From lol to kinda - the use of abbreviations in online chatrooms

(1)

From lol to kinda – the use of abbreviations in online chatrooms

Anniina Savolainen, 241742 MA Thesis Advisor: Mikko Laitinen University of Eastern Finland April 2017

(2)

ITÄ-SUOMEN YLIOPISTO – UNIVERSITY OF EASTERN FINLAND

School

University of Eastern Finland Author

Anniina Savolainen Title

From lol to kinda – the use of abbreviations in online chatrooms

Main subject Level Date Number of pages English language and culture Pro gradu -tutkielma x 13.04.2017 110

Sivuainetutkielma Kandidaatin tutkielma Aineopintojen tutkielma

Abstract

The aim of this study was to look into the use of abbreviations in an online chatroom, and investigate possible changes that may take place in their use under a six-month period. There were two sets of data that were analysed, the first one from winter (Data 1) and the other from summer (Data 2), but each gathered over a period of two weeks and for one hour per day. The investigator did not personally participate in the chatting that took place, and the data have been anonymized. The material was gathered from a chat site known as E- Chat.

The two data were analysed and compared for changes in seven categories. These categories were Initialisms (lol for “laughing out loud”), Apostrophe-frees (dont for “don’t/do not”), Clippings (bday for “birthday”), Letter-number homophones (u for “you”), Phonetic spellings (thanx for “thanks”), Contractions (wanna for “want to”) and Word-value characters (f for “female”). They were largely based on Bieswanger’s (2007) study on text messages. The results of each data were then compared to each other in terms of how often abbreviations occurred per turn in these categories. The counting was carried out manually and partially with the help of the AntConc program. The differences between the data were compared through the help of the log likelihood calculations.

The results showed that Initialisms has been the most frequent category of abbreviations in both sets of data, with Letter/number homophones as the second most used category. Word-value characters was the least frequent category. The categories are in the same order in terms of how much they are used in both sets of data. However, when looking at the sheer numbers in terms of how much the categories are used in each data, most of them have undergone significant changes between the two data sets. However, the changes were not significant for all of the categories, and when looking at all of them together, the changes were not very highly significant.

This may imply changes need more time to fully take root, although they can be seen starting to happen even within six months.

Therefore, a follow-up study could include studying changes in longer term, or with larger data.

Finally, there were some patterns in the use of abbreviations that can be noted. Something that became increasingly clear was that certain abbreviations were very common and used over and over again. These abbreviations also tended to belong to the categories that became the most frequently used. This may indicate chatting resembles speaking more than writing as a language form, as speech also includes a lot of repetition. Some of the very common abbreviations, such as lol, may also be losing their old meanings and be gaining new ones, as lol was often used even when something was not especially funny.

Keywords

Abbreviations, chatroom, online language, Netspeak, short-term changes in language

(3)

ITÄ-SUOMEN YLIOPISTO – UNIVERSITY OF EASTERN FINLAND

Osasto

Itä-Suomen yliopisto Tekijät

Anniina Savolainen Työn nimi

From lol to kinda – the use of abbreviations in online chatrooms

Pääaine Työn laji Päivämäärä Sivumäärä Englannin kieli ja kulttuuri Pro gradu -tutkielma x 13.04.2017 110

Sivuainetutkielma Kandidaatin tutkielma Aineopintojen tutkielma

Tiivistelmä

Tutkimuksen tavoite oli selvittää lyhenteiden käyttöä chatissa, sekä tutkia mahdollisia muutoksia joita saattaa tapahtua niiden käytössä puolen vuoden aikavälillä. Analysoitavana oli kaksi dataa, ensimmäinen talvelta (Data 1) ja toinen kesältä (Data 2), mutta molemmat oli kerätty kahden viikon ajalta ja tunnin päivässä. Tutkija ei itse osallistunut chattailyyn, ja data on tehty nimettömäksi myös chattiin osallistuneiden osalta. Materiaali on kerätty E-Chat -sivustolta.

Tutkittavana olleet kaksi dataa analysoitiin ja niissä tapahtuneita muutoksia verrattiin seitsemän kategorian kohdalta. Nämä kategoriat olivat Alkukirjainlyhenteet (lol / “laughing out loud”), Heittopilkun poistavat (dont / “don’t/do not”), Leikkeet (bday / “birthday”), Kirjain/numero homofoonit (u / “you”), Foneettiset kirjoitusasut (thanx / “thanks”), Tiivistykset (wanna / “want to”) ja Sananarvoiset merkit (f / “female”). Kategoriat perustuivat suuresti Bieswangerin (2007) tutkimukseen tekstiviesteistä. Kummankin datan tuloksia verrattiin sittemmin toisiinsa sen perusteella, kuinka usein lyhenteitä esiintyi per vuoro jokaisessa kategoriassa. Nämä tulokset laskettiin manuaalisesti ja osin AntConc -ohjelman avulla. Datojen välisiä eroja verrattiin log likelihood -laskujen avulla.

Tulokset osoittivat, että Alkukirjainlyhenteet oli yleisimmin käytetty lyhennekategoria molemmissa datoissa, ja Kirjain/numero homofoonit olivat toiseksi yleisin. Sananarvoiset merkit oli vähiten käytetty kategoria. Kategoriat esiintyivät samassa järjestyksessä niiden yleisyyteen nähden molemmissa datoissa. Tästä huolimatta katsottaessa numeroita lyhenteiden kategorioiden käyttöön liittyen, suurin osa kategorioista on läpikäynyt merkittäviä muutoksia kahta dataa verrattaessa. Muutokset eivät kuitenkaan olleet merkittäviä jokaiselle kategorialle, ja kun niitä kaikkia tarkastelee yhdessä, muutokset eivät olleet erityisen tilastollisesti merkitseviä. Tämä voi merkitä sitä, että muutokset tarvitsevat enemmän aikaa todella tapahtuakseen, vaikka niitä ne alkavat näkyä jo puolen vuoden aikana.

Tästä syystä jatkotutkimus voisi esimerkiksi käsitellä muutoksia pidemmällä aikavälillä, tai suurempaa dataa tutkien.

Lopuksi, lyhenteiden käytössä oli tiettyjä huomattavia jatkuvuuksia. Kävi nopeasti ilmi, että tiettyjä lyhenteitä käytetään hyvin usein, uudelleen ja uudelleen. Nämä lyhenteet kuuluivat yleensä niihin kategorioihin, joista tuli tavanomaisimmin käytettyjä. Tämä voi viitata siihen, että chattaily muistuttaa puhetta enemmän kuin kirjoittamista kielimuotona, sillä puheessa esiintyy myös paljon toistoa. Jotkut erityisen yleiset lyhenteet, kuten lol, voivat myös olla menettämässä vanhoja merkityksiään ja saamassa uusia tilalle, sillä lol - lyhennettä käytettiin usein myös yhteyksissä, joissa ei tapahtunut mitään erityisen hauskaa.

