Dimensions of web registers: an exploratory multi-dimensional comparison

In this paper, we use corpus-based, multi-dimensional analysis ( Biber, 1988 ) to investigate the structural and functional linguistic variation in online language represented by blogs, micro-blogs, workplace e-mails, discussion posts and reader comments, and online newspaper and opinion columns. The primary goal of this study is to identify the statistically co-occurring features of online language and compare their distributions across web registers. The data used in our analysis came from an exploratory corpus of online texts (containing approximately 16,501,785 words) collected from various public domains from 2006 to the present. A combined automated and manual collection of texts was conducted across registers and the resulting corpus was grammatically tagged using the Biber tagger. Results show that the four primary functional dimensions of web registers are: (1) Personal Narrative Focus versus Descriptive, Informational Production; (2) Involved, Interactive Discourse; (3) Complex Statement of Opinion; and (4) Past versus Present Orientation.