📄 Peters, Pam, Peter Collins & Adam Smith (editors). Comparative Studies in Australian and New Zealand English: Grammar and Beyond. Varieties of English Around the World G39. Amsterdam/Philadelphia: John Benjamins 2009. 406 pp. ISBN 978 90,272 4899 2

Summary

This is Elisabeth Gustawsson's review of an edited volume (Peters, Collins & Smith, 2009). The volume collects corpus-based research on the grammar and usage of Australian English (AusE) and New Zealand English (NZE). It has two aims. The first is to map where the two antipodean varieties agree with each other and where they differ. The second is to test for endonormativity, meaning whether each variety has settled into its own national standard rather than following British or American norms.

There are 19 chapters arranged in five sections, moving from morphology through verbs, nouns and clauses to discourse. Most chapters rely on corpora such as ICE-AUS, and some add reader surveys from Australian Style. The reviewer judges both aims achieved. The volume finds features the two varieties share against northern-hemisphere English, features where they split, and several signs of local norms forming. She considers it a valuable reference work, though too advanced for undergraduate teaching. Its main gap is the lack of diachronic comparison, because no AusE/NZE counterparts to LOB/FLOB or Brown/Frown exist yet.

Key points

• The volume has 19 chapters by 21 contributors in five sections, framed by a Prologue (Peter Collins) and an Epilogue.
• Shared AusE/NZE features: non-numerical quantifiers like heaps of, non-reflexive myself, and -t past forms (burnt, leapt, learnt, spelt).
• Divergences and local norms:
  • The two varieties diverge in modals and quasi-modals and in no in negative collocations.
  • Signs of endonormativity appear in -ie hypocoristics, the progressive, and zero complementisers (leaving out that before a complement clause following a verb such as decide).
• Gender-neutral job titles: AusE leads British, American and NZE English in using terms like spokesperson and firefighter.
• Survey and corpus findings:
  • 56% of Australian Style readers chose the present perfect over the preterite in a "just" sentence.
  • Bloody turns up more often among London teenagers (COLT) than among Australians (ICE-AUS).
  • Quasi-modals (have to, be going to) occur more often in speech than in writing.

Interesting ideas

• I1. Corpus frequencies are cross-checked with speaker surveys, so usage counts and stated preferences can confirm or contradict each other.
• I2. Several authors warn that a frequency can reflect genre or writers' perceptions rather than real usage. One example is possessive me in fiction, which may appear because authors see it as vernacular.
• I3. Varieties are compared using matched reference corpora (COLT vs ICE-AUS), and the reviewer notes that diachronic claims need corpora built to the same design, like LOB/FLOB.
• I4. The volume lists concrete regional morphosyntax such as -t past forms and zero complementisers, which any processor of these varieties must handle.
• I5. Australian speakers prefer the perfect where others might use the preterite, and Hundt's chapter covers concord with collective nouns. Both show that surface grammar can vary while the meaning stays the same.

For Mova

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

• I1 (learn): Silver-data agreement between witness parsers could be checked against a small set of human acceptability judgements, which would catch cases where parsers agree on something speakers reject.
• I2 (learn): Evaluation sets should be split by register (fiction, biography, speech) so that a gain on one genre is not read as general improvement.
• I3 (learn): Treebanks for regression testing could be built as matched pairs that differ in one dimension only, which makes negative controls meaningful.
• I4 (tree): The parser should attach a ccomp correctly with no that marker, and the en crate should lemmatise burnt/learnt/spelt to their base verbs.
• I5 (world): Perfect and preterite forms with just should yield the same event ordering, and a collective noun with plural agreement should still resolve to a single group entity.