Onomatopoeia and metonymy

2. What onomatopoeic forms are there in English?
When asked to do so, speakers of English usually have no trouble rattling off a couple of onomatopoeic words, such as baa, click, swoosh or ouch. However, the question of what exactly is an onomatopoeic word becomes all the more intriguing if, on the one hand, comic book “words” that imitate sounds are also included. For instance, a quick browse through Garfield comics yields examples such as ba-going (for the sound that an oven makes when the food is ready) or patta-patta-patta (the sound that Garfield’s paws make on the floor while running). Such novel coinages often do not go into wider circulation – i.e., they remain one-off formations that are lost outside of the comic strip they appeared in. On the other hand, there are also words that reference books such as the Oxford English Dictionary (henceforth OED) cite as having onomatopoeic (imitative) origins, but which – over time – have mostly lost their onomatopoeic flair due to lexicalization. Thus, it can prove to be difficult to identify the exact sound similarity in words such as bounce, crisp or flap (all of which are listed in the OED as “imitative”).
The difficulty in defining what onomatopoeia are comes to the forefront on a closer inspection of the list of entries that the OED identifies as having an “imitative” origin. In June 2021 the dictionary listed 1,084 such entries (which, needless to say, is a mere fraction of the collection of words that can be found in the OED), which includes words that are a) “imitative”; b) “echoic”; and c) “onomatopoeic”. At this point it needs to be stressed that the OED does not clearly differentiate between “imitative”, “echoic” or “onomatopoeia”. The entries in the list are often classified as being imitative or echoic, meaning that the entries are imitative/echoic of a sound, but the exact nature of this “imitation” is often left unspecified. The OED in fact uses “onomatopoeic” ambiguously: 1) it can refer to a word that imitates a particular sound, as in the case of put-put, the etymology of which is provided as “to make a rapid intermittent sound characteristic of a small internal combustion engine ”; and 2) it is also used in reference to conventional sound symbolism, as in the case of brattle, the etymology of which is provided as “this and its verb are onomatopoeic, probably with association of break, brast and rattle ”. Despite these problems, the OED’s list can still be regarded as the most authoritative list of words in English with an imitative (onomatopoeic) origin.²
The items in this list range from the simplest interjections,³ such as ai (“an exclamation of surprise, regret, pain, etc., often used in imitation of foreign speakers of English”) or poof (“expressing contempt, disgust, or distaste as conveyed by the emission of a short sharp puff of breath”), through “classic”, repetition-based onomatopoeia “imitating” the sounds of animals, such as ruff (“imiative of a dog’s bark”) or weet-weet (“an imitation of the cry of certain birds, esp. the sandpiper and chaffinch”), all the way to more conventionalised (and etymologically obscure) examples, such as morepork (“the boobook owl”)⁴ or chuck-will’s-widow (“popular name in U.S. of a species of goat-sucker (Caprimulgus carolinensis)”). Over the centuries, the number of onomatopoeic formations has steadily increased in number, peaking in the first half of the nineteenth century (with a total of 160 onomatopoeic words) – see Figure 1.
The OED’s list provides a fascinating insight into a number of the problematic and challenging aspects of onomatopoeia. A quick glance at the “Etymology” of the entries demonstrates the fairly subjective and overall challenging nature of deciding what counts as onomatopoeia or a word of an imitative origin (see Campbell 2005, p. 272 for a similar opinion; cf. Körtvélyessy 2020). See, for example, the etymology of tik, the most recent onomatopoeic addition to the dictionary (“the drug methamphetamine; esp. crystal methamphetamine”), which is provided as “perhaps imitative of the distinctive popping sound made when the drug in crystal form is lit and smoked” Many of the entries have a question mark beside the explanation, as in the case of flod (“to walk slowly”), the etymology of which is provided as “?Onomatopoeic”, clearly indicating the uncertainty of the lexicographer; and see especially the etymology of the now obsolete chymer (“to shiver”), the etymology of which is given as “Apparently onomatopoeic (unless an error)”. This doubt is also pronounced in the prolific use of hedges, such as apparently, possibly or probably in the etymological section of the entries, as in the example of brash (“probably ultimately imitative of a humming, murmuring, or rumbling sound”; emphasis added).





Figure 1. Timeline of onomatopoeic formations in the OED as of June 2021 (own compilation).

Nevertheless, what unites all of the entries in the OED’s list is that they are felt or considered to echo or copy sounds that we hear in our environment, which might be a good starting point for a definition of what onomatopoeia are. Yet they do this to varying degrees: ruff refers to the sound itself that a dog makes, while chuck-will’s-widow is the name of a bird that makes a particular sound that echoes the word form, and the meaning is thus based on a metonymical relationship between the sound and the bird producing the sound.⁵ Thus, it is a further problematic aspect to what degree onomatopoeia echo the sound in question: in other words, to what degree is chuck-will’s-widow – or ruff for that matter – similar to the sound that is actually produced. Last but not least, there is also the question of conventionality: while the OED lists both mph and mphm as alternatives (with the meaning of “expressing disapproval, doubt, or dissatisfaction”), neither one of these conforms to English spelling rules.

Judging by the entries listed in the OED, onomatopoeic forms are a quirky set of words (or word-like units)⁶ that are reminiscent of a particular sound; they represent various degrees of lexicalisation; they do not necessarily have a single form; and they do not necessarily follow conventional word-formation patterns. They extend over several of Hinton et al.’s (1994) original sound symbolic categories: corporeal sound symbolism (sound symbolic patterns typically used for expressing the speaker’s emotional or physical state, such as mph), imitative sound symbolism (words and phrases that echo or copy the sounds in our environment or the sounds that animals make, such as ruff), and possibly conventional sound symbolism as well (i.e., phonesthemes), implying that there may be many more such items in English than what the OED’s limited list suggests. In fact, there is considerable ambiguity concerning the term onomatopoeia in the literature, and it is not uncommon to use this term in reference to phonesthemes (i.e., conventional sound symbolism), as for example in Hughes (2000). In the present paper, however, and following Benczes (2019), onomatopoeia will be primarily understood to cover corporeal and imitative sound symbolism.