Time: 2:00am - 5:00pm Venue: People's Palace LG01
9:00am: Rebecca Starr
The multimodal construction of affective stance in ASMR performance media
9:45am: Marisa Brook
Internal layers and external forces in defining envelopes of variation
10:30am: Betsy Sneller
Social meaning and language change: A multifaceted approach
11:15am: Sophie Holmes-Elliott
Change spotting: diachronic and synchronic approaches to understanding language change
Over the past decade, a novel genre of performance has arisen among online communities devoted to the experience of Autonomous Sensory Meridian Response (ASMR), the tingling sensation felt by some individuals upon exposure to soft voices and other stimuli. The explosion in popularity of ASMR media provides sociolinguists with an unusual opportunity to observe the cross-linguistic social construction of a newly-enregistered sensory phenomenon that centers around the experience of the human voice. While prior research indicates that ASMR media is primarily consumed for relaxation and is not generally perceived as sexual (Andersen 2015; Barratt & Davis 2015), sexually-charged videos purporting to be ASMR have grown popular in certain regions, including mainland China. In 2018, the Chinese government issued a ban on ASMR media, citing the concern that ASMR was being used to covertly disseminate sexual content. Similar issues surrounding the nature of ASMR have arisen in the West, with online platforms beginning to restrict some ASMR media due to (mis)identification of their content as sexual. The question of how these performances are semiotically constructed, therefore, is not only of academic interest, but also of considerable practical relevance, as the work of online content creators, and particularly young women, is increasingly policed.
Focusing on the case of ASMR in China, this study draws on evidence from the acoustic analysis of suprasegmental features in combination with content analysis to argue that, rather than ASMR media being a vehicle for covert sexual content, conventional and sexual ASMR performances constitute two separate genres in which performers construct distinct personae to achieve different aims. While all performers index intimate affective stances, conventional ASMR performers construct the persona of a calm, gentle professional, while sexual performers evoke the childish, amateur camgirl and sajiao (‘whining’) persona. Chinese content creators are found to juxtapose local and non-local semiotic resources, illustrating the evolution of online performance genres via transnational cultural flows. The phenomenon of ASMR also highlights the sensual nature of the voice, and underscores the necessity of incorporating sensuality into theorizations of sociolinguistic meaning.
The conventional envelope of variation is a central concept in variationist sociolinguistics; Aaron (2010:2) defines it as "the overlapping space in which all of the variants under consideration may occur." For instance, in a study of present-day English quotative verbs, examples in passivized contexts – e.g. I was told, "This is the best!" – are excluded from the envelope of variation since quotative go, quotative say, quotative be like do not have direct objects and thus cannot be grammatically forced into a passive. Variables on the morphosyntactic level have become well-accepted (Lavendera 1978, Dines 1980, Sankoff and Thibault 1981, Romaine 1984; see also Pichler 2010, Hasty 2014) based on functional equivalence and/or near-synonymy (Sankoff and Thibault 1981, Tagliamonte 2006a:75-78). Conventional approaches to the envelope tend to focus on exhaustively identifying everything within the functional space (see e.g. Gudmestad et al. 2018:65). However, recent work suggests multiple reasons for greater nuance. One is influence from forms that overlap phonologically and semantically but belong to different variables (Aaron 2010, D'Arcy 2015, Dinkin 2016). Another is that defining functional equivalence in a meticulous, fine-grained manner can reveal internal layering (Wiltschko et al. 2015). I extend this line of thinking by arguing that although the fundamental concept of the envelope of variation is not misguided, there is ample room for gradience both within the envelope and outside it. To this end, I examine two case studies from Canadian English as represented by several large corpora of vernacular speech (Tagliamonte 2003-2006, 2006b, 2007-2010, 2010-2013, 2014; Tagliamonte and Denis 2014). One of these (Brook 2018) is a syntactic chain shift after the verb seem, suggesting nestedness of two or even three layers of variation, each of which spills over into the next towards the end of a change. The other (Brook to appear) is a deliberate micro-study of a tiny corner of a traditional envelope of variation (epistemic markers) demonstrating the potential inherent in focusing on only two variants of a variable. Although this approach demands considerable caution in order not to violate the Principle of Accountability (Labov 1972:71-72), it allows signals of subtle localized variation within a larger envelope to emerge. Both of these sets of results suggest that envelopes of variation on the morphosyntactic level are not the flat, inert structures that variationist practice sometimes accidentally ends up treating them as. Rather, they are dynamic, responsive units with considerable (and potentially informative) grey areas rather than rigid edges. These findings open the possibility of new approaches to spotting and evaluating co-variation within a variable, and/or that of quantifying how closely related two morphosyntactic structures are within the grammar of the speech community.
