The length of individual messages was calculated using the standard Microsoft Word 'word count' function. This was a somewhat crude calculation since it was unable to distinguish lexical items conjoined by a punctuation mark (e.g. i'll be there later today.what time are u coming?); however, in its favour, it did include individual-character lexemes such as the 'u' in the same example. On this basis, the average length of text-messages was approximately 14 'words'. Compared with the average length of turns in online chat (i.e. six words - Werry, 1996), the messages of participants were certainly longer which was to be expected from the kind of asynchronous communication afforded by SMS. However, given the standard restriction imposed on the length of text-messages (i.e. usually about 160 characters, including spaces), it was also interesting to note that the average length of participants' text-messages was only 65 characters (Md = 55, Mo = 13, 23, 39), although with quite a lot of variation (SD = 45). While much is made about the technologically imposed need for brevity in SMS, our participants' messages seldom used the space available . As such, the length (and abbreviated linguistic forms) of messages would therefore seem instead to be a function of the needs for speed, ease of typing and, perhaps, other symbolic concerns. Indeed, as others have noted elsewhere (for Finland: Kasesniemi & Rautiainen, 2002; for Germany: Rössler & Höflich, 2002), young people appear increasingly to be employing SMS for more dialogic exchanges - especially when the costs are lower as is the case in Finland. In this sense, therefore, the language of SMS starts to look much more like the 'interactive written discourse' of a conventional CMC niche like IRC (Werry, 1996:48). We return to this point later.
With obvious implications for linguistic practice, it is worth noting here that some mobile phones enable 'predictive text'. Users need only press once on the keypad number corresponding to the letter and, as long as the desired word is already stored, the phone should recognize and complete the word automatically. When asked by Brown (2002), however, only about half (55%) of our participants reported using this facility - mainly because it was thought to be quicker and easier. On the other hand, reasons given by those 37% who said they didn't use predictive text included, in order of priority, that it was too difficult to use, they did not actually have the facility to start with, it was annoying, it did not choose the right words, it was slower to use and did not facilitate the need for abbreviations.
Following the kind of typology offered by Shortis (2001), in Table 1 (PDF version for download) is listed every different example found in participants' transcribed messages of what might reasonably be regarded as non-standard orthographic and/or typographic forms. For the purposes of descriptive overview rather than quantification, these are shown organised into the following broad categories: (1) shortenings (i.e. missing end letters), contractions (i.e. missing middle letters) and G-clippings and other clippings (i.e. dropping final letter), (2) acronyms and initialisms, (3) letter/number homophones, (4) 'misspellings' and typos, (5) non-conventional spellings, and (6) accent stylizations. [note 8]
Heavily abbreviated language is of course also a generic feature of interactive CMC niches like IRC's online chat and ICQ's instant messaging and we were not surprised to see that 82% of participants had already reported using abbreviations in their text-messages, especially the women (F = 89%; M = 57%) (see Brown, 2002). However, in looking at their actual text-messages, only 1401 examples of abbreviations were found - about three per message - which meant that abbreviations in fact accounted for less than 20% (18.75%) of the overall message content. As we discuss shortly, this initial finding immediately appears to run counter to popular ideas about the unintelligible, highly abbreviated 'code' of young people's text-messaging.
In the same vein, only 509 typographic (as opposed to alphabetic) symbols were found throughout the entire corpus - almost all of which were simply kisses and exclamation marks usually in multiple sets (e.g. xxxxxx and !!!!!). There were also only 39 instances of emoticons (e.g. :-) ) (See Table 2 - PDF file).
