📄 Review of Nelson, Cecil L., Zoya G. Proshina & Daniel R. Davis. 2020. The Handbook of World Englishes. 2nd edn. Hoboken, N.J.: Wiley - Blackwell. 816 p. + xxx p.

Summary

In this book review, Ekaterina S. Lebedeva assesses The Handbook of World Englishes in its second edition (2020), whose editors are Zoya G. Proshina, Daniel R. Davis and Cecil L. Nelson. The original 2009 edition was produced by Braj B. Kachru, Yamuna Kachru and Cecil L. Nelson. The new edition revises and expands that volume to address criticism and developments that appeared after it.

The review walks through the book's nine parts:

• the history of English varieties across Kachru's Inner, Outer and Expanding Circles;
• contact linguistics, pidgins and creoles;
• acculturation, speech acts and genre;
• bilingual creativity and intelligibility;
• grammar, standards and corpus work;
• ideology and identity;
• media, advertising and commerce;
• policy, pedagogy, testing and lexicography;
• the paradigm's future.

Lebedeva concludes that the handbook is a landmark reference. It treats English as pluricentric and gives most of its space to non-native varieties. Its contributors come from all three Circles, and it opens many new research questions. The review matters as an orientation map for anyone who handles English produced far beyond Britain and North America.

Key points

• The second edition has 816 + xxx pages in 9 parts. The first edition (2009) was led by Braj B. Kachru.
• Pages on Outer and Expanding Circle varieties outnumber those on Inner Circle varieties by roughly 2.5 times. This mirrors the fact that non-native varieties outnumber native ones.
• The history is organised as successive "Diasporas":
  • the British Isles origins;
  • the Inner Circle (USA, Canada, Australia, New Zealand, Caribbean);
  • South and South East Asia and Africa;
  • Chinese English and Russian Englishes as a "Fourth Diaspora".
• Named contributors include:
  • Rajend Mesthrie (contact), Salikoko S. Mufwene (pidgins/creoles) and M.A.K. Halliday (standard vs global language);
  • Yamuna Kachru (speech acts, nativization) and Vijay K. Bhatia (genre);
  • Alexandra A. Rivlina (bilingual creativity) and Larry E. Smith & Cecil L. Nelson (intelligibility);
  • Gerald Nelson (corpus linguistics), Tej K. Bhatia (advertising), Margie Berns, Aya Matsuda and Kingsley Bolton.
• Variety descriptions cover phonology, lexis, syntax, culture and pragmatics. A recurring theme is English's growing role in education, tourism, trade and advertising.

Interesting ideas

• I1. English should be treated as pluricentric, with several legitimate norms rather than one correct standard and deviations from it. Labelling non-Inner-Circle forms as errors throws away information.
• I2. Mesthrie warns that literary texts by creative writers may not faithfully represent how a community actually speaks. Corpus genre therefore shapes conclusions about a variety.
• I3. Speech acts and rhetorical strategies differ across the Outer and Expanding Circles, through nativization of English and Englishization of local languages. Politeness and indirectness conventions are therefore culture-dependent.
• I4. Gerald Nelson's chapter argues that variety description should rest on comparable corpus evidence, which matches Mesthrie's call for a shared comparative database.
• I5. Bilingual creativity and advertising often mix English with other languages at several levels (words, morphemes, script), producing deliberate language play.

For Mova

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

• I1 (tokens): Lexicon and morphology entries could carry a variety tag, so that forms typical of Indian, Nigerian or Russian English get valid lemmas and tags instead of "unknown/misspelling" fallbacks.
• I2 (learn): When silver data or evaluation sets are built for a variety, fiction should be kept separate from conversational or news text, so parser quality on "Variety X" is not measured on unrepresentative literary prose.
• I3 (frames): Speech-act and implicature rules in prag could be parameterised by cultural or variety context, so a request or refusal phrased by local conventions is not misread as rude, literal or ironic.
• I4 (learn): A comparable multi-variety test suite could serve as a robustness control, exposing whether parser or reader accuracy collapses outside the Inner Circle.
• I5 (tokens): The tokenizer and tagger could explicitly mark mixed-language or hybrid tokens as foreign/code-mixed rather than forcing an English analysis, keeping downstream trees honest and explainable.