Avainsanat

Lyhenteet, chatti, nettikieli, lyhyen aikavälin muutokset kielessä

(4)

1. Introduction

As the Internet has become something commonplace that is accessible to most households around the world, its influence on many things has increased. Its effect on language is no exception. Already in 2001, David Crystal was certain that the “biggest language revolution ever” (2001: 241) was about to begin thanks to the World Wide Web and its influence on language. Indeed, as stated by Baron (2008: x) as well, due to the explosion of language on the Internet in the 21^st century, more and more people are getting involved in this language revolution. Crystal (2001: 241) also believed that the influence of the language online could have a larger impact on language in general than its predecessors: speech and writing. He theorised that online communication may even become a linguistic norm for people; the main source for language change over face-to-face interaction.This prediction seems to have become reality, since Warschauer (1997: 6, quoted in Shekary and Tahririan 2006: 558) states, referring to online chatting, that the “historical divide between speech and writing has [finally] been overcome with the interactional and reflective aspects of language merged in a single medium”. Warschauer (1997: 6) also noted that online writing has “unleashed the interactive power of text-based communication”. Therefore, due to its growing impact, the language of the Internet and online chatting – or “Netspeak” as Crystal (2001: 17) generally calls it and as I shall refer to it – is an important field of study in linguistics. The importance of the language used on the Internet grows stronger all the time and constantly influences the way many use language in their daily lives, especially young people.

This study will focus on the use of lexical shortenings and abbreviations of the English language in real-time online chatrooms. The study concentrates specifically on how and to what degree abbreviations used in online chatrooms can be placed into seven different categories: Initialisms (lol for “laughing out loud”), Apostrophe-frees (dont for “don’t/do not”) Clippings (bday for “birthday”), Letter-number homophones (u for “you”), Phonetic spellings (thanx for “thanks”), Contractions

(6)

2

(wanna for “want to”) and Word-value characters (f for “female”). These categories are mostly based on Bieswanger’s (2007) study on text messages, and will be further explained in the Abbreviations section (Chapter 4). The material for this study was collected over a period of two weeks at a time at two separate occasions for one hour at a time at a chatroom website known as E-Chat. This study is also a continuation of the BA thesis I completed in 2014. The aim of the study is to figure out how much the different categories of abbreviations are utilised in terms of their use, as well as to observe the changes that can be detected between the two sets of data used for the study, and how the passing of time may have affected the data in comparison to one another. The research questions can therefore be summed up as:

(1) What are the frequencies of the seven categories in the chatroom data?

(2) What changes have taken place between the six months taking place between the collecting of the data?

The idea to study abbreviations specifically was chosen since Crystal (2001: 84) points out that abbreviating is one of the unique features of Nestpeak. The reason for studying them in online chatrooms mostly lies in my own interest in the language of chatrooms based on my personal experiences with them. Additionally, there seems to be a lack of studies concerning online chatrooms.

There have been some studies on chatting which takes place in controlled environments (Freiermuth and Jarrell 2006, Shekary and Tahririan 2006, Ferreira da Cruz 2008, Sauro and Smith 2009), usually for the purpose of examining how well chatrooms can be utilised as a way of teaching students a foreign language, but there do not seem to be many studies based on actual online chatrooms where people can freely talk about various topics. Mattiello (2013: 65) indicates that abbreviations also have gained relatively little attention in literature so far, although Leijten et al. (2006: 68) have studied playfulness and ludicity with the use of language use in online chatting. Regardless, it seems reasonable to study this area of the English language. Even though, as mentioned, the topic has

(7)

3

already been studied for my BA thesis, this study expands it to see if the results change when there is more material to examine.

This thesis paper consists of nine different main chapters, including this introductory section. In the second chapter I discuss chatting, further explaining what it refers to, how it is done and what type of language is typically used in chatrooms. In the third chapter I attempt to define what can be meant when we refer to “words”, in order to be able to understand what we mean by “abbreviations”. This chapter discusses multiple points of view about what the meaning of the term “word” can be, as well as goes over some of the key methods in which words can be constructed. The fourth chapter deals with abbreviations and their categorisations, as well as how they can be used in language. The fifth consists of looking into earlier studies, while in chapter six I explain the methods for conducting this study and explain how I gathered the material. In the seventh chapter I go over the results discovered in the study, and discuss them further both in general and in the context of the earlier studies mentioned in chapter five in chapter eight. Finally, in chapter nine I offer some final concluding remarks for this study and offer a few possible ideas for further studies.

2. On chatting

2.1 Introduction to chatting

Forms of communication on the Internet can roughly be divided into two dimensions: synchronicity and audience scope (Baron 2008: 14). Synchronicity refers to whether the communication in question happens in real time (synchronous) or whether the recipients of the communication acts are only expected to receive and react to them at an occasion that is convenient to them (asynchronous).

Audience scope, instead, refers to whether the communication is intended to be received by a single person (one-to-one) or by multiple people (one-to-many). Online chatting is a synchronous one-to- many type of communication. However, chatting as a form of communication is synchronous only to

(8)

4

the degree as written communication on the Internet can be. Baron (2008: 14) claims that the only really synchronous communication is such where people are able to interrupt each other in the middle of another person’s turn – such as in the cases of face-to-face interaction or telephone conversations.

Still, chatting is unique in that it contains features of both speech and writing, as it is done in real- time like oral communication, yet is created through writing (Shekary and Tahririan 2006: 557).

Chatrooms indeed contain types of conversations that are created through writing, and they occur in real time and can have multiple people participating in them simultaneously (Ferreira da Cruz 2008:

101).

Chatting on the Internet was also basically created in order for people to be able to converse with multiple other people online in real time. Early pioneering programs for this purpose included Turoff’s EIMSARI and later in the 1980s UNIX-based “talk” programs, which allowed the same functions and uses as chatrooms today; being able to participate in synchronous messaging simultaneously. However, chatting in general as we have come to understand it was created in 1988 by Jaakko Oikarinen, a student at the University of Oulu. He wrote a program that came to be known as Internet Relay Chat (IRC), which was intended as an improved version of the UNIX, and over the course of a few years gained a wide audience. It would also serve as the model on which later chatting programs would be based (Baron 2008: 22–23). According to Yus (2011: 151), chatrooms have become one of the most popular forms of online communication.

