A Reader in Sociophonetics

The Sociophonetics of Prosodic Contours on NEG 155

(3) I don’t wanna deny them their rights!—SWB 2709n.^17

Hedged, self-corrective or self protective tokens were coded, but in the runs
reported here, these factors were discarded from the Goldvarb analysis. The
analysis of these turn footings will be presented elsewhere, and only Sup-
portive, Remedial, and Informative footings are included in the analysis
reported here.

Table 5.5 Turn-Footings Coded in Study

CODE RECODE SIGNIFICANCE SAMPLE SENTENCE SOURCE

IIInformative the surveillance system is not that

sinister.

BUR News

SSSupportive I agree, they don’t write anything

like they use’to.^18

SWB 2281

RRRemedial He simply doesn’t know what he’s

talking about.

K/N

C -Self-Correct I don’t know- I don’t know the

immigration laws.

swb 2709

P -Self-Protect Y’know, I don’t wanna deny them

their rights!

swb 2709

H - Hedge If I’m not mistaken... ALL

Speci¿ c turn-footings coded in this study; after initial analyses, the ¿ nal analysis discussed

here includes only the ¿ rst three footings, with the others excluded from the analysis.

Other independent factor groups were signi¿ cant for one language or another.
However, those signi¿ cant as independent variables for at least two of the lan-
guages are those discussed here: Corpus/Situation/Stance (News, CallFriend),
Footing (Supportive, Remedial, Informative) Morphology (Full, Cliticized),
Sentence Position (End, Other), and speaker characteristics (gender, region).^19

4. The analysis

Once all the tokens were coded, and those tokens not included in this analysis
were discarded, a Goldvarb statistical analysis (Sankoff et al. 2005) deter-
mined the degree to which one or another factor group inÀ uenced the likeli-
hood of prominence. Table 5.6 displays those variables which the Goldvarb
showed to be a signi¿ cant inÀ uence on prosodic variation of NEG tokens,
which are discussed here. The Goldvarb weights are found on Table 5.7.