Home » Media & technology

Variations in IM Speech

Written By: nicoletta on December 9, 2009 1 Comment

Hi everyone. I’m another student from French 24 writing about my final project for the class.  Over the course of the semester, we were asked to read Naomi Baron’s Always On. In her book, Baron discusses in some degree of detail the ways in which language is utilized and modified in various forms of communication, including instant messaging.  The “IM Speech” chapter of the book touches upon several of the more common and, to many of us, more obvious idiosyncrasies of IM conversation, such as standard abbreviations (lol, brb); emoticons, spelling and punctuation errors, and transmission length within a single utterance.  As a frequent user of instant messaging, however, I found this section of Baron’s book to be very simplified account of a much more complex linguistic phenomenon.  For my project in this course, I decided to take a closer look at the ways in which different people expressed themselves using instant messaging, identify variations, and try to guess where these differences come from.  Do do this, I chose four friends who use IM often and began keeping an archive of all of our conversations.  The entire conversation logs will not be posted here, but what I observed generally was a great deal of variety, creativity, and individual expression within IM speech. Rather than simply being a code for actual speech, IM speech has many unique characteristics that cannot be replicated in spoken language.  For example, one transmission from a friend is as follows:

Ha! Good day –> bad day –> good day

Another set of transmissions looked like this:

1: ohaider

2: ohai2u

1: howru

2: <33

Such examples appeared frequently and consistently in my conversation logs.  In summary, I believe my project to be a small piece of evidence for the existence of virtual idiolects, which are just as various, complex, and unique as their spoken counterparts.

Tags:

Digg this!Add to del.icio.us!Stumble this!Add to Techorati!Share on Facebook!Seed Newsvine!Reddit!

One Response to “Variations in IM Speech”

  1. Youki on: 10 December 2009 at 7:56 am

    very interesting! I agree, IM language is very complex and usually treated too superficially, as though a dictionary-type translation is sufficient to explain the cultural aspects of the language. Have you thought about how memes influence IM speech? By memes I mean the tendency to replicate certains forms of speech that originated externally from that IM speech community, like the set of “hai” phrases in your data. I see it quite often, and it’s really interesting the different types of identities it projects. For example, I have friends who primarily speak Japanese/Chinese and use phrases like “hai” (hi) and “bai” (bye) because it feels more natural to them and it projects a type of “non-native speaker of English” identity. I also have English-speaking friends who use it in a more comical sense, such as saying, “hai guyz, whats going on here?” as a sort of hedging device to lessen the impact of the query (sort of like, “don’t take my question too seriously”). And if you read gaming forums you’ll see it used in a hundred different ways. In each case, it’s not just about the language that is being spoken from a structural perspective, but also about the virtual identities that are being constructed (and co-constructed).

    You mention virtual idiolects, but what do you think of virtual dialects? The term “ohaider” may have a specific meaning within your community of IM speakers, but it’s likely that people are drawing from forms of speech they’ve seen in other speech communities. How did they learn to speak that way, and how did it integrate itself into your IM speech community? I’m curious what you find out.

Leave a Reply:

You must be logged in to post a comment.

  Copyright ©2009 Found in Translation, All rights reserved.| Powered by WordPress| WPElegance2Col theme by Techblissonline.com