M1: YO YO YO HESS WOZ UP IN DA HOOD?!HOW IS MAZZAS?WHEN U GOIN BACK?LOVE ME X
There were also relatively few (n = 73) examples of language play using letter-number homophones (e.g. Gr8 'great', RU 'are you'), which, in popular representations at least, have become the most definitive feature of text-messaging (see, for example, Image 1 - PDF download). Like many of the paralinguistic and prosodic cues found in IRC by Werry (1996), a much more frequent type of language play in the test-messages was found in the form of accent stylizations or phonological approximations such as the 'regiolectal' (Androutsopoulos, 2000:521) spelling novern for 'northern' and those in [M1] above. In addition there was a range of onomatopoeic, exclamatory spellings (e.g. haha!, arrrgh!,WOOHOO!,t'ra, Tee Hee, Oi oi savaloy!, yeah, yep, yay!, rahh, ahhh, mchwa!, eh?, and woh!) and a couple of other typographical-cum-linguistic devices for adding prosodic impact (e.g. quick quick, wakey wakey, wotcha, and yawn…). Unless used in marked isolation, it was not possible to determine if the use of capitalization such as in [M1] was used deliberately for prosodic effect or if, as we suspected here, it was the sender's personal style preference to send all their messages in capital letters anyway.
Finally, as a passing reposte to journalist John Humphreys (see above), there were in fact 192 apostrophes used across the 544 messages (e.g. we're, she's, can't, I'm, it's). Accounting for about one in every three messages (or 35% of them), their occurrence seems surprising given the technological imperative for speed and ease of 'typing'. Without anything to compare it with, we do not wish to make any serious claim for this figure, except perhaps to say that, as far as the supposedly solecisitic participants in the current study were concerned, it does not appear that the apostrophe is quite dead!
Language is always multifunctional and always dependent on context for its meaning. As such, it was not always possible to be certain of the meaning of some participants' messages and even less so the communicative intent with which they were sent. In looking to code their text-messages, however, we identified what we regarded as the primary functional orientation of each message; on this basis, individual messages were assigned to nine broad categories, including an additional category for chain messages. The multifunctionality of the messages was also retained to some extent by coding them in terms of more than one category where relevant (n = 121 messages, 22%). These functional categories are not strictly exclusive of each other and present a largely descriptive overview.
In order to render this process of categorization as explicit as possible, brief descriptions of each category are given here together with two or three example messages. Throughout the paper, original messages are indicated in a different font and colour, and, for ease of reference, are numbered consecutively. All messages have been anonymized.
Messages in this category dealt primarily with the exchange of practical details or straightforward requests for information.
M2: Where's sardinia?Answer me quick hun! xx
M3: Put money in ur account
Messages in this category dealt primarily with more solidary information exchanges or requests for personal favours.
M4: I Passed
M5: I'm not feeling v well can you get the lecture notes for me please
Practical Arrangement Orientation
While perhaps implicitly recreational, messages in this category dealt primarily with plans to meet or the coordination of shopping and other household expeditions.
M6: Where shall i meet you tonite?what time?See u soon love me x
M7: Wanna come to tesco?
Social Arrangement Orientation
Also about plans for meeting up, messages in this category were explicitly about recreational planning such as going out together for the evening, going to the cinema and other social arrangements.
M8: R WE DOIN LUNCH THIS WK?CHE
M9: Hello.Me and laura want2go2jive2moro.Does u want us 2 buy tickets
Messages in this category were non-specific, usually very brief and often flippant; many of them were little more than simple, friendly greetings.
M10: Yo man whats de goss
M11: morning,how are you today? xxjtxx
Friendship Maintenance Orientation
Messages in this category dealt primarily with 'friendship work' such as apologies, words of support and thanks.
M12: Happy Birthday, i hope you are having a good one,see you in a few days.Love Duncan x x x x
M13: Don't worry bout exam!Just had hair cut & look like a ginger medussa!Arrgh!
Usually more so than the Friendship Maintenance category, messages in this category dealt primarily with romantic expressions of love, intimacy and affection.
M14: R u bak already khevwine?!i am not comin 4 anuva 2 wks,but khevwine, u r the sexiest thing since sliced bread!c & sexia then sliced bread!oh my luv.I miss u so!x
M15: Each time ur name appears on my phone i smile like this :)
Messages in this category had explicit sexual overtones.
