2. Historically, ‘onomatopoeia’ has been used in a broader sense (see for example a previous post on early sources on African ideophones). The sound-only connotation of ‘onomatopoeia’ is the main reason for advocating the more general term ‘ideophone’.
  3. Diffloth 1976:249
  4. Samarin 1970:55
  5. See Lévy-Bruhl (1910) Les Fonctions Mentales dans les Sociétés Inférieures (translated as How Natives Think, 1926).

15 thoughts on “Three misconceptions about ideophones

  1. Thanks for this. I’ve been following the blog for months just out of curiosity but always had a sneaking suspicion I didn’t really understand what an ideophone was. I’m having a hard time combining the definition above, “evoke sensory events” with the one from the other post, “describes a predicate, qualificative or adverb in respect to manner, colour, sound, smell, action, state or intensity” and from earlier…

    “They are phonologically marked by deviant phonotactics and all sorts of co-occurrence constraints; they have special morphology (often iconic, e.g. reduplication, lengthening); there is some syntactical ‘aloofness’ to them (can’t be negated, are often less well integrated into the sentence); and they are semantically very specific, typically evoking experiences as a whole rather than encoding some attributes of objects of events.”

    …to come up with a set of rules I can apply to determine whether something’s an ideophone or not. In Mandarin, for example, what would you need to know about…

    “pitch black”

    …to know whether it’s an ideophone or not? It’s got (almost?) reduplication. It certainly is evoking a sensory event. I’m pretty sure it can’t be negated. But is it really an ideophone and how would I know?

  2. Hah, one point I saved for a followup post was actually ‘Ideophones are a separate word class / grammatical category’. They don’t have to be — Somali ideophones for example are a subset of nouns, while Hausa (Newman 1968) has them sprinkled over different word classes.

    As always, the case has to be made on a language internal basis. I know next to nothing about Mandarin, but what you’ll want to find out is whether you can identify not just a few isolated items, but a form class based on phonological, morphological, syntactic, and semantic properties. If not, it would be difficult to make a case for ideophones in Mandarin.

    I’m not aware of work on ideophony in Mandarin, but you may be interested in Adams Bodomo’s work on Cantonese ideophones; see for example his 2006 article in the Proceedings of the 35th ACAL. I would be very interested in your comments on his argumentation.

  3. Oh, and regarding your comment on having a hard time combining the different definitions, I think the main outlier in the fragments you’re citing there is Doke’s definition, which was devised with only Bantu languages in mind, and even then not without problems (for rather insensitive to the semiotics and semantics of ideophony, as I’ve tried to outline in On the history of the term ‘ideophone’).

    Or is there another inconsistency that is bugging you?

  4. Thanks, Mark. Very helpful. The paper looks promising — I’m guessing there are parallels to Mandarin. Maybe I’ll cross-query on my blog and see if anyone knows of something that’s been done in Mandarin.

    If I get more info I’ll try to return here and let you know what I find.

    For the second point, I think I’m clearer now. That was the inconsistency I was thinking of.

  5. happy to see that ‘twiddle’ is getting more press as an ideophone… my thumbs were getting jealous. As I was thinking about these misconceptions I wondered what the kernels of truth might be that led the mistaken parties to have gone astray. 1. The depiction of non-audible sensory experiences by linguistic means is inherently problematic in a culture which sees synesthesia as a “neurological condition.”
    2.It is statistically true that there are fewer ideophones in the SAE’s I know, though they still have plenty
    3.Ideophones seem to express an almost playful element of language, which I don’t think is primitive, but it seems connected to an idea of holding language loosely, or not taking one’s own language entirely seriously. The performativity of some ideophones is such that they seem to be self-referential with hyperbolic and humorous effect (almost like a hipster ;) which could to an outsider seem primitive, because they are missing a whole level of subtlety which goes straight over their heads… again perhaps the cultural perception is the dominating factor, and maybe the cultures which use many ideophones do not value or tolerate self-importance in speech. Just musings, but I’d love to here more on syneshesia…

  6. Khawaji, great points as always.
    To 1 and 3 I’d add that mimesis is associated with childishness and simplicity in SAE cultures (as noted by Childs and Nuckolls); that’s one reason that people have difficulty taking pervasive ideophony serious. See also the Max Müller quote at the right top of this page: ‘they are playthings, not the tools of language’.

  7. Pingback: Wordcatcher Tales: Hokahoka, Hougan, Hiiki « Far Outliers

  8. Pingback: Wordcatcher Tales: Hodohodo, Czechia, Kanakysaurus « Far Outliers

  9. Highly inflective languages tend, on average, to have many fewer own-class ideophones than agglutinative or isolating languages. That not all isolating languages are equivalent may also have to do with the fact that fusion also seems to limit ideophone inventories, and some of the isolating languages have undergone a lot of fusion on their way to their current state.

    One would be hard pressed to find more than a small handful of ideophones in strongly polysynthetic languages like Mohawk or Yupik Eskimo, and the ones they have are simple onomatopes.

    Navajo ideophones, by and large, are also no longer phonosemantically transparent, which indicates they have moved pretty far from protypicality.

    Interestingly, complex ideophones in languages that have very many have internal structure reminiscent of serial verbs- and many (most?) of the languages that have these also have real serial verbs. And strongly polysynthetic languages, according to Sasha Aikhenvald, don’t particularly like serial constructions either.

    In some Niger-Congo languages ideophones can be derived FROM verbs and/or vice versa- roots often have suffixal expansions before the usual derivational suffixes- old serial morphemes?

    OTOH, in other languages where such interaction is not living one can still clearly see some sort of remnant of process, where some ideophones are identical to intransitive verbs, or verbs have similar form/meaning mapping at the level of individual phonemes to ideophones, despite non-identity at the level of the string. This can be seen for instance in Ijoid.

    If Niger-Congo was ancestrally SOV (and crosslinguistically, and geographically the languages with the largest transparent ideophone systems appear to be SOV), then it might serve as a testing ground for hypotheses relating to ideophone lexicalization and opacification in the face of historical phonological and morphosyntactic changes which disrupt the symmetries of consonantal, vocalic, and tonal inventories.

    How do ideophones react when verb serialization stops working, or vowel harmony dies? What about reduction of syllable structures?

    Jess Tauber

  10. Jess, thanks for this rich comment. That’s a lot of observations and hypotheses crammed into a few paragraphs.

    Actually, on that very last point (how do ideophones react when vowel harmony dies) I should be able to say something in the future. Siwu has lost vowel harmony, but several of its close relatives (Santrokofi, Lelemi, Likpe) still have a working vowel harmony system. I’m planning to gather some comparative data on ideophones from Siwu neighbours to find out the extent to which ideophone inventories are shared (or related).

  11. Pingback: Oh no! Ideophones are not response cries! — The Ideophone

  12. Pingback: Ideophones around the web — The Ideophone

Leave a Reply

Your email address will not be published. Required fields are marked *

nine − 6 =