Naturalistic data – whether focused on rich ethnographic data or larger-scale corpus data – have been the cornerstone of sociolinguistic research since its inception in the 1960s, providing the primary data for sociolinguistic theory. Building on this foundation, more recent advances in computational and experimental approaches have offered additional perspectives and nuances to the theories drawn from naturalistic data. Through the combination of traditional, experimental, and computational approaches to specific sociolinguistic hypotheses, my work aims to provide a multidimensional view into longstanding sociolinguistic questions.
In this talk, I begin with an analysis of ethnographic research conducted in a neighborhood in South Philadelphia, demonstrating that white men who are in regular hostile contact with their African American neighbors have adopted the African American English (AAE) feature of (TH)-fronting (the pronunciation of /θ/ as /f/, as in “bofe” for “both”). This result is in itself somewhat surprising, as white speakers using features of AAE typically exhibit a positive affiliation towards Black folk or towards Black culture (e.g. Cutler 1997, Fix 2010, Sweetland 2002), while my participants exhibit overtly hostile attitudes toward their Black neighbors. I argue that, for these white speakers, (TH)-fronting is underspecified for ethnolect and is therefore available for straight-presenting white males to borrow as an index of “tough”. These findings have the following implications: (1) higher-order indexical variants are more likely to spread across dialect boundaries than first-order variants and (2) while the nature of indexicality is fluid and can vary between first and higher-order indexicality, these results suggest that it is possible for a higher-order index to eclipse the first order indexical meaning. In the second half of my talk, I present the results of a laboratory experiment designed to test the central hypotheses drawn from the Grays Ferry ethnography. Using an artificial language game (Roberts 2010), I demonstrate that borrowing across dialects propagates at a higher rate when a linguistic feature is a higher-order index than when it is a first-order index. In other words, a feature was borrowed more readily when it was both alienable (when it was associated with “tough” players rather than another species of player) and when it was socially relevant (in this case, when fighting was an option in game play). Additionally, we find that variants that were either alienable or socially relevant – but not both – were treated like first-order variants, suggesting that the difference between first-order indexicality and higher-order is categorical rather than gradient.
Taken together, these studies demonstrate how a diverse methodological approach enriches our understanding of sociolinguistic processes by enabling the joint generation, modeling, and testing of novel hypotheses.
Through the instrumental analysis of linguistic variation, quantitative sociolinguistics enables the observation of language change. Tracking patterns of change affords key insights into the nature of the linguistic system, language development and learning, and the organisation of broader social structures and processes. In this talk I present findings from diachronic and synchronic approaches that contribute to these issues.
My diachronic approach combines real and apparent time methodologies to examine the role that children’s developmental shifts play in ongoing change. These shifts involve a process of ‘vernacular reorganisation’ where young speakers move away from the parent-oriented models of early childhood and towards the peer-oriented models of adolescence, until they finally settle on their relatively stable adult systems. By tracking the developing systems of young speakers as they change in real time, it is possible to probe the limits of vernacular reorganisation, and how this in turn constrains patterns of change. For instance, to what extent can speakers reorganise the first acquired language? In other words, how plastic are young speakers’ linguistic systems? My findings indicate that while both superficial and deeper elements are susceptible to change, the exact nature of the modification depends on the broader patterning within the community. As well as informing debates surrounding the maturational nature of linguistic plasticity, these findings contribute to questions concerning the incrementation of language change, both in the increase of innovative frequencies and at a deeper structural level within shifting variable grammars across generations.
In addition, I take a synchronic approach to examine linguistic variation as it patterns across speakers and settings, using data from reality television. Here, I present findings that demonstrate how speakers’ use of variation across interactional goals creates a process of stance accretion. Furthermore, these results demonstrate the emergence of change through the calcification of sociolinguistic style. This research contributes more broadly to understanding the actuation of socially meaningful change.
Finally, I present results from recent work exploring the link between articulatory setting and the embodiment of sound change. These findings help explain how linguistic features spread across communities as they become attached to salient, recognisable and, crucially, imitable personae. In doing so, this analysis sheds light on the diffusion of change.