Language constructing indentity

The interaction between phonetic characteristics of voices and social responses to those voices is fascinating, and this week’s papers explored them in different ways.

            The King-Sumner paper showed that phonetic characteristics of voices affect interpretation of semantic meaning; in one of the experiments, for example, there was a significant difference in the top word-association that differed based on speaker voice. The format of the experiment reflects characteristics of the way people process voices without seeing faces, a situation that arises often enough for these neurological judgments (happening in less than 300ms) to be relevant.  The speed of these changes in how we interpret words and the unconscious nature of our listening makes me wonder: to what extent are we able to control the stereotypes and biases that result from hearing a certain speaker? Using a word pair from the paper, would I be able to choose not to associate “infection” with the word “yeast” if I hear the word said in a female voice?

            “The Social Weight of Spoken Words” shows that social representations are activated by voice cues “fast and early,” meaning they aren’t necessarily built up with experience that is numerous or deliberate (238). Interestingly, the paper notes that certain accents that are less common are less attended to (like a New York Accent), whereas the contents of “a prestigious accent (e.g. Southern Standard British English)” are remembered more exactly (239). My intuition is that a British English accent isn’t necessarily more commonly heard but may be more attended-to because it is distinctive. At what point does the distinctiveness of an accent stop increasing the amount the voice is attended to and become a deficit to how well-attended the voice is? At what point to non-standard accents begin to hinder listener interpretation?

            The Podesva paper examines a rich dataset of Condoleeza Rice’s speech and performs a sociophonetic analysis of speech patterns, attempting to identify different aspects of her identity (southern, western, African-American, conservative) within her pronunciations. The idea that phonetic characteristics within a speaker change how well they are attended to is strongly demonstrated in Rice’s speech, whether her phonetic patterns are conscious or not. Podesva found that Rice exhibits few characteristics typical of southern, western, and African-American vernacular speech (like no pin-pen merger) but instead uses elements of Standard African-American speech (vocalized r, glottalized d) and has a relatively high PVI and rates of –t release, both of which mark careful, enunciated speech. Rice’s speech reflects an identity she wants to create for herself: that of an educated, professional African-American with standard speech that appeals to a wide range of Americans. As Podesva put it, the phonetic qualities of her speech are “a symbolic embodiment of a particularly educated construction of the American Dream.” Although we know the biases experienced when we hear speech are to some extent unconscious, how conscious is the way we modify our speech? If Rice put no thought into her manner of speaking at all, would she exhibit more southern or western characteristics like pin-pen merger? How much control do we have over how our speech reflects our identity?