2.2 The practice of chatting

When joining a chat, participants simply enter a so-called “channel”, of which there are usually a few that people can choose from, based on the topics. Generally, some of those topics are more specific (#TEENS 2016, Gay / Bi Room) and some are more general (Just Chat, *HAPPY.CHAT). This is likely due to chatrooms not only being places for online conversations, but also places where people can create communities they can bond with (Yus 2011: 151). Users log into these channels by using

(9)

5

nicknames and are in this way able to cover up their real identities, including their age, gender, location and other background information. This has been, indeed, mentioned as one of the features that Korpi-Hallila (2005: 6) sees as being unique about chatting; the participants generally have no idea whom they are actually talking to. While the conversation takes place in real time, the chatters can usually scroll back through the conversation to be able to see and respond to earlier messages (Baron 2008: 23).

As one joins a chat, they will become a part of conversation threads that are already in progress (Yus 2011: 161). The chatter thus comes across multiple messages, some of which are addressed to everyone in the chatroom, while others are to specific chatters. The chatter will then have to decide which threads they would like to take part in (Yus 2011: 161). A model has been suggested of two stages for how this happens. The first stage is called the information-acquisition stage, where the chatters first view the screen where the current conversations are happening. At this stage, the chatters need to solve which lines of text belong to which conversation thread, understand in which context they were written and type their own messages as responses to them. This must usually be done fast, as the messages can disappear from the chatroom very quickly, buried under newer ones. However, some of these messages remain in the chatter’s short term memory usage, leading to situational awareness. After gaining it, the new chatter will need to decide in which conversation they would like to take part in. As they do this, the second phase called the stage of information output is taking place. In this stage, the chatters decide what they are going to type. In order to do this, they need to look deeper to understand some of the earlier messages. During this process, what these older messages have conveyed will likely remain with the chatter for a longer period of time, becoming stored in their long-term memory. Meanwhile, the information gained about the earlier messages is used to compose what the chatter wishes to say, together with the knowledge on how to use the chatroom interface properly and to convey a message in an appropriate manner (Yus 2011: 161–162).

(10)

6

People tend to use chatrooms to communicate with other people regardless of either one’s location, and to attract the attention of other chatters, much like we do in the physical, “real” world as well (Yus 2011: 154). In order to do this, various tactics are generally used and the typed texts altered as much as is deemed necessary to fulfil this purpose. However, there is the risk of being misunderstood if too many of these tactics are utilised. Indeed, chatters may need to exercise much thought to make sense of what the others in the chatroom are attempting to communicate. In order to keep the misunderstandings to a minimum in these cases, the new chatters need to learn certain norms expected of them if they want to properly become a part of the conversation taking place. The methods used are also utilised in order to create a sense of solidarity between those chatters that can understand what is being conveyed. Indeed, it has been said that chatters take part in this type of communication precisely because they can use these conventions to communicate in chatrooms (Yus 2011: 156).

Some of the conventions of chatting people will need to learn to navigate are, for instance, the lack of being able to express what they are trying to say as properly by using a keyboard as they would if they were talking face-to-face to someone in their presence. However, Yus (2011: 156) claims that chatters are in fact satisfied with what they have, as they are freed from some of the pressures of face- to-face interaction by the anonymity of being able to hide behind a nick (nickname, screen name).

This allows them to express themselves more freely, accurately and spontaneously, as well as to possibly play around with multiple identities. The chatters may even be people who shy away from real life conversations and to online ones, which allow them more control over how people might perceive what they are saying (Yus 2011: 161). Indeed, as Leijten et al. (2006: 67) state, the success of online chatting tells that it must have some ways that people can understand the threads of ongoing conversations. For this, some formal regularity must exist to the attending parties. Without any understandable structure, successful communication cannot take place. This structure is not stable in chatrooms, however, and new areas of conversing are tried out and their meanings are negotiated

(11)

7

frequently. This leads to there sometimes being misunderstandings or confusion about new meanings between parties, especially if some of them have lacking previous knowledge on how language and talking on chatrooms work.

Each chat community also has its own culture and set or rules its users need to follow. According to Leijten et al. (2006: 73), the people of chatrooms can generally be divided into three groups: an “in- group”, an “out-group” and an “inter-group”. The “in-group” consists of those chatroom members who are familiar with the chatroom culture and often also share some history with other chatroom members, knowing how to act in chatrooms and being accepted members of the group. They have relationships amongst themselves and can take on several roles online, which may or may not be consistent with who they are in real life. Newer users, or “newbies” are those in the out-group, while the inter-group consists of those who are attempting to become accepted members of the in-group.

The older members may or may not ridicule newer members for their ignorance in the public of chatrooms, although they may also privately assist them. There are clear power relations at play here, including the in-group having a type of secret code, but also the legitimacy, which allows them to deviate from that code. In the meantime, those who wish to become a part of the in-group need to learn how to manage and understand the rules to the game and the language use. Newbies may, for instance, hesitate to creatively use diminutives nicknames, fearing they are intruding on unknown territory, but the in-group often uses them as they all know each other and are friends. At the same time, if the newbies try and apply the methods used on another chatroom to the one they are currently using, they may be ridiculed for it by the veterans. However, once they have become accepted members of the group, they are the ones who are free to ridicule the new “newbies” on the chatroom (Leijten et al. 2006: 73–75).

(12)

8

2.3 Language of chatrooms

Chatrooms often have linguistic strategies specific to their users, both indicating group membership and making it easy to spot those who are new to chatting (Yus 2011: 151). Katamba (2005: 168) has indeed stated that there are subvarieties of language “that are associated with particular occupations”.

They can be realised through their specific jargon, referring to the strange words and word-like expressions that they use. Nowadays, this type of language use is even an often-used source for the creation of new words into a language, as it generally has a very specific purpose. It allows people belonging to a subgroup to discuss what their field specialises in with the lexical items that belong to their jargon. This type of jargon can then sometimes enter the more widely-used language, especially when whatever the jargon-using group does is seen as fashionable or trendy, or as such that is has an impact on the lives of the rest of humankind. For instance, much of the abbreviations in this study have become a part of the everyday language as well. Related to jargon, and another source of new words in a language is slang (Katamba 2005: 169). This is “the term used to describe a variety of language with informal, often faddy, non-standard vocabulary”. Although the general language can gain multiple words from slang, many of them do not survive for very long periods of time. Slang also utilises old standard forms of words in new and creative ways. There are several reasons for why one might want to use slang (Katamba 2005: 169). These reasons include:

“the desire to experiment with using language “poetically” or creatively for pleasure;

the desire to be secretive; the desire to be expressive; the desire to use language as a badge of group membership so as to express intimacy with those inside the group and to exclude those who are not (technical jargon also does this job); to indicate that one is casual and relaxed etc.”

