Skin-tone modified emoji and first-person indexicality

ABSTRACT

Beginning in 2015, people have been able to transform many emoji – typically 18 byte, 12 × 12-pixel images inserted into digital text – with “skin-tone modifiers.” Racialized aspects of self-presentation have a long history of being marked in various ways in semiotic practice. However, this article argues that, parallel to other systems of social indexicals like honorifics and gendered speech, skin-tone modified emoji represent a robust example of the complex ways language and culture are bound together dialectically. Based on the views of 451 anglophone American respondents to a survey, I demonstrate that the selection of emoji – even yellow emoji – can appear as a social and political choice for certain speech communities. For these individuals, the addition of skin-tone modifiers in the emoji set may remove the possibility of remaining outside this system of author identification when using signs that have the potential to bear such modifiers.

KEYWORDS:

• emoji
• racialization
• speaker-focal indexicality
• multimodality
• digitally-mediated communication

Acknowledgements

Versions of this paper have benefited greatly from the close readings and generous feedback of Lisa Bastarache, Ella Butler, Hannah Chazin, Catherine Egenberger, Patrick Haltom, Maira Hayat, Rachel Howard, Britta Ingebretson, Sharese King, Giovanni Ricci, Adam Sargent, and Anna Weichselbraun. I would also like to thank Andrea Shin and all of my survey respondents for their generous help with this project. This paper could not exist – to be sure – without the dedicated care I received as a graduate student and young scholar from the newly ascended master, the inimitable Michael Silverstein. All errors, of course, remain my own.

Disclosure statement

No potential conflict of interest was reported by the author.

Notes

1 Discourse on digital blackface focuses in particular on the sharing of animated GIFs, which – though an important feature of digital communication – is outside the scope of this paper.

2 STME may be used as resources to perform racialized identifications, with individual users selecting from different sets across interactions to accomplish different performative tasks. This is discussed in more detail in Section 4.4.

3 Data obtained through online surveys and AMT recruitment has been shown closely to approximate the reliability standards met by offline surveys (Shank 2016).

4 Note, however, that respondents did encode the perceived gender of the authors in their glosses, as this is a dimension of social identification that is grammaticalized in the (spoken and written) English language. Additionally, in blinded tasks, respondents habitually imputed the genders of emoji to their authors. In short responses, the majority of participants in the survey (83%) said they had never used an emoji whose gender did not correspond with their own, with the majority of those saying their reason was that it didn’t conform with their biographical identity (48%) or that it had never occurred to them to use a non-concordantly gendered emoji (44%). While these data support the race-related conclusions of this paper, gendered emoji could be a useful site for further investigation of the topics discussed here.

5 I borrow the term epicene from the linguistic analysis of gender as I see a parallel discourse around “epicene” gender pronouns in English, for example, singular they (e.g., Silverstein 2003).

6 In blinded tasks, respondents rarely imputed Asian or Hispanic/Latinx identification, unless the respondent him- or herself identified as Asian or Hispanic/Latinx, respectively. This appears to be a further embedding of a broadly circulating American racial ideology called the “bipolar model” of racialization (Omi and Takagi 1996). Under this model, individuals either conceptualize Black and White as the essentialized poles of a racial continuum or they simply overlook all other potential racial identifications. This model is significant in considering which assumptions are in play in the interpretation of STME and may explain what seem to be effects of emblematization, in which T1-2 and yellow emoji are conventionally linked to White authors while T5 and T6 emoji are conventionally linked to Black authors, separate and distinct from an iconic representation of phenotype. The unidimensionality of the STME from light to dark, T1-2 to T6, may be tracked onto this bipolar model for certain readers.

Additional information

Notes on contributors

Colin Michael Egenberger Halverson

Colin Halverson, Ph.D., is a linguistic anthropologist at Indiana University–Purdue University Indianapolis and a faculty investigator at the Center for Bioethics in the IU School of Medicine. His work focuses on ethics and communication, with a particular focus on semiotics and indexicality. He is a graduate of the University of Chicago, where his dissertation investigated the semiotic ideologies informing doctor – patient communication in a medical genomics clinic in the American Midwest and the ways in which commensuration of these different forms of talk and different life worlds was accomplished through a process of “simplification.” He has also published on the morphosyntax of gene nomenclatures and other aspects of gene naming from an historical and historical linguistic perspective.