📄 Patrizia Anesa, Lexical Innovation in World Englishes: Cross-fertilization and Evolving Paradigms

Summary

This is a book review, not an original study. Li and Ye assess Patrizia Anesa's monograph (Routledge, 2018, 174 pp.), which examines how new words arise and travel among varieties of English worldwide. Anesa argues that lexical change in World Englishes should be read through discourse and social context, not described only in structuralist terms. The book has three parts and seven chapters. It first revisits models for classifying varieties, along with the contested ideas of nativeness and standardization. It then sets out a typology of word-formation processes and a corpus-based method that pairs automatic or semiautomatic extraction with lexicographic sources. Finally, eight case studies cover varieties in Africa, Asia and the Caribbean. The central claim is that varieties keep borrowing from one another, so the boundaries between them are porous, and innovation includes sharing words as well as coining them. The reviewers praise the corpus grounding. They add that counting frequencies is not enough and that qualitative study of convergence and divergence is still needed.

Key points

• The book contains eight empirical case studies. The South African material is grouped as three varieties. The Asian chapters cover Singapore, Hong Kong, the Philippines and India.
• The processes analysed are composites, abbreviations, blends, borrowing, conversion, semantic shift and eponyms. The book stresses that these categories often overlap.
• Neologisms are traced along a lexical life cycle that starts with nonce-formations. Productivity is shown to be limited by pragmatic, grammatical, semantic and conceptual factors.
• Methodologically, the book favours automatic or semiautomatic detection of neologisms over purely manual analysis. Corpora are checked against dictionaries to track codification.
• The reviewers criticise a heavy reliance on frequency and distribution data. They call for qualitative work on how varieties converge and diverge.

Interesting ideas

• I1. A new word passes through stages, from a one-off nonce use to an established and codified item, so its status should be recorded as a stage rather than a yes/no "known word" flag.
• I2. Word-formation types have fuzzy borders: one form can be both a blend and an abbreviation, or both a borrowing and a semantic shift.
• I3. Automatically extracted candidate neologisms are checked against lexicographic resources to separate truly new items from merely rare ones.
• I4. Varieties are treated as an ecosystem in which variety labels are unstable and words migrate into a common repertoire.
• I5. The reviewers warn that frequency counts alone cannot establish overuse or underuse, so qualitative patterns are needed to back up quantitative claims.

For Mova

How these ideas could serve Mova, our small open language model — its files, stages and tests:

• I1 (tokens): The lexicon in crate en could tag out-of-vocabulary forms by life-cycle stage, such as a single occurrence, recurring, or dictionary-attested. This would let the tagger explain how much it trusts its lemma guess.
• I2 (tokens): Morphological analysis of unknown words could produce several ranked formation hypotheses, each with its evidence. Forcing one label would hide real ambiguity.
• I3 (learn): Silver-data pipelines could flag tokens missing from WordNet or dictionaries as neologism candidates. These candidates could then go to an LLM teacher tier for verification rather than being treated as tagging errors.
• I4 (world): Compiled lexical knowledge could store regional origin as a soft, revisable attribute of a word sense rather than a fixed partition. Entities and concepts named with variety-specific words could then still link to shared ontology nodes.
• I5 (learn): Evaluations could pair aggregate metrics with targeted qualitative checks on texts from different English varieties. This would show whether parser or frame errors cluster on variety-specific vocabulary.