At one level, it is extremely easy to define the linguistic identity of e-mail as a variety of language; at another level, it is surprisingly difficult. The easy part lies in the fixed discourse structure of the message – a structure dictated by the mailer software which has become increasingly standardized over the past twenty years or so. Just in the same way as we can analyse the functionally distinct elements that constitute a newspaper article (in terms of headline, body copy, illustration, caption, etc.) or a scientific paper (in terms of title, authorship, abstract, introduction, methodology, etc.), so we can see in e-mails a fixed sequence of discourse elements. They will be so familiar to likely readers of this book that they need only the briefest of expositions. The difficult part, to which the bulk of this chapter relates, lies in the range of opinions about the purpose of e-mail, as a communicative medium, and about the kind of language which is the most appropriate and effective to achieve that purpose. With around 80 billion e-mails being sent every day (in 2005), a consensus seems unlikely, especially when age, sex, and cultural differences are taken into account. At the same time, it ought at least to be possible to identify what the parameters of disagreement are, to develop a sense of the range of linguistic features which any characterization of e-mail would have to include.