I am intending to conduct a critical linguistics research using corpus linguistics as a method, particularly, the analysis of semantic preference analysis to word-pair patterns (henceforth collocates).

The reason to why I use semantic preference in contrast to semantic prosody is that McEnery & Hardie (2012) has pointed out that semantic prosody research results are often highly influenced by intuitive knowledge of its researcher about reality, and moreover it only focuses as to whether a collocate has "negative" or "positive" aura. In contrast, semantic preference seeks to see collocates on the basis of their semantic attributes (varied categories).

However, Hunston (2002) explains that most of critical linguistics research are done by deploying semantic prosody analysis, therefore an "implicit meaning" of a discourse can be shown to light and can be categorized based on its "negative" or "positive" association.

So, in short, can I used semantic preference analysis in critical linguistics?


  • Prosody semantics is merely the study of how meaning is created using stress and intonation. And semantic preference analysis is just that: Analysis of this word over that word. So, you will have to use an oral corpus and learn how to use stress and intonation marks to show meanings, and probably recordings of that oral corpus as well.
    – Lambie
    Jan 21 at 20:51
  • Hi, thanks @Lambie for the reaction. Was busy with life and upon researching the topic turns out that using semantic preference for CDA is acceptable and have been conducted before. However I ended up sticking with the notion of discourse prosody as introduced by Stubbs for corpus-based semantics research. Thanks for the addition!
    – pindakazen
    Feb 11 at 10:59
  • I don't think you understood what I said: blog.oup.com/2023/03/semantic-prosody hen linguists talk about prosody, the term usually refers to aspects of speech that go beyond individual vowels and consonants such as intonation, stress, and rhythm. Such suprasegmental features may reflect the tone or focus of a sentence. Uptalk is a prosodic effect. So is sarcasm, stress, or the accusatory focus you achieve by raising the pitch in a sentence like “I didn’t forget your birthday.”
    – Lambie
    Feb 11 at 14:39
  • My main point is that prosody involves speech and you have to use files taken from speech, not files taken from writing. I never said that semantic preference is not relevant to CDA. Your corpus will be of spoken language because you can't do it with writing.
    – Lambie
    Feb 11 at 14:41


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Browse other questions tagged or ask your own question.