Palatal is for happiness, plosive is for sadness: evidence for stochastic relationships between phoneme classes and sentiment polarity in Hungarian

Abstract

The past couple of decades have seen a substantial increase in linguistic research that highlights the non-arbitrariness of language, as manifested in motivated sound–meaning correspondences. Yet one of the challenges of such studies is that there is a relative paucity of data-driven analyses, especially in the case of languages other than English, such as Hungarian, even though the proportion of at least partially motivated words in Hungarian vocabulary is substantial. We address this gap by investigating the relationship between Hungarian phoneme classes and positive/negative sentiment based on 3,023 word forms retrieved from the Hungarian Sentiment Lexicon. Our results indicate that positive polarity word forms tend to contain more vowels, front vowels, continuants, fricatives, palatals, and sibilants. On the other hand, negative sentiment polarity words tend to have more rounded vowels, plosives, and dorsal consonants. While our analysis provides strong evidence for a set of non-arbitrary form–meaning relationships, effect sizes also reveal that such associations tend to be fairly weak tendencies, and therefore sentiment polarity cannot be derived from the relative frequencies of phoneme classes in a deterministic fashion.

Statistical Relationships Between Phonological Form, Emotional Valence and Arousal of Spanish Words

“…Beyond sound symbolic relationships, there are non-arbitrary relationships between sound and meaning that manifest as statistical regularities more extensively within languages, variously referred to as phonological form systematicity or typicality (see Dingemanse et al, 2015;Haslett & Cai, 2023). 1 These too have recently been investigated for emotional valence and arousal (the degree to which the activation or intensity of a word's referent is calming or exciting) at the level of phonetic features (e.g., place and manner of articulation, voicing; Adelman et al, 2018;Benczes & Kovács, 2022;Calvillo-Torres et al, 2024;de Zubicaray et al, 2023;Kambara & Umemura, 2021;Louwerse & Qu, 2017). In English, positively valenced words tend to have more bilabial and velar sounds in their initial phonemes and more labiodental final phonemes, while negatively valenced words are more likely to comprise more stops and fricatives and have a stressed syllable in addition to more nasal sounds in their initial phoneme (e.g., de Zubicaray et al, 2023;Louwerse & Qu, 2017).…”

“…In German, words expressing high arousing concepts tend to comprise short vowels, voiceless consonants, and hissing sibilants (Aryani et al, 2018;Ullrich et al, 2016). In Hungarian, a non-Indo-European language, positive valence instead tends to be associated with more fricatives, palatals and sibilant sounds and negative valence with plosives (Benczes & Kovács, 2022). However, it should be acknowledged these statistical regularities explain a relatively small proportion of variance in affective ratings (i.e., a few percent; e.g., Benczes & Kovács, 2022;de Zubicaray et al, 2023).…”