Heteroglossia - two examples

Example 1
Hoe maakt u het Met Gamma Gamma	Relying on conversation analytic concepts, this advertisement can be said to consist of a adjacency pair with a first and a second. Heteroglossia occurs because the advertisement creatively draws upon two possible contextual insertions of the first, "hoe maakt u het?". One is part of a routine greetings exchange ("how are you?" - the most straightforward context for the utterance, one which is routinely replied to by "fine"), whereas in the second context, the first counts as a request for information ("how do you (manage) to make it?"). Although the sequence which is realised is the second "how do you make it? - With Gamma", the first is likely to be read initially as a greeting. So, the advertisement plays with a discrepancy between pro-active and retro-active logic in a conversational sequence. Such is made possible because the utterance "hoe maakt u het?" can be read in two ways and this goes together with two different sequence types. Note, finally, that a reversal of the order results in a perfect first of a greetings pair, i.e. "Met Gamma. Hoe maakt u het?" - a standard telephone greeting when answering the the phone ("This is Gamma. How are you?").
Data recorded on 20 October 2000, outside entrance of local Gamma branch in Gent. Gamma is a DIY-chain.

Example 2
[...] it turned out that there was a further four weeks before the child was taken to hospital for an appointment on being taken there erm (1) the hospital felt that this was a clear picture of failure to thrive (2) the child was as I recall off the top of my head I think it was two and a half kilos underweight was very dehydrated and in fact had the situation been left for longer the child would have died the child was admitted and what then happened was that the mother [...]	This bit of data has been taken from a situation where a social worker is being interviewed by a senior colleague about a particular case where a baby during a routine check was diagnosed as failing to acquire the appropriate weight for its age. The interview is part of a policy review in the domain of child care. There are two strings of speech representation in this excerpt (numbered 1 and 2). The two strings are rhetorically implicated in a structured display of evidence: (1) gives the conclusion ("a clear picture of failure to thrive"), while (2) makes available the detailed observations which warrant this particular conclusion ("the child was (...) was two and a half kilos underweight (...)"). The reported speech therefore counts as a rhetorical device which must help persuade the listener that the institutional intervention was appropriate but it also shifts the responsibility for stating the seriousness of the situation away from the social worker. In the first string, the diagnosis "a clear picture of failure to thrive" is attributed to a voice of medicine, referred to collectively as "the hospital". By introducing the medical voice, the social worker evokes an interprofessional working relationship with a division of responsibilities in the domain of child care and corresponding institutional action (medical diagnosis versus assessing the social circumstances). Within the context of the policy interview, the attribution helps justifying the classification of the child as being potentially at risk. Additionally, note how the specific way in which the reported medical voice is being projected invests it with credibility. This is especially clear from the second string of speech representation. Social work voice and medical voice become one: although clearly citing medical facts, he attributes them to his own memory ("as I recall off the top of my head"). One or two implications of this analysis: Whenever one represents what someone else has said or written down, a network of social relationships comes into play, in this case: social work and the medical work, but also the relationship between the social worker/reporter and the interviewer/reportee. While grammatical/stylistic studies of reported speech will tend to be mostly concerned with the classification of the speech representation strings as belonging to a particular category type (e.g. direct speech, indirect speech, free indirect speech, etc.), such taxonomies may still say very little about why people appeal to other voices in the first place or about the functions which such reporting fulfils in a particular context of use.
Example and analysis borrowed from Baynham & Slembrouck (1999: 439-440 )