TopAs Squires (2010) puts it, much discussion of computer-mediated communication has started from the idea that there is a specific type of language associated with communicating online. Crystal (2001) used the term netspeak to describe communication in online environments, and this has led to a perception among commentators, especially in the media, that there is a special language used online. This is a massive over-simplification of the situation, however, and the reality is that every communicative situation is unique, plus that the non-standard features of CMC are seen in other types of language use, as Crystal (2001, 2008) himself argues. A similar point is made by Tagliamonte and Denis (2008) and Squires (2010).
No matter whether we communicate in speech or writing, by writing an email or talking with friends, we adapt our language to that specific situation. We as individuals are different from one communicative act to the next, and so we may change the language we use from one situation to the next, even if these situations appear to be the same. We all can use many different types of registers of language. We speak and write more formally when applying for a job or in an interview, and less formally when talking to friends or writing a diary. Agha (2005: 47) describes three aspects related to registers. Firstly, there are the characteristics of the repertoire itself, i.e. its linguistic features; then, the social range, which consists of the stereotypical social situations in which the register is used; and finally, the social domain of the users of the register, which sets out who can use that register.
When characterising CMC as a register, Thurlow (2003: §4.2.1) refers to the sociolinguistic maxims of CMC, which are: brevity and speed, paralinguistic restitution (i.e. emoticons which make up for the lack of body language and intonation), and phonological approximation. Thus, reduction is one manifestation of the need for efficient, fast communication. For Androutsopoulos (2007), networked writing, which is what he calls computer-mediated communication, is shaped by four conditions. One is that it is vernacular (typically located outside institutional settings like education). Next, it is interpersonal and relationship-focused (not subject-oriented). Then, it is unplanned and spontaneous; and finally, it is dialogical and interaction-oriented. Thus, computer-mediated communication varies according to the context, and therefore, one instance of CMC may look very different from another, precisely because of the context.
Bakhtin (1981, 1984) writes about the voices that are encoded in language, and these can be either individual or social. Agha (2003, 2005, 2007) went deeper into the idea of social voice, and discussed the enregistering of uses of language. These are particular stereotypes of how language is used in particular social situations, and are “… socially recognized (or enregistered) as indexical of speaker attributes by a population of language users” (Agha, 2005: 38). It is crucial to remember here that these are only stereotypes. As with all language use, we do not always “play by the rules” of how we should speak or write. We can decide to follow the standard conventions, or can choose to be outsiders and deliberately break them. We can even overuse particular features of registers to mock them (see, for example, Squires’, 2010 and Lange’s, 2015 examples of meta-pragmatic comments on netspeak). There may be some conventions that are associated with using language online, but users can choose freely whether or not to follow these conventions depending on the situation and their purpose. In work on Complexity Theory (Larsen-Freeman & Cameron, 2008; Larsen-Freeman, 2012), individual variations of this type are described as the norm in communication, especially an expression of individuality. In terms of computer-mediated communication as a register, much has been written on its linguistic features. However, the process of actually setting such norms of language use have really not been explored.