This is related to the use of Netspeak as well.

(13)

9

Although nowadays there are also chatrooms available where one can enhance the experience by using web cams or sound, there are many people who still prefer the original text-based messaging format of chatrooms (Yus 2011: 151). These chatrooms require the mastering of how to chat. The new chatters will need to learn the proper etiquette and manner of speech in chatrooms before they can become proper members of the chat community. Some of these linguistic strategies which one needs to know how to interpret and use include oralisation of text, emoticons and, indeed, abbreviations (Yus 2011: 152). A large part of this is a type of playfulness that characterises interaction, which Leijten et al. (2006: 67) refer to as ludicity. It also generally indicates creativity in the people who use it in language in creative and imaginative ways (Leijten et al. 2006: 67), and it can be a part of a type of performance for others. It can also simply be out of the joy of playing around with language, or a means to a social or some other end. It is a natural part of conversation, but what makes it special in chatting environments is that aside from social aspects, it also takes on some competitive ones as well. People who come up with the ways to use language that are picked up by other people as well are sort of “winners” in this scenario (Leijten et al. 2006: 68), the wittiest or most outrageous of chatters who arouse the most attention (Leijten et al. 2006: 71).

Of the creative and playful ways of using the language that were found in their study, Leijten et al.

(2006: 69) name as quasi-phonetic spellings (thanx, sux), colloquial usage (gonna, hafta), emoticons () and abbreviations and acronyms (lol). The ludicity used can influence any part of language:

phonology, morphology, orthography, syntax or lexis. As can be seen, it often includes the use of non-standard forms of language, adding to the informality of the situation. Indeed, it is often spontaneous and scarcely edited. There is often interruption or subversion of conversation threads, leading to their decay, as well as so-called “stage directions” where the chatters indicate actions that they are committing outside of the chat. They are usually designated in third person and are often marked by an asterisk or angle brackets (*Nina throws her hands in the air*). As stated before, there

(14)

10

is a type of competition between chatters, over the attention of other chatters. This can sometimes escalate too far, in which case the moderators of the chatrooms sometimes have to step in. New members need to learn what is crossing a line fairly quickly in order to not be excluded from chatrooms (Leijten et al. 2006: 73).

2.4 Chatroom-specific research questions

Yus (2011: 157) calls chatrooms “transit places” where people can interact with others in a mostly casual manner. They can interpret what others are saying in the ways they want, and contribute to the conversations the way they wish. There are, however, some substantial differences between spoken interaction and chatroom conversations. Firstly, there is the synchonity, already discussed above.

Chatrooms are only as synchronous as the rigid succession of messages as during their arrival on the chatroom screen allows them to be. Chatrooms do suffer from succession in time and space, of chatters not being able to create and post messages quite quickly enough, despite the fact that the chatters using them are online at the same time. This makes the communication over chatrooms slightly less natural than with face-to-face conversations. Secondly, there is a lack of context to aid the chatters, without vocal or visual information of face-to-face interactions. Added to that is the speed at which the chat messages can disappear from the room at the top of the screen. Thirdly, in real life conversations people usually have “turns” for when to speak. In chatrooms, it is the sequencing of the “software that manages the interactions” which makes turns. All utterances are thus sequenced in chatrooms, even when they are typed at the same time. (Yus 2011: 157–159)

Fourthly, as mentioned above, there is also the issue of multiple interactions, where many different conversations are taking place at the same time in the same room. The chatters need to make sense for themselves which parts of the chat belong to which conversation. This can also cause trouble with the synchronity, as new conversations may be sprung up and old ones left behind before someone can properly manage to contribute to the earlier one. One way to handle this is to address a certain user

(15)

11

directly by their chatroom nickname. This also helps to keep track of which conversation each chatter is participating in. Herring (1999), for instance, studied the coherence of turns and threads in chatrooms, and found that much of the interaction in chatrooms suffers from fragmentation, which tends to make the chatters disinterested as they have to juggle multiple lines of conversation. It was also possible to see the tendencies towards multiple conversations taking place at once in the data for this study. In the example below, one can spot four different threads of conversation (marked here with the use of italics, emboldening and underlining):

(1)

Master Runner (21:48)

moony just because u still young that's why :P paul.plz

(21:48)

moony where is letsdance doctor rollins 124

(21:48)

thanks you all i love each and everyone one of my friends you all changed my life for the better you all are so incredible nice and all have lovely personality <3 <3

Chino (21:48)

moony , Chess , so do we have the right to judge others according to ur premis

?

moony98 (21:49)

loooool runner you are young also!

(December 30^th, 2015)

Fifthly (Yus 2011: 159), there is the earlier-mentioned fact that messages can disappear from the chat rather quickly, when multiple chatters are taking part in the conversation actively. This is sometimes known as the scroll factor. There thus needs to be a quick handling of the information on chatrooms.

This has, for instance, the consequence of keeping the messages short, something with which using abbreviations certainly helps. Indeed, Clipped messages is mentioned as the sixth component that

(16)

12

makes chatroom interactions different from face-to-face ones. The messages of chatrooms are often divided into chunks that occur in different turns taken by the chatter, sometimes with messages from other participants in-between the messages.However, despite these limitations, people can learn to navigate chatrooms and also gain the ability to be effectively playful with language.

3. What is a word?

Before we begin to talk about abbreviations, it is important to understand what we mean when we speak of words in general. We often do not give words much thought (Plag 2003:1), taking them for granted. However, we all need to know and use words in order to learn a language. According to Plag (2003: 4), we humans have on average 45,000 to 60,000 word types of a language stored in our minds.

This is called our mental lexicon. Indeed, as stated by Bauer (1983: 7), it has been for a long time a major challenge to actually define what a word is, as there tend to be linguistic units which people still refer to as words although they are not covered by the definitions given for what a word is supposed to be (Bauer 1983: 8). Regardless of these difficulties with definition, it is important in many ways to operate with the assumption that words exist. All speakers of language, even illiterate ones, can usually give an estimate for what a word is, although they may have difficulties with separating words into smaller units. Another reason to believe words to be a part of linguistic repertoire, morphological conditioning always takes place “within the word”. This refers to, for instance, the irregular plurals of words being determined by the word itself, occurring only with that word. For instance, the plural suffix of -en is unique to the words ox and child, not occurring with any other word.