M16: Read ur email-thought waz gonna burst so horny xxxxxx
M17: Your wish is my command!I promise to be a better hostage next time.Sweet dreams princess.xxx
Typically, chain messages are comparatively longer epigrams, jokes or word-plays which are passed on from messager to messager.
M18: I believe friends are like quiet angels who lift our feet when our wings forget how to fly!send to 4 friends and sont send back and see what happens in 4 days
M19: sex is good,sex is fine,doggy style or 69,screwin 4 free or getting paid,everyone loves getting laid,so spread ur legs,lay on ur back,lick ur lips & text me back!
In Figure 1 (PDF version for download), all the messages in our corpus are shown distributed in terms of these primary functional orientations. Even though, theoretically speaking, it is impossible to separate 'doing sociability' from information exchange (Jaworski, 2000:113) for analytical convenience it can be revealing to compare the relative weighting of 'relational' and 'informational' dimensions of communication in participant responses (cf. Thurlow, 2001a). With the 'transactional' or 'interactional' orientation of text-messages tending to be either foregrounded or backgrounded, it is possible to locate each of the functional categories along a continuum according to the relative degree of relational intimacy conveyed by each as in Figure 2 (PDF version for download).
On this basis, and relying on the Informational-Relational category as a notional midway point, initial content analyses of participants' 544 messages thereby reveals how at least two thirds of their messages were explicitly relational in their orientation, ranging from making social arrangements, friendly salutations, friendship maintenance, to romantic, flirtatious and openly sexual exchanges. In fact, recognising the possibility that chain messages too have a relational orientation (see below) and that many of the messages dubbed 'practical arrangements' may well represent a more implicit social arrangement, the amount of explicitly transactional or 'informational' messaging was relatively small - as little as 15% of all the message codings. As a way of 'unpacking' a little further the decidedly social-relational orientation of participants' texting, we would like briefly to pick out for discussion some of our initial impressions of the general tone and content of their messages. This is done largely as a means of contextualizing the discussion about SMS language which follows.
Within the general category of friendship maintenance, were found a range of messages of apology, thanks and support (e.g. M12 and M13 above). However, we also identified a number of instances where text-messages were being used by friends to stay in touch while apart and also as means of resolving (e.g. M20) - and, possibly, instigating (e.g. M21) - conflict:
M20: u stupid girl,why ru upset & worried?i'm not in a mood or stressed so u shouldn't be + def don't b scared of me-i'm a softy!cu in a bit x
M21: Olly's brought up the house again!Wanker!He's said he reckons you + him'll "come to blows" by the end of the year.He'll fucking die!
It is these types of messages which most clearly indicate the way in which participants appeared to rely on text-messaging to facilitate relational maintenance and social intercourse, and to complement their face-to-face interactions. It is not only through the symbolic (or metacommunicative) exchange of messages that they sustain their relationships, but young people also use textingstrategically to manage a wide range of friendship concerns and issues.
Another strong impression formed throughout our reading of the messages was of an overriding jocularity or teasing tone. Although humour is generally very difficult to discern by third parties - not least given that it is intensely context-dependent - there were nonetheless numerous instances where the messager's intent was very clearly humorous.
M22: Simon said you didn't come home last nite.U dirty stop out
M23: You are a drunken fool with a bad memory
Specifically in conjunction with those messages categorised as 'salutory', we believe that humour helps to fulfil the generally phatic (cf. Malinowski, 1923) function of text-messaging by which an almost steady flow of banter is used in order to maintain an atmosphere of intimacy and perpetual social contact. In this sense, text-messaging is small-talk par excellence - none of which is to say that it is either peripheral or unimportant (see Coupland, 2000).
