To the top

Page Manager: Webmaster
Last update: 9/11/2012 3:13 PM

Tell a friend about this page
Print version

Analysing unexpected form… - University of Gothenburg, Sweden Till startsida
To content Read more about how we use cookies on

Analysing unexpected forms of potential question tags in the spoken component of the British National Corpus

Conference paper
Authors Karin Axelsson
Published in Pre-conference workshop on errors and disfluencies in spoken corpora held in conjunction with ICAME 30, Lancaster U.K., 27 May 2009
Publication year 2009
Published at Department of Languages and Literatures
Language en
Keywords tag questions, spoken language, corpora, errors
Subject categories English language


In my thesis project, I compare the use of ‘canonical’ question tags in fiction dialogue and spoken conversation, using data from a BNC fiction subcorpus and the demographic part of the spoken component of the BNC. I first searched for and analysed question tags in fiction, finding it quite unproblematic to apply a conventional definition of question tags (N.B. anchor + question tag = tag question): the tag subject should be co-referent with the anchor subject, the operator should be the same as in the anchor or a form of do if there is a lexical verb in the anchor, and the tag operator should have the same tense and number/person properties as the anchor verb: (1) Well, these things happen, don’t they? Using the same definition in the demographic part turned out to be more problematic, as some phrases which seem to be intended as question tags do not meet the criteria in the definition. These examples can be divided into three categories. The first category is invariant question tags. The tag has a canonical form, usually isn’t it or innit, but it is connected to an anchor with a subject with other number/person properties and/or a finite other than is: (2) you pay much more though isn’t it? The second category may be called modified question tags. These tags have an unexpected form but are logical as there is a plausible alternative wording of the anchor with the same meaning which would make the form of the question tag acceptable. This conforms to the hypothesis that we tend to remember the content of speech rather than the exact wordings. Some examples: (3) But that can be the same anywhere couldn’t it? [alternative: could] (4) Oh it [i.e. the price] was sixteen nineteen nine were they? [alternative: they were] (5) Oh that’s why I came up wasn’t it? [alternative: was] (6) [about a photo] this is at the zoo was it? [alternative: was] The third category is erroneous question tags, i.e. tags where the speaker makes an error in the wording, but where it is still clear that a question tag is intended, as in example (7), which was uttered by a four-year-old girl: (7) you count, one, two, three, four, don’t it? Instances in these three categories are included as question tags in my study. Excluded are question-tag-like wordings involving a new proposition, as in (8) where the change of operator adds a new meaning when the speaker asks for advice: (8) I won’t open another one, should I? and disfluencies, where there seems to be no logical connection to the preceding discourse: (9) Big fancy does are you? Unexpected forms of potential question tags thus require manual analysis; the context is very important here, especially the reaction from the addressee. However, a small number of instances are still problematic. This is sometimes due to the lack of prosodic marking, which can make it difficult to decide whether a tag-like wording is really a question tag or the beginning of a new question.

Page Manager: Webmaster|Last update: 9/11/2012

The University of Gothenburg uses cookies to provide you with the best possible user experience. By continuing on this website, you approve of our use of cookies.  What are cookies?

Denna text är utskriven från följande webbsida:
Utskriftsdatum: 2019-12-07