Another argument is that “the ordering of elements within a word is frequently independent of, and even compulsorily different from, the ordering of strings of words”. This refers to, for example, how in English the genitive can be shown either by the suffix –‘s or by the word of. –‘s occurs as a part of the possessor word, while of occurs before the possessor. This argument can also be used with the

(17)

13

comparatives of adjectives, where the comparative can be shown by the use of separate word more, occurring before the comparative adjective, while the comparative suffix -er takes place at the end of the comparative adjective (Bauer 1983: 8–9). These are all sufficient arguments for the existence of linguistic units called words (Bauer 1983: 9). However, it may depend quite a bit on the language in question on what counts as a word. What might count as “words” in other languages may become long phrases in, for example, English. Therefore, whatever we refer to as words do not seem to mean the same thing in each language. In this sense, it may not be possible even to find a definition for a word that can encompass all languages. However, because this study focuses on the English language, I will only focus on the definition of words in English. So, what are these units we call words when it comes to English, exactly? The answer is not as straightforward as one might at first assume.

3.1 Orthographic word

However, all of us do not seem to share the exact same definition for what a word is. When we think of words, we may not always share the same thoughts on what that means for us individually. We can, for instance, speak of word-forms, which refer to the words’ represented physical form in speech or writing (Katamba 2005: 11). Indeed, often, the first definition that comes to mind when thinking of words is the one where they are thought of as units of the writing system (Plag 2003: 4). This is called the orthographic word. This type of word could for instance be defined as “an uninterrupted string of letters which is preceded by a blank space and followed either by a blank space or a punctuation mark”. Katamba (2005: 12) defines orthographic words as units that have spaces surrounding them on both sides. This type of definition may at first seem like an encompassing definition that can be easily applied, and tends to be in agreement with our first thought of what a word is. However, this definition cannot always be applied and things are not as straightforward as they seem (Plag 2003: 5). For instance, with genitives such as “Mary’s”, if apostrophes are seen as punctuation marks, this would constitute as two (orthographic) words. Otherwise, this would be one word. There are also hyphenated words, which would count as two orthographic words if the hyphen

(18)

14

is seen as a punctuation mark. Katamba (2005: 12) states that the definition of these kinds of complex words varies, depending on “how transparent the compound nature of a word is”.

There are also other cases where the rule of the thumb of orthographic words is not adequate (Katamba 2005: 14). How should we, for instance, categorise words such as boyfriend or apartment building? If they are seen as orthographic words, boyfriend would be one orthographic word while apartment building would be two (Plag 2003: 5). However, the orthography in cases like these is often variable. Boyfriend can also be spelled boy-friend or even boy friend, and these different types of spellings are not rare. There is a certain flexibility in how different words are written down (Katamba 2005: 14), and this complicates the attempts to define what a word is, as the definition should not be dependent on the ways different people spell or write certain words (Plag 2003: 5).

Another problem with this type of word definition is that orthographic words do not always coincide with the way we instinctually view words (Plag 2003: 5). Many of us would probably view boyfriend as a single word, created out of two words; in other words, a compound. However, compounds themselves are a tricky case. If they should be seen as a single word, they should be spelled as a single word as well. This is however not always the case, as can be seen with the above example of apartment building. When it comes to the hyphenated compounds such as boy-friend, words that can be seen as having lost their status as compounds, hyphens are growing rarer, and is mostly used with words that are newly coincided. However, the agreements on what established words that have become common can vary greatly (Katamba 2005: 13). With certain words, even versions that either lack or maintain a hyphen can be acceptable, as in the case of first-rate and first rate. Hyphenation can also be used to show that a concept is being treated as a single word, when it can only be expressed by a string of words. Simple to serve, for instance, can become simple-to-serve when used in this manner.

(19)

15

Based on this discussion, defining words based simply on orthography does not seem to be enough.

The orthographic words also only discuss words in writing, when they naturally also exist in speech (Katamba 2005: 15). Simply discussing orthographic words therefore has the implication that illiterate language users may not understand what a word is, which is a false assumption (Plag 2003:

5). Some of the other major ways that have been brought up when discussing our first instincts on how a word could be defined include: by a word’s sound structure (phonologically), by its internal integrity, by its meaning (semantically) or by the use of sentence structure in relation to the word (syntactically). I will discuss them below in chapters 3.2 and 3.3.

3.2 Words and phonology

When looking at how words are spelled and comparing them to their sound structure, one might assume words to be units of speech that are surrounded by pauses (Plag 2003: 5–6). However, when listening to naturally occurring speech, one can quickly come to realise this is usually not the case (Plag 2003: 6). People do not create forced, distinct pauses between each word which occur in writing (Katamba 2005: 14). Words instead come out as a string in real speech, and even overlap. It could maybe be said that words are surrounded by potential pauses, but this has its own set of problems, as people also make pauses between syllables, not merely words (Plag 2003: 6). It is therefore more difficult to differentiate between words in speech than in writing. However, there are methods that can be used to do this (Katamba 2005: 15). After all, people are still able to recognise words in speech, and even without pauses, there is another way through which the sound structure can help us define what a word is and understand it. What helps with this is the distinction between content words and function words. Content words include nouns, verbs and adjectives, holding inside of them most of the cognitive meaning of a sentence. They tell us what has happened, to whom and how. Function words are those that do not belong to these categories, including prepositions and articles, for instance.

(20)

16

The reason why we need this differentiation in this case is that the content words in English have one syllable that is more prominent than the others due to it receiving the main stress of the word (Katamba 2005: 15). It can only fall on a single syllable in a word, and is part of the inherent structure of the word (Katamba 2005: 16). In English, as in most languages, words are the parts of language that are “crucial for the occurrence and distribution of stress” (Plag 2003: 6). Spoken alone, each word only has a single main stress, with the stressed syllable being the most prominent one in a word, the one that is loudest, has highest pitch and longest duration. This stress does not change among speakers (Katamba 2005: 16). When considering street and lamp, with the compound street-lamp, only the first syllable is stressed, whereas with street lamp, both words gain stress. When looking at words in this manner, although in orthographic terms apartment building is two words, in terms of stress it is a single word (Plag 2003: 5). This type of word definition seems to be closer to our instinctual thoughts of what a word is. What causes difficulty with this definition, however, is that as said before, not all words gain stress. In this way, even the stress-based definition does not cover all words.

Stress is not, however, the only phonological clue when it comes to words (Katamba 2005: 16). There are also rules in English as in other languages which dictate the positions and order in which sounds can take place and which sounds can be combined together. A phonological word must follow these regulations as it is spoken. For instance, the consonant sound of n cannot occur as the first sound of the word, although it can occur elsewhere (Katamba 2005: 16–17). Certain consonants also cannot occur in certain combinations together, such as stop-plus-nasal sounds including tm (Katamba 2005:

17). This is more prominent in the pronunciation of English, as the relationship between how words are spelled and pronounced in English is not always simple, but even within orthographic words we can usually tell when a word by its spelling conventions does not belong to English. For instance, a

(21)

17

word such as zvroglen would quickly be dismissed by most speakers as belonging to the English language (Katamba 2005: 18).

