Orð og tunga - 01.06.2014, Page 86

74
Orð og tunga
ferent from other Facebook communication, it constitutes a suitable
corpus for linguistic observation. Because groups consist of a limited
number of members that interact with each other in various com-
munication "pairs", it provides a well-defined and yet multifaceted
source for the collection of data. Private groups moreover allow for
data collection that complies with ethical standards. Their members
are generally more active and their number lower so that it becomes
easier to obtain the permission of all group members to use the data
for research purposes.
3 Data and Methodology
The rest of this paper focuses on a case study concerning characteris-
tic properties of Icelandic CMC in order to depict and analyze CMC-
inherent properties and to define it as a potential new variety of writ-
ten Icelandic.
3.1 Corpus
The corpus introduced here is an active private Facebook group4 with
24 participants aging between 21 and 31. Three members are non-
native speakers of Icelandic, and have therefore been excluded from
the analysis. This leaves 562 posts and comments, published between
June 19th, 2012 and April 25th, 2013. The total number of analyzable
tokens is 5,310. It is important to note, that the group members know
each other personally which might influence the interaction.
Although the corpus is rather small, it allows an initial overview of
the diversity and the potential of Icelandic CMC.
3.2 Methodology
In order to facilitate the analysis, the corpus has been downloaded
and sorted into utterances and further into lists of tokens. The lists
have been analyzed according to the themes proposed by Androut-
sopulos (2007): conceptual orality, semiotics of compensation, linguistic
economy and graphostylistics. In so doing, properties corresponding
4 The members have been asked for permission to use the data for research pur-
poses. They have agreed without exception.