Journal Summary on Computer Mediated Communication (CMC)
This summary is based on an article entitled ‘Individual Usage: a corpus-based study of idiolects’ by Michael Barlow from University of Auckland, New Zealand. Idiolect is a popular idea among linguists distinguishing alarge gap between the familiarity of the concept and lack of empiricaldata on the phenomenon. Insufficient data to properly consider the issue of regularity and stability of individual grammars, although there are signs of a surprising consistency over a period of a year or so is a problem to this study and many other studies pertaining to this issue. As such, studies done can only be regarded as a representation of a very minute example of the real-time situation. Hence, results of this study cannot be made as suggestion as to whether the idiolects or sociolects are the primary unit of language. Undoubtedly, the variation of individual grammars are quite extensive and it seems unlikely that the view of an idiolect as an individual's share of the sociolect can be upheld, taking into account the lexical and syntactic usagefound in the analysis presented.
This study examines the speech of five White House Press Secretaries and show that individual differences involve a wide range of core grammatical constructions instead of just the use of a few idiosyncratic phrases. Results obtained suggest that idiolectal patterns are consistent and valid for a certain period of time.
The working hypothesis for this investigation is that the variation that distinguishes individual speakerslies in the profile of the central components of lexico-grammar, and not only in some idiosyncratic peripheral phraseology. Taking a cognitive sociolinguistic position and exploring the issue of variation in idiolectsempirically from the standpoint of usage-base account of language. Frequencies of use of lexical and syntactic patterns show that individual patterns appear to be stable over a period of at least a year or two, individual patterns of variation are maintained despite differences from general usage within thecommunity, inter-speaker variation is typically greater than intra-speaker variation based on samples taken over different time periods or idiolectal variation is based on core aspects of language and not on peripheral idiosyncrasies.In order to reduce the likelihood of priming from one day to the next, transcripts from consecutive dayswere not chosen, giving a time interval of two or three days between speeches. The five press secretaries chosen for the study are: Mike McCurry (1994-98), Ari Fleischer (2001-03),Scott McClellan (2003-06), Tony Snow (2006-07), and Dana Perino (2007-08). The choice was madeon the basis of the length of their tenure since one goal of this study is to work with reasonably largesamples of idiolects taken over the course of several months at a minimum. The sampled data coversaround a year or more of speech output for each speaker.
Data were analysed by collecting bigrams, which are used as indicators of grammatical patterns and distinguishes one speaker from another.The outline of the structure of the discourse is retained only with respect to turn taking. The actualcontent of the speech of the reporters or other administration officials has been removed. Thus the datahas only one use: the analysis of the utterances of each press secretary. All interactions not involvingthe press secretary are indicated by <Q>, even in those cases where the contribution is a statement by aWhite House Official and not a question. Eliminating the contributions of the interlocutors makes thengram and concordance analysis quite simple and robust.The speech samples were split into text files containing 200,000 words each. For two speakers there is just a single 200,000 word file, but for the other speakers there are three, four and six such files,enabling comparisons between the speech of one individual at different times and the speech of different individuals.The files were tagged for POS using the CLAWS7 tagset and for semantic tags using the USAS tagset.The tagging was performed using the tools at the Wmatrixsite,http://ucrel.lancs.ac.uk/wmatrix/,(Rayson 2008). The files were analysed using the concordance program MonoConc Pro and thecollocationor n gram analyser Collocate.
Through the study, we can say that speech pattern is recognisable and differences in idiolects are not due to a few idiosyncratic phrases, but are due to differences inthe preferences in the use of grammatical constructions in addition to differences in the use of common phrase. The use of CMC in data collection and analysis help linguists and researchers to support the view that frequency of use of constructions is an integral part of idiolectal grammars.
No comments:
Post a Comment