3.3 Other ways to define a word

Another definition method to consider for words is the integrity criterion, which sees words as invisible units “into which no intervening material may be inserted”. Therefore, any type of modification of words must happen at the edges of the word, but never inside of it (Plag 2003: 6).

This can be seen, for instance, in the use of the suffix -s with plurals and the use of prefix un- for negation. Indeed, these types of modifications always occur around the word they are modifying.

However, there are also cases where the modifications can take place within the word as well. For instance, the plural for daughter-in-law is daughters-in-law rather than daughter-in-laws. If we consider this to be a single word, the indication of its plurality occurs in its middle, rather than at the end. There are also some cases where curse words are added in the middle of other words, such as in the case of abso-bloody-lutely. Here, the word bloody is inserted in the middle of the word absolutely, rather than occurring at the edges of it. To gain the same effect, bloody cannot in fact in this case be placed at the edges of the word. Therefore, it seems there are some exceptions to this particular word definition rule as well (Plag 2003: 7).

“The semantic definition of “word” states that a word expressed a unified semantic concept” (Plag 2003: 7). This is indeed true for most words, but does not serve quite far enough to be a definition for what a word is and is not, as not every semantic concept can be explained by a single word in every language. Take for instance the concept of the smell of fresh rain in a forest in the fall. This is a single unified concept, but it cannot be expressed as a single word in the English language. Therefore, “a word may always express a unified concept, not every unified concept is expressed by one word”.

This way of defining words is not therefore very adequate. Even the definition of what unified

(22)

18

semantic concept means is rather vague here, making this definition of what a word is even less useable (Plag 2003: 7).

We can also distinguish between word-forms and words as vocabulary items in English (Katamba 2005: 18). This differentiation is important, as often when we talk about words, we are thinking of something more abstract rather than word-forms. These are referred to as lexemes, or vocabulary items. Words in a dictionary, for instance, are listed with this in mind. This happens in the cases of different word forms, for instance write-wrote-written. These three would not gain their own entries in a dictionary, but rather the first item would likely have its own entry, with the others mentioned underneath it (Katamba 2005: 18–19). In the cases of words that follow the general patterns of English, the other forms will likely be left out altogether, such as in the case of wash-washes-washed and so on. Instead, the grammar of the language is expected to provide the rules of how the words are used. The same goes for regular and irregular nouns and adjectives. Lexeme here is “an abstract entry that is found in the dictionary and that has a certain meaning. Word-forms are the concrete objects that we put down on paper (orthographic words) or utter (phonological words) when we use language”. (Katamba 19–20). Lexemes and word-forms relate to one another through the means of realisation, representation or manifestation. The lexeme write, for instance, can be portrayed in any of the forms that belong to it, such as wrote. These forms are the ones that happen in speech and writing as manifestations of the lexeme write.

There is also the syntactic way of defining words. In this, words are seen as “syntactic atoms, i.e. the smallest elements in a sentence” (Plag 2003: 7). They belong to syntactic classes, such as nouns and verbs, which are referred to as parts of speech, word classes or syntactic categories. The place of a word in a sentence is dependent on the syntactic rules of language, which refer to words and the

(23)

19

classes they are a part of. The, for instance, is a part of a class called articles, and the rules of English language determine where in a sentence such words can occur. With these thoughts in mind, we can test whether a word is a word by seeing if it belongs to a word class in this manner. Thus, if a word follows the rules of nouns, it can be counted as a noun, and therefore, it is a word. Additionally, only words or groups of words can be moved around in a sentence, furthering the criteria for word definition.

We can additionally look at words from a grammatical point of view (Katamba 2005: 21). Words have their own vital role in syntax, as sentences consist of different arrangements of words. “A word, in the sense of a lexical item with a certain meaning plus certain syntactic and morphological properties, is referred to as a GRAMMATICAL WORD”. Words generally need to have certain properties in order to be used in syntactic contexts (Katamba 2005: 21–22). Word-forms can have different types of grammatical usages in sentences, for instance when they occur in different tenses.

There are, however, differences in the appearance of certain tenses. Compare, for instance, the words play-played-played and take-took-taken. We can see that neither the past tense nor the supine look the same, and that with take, they do not even look the same as each other. This type of use of the same word-form for the realisation of “two (or more) distinct grammatical words that are represented separately in the grammatical representations of words, belonging to some other comparable lexemes” is referred to as syncretism by linguists. This is one method to tell apart word-forms and grammatical words. It lets us know that words of the same lexeme that have the same forms can still have different meanings. This can also be seen in verbs such as hit and cut, which can serve as both present and past tenses of the word (Katamba 2005: 23). This type of syncretism is not limited to verbs either, but can also be seen in cases of nouns such as sheep, which can be both singular and plural.

(24)

20

According to Katamba (2005: 11), Leonard Bloomfield, an American linguist, stated in 1926 that “a minimum free form is a word”. He therefore saw word as “the smallest meaningful linguistic unit that can be used on its own. Its form cannot be divided into any smaller units that can be used independently to convey meaning”. (Katamba 2005: 11). Words are generally treated and seen as the minimal free forms that can be used as tools of grammar in a language (Katamba 2005: 23). By this free form we refer to a unit that “can stand on its own and act as a free agent; it is an element whose position in a sentence is not totally dictated by other items”. To understand what is meant by

“freedom” in this context, we need to consider two ancillary ideas: positional mobility and stability.

An example of this would be the word child, which cannot be divided into bits that would alone convey meaning. Instead, the word childish can be divided into the bits of child- and -ish. Child in this case has a meaning on its own, but -ish does not (Katamba 2005: 11). To understand what is meant by “freedom” in this context, we need to consider two ancillary ideas: positional mobility and stability (Katamba 2005: 23). Words may not be the smallest grammatical units in sentences, but they are generally treated as unanalysable units when it comes to sentence formation. Words can usually appear in different places in a sentence and still make sense together. However, the elements that make up words cannot be moved around in this manner. Elements of word cannot be transported from one place to another in a sentence according to syntactic rules, when moving words around generally only creates a different type of emphasis or meaning in the sentence.