Beyond their notable sexual content, the chain messages might also be regarded as a form of 'gifting' (Ling & Yttri, 2002:159), whereby messagers - especially so amongst younger teenagers - forward these stock sentiments and saucy jokes not only to communicate some desired aspect of identity, but also as means of social bonding through (potentially) shared humour and taboo breaking. As such, although apparently transactional in content, chain messages are clearly more relational in function. Although there were only a handful altogether in the current data-set, what sexual jokes were found were almost always reported by male participants which would not be atypical of the often (hetero-)sexualized nature of young men's conversational discourse (Edley & Wetherell, 1997; Cameron, 1997).
Allowing also for the sexual tone of many of the chain messages, a striking number of the messages oriented around romantic and, occasionally, sexual themes - either as subject matter (M24 below) or interactional goal (M17 above).
M24: HAD SEX!
It is in this way, that SMS is seen to afford an interesting mix of intimacy and distance not unlike various other CMC niches such as IRC, IM and, to an extent, email. The technical rapidity and ephemerality of text-messaging seems to bring with it a relative anonymity even though, unlike the CMC of much online chat, the sender and receiver are invariably revealed to each other through caller/number display. Nevertheless, it is this kind of 'recognised anonymity' which might explain the relative licentiousness or flame-potential of some of the messages reported by participants (see O'Sullivan & Flanagin, 2003 , for a discussion these issues in internet CMC). The face-saving potential of this type of anonymity, was also borne out by Brown's (2002) finding that as many as 52% of the participants reported having sent a text-message to say something they wouldn't ordinarily have said face-to-face.
Within the general message category 'Practical Arrangement' was an important sub-grouping of messages which exemplifies precisely the kind of interpersonal co-ordination discussed by Ling & Yttri (2002) and which they refer to as 'hyper-coordination'. What is meant by this is the type of mundane, micro-level coordination involved in redirecting trips already started (e.g. 'I need to pick up some milk; can we meet at the store instead?'), letting people know that you're going to be late (e.g. 'I'm held up in traffic but will be there in ten minutes hopefully') or confirming exact timing and location (e.g. 'I'm walking up the high street right now - are you still waiting in front of the post office?'). From the current data-set, examples included:
M25: C u in 5 min x
M27: Where r u?We r by the bar at the back on the left.
It is this finely-tuned arrangement-making which demonstrates one of the clearest instances of mobile telephony's shaping a new, distinctive style of social interaction; Ling & Yttri (2002:144) propose that this type of mundane, micro-level organising allows for both the 'structuring and rationalization of interaction'. Certainly, it would seem from our corpus that a high premium is placed by young people on such continual accessibility and connectivity - or what Katz & Aakhus (2002) characterise as 'perpetual contact' - and that, once again, this is done primarily in the service of social intercourse.
Related to this sense of perpetual contact, and as another example of how text-messagers capitalize on technological affordances (more on this point below), some participants' messages revealed a level of contact which was so continual to the extent of being actually co-present:
M28: Who the girls your with is it one of your adoring fans?
M29: Have you had a shower today as i'm sure I can smell u from here!(Teehee)
In both these instances, where sender and receiver are apparently within viewing distance of each other, users are able to interact covertly, enabling an immediate, and potentially very intimate, form of communication. The subversive potential in this kind of secret messaging is seen even more clearly in M30, another co-present text-message, where sender and receiver appear to be sitting in the same lecture but are able to contravene interactional norms undetected. [note 9]
M30: How r u sweetie?Why am I doing this subject?It's just so boring!cu soon xxx
It is this 'culture of concealed use' (Ling & Yttri, 2002:164) which again makes apparent how and why text-messaging has come to be stitched so seamlessly into the social fabric of young people's lives; by no means necessarily replacing face-to-face interaction, mobile phones and SMS enhance communication in ways which allow for multiple (or even parallel) communication events, offering an attractive combination of mobility, discretion, intimacy and, indeed, fun - illicit or otherwise.