📄 DOWNING, A., y LOCKE, P. (1992): A Universitv Course in English Grammar.

Summary

This is Félix Sanz González's Spanish-language review of Downing and Locke's A University Course in English Grammar (Prentice Hall, 1992). The reviewer first places the book in a line of English reference grammars: Quirk et al.'s structural grammars, and Leech and Svartvik's communicative grammar, which added the missing focus on language use. Even the 1985 Comprehensive Grammar, in his view, could not settle every need. He then argues that advanced university learners need one usable book that combines systematic description, reference depth and tools for discourse analysis, and that Downing and Locke supply it.

The authors, both at the Universidad Complutense de Madrid, build on Halliday's systemic-functional grammar. They treat each clause as carrying three kinds of meaning at once: experiential, interpersonal and textual. They use the clause complex, rather than the written sentence, as the largest structural unit. The book is organized into 60 teaching modules with exercises, and the reviewer praises it highly. His one complaint is the lack of a bibliography. He values the book as a bridge from grammar to text analysis.

Key points

• The book has XX + 652 pages in three blocks:
  • one chapter introducing the basic concepts;
  • six chapters on clause and clause-complex structure (e.g. "The skeleton of the message");
  • six chapters on groups (verbal group; tense, aspect and modality; nominal group).
• The chapters are split into 60 modules. Each module is a self-contained teaching unit that opens with summary points.
• Situations are modelled as the speaker's conceptualization, not as external reality. Their components are processes, participants, attributes and circumstances (p. 3).
• "Sentence" is defined orthographically as the stretch between full stops (p. 19). The clause complex can extend beyond it, especially in speech, and is illustrated with an advertising text (p. 276).
• For comparison, the review gives the size of Quirk's grammars: GCE (1972) has 14 chapters in 1,120 pages, and the 1985 Comprehensive Grammar has 19 chapters in 1,779 pages.

Interesting ideas

• I1. Each clause is analysed through three independent function layers: syntactic (Subject/Object), semantic role (Agent) and informational (theme). Each layer has its own structure, and together they make up one realization.
• I2. The written sentence is treated as only a spelling convention, while the clause complex is the real structural unit, because pauses in speech do not reliably mark structure.
• I3. Syntactic variants with marked or unmarked theme can share a propositional meaning while differing in communicative significance, so each fits a given point in discourse differently.
• I4. The "situation" a text describes is the speaker's conceptualization, made of processes, participants, attributes and circumstances.
• I5. The exercises mix closed tasks with an answer key, open text-analysis tasks, and production tasks that ask the learner to write a text with the same features as the one analysed.

For Mova

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

• I1 (frames): prag could keep separate, parallel annotations per clause for roles, speech act/mood and theme/given-new, instead of merging them into one frame record.
• I2 (tree): sentence splitting could be treated as provisional, with links between clauses allowed across full stops, which would help with speech-like or advertising text.
• I3 (back): when generating from the meaning graph, word order and theme could be chosen from what is already given in the co-text, rather than always using canonical SVO.
• I4 (world): the world of the text could be modelled explicitly as the narrator's construal, which leaves room for viewpoints that conflict with compiled knowledge.
• I5 (answer): student evaluation could pair keyed tests with "imitate this text" tasks, checking whether a generated text reproduces the analysed structures.