In this way, words appear as grammatical units with stability and internal cohesion (Katamba 2005:

24). The form of words is fixed and cannot be altered. When they are moved around in a sentence, they move as a block. This is the reason for why words are considered the minimal units of sentences, and intended to separate words from phrases in sentences. Phrases, as well as words, can exist on their own and can be moved around in the sentence, but they are not the smallest units of sentences, as they also consist of smaller units, words. Phrases also are not internally cohesive the way words

(25)

21

are, but can gain other words within them. My old house can easily become my very old beautiful house, where more words can be inserted into the phrase. However, even the assumption of the grammatical word as the smallest independent unit in a sentence can be problematic in certain cases, generally once again with compounds. Compound words can be said to not be the smallest possible independent unit, bur often rather consist of two or more of those smallest independent units.

Therefore, it is preferable to define words as units “on which purely syntactic operations can be performed”. This definition applies even with compounds (Katamba 2005: 24–25). One cannot manipulate even the compound word’s insides (Katamba 2005: 25). Syntactic rules cannot be applied separately to words creating the compound. For instance in the case of the compound word wheelbarrow, it can be modified by the adjective big to become big wheelbarrow, the adjective cannot be inserted between the two nouns wheel and barrow that make up the compound.

3.4 Concluding remarks

Looking through all these possible definitions for words, it can be seen that defining a word is a tricky business and not an easy task to do by any means, despite the initial instinct that it should be.

However, in most cases, by judging words through the stress criteria, the integrity criteria and the syntactic criteria together can lead to adequate results on defining what a word is. Through them, we can define words as:

(1) entities that have a part of speech specifications (2) syntactic atoms

(3) (usually) words only have one main stress

(4) (usually) are invisible units (nothing to be placed in the middle) (Plag 2003: 8)

Katamba (2005: 25) states that in order to avoid ambiguity when it comes to defining words, we need to separate between the types of words we are handling at the current moment:

“1 a word-form (i.e. a particular physical manifestation of at least one lexeme); in speech it is called a phonological word and in writing an orthographic word;

(26)

22 2 a lexeme (i.e. a vocabulary item);

3 a grammatical word (i.e. a unit of grammatical structure that has certain morphological and syntactic properties)”.

However, it is difficult to tell when a language has gained a new word (Harley 2005: 91). It is rare that anyone simply coincides a whole new word on the spot, and even rarer for such single-

handedly created words to become a part of the general uses of language. Most of the “new” words that come to existence are created through the many strategies that can be utilised to alter words that already exist. One of these strategies has been stated to be abbreviation. Yule (2010: 56, 58) sees Clipping and Acronyms, which are ways to abbreviate, as ways of creating new words. I will discuss abbreviations further next.

3.5 Word creation

When we start looking at words and how they are formed or brought into a language, we quickly discover that there are multiple ways for words to come to be (Yule 2010: 53). In many cases the words we now commonly use were once considered misuses of language, and this still sometimes happens when new terms are coined to describe new or even old things. It is more productive to simply see this process as the constant revolution and change of language. I will present some of the ways words can be coined or brought into a language below.

Coinage, which is the process of creating completely new words, is the rarest way for word

formation in English. It is most often utilised with trade names or commercial products, which later become commonplace terms. Examples of this include aspirin and nylon, but also Kleenex and granola (Yule 2010: 53-54). This way of creating words also includes words that originate from the names of people or places, referred to as eponyms. They include words such as sandwich and jeans.

(27)

23

A far more common way of bringing new words into a language is borrowing, which refers to the adoption of words from other languages. In English, borrowed words include croissant (from French) and piano (Italian). Other languages also borrow words from English. Something related to borrowing is also referred to as loan-translation or calque, where the original elements of a word are translated into the language it is entering (Yule 2010: 54–55). For instance, the English phrase moment of truth is seen as having come from the Spanish phrase el momento de la verdad (Yule 2010: 55).

Another common way of creating words, alluded already above somewhat, is compounding. This is the process of joining two or more words together to create new ones. It is especially common in English and German, but not as much in French or Spanish. Some of the English compounds include bookcase and good-looking. Somewhat related to compounding is the process of blending, which also has the combining of two different forms, but usually in a way that only takes the

“beginning of one word and joining it to the end of the other word”. An example of this would be smog, which is created from “smoke” and “fog”. Clipping resembles blending in that it is also a process of losing some elements of original words, but to a further degree. Clipping “occurs when a word of more than one syllable is reduced to a shorter form, usually beginning in casual speech”.

This includes for instance ad for “advertisement” and flu for “influenza”. British and Australian English include a special form of Clipping, referred to as hypocorisms, where a longer word is clipped into a single syllable, and then -y or -ie is added as an ending to that clipped syllable. An example of this would be movie (“moving pictures”). Another type of reduction process to create words is known as backformation. This refers to the reduction of a word of one type (typically a noun) to become a word of another type (typically a verb). An example of this would be the verb televise, being a reduced form of the noun television (Yule 2010: 57).

(28)

24

Conversion is a way of changing the function of the word to other purposes. For instance, a noun becomes used as a verb. This is obviously rather similar to backformation, except that the words do not change in appearance. Nouns that have become verbs include bottle, while verbs that have become nouns include spy. Both can also become adjectives. Yule (2010: 58) refers to Acronyms as a way of forming words from the initial letters which form a set of other words. They can either be pronounced by saying each separate letter, such as CD (“compact disk”), or as single new words, such as NASA. Although these examples have kept their capita letters, many acronyms also lose the capitalisation and become everyday words, such as in the case of laser (“light amplification by stimulation emission of radiation”). The most common way of creating new words into English is, however, derivation. It is accomplished by the use of affixes in words, such as un-, pre- and -ness, which can be used to adjust words. In this way, the word happy can, of instance become

unhappiness.

Although the word formation processes have been listed separately here, it is important to note that they can also come together to form words (Yule 2010: 60).

4. On abbreviations

4.1 Abbreviations in general

Netspeak can be considered to be a new form of language with its own various unique features. One of these features, perhaps even the most obvious one, is the lexicon exclusive to the language of the Internet. As mentioned by Crystal (2001: 81–82), there are many words and phrases that are used in various situations exclusively on the Internet, making it a very creative lexical domain in present-day English. It involves all major lexical processes, or methods of word-creation, one of which is abbreviation. Abbreviation, according to Crystal (2001: 84), is “one of [Netspeak’s] most remarked features”. Poplavskaia and Svitsun (2010: 186) also point out that while Netspeak is generally formed by comprising features of spoken and written forms of language, it uses abbreviations extensively.

(29)

25

This is generally done in order to save time, space and effort on the part of the chatters, and also to express “some emotional condition”. Chatting can indeed be very colloquial (Ferreira da Cruz 2008:

100). The setting of a chatroom has also been compared to a bar: people can ignore interlocutors, enter and depart the conversation whenever they see best as well as maintain multiple conversations simultaneously. Communication of this type can however also be very fragmentary, and to make it simpler, there are special resources that can be used, such as smileys (Jenks 2009: 29).

