Conversation Analysis of Internet Relay Chat……………….……………… 56 4.1 Address Forms……………………………………………………………56 4.1.1 Theoretical Background……………………………………………

Home , Flaming (Internet)

Masarykova univerzita v Brně

Filozofická fakulta

Katedra anglistiky a amerikanistiky

Studijní program: 7301 V Filologie

Studijní obor: anglický jazyk

Language of Internet Relay Chat

disertační práce

Školitelka: PhDr. Jitka Vlčková, Ph.D.

Brno 2006 Mgr. Šárka Kašičková

Prohlašuji, že jsem disertační práci vypracovala samostatně a v seznamu pramenů a literatury uvedla veškeré použité informační zdroje.

2 Acknowledgements

I would like to express my thankfulness to my supervisor, Ph.Dr. Jitka Vlčková, Ph.D. , for her guidance, valuable feedback and support that she provided throughout this work.

For advice and assistance with statistical methods, I would like to thank my colleague Ing. Vladimíra Valentová, Ph.D.

For correction of grammar and spelling I express thanks to my colleague Gladice Hughes, M.Sc.

Finally, I would like to thank my partner Jan and my family for their moral support. Without their patience and tolerance this thesis would have not been completed.

3 Summary

This work makes an inquiry into the conversation practises on Internet Relay Chat (hereafter IRC) that are determined by specific conditions of an Internet chatroom. The specific conditions are anonymity, the lack of audio-visual cues and an exclusive text-based interaction. In this respect, it focuses on those conversational elements that are regarded pivotal for successful communication. These conversation techniques are address forms, opening sequences, closing sequences and turn-allocation strategies.

This empirical study of IRC language sets the aim of investigating the impact of age on the given conversation strategies on IRC and to answer the following research question: To what extent are specified conversation techniques used on Internet Relay Chat affected by the variable of age?

Two main hypotheses are verified throughout the analysis: H0 claims that conversational techniques used on IRC are independent of age, while H1 states that the form of communication on IRC depends on age.

Furthermore, three representative age groups have been selected and particular attention has been paid to their performance in a chatroom. The age groups under investigation are as follows – Teens (13-19 years old), 30+ (30-59 years old) and 60+ (over 60 years old). A substantial corpus containing 2,382 IRC messages was collected in years 2004 and 2005 from various chat providers. Each age group contains an equal number of 794 messages.

The conversation analysis presented in this work, then, draws on two sources: the research done in spoken conversation: address forms (Adler, 1978; Brown and Ford, 1964; Little and Gelles, 1975); opening sequences (Schegloff, 1968; Schiffrin, 1977); closing sequences (Schegloff and Sacks 1973; McLaughlin, 1974); turn- allocation strategies (Sacks et al., 1974; Zimmerman, Boden, 1991; Schegloff, 2000) and the research in Internet language (Herring, 1996; Werry, 1996; Rintel et al. 2001; Panyametheekul, Herring, 2003 and others).

The statistical comparison of the age groups reveals quantitatively and qualitatively varied distribution of individual elements across the age groups and their different proportions. Thus, the data obtained via the analysis also points to distinct conversation techniques and „rituals‟ employed by the groups in their discourse.

4 The thorough analysis and the results can be found in the practical part of the thesis - chapters 4.1 Address Forms, 4.2 Opening Sequences, 4.3 Closing sequences and 4.4 Turn-allocation Strategies. The whole corpus is attached to the dissertation in a form of an Appendix.

The main contribution of the thesis to the field of linguistics and conversation analysis is to be viewed as follows:

1. It introduces a rather novel and non-traditional view to the conversation techniques of synchronous Internet chatting related to the variable of age.

2. It demonstrates the variation of conversation methods of individual age groups.

3. Address forms – It has been found that they are closely linked to the level and quality of communication on IRC. In Teens chatting resembles individual „shout outs‟ without any specific coherence and a very limited number of attempts for real conversation, therefore we cannot expect a high number of address forms. On the contrary, 60+ IRC participants develop a more sophisticated type of conversation that resembles spoken exchanges where nearly every participant entering the chatroom is addressed and greeted. In this conduct, I observe a kind of ritual or a norm of behaviour that every participant entering the chatroom should follow to be successful in an interaction.

4. Opening sequences – It has been found that greetings and AJS summons hold the first positions from the point of view of frequency and efficiency. This means that the standard opening strategy consisting of a range: AJS → greeting → question → contact advertisement represents the „ritual of access‟ on IRC regardless the age.

5. Closing sequences – I observe the following sequence: ALS without any previous contribution → ALS with a previous contribution → continuity statements → pre- closing sentence+continuity statements+welfare concern. However, when age is taken into account, the results show that the group 60+ follows the rules resembling a spoken interaction more than the other two groups. This means that the closing sequence in 60+ is: continuity statements → pre-closing → welfare concern → saying goodbye. Closing sequences in 60+ represent 65 per cent of the total occurrence rate. The results of my analysis suggest that the older the chat participants, the more important and frequent it is, after establishing a certain kind of relationship inside a chat room, to reaffirm it with continuity statements, well-wishing or creative cyber closings, group 30+ and 60+. Provided that there was no need for chat participants to act in this way, there would be no need for good-byes either, and the conversation could end with

5 a vague pre-closing topic, termination section acknowledged with ALS or with no closing at all as in the majority of cases in the Teen group.

6. Turn-allocation strategies - My findings based on quantitative and qualitative analyses point out that the most used strategy is Strategy A, the current speaker selects the next speaker, 75.5 per cent regardless of age. In other words, this one bears the most resemblance to the circumstances of face-to-face conversation, with a limited number of overlapping contributions. Still, I claim that the transfer of turn-taking skills from a face-to-face setting to the IRC environment is problematic. It is important to mention that one of the most specific characteristics of IRC is its semi-permanence. This means that previous utterances, or turns, are available for direct review by conversational partners. This feature affects conversational practises in many ways. The prevalence of Strategy A may be the consequence of this effect. IRC, serving as a record of participants‟ input, may also help suppress some problems associated with the increase of a group size. Thus, the circumstances described above provide users with a flexibility that is not possible in face-to-face conditions.

7. The analysis verifies the hypothesis H1 and confirms that that age has a medium- strong affect on the conversational practises of chatroom participants in all four conversational denominations.

8. Apart from quantitative and qualitative methods, the work introduces another statistical method of chi-square test for independence that verifies strength of dependence of individual variables.

The study is expected not only to be an asset for understanding the principles of synchronous Internet chatting, albeit this is its central objective, but also to provide some valuable hints as regards the rules, norms and rituals governing Internet Relay Chat in general. Thus, the analysis may serve as a helpful aid for chat providers working to improve the IRC interaction programmes, and where age becomes one of the main denominators.

6 Shrnutí

Tématem této práce jsou konverzační metody při synchronním chatování na Internetu, které jsou podmíněny specifickými podmínkami chatovací místnosti. Tyto specifické podmínky jsou anonymita, nedostatek audio-vizuálních podnětů a interakce založená výhradně na písemném projevu. Práce se tudíž zaměřuje na takové konverzační prvky, které jsou považovány za nezbytné pro úspěšnou konverzaci. Jsou to formy oslovení, sekvence zahájení konverzace, sekvence ukončení konverzace a konverzační strategie “turn-taking”.

Tato empirická studie jazyka synchronního chatování na Internetu si klade za cíl prozkoumat vliv věku na dané konverzační strategie a odpovědět na následující výzkumnou otázku: Do jaké míry jsou konverzační techniky užívané při synchronním chatování na Internetