This special issue is a follow-up to the 4th ISMo (International Symposium of Morphology) conference, held in Nancy in September 2023. It includes papers focusing on word formation in a variety of languages – French, Italian, Hebrew, Chinese – that fall into three typological groups (Bickel & Nichols, 2013), with Chinese at one end (an isolating language with no inflection and an exclusively concatenative derivational system), Hebrew at the other end (a language with root-and-pattern morphology), and French and Italian in between (two concatenative and inflectional languages).
The topics of these articles are equally diverse: verb formation from borrowings (see, among others, ten Hacken & Panocová, 2020; Wohlgemuth, 2009); productivity of suffixations that form ethnonyms (Thuilier, Tribout & Wauquier, 2023); polysemy of newly coined derived verb (Plag, 2000; Lignon, 2013; Rainer, 2014; Bonami & Thuillier, 2019; Salvadori & Huyghe, 2022); nominal compounding (Lieber & Štekauer, 2009). These topics are approached from specific angles: affix rivalry (Huyghe & Varvara, 2023); competition between morphology and syntax (Dal & Namer, 2003; Ackema & Neeleman, 2004; Padrosa, 2007); contrastive derivation (Lefer, 2011); comparison between diachrony and synchrony (Rainer, 2008; Rainer, Dressler, Kastovsky, Luschützky & Peters, 2010).
This special issue provides an overview of current research in constructional morphology: description of neologisms in Semitic (Laks) and Sinitic (Zhu) languages, from the perspective of morphology-syntax competition, for the former, and from that of polysemy and comparison with French, for the latter; experimental analysis of the rivalry between affixes that form demonyms in French (Huygevelde, Kayirici, Bonami & Hemforth); diachronic corpus analysis of attributive‑appositive compound nouns in Italian (Radimsky & Micheli).
It also provides insight into recent, and original, approaches to describing word formation: the hypotheses defended by Huygevelde et al. are supported by the statistical approach of Bonami and Thuillier (2019); Zhu develops a hybrid model inspired by cognitive grammar (Langacker, 1987); Laks defends a word-and-pattern perspective (Ussishkin, 2006), while Radimsky and Micheli refer to relational morphology (Jackendoff & Audring, 2020), inspired by Jackendoff’s Parallel Architecture (2007) and very close to construction morphology (Booij & Audring, 2015).
Lior Laks’ paper, the first in this issue, focuses on the competition between Hebrew borrowing-based morphological and periphrastic verb formation. The verbs were collected online. Their formation involves non-concatenative patterns (Ussishkin, 2005; Bat-El, 2011; Aronoff, 1994), because new verbs must conform to existing patterns. Laks (2018) argues that the selection of these patterns is based on the interplay between morphophonological and semantic-syntactic criteria, the former partly explaining the competition between morphological and periphrastic structures because word formation from longer than two syllabled loanwords is rare (Bolotzky, 1978). In this paper, Laks’ analysis is that these criteria block word formation when structural transparency is low and when loanwords are suffixed. Conversely, low semantic transparency tends to block periphrastic formation because there may be no alternative to express the intended meaning.
The second article, co-authored by Marie Huygevelde, Ridvan Kayirici, Olivier Bonami and Barbara Hemforth, examines the phonological motivations that partially explain the distribution of demonym forming suffixes in French. The authors interviewed about 70 participants in order to experimentally test whether the processes preferred by speakers correspond to trends observed in the established lexicon with regard to the phonological properties of the toponymic bases. The authors focus on the distribution of the four most productive exponents in contemporary French: -ais, -ois, -éen, and -ien, selected from the large number of suffixes available in French that may be used to form denonyms, such as -ain (Toulouse > Toulousain ‘Toulousian’), -ard (Savoie > Savoyard ‘Savoie inhabitant’), -eau (Touraine > Tourangeau ‘Touraine inhabitant’), -ite (Yemen > Yemenite ‘Yemeni’), -ote (Chypre ‘Cyprus’ > Chypriote ‘Cypriot’), etc. The study builds on Thuilier et al. (2023), who found that the relative productivity of these affixes in the attested lexicon is determined by phonological and geographical conditions. The present paper refines the results of this study and offers a statistical measurement of the extent to which current speakers conform to these conditions.
The article by Jan Radimský and Silvia Micheli focuses on the productivity of attributive-appositive NN compounds (ATAP NN) in Italian. The study is carried out using Google n-grams, from which the authors extracted a dataset of almost 3,000 compounds. It falls within the framework of Construction Grammars and, more specifically, Relational Morphology (Jackendoff & Audring, 2020a, 2020b). The authors show that the productivity of Italian ATAP NN compounds dates to the first half of the 20th century. They use the families formed by each of the two nouns N1 and N2 of the compounds to study ATAP NN productivity diachronically (Traugott & Trousdale, 2013). They argue that part of the productivity of the compounds is accounted for by the inclusion of an X-N2 patterns in the Italian construction (where N2 stands for the second nominal component), which allows the development of N2 families, i.e., families of compounds sharing the same N2. An original contribution of the study is the finding of the existence of N1-X type patterns, which also contribute to the increase in the number of ATAP NNs in Italian. The authors show that the two types of families are strongly intertwined in the sense that N1 families tend to grow by adding compounds that include N2s that already have families in the lexicon.
The final article, by Jiahiu Zhu, examines causative relations expressed by the verb suffixes -iser in French and -化 [huà] in Chinese, proposing a model that draws on principles from Cognitive Grammar (Langacker, 1987), Construction Grammar (Goldberg, 1995), and Causal Event Structure (Croft, 2012). The study is based on a corpus of 6,900 causative constructions in French and 7,200 causative constructions in Chinese, all of them first attested after 2000. Zhu shows that the polysemy of these constructions covers similar uses of causative relations in both languages, despite their typological differences. She groups these uses into six schemas that reflect the distribution of verb interpretations in both French and Chinese.