1In many studies in lexicology, the description of the lexicon of a given language is traditionally considered from two complementary points of view: the level of form, which concerns lexical morphology, and the level of meaning, which is partly dealt with by lexical semantics. Thus, from a Saussurean perspective, a lexical unit is a two-sided entity, combining a signifier (form) and a signified (meaning). The “auditory” aspect of words, which is an integral part of the signifier, is most often reserved for phonological or phonetic studies, which have long been autonomous research fields. However, it is clear that lexicology and phonology are intertwined, to the point that, if lexicology is to truly deal with all the aspects of lexical units (form, meaning and use), it cannot do so without phonology.
2In the field of diachronic linguistics, the various processes that have contributed to the diffusion of the pronunciations of specific words or of sets of words are relevant to the interface between lexicology and phonology. Particularly interesting cases are the processes of lexical diffusion (when the unit of phonological change is the word, or rather its root, and not the phoneme; see Labov [2010: 260]) and what they tell us about the lexicon and lexical storage.
3The question may of course be considered from the perspective of word formation processes (e.g. clipping, the association of a suffix and a base according to the stress pattern of the base). Morpheme or word boundaries and boundary strength are also strongly related to phonology and the phonological behaviour of words (e.g. Oh & Redford [2012], Ben Hedia & Plag [2017]). Phonological data sometimes informs us about the way the lexicon is structured. Bell & Plag [2013] have shown that the stress patterns of noun + noun compounds depend on lexical factors such as their spelling, the frequency of the compound and its constituents, and the informativeness of the second element. Many studies in phonology have also revealed distinct behaviours for historical prefixed words such as contain, resist or submit, which suggests that these words are treated as complex entities despite the lack of clear semantics and productivity of the constituents involved (see the overview in Dabouis & Fournier [to appear]). This is corroborated by studies in psycholinguistics (see Forster & Azuma [2000], Ktori et al. [2016], McKinnon et al. [2003] and Rastle & Coltheart [2000]). Phonological identity between morphologically related words may also inform us regarding the organization of the lexicon. Apparent faithfulness effects (e.g. stress identity) between a complex word and a member of its morphological family that is not its morphological base (e.g. remédiable cp. rémedy but remédial) are at the core of a controversy between stratal-cyclic models of phonology (Bermúdez-Otero [2018]) and models using faithfulness between different surface forms to capture such effects (Steriade [1999]; Breiss [2021]).
4Sometimes, the association of the signified and of the signifier is not entirely arbitrary. It may even be downright iconic, as may be the case with sound symbolism and phonesthemes (Jespersen [1922], Firth [1930]). Sounds may therefore carry meaning at the morphemic or sub-morphemic level (Bottineau [2022]).
5Borrowings constitute an area of the lexicology/phonology interface as most borrowings have their morphological, semantic and phonological specificities and/or have been adapted to the host language on the semantic, morphological and phonological levels. Dabouis & Fournier [2022] hypothesize that the English lexicon is organized into subsystems sharing a number of morphological, semantic and phonological properties, and that these properties are related to real etymology (e.g. French words have a similar behaviour) or perceived etymology (see Wells [2008] who notes “Not a real French expression” for the entry bon viveur). There may also be an interplay between spelling, pronunciation, and the process of borrowing. In diachronic linguistics, the diffusion of borrowings and the diffusion of the phonological forms and stress patterns of lexical items are of course intricately linked to phonology and the part it plays in lexical storage.
6The interface may be situated on a general abstract level and determine the ways in which phonology theoretically influences morphology and semantics. Phonological theory may help the improvement of our understanding of lexical acquisition.
7The list above is by no means exhaustive. This issue of Lexis aims to explore some of the ways in which the strong, even consubstantial, link between lexicology and phonology manifests itself.
8The first paper by Pierre Fournier, entitled “The impact of source languages on the stressing of loanwords in English”, focuses on stress placement in loanwords from a variety of different languages. He questions the view that loanwords would obey the same stress rules as those of the rest of the lexicon and evaluates the extent to which the position of stress in loanwords in English reproduces the position of stress for the corresponding words in source languages. He also seeks to evaluate the hypothesis that English has a preferential stress pattern for loanwords: penultimate stress. The study is conducted on a pronunciation dictionary-based dataset of 768 loanwords, and the results show both a strong tendency for loanwords to be stressed on the penult and to preserve the position of stress found in the source language. However, while penultimate stress and antepenultimate stress are quite regularly preserved, final stress is often not.
9Then, Semra Baturay-Meral’s theoretical and empirical contribution, entitled “Phonological templates and the lexicon”, develops a new way of identifying word boundaries and explaining phonological processes that aim to differentiate word bases from affixes (suffixes and prefixes). She argues that traditional “extra-phonological items” (diacritics, brackets, etc.) bring arbitrariness to phonological theory when it comes to accounting for phonological processes. Instead, her New Template Model aims to demonstrate that word bases, suffixes and prefixes are stored in the lexicon in relation to the following phonological templates (where O stands for “onset” and N for “nucleus”): the base template, ONO (always begins and ends with an onset), the suffix template NO (starts with a nucleus and ends with an onset) and the prefix template ON (starts with an onset and ends with a nucleus). With this new model, Baturay-Meral argues that phonological processes can be explained non-arbitrarily considering that base-initial and base-final onsets (O) serve as “natural boundary markers” and do not require the use of extra-phonological items to account for morphological boundness.
10In the third paper of this issue, Maël Farina’s contribution, entitled “Groaning and Grunting: Investigating Sound Correspondences in The English Lexicon”, offers a corpus-based analysis of phonesthemes, i.e. meaning-carrying sounds that are emotionally expressive and that convey some degree of symbolism. Phonesthemes carry phonetic similarity between the forms, which can be exemplified with the /gr/ sequence in words such as grumble, groan, grunt, grieve and grudge, where /gr/ relates to a “form of complaint”. The study combines a lexicographical analysis of the Oxford English Dictionary (OED) and a corpus analysis of the Oxford English Corpus (OEC) with the help of the distributional tool Sketch Engine. Farina investigates the semantic traits associated with the phonestheme /gr/, its semantic characteristics and collocational behaviour based on the sample words groan, grunt, grudge, grip and grasp. Farina’s contribution shows the key role played by surrounding collocates and context in the semantic behaviour of phonesthemic words, especially so with the phonestheme /gr/ that displays various submeanings in context.
11In her paper entitled “Rah-rah! Investigating the variation in phonosemantic motivation in a set of iconic nouns expressing the concept <enthusiasm energy vitality>. A diachronic semantic approach”, Chris Smith examines phonosemantic motivation in a set of 16 nouns belonging to the field of enthusiasm, energy or vitality. She explores their emergence and semantic development. She also sets out to determine their degree of similarity, some of these nouns being clearly imitative (e.g. zip, dash), while others are non-imitative (e.g. pizzazz). The approach is multi-method, using the Historical Thesaurus of English to generate the nouns to be included in the study, and the Corpus of Historical American English as well as the Oxford English Corpus to investigate their distributional behaviour. It is also multi-factorial, in that it integrates etymology, morphology and semantic associations with other words in the nearby context or in the metapragmatic context. One of Chris Smith’s findings is that although the words under study followed different pathways of emergence of the idea of “enthusiasm/energy/vitality”, and may exhibit different distributional behaviours, there is a tendency towards the clustering of sound imitative effects in a given context. For example, the repetition of initial consonants is particularly striking in some coordinated nominals such as zip and zest or pep and punch.
12Finally, in “What a difference a digraph made: phonetic spelling and the assimilation of the word ‘armada’ in Early Modern English”, Laetitia Sansonetti aims to examine the interplay between spelling, pronunciation and borrowing in Early Modern English, a period in which many words were imported. Writings displayed significant orthographic variation at the time, as spelling and pronunciation were not fixed. The case of armada is particularly interesting because the word captured the collective European imagination after the 1588 episode of the “Spanish Armada”. The spelling of the word varied a great deal, encoding certain pronunciations that reflected cultural, social or political concerns. The article highlights the complex dynamics at work in borrowings from Romance languages throughout Early Modern English.