Poplavskaia and Svitsun (2010: 186) additionally note that the term “abbreviation” is derived either from the Latin words abbrevio and brevis, meaning “short” or from Italian abbreviatura, meaning

“shortening”. Markus Beiswanger (2007: 3–4) defines shortening as “a neutral term to cover all forms of lexical shortening strategies. Shortenings in the sense of this study are all lexical forms that are made up by fewer characters than the full from of a word or a combination of words”. Mattiello (2013:

64) uses the term abbreviation as an umbrella term for Clippings, Acronyms and Initialisms, which according to her “share the same abbreviatory mechanisms, obtaining new word-forms by shortening existing lexemes”. Looking at it this way, this study uses “abbreviation” and “shortening” as synonyms which denote the same thing. Mattiello (2013: 65), however, also states that some people prefer to use “shortening” in the place of “abbreviation” when referring to the same thing. Still, certain people only utilise the term “abbreviation” for specific cases, such as for words that are pronounced as their initial letters. In these cases, “shortening” has been the preferred umbrella term (Mattiello 2013: 67). Here, however, although the terms vary from person to person, they are considered each other’s synonyms.

Abbreviations have, according to Mattiello (2013: 64), been used since the ancient times, but have become especially prominent in the modern times. This is likely due to the need for more efficient

(30)

26

vocabulary for different uses and also to establish familiarity or intimacy between people who use them. According to Mattiello (2013: 64), abbreviations in the modern era, especially Clippings, are generally used in less formal contexts, and denote the feeling of familiarity between speakers. Indeed, many Clippings are forms of specific slangs, indicating for instance family members (hubby for

“husband) and teenagers (disco for “discotheque”). It has been taken further with, for instance, SMS and chatting. Abbreviating indeed uses various methods to create new words or variants of new words as well as phrases for people to express themselves with. The degree and manner of shortening, however, can vary greatly from one abbreviation to another, and this is what creates the different categories of abbreviations.

4.2 Categories of abbreviations

For the purposes of this study, I will be using the categories of abbreviations used by Bieswanger in his 2007 article ‘2 Abbrevi8 or not 2 Abbrevi8: A contrastive analysis of different space- and time- saving strategies in English and German text messages’. These same categories were also used in a successful manner in Karoliina Koljanen’s BA essay, Gender Differences in the Use of Lexical Shortenings in Twitter Messages (2013), as well as my own BA essay, so I have found them suitable for my purposes. They are far-reaching and seem to cover many different types of abbreviations.

Crystal (2001: 84–86) denotes only two categories for abbreviations, Acronyms and Letter-plus- number combinations, perhaps because in 2001 Netspeak was still such a new phenomenon, and this did not feel like it would cover enough abbreviation types. In her study of Instant Message conversations, Baron (2008: 59–62) also divided the abbreviations she had found into only two categories: Acronyms and Contractions. Poplavskaia and Svitsun (2010) also have many categories similar to those of Bieswanger (2007). However, as Bieswanger’s categories seemed to lack nothing, they were the ones I ended up using for the purposes of this study, despite the fact that his article focused on the language of text messages rather than that of chatrooms. Grace et al. (2015: 81) have named some similar categories in their study of the use of what they call “textisms”, such as

(31)

27

contractions, shortenings and single homophones, but these do not seem specific enough and not all of the rest of them are really related to abbreviations.

Bieswanger has placed abbreviations into six different categories: Initialisms, Clippings, Letter- number homophones, Phonetic spellings, Contractions and Word-value characters. To these, I have added the category of Apostrophe-frees, making this study have seven categories of abbreviations in total. Initialisms are abbreviations where only the first letter of the words in a phrase are used, such as in the case of ROFL (“rolling on the floor laughing”). Clippings refer to dropping or deleting letters of any part of a certain word, making all the arrangements shorter than the original word, but also preserving some of its original letters while not adding any extra ones. An example of this would be bday (“birthday”). Letter/number homophones refer to when letters or numbers with identical pronunciation replace a word or a part of word, such as when the letter u is used for the word “you”.

Phonetic spelling is related to this, indicating words that are spelled differently from their actual, official forms in a way so that the pronunciation pattern remains the same, including such cases as nuff (“enough”). Contractions refer to forms such as gonna (“going to”) that consist of two words that are combined by leaving out medial letters and space. Word-value characters refer to characters and letters representing whole words, with examples such as f referring to “female”, usually when one first introduces herself on a chatroom. Finally, my own category of Apostrophe-frees refers to abbreviations which, as the name states, are missing the apostrophe (‘) in places where it should be.

This refers to such cases as where “can’t” is spelled as cant. I have put the categories and their examples in the chart below.

Table 1: The categories of abbreviations

Category Explanation Example

Initialisms uses the first letters of words

lol for “laughing out loud”

(32)

28

Clippings drops or deletes parts of words

pic for “picture”

Apostrophe-frees lack the apostrophe (‘) where it should be

dont for “don’t/do not”

Letter/number homophones letters/numbers with the same pronunciation replace

the word

u for “you”

Phonetic Spellings words are written as they are pronounced

ya for “yeah”

Contractions two words combined into one, lacking the space

gonna for “going to”

Word-value characters characters stand for whole words

f for “female”

4.3 Further on the categories

Some of these abbreviation categories have been talked about more widely in literature than others, with varying definitions on what belongs to which category. Harley (2006: 96) sees Initialisms as

“extreme clipping; using the initial letters of the content word in a phrase to stand in for the whole phrase”. She claims this method has been used to create words in English for some time now, some Initialisms such as C.O.D (“cash/cost/collect on delivery”) originating already in the 1800s, but it has only really taken off as a means of word-formation in the latter half of the 20^th century. According to Harley, Initialisms also has two subgroups: Acronyms which are Initialisms that are pronounced as a united phonological word “according to the spelling conventions of English”, such as in the case of AIDS (“Acquired Immune Deficiency Syndrome”), and Abbreviations where “the letters are read one at a time”, such as in the case of MS (“Multiple Sclerosis”). This same split is used between Initialisms and Acronyms by Mattiello, her using the term Initialisms to replace Abbreviations here (2013: 83).

Mair (2006: 61) has used these same terms as Mattielo, although he later also seems to include more Clipping-like words (such as quango for “quasi non-government(al) organization”) into the category of Acronyms. However, there are also borderline cases which can accept both ways of pronouncing (Mattiello 2013: 83). This is sometimes determined by the use of dots or capital letters (ASAP or a.s.a.p) (Mattiello 2013: 84). However, this split was not taken into account with this particular study,

From lol to kinda - the use of abbreviations in online chatrooms