Revising an implicational hierarchy for the meanings of ideophones, with special reference to Japonic

Abstract

An elicitation task was conducted with speakers of Japonic varieties to investigate whether stimuli of varying sensory modalities (e.g. audio, visual, tactile etc.) were more or less likely to elicit ideophones or iconic words. Stimuli representing sounds, movements, shapes and textures were most likely to elicit ideophones, and this is posited to reflect the relative ease or naturalness with which these domains can be mapped iconically to speech. The results mirror macro-level patterns of linguistic diversity, as these are also domains in which ideophones are most widely attested cross-linguistically. The findings call for the revision of a previously constructed implicational hierarchy for the semantic development of ideophone systems, adding to it the categories of FORM and TEXTURE as domains in which ideophones are most likely to develop, after SOUND and MOVEMENT. Independent evidence for the revised hierarchy comes from a semantic analysis of the elicited ideophones, where it was found that domains early in the hierarchy were more likely to be sources for semantic extension, while later domains were more likely to be targets. These findings are expected to be replicable for other languages, and offer exciting new directions for research into the semantic typology of ideophones.

A Cross‐Modal and Cross‐lingual Study of Iconicity in Language: Insights From Deep Learning

“…In the ideophonic lexicon – i.e the portion of the vocabulary that includes marked words depicting sensory imagery (Dingemanse, 2012 ) – auditory terms are the most prominent class. They occupy the highest rank in the cross‐linguistic implicational hierarchies developed by Blasi, Dingemanse, Lupyan, Christiansen, and Monaghan ( 2015 ) and revised by McLean ( 2021 ), meaning that if a language does not develop auditory ideophones, it will not produce ideophones related to the other senses. Nonetheless, we chose to focus on vision since we were interested in an analogical iconic mapping that involved a cross‐modal link.…”

A semantic map for ideophones

“…Especially in relation to ideophones, it has been understudied (Dingemanse, 2012), but see the work of Samarin (1967;1970a). This type of research has largely been composed of lexicographic works (Nuckolls, 2016), narratives (Nuckolls, 2010), and language-specific grammars (Akita 2009), although surveys are gaining traction, e.g., folk definitions and sorting tasks based on them (Dingemanse and Majid, 2012;Dingemanse, 2015;Kroeger 2016) and elicitation tasks McLean, 2020).…”

“…In relation to Dingemanse's Implicational Hierarchy, then, it becomes clear that we first need to lay out the different sensory domains that are present in ideophones across languages, and only then are able to study the different extensional routes between these domains on a language-specific level, as has been done in a seminal article on the extension of sensory vocabulary by Williams (1976) and recently revisited in the growing field of sensory linguistics (Winter, 2019). In the field of ideophone studies such extensions have been studied for Japanese (Lu, 2011), Middle Chinese ( Van Hoey, 2015), Pastaza Quechua (Nuckolls, 2019), and a wider survey of Japonic languages (McLean, 2020), to name a few. We will revisit this point in the discussion (section 7).…”

“…In other words, while Dingemanse's Implicational Hierarchy (10) predicts that languages with a mimetic lexicon containing visual patterns will have depictions of motion and sound as well, it makes no predictions about the sizes of those domains. Nevertheless, in some languages the hierarchy corresponds with the frequencies of ideophonic categories, e.g., Kammu (Svantesson, 2017), or is revised to correspond with the frequencies (McLean, 2020).…”

“…The idea of two extremes of iconic affordance (imagic and diagrammatic) has been kept. Next, we have, inspired by other cross-linguistic approaches (Lu, 2006;Akita, 2009;McLean, 2020), adhered to the basic outline of the hierarchy but divided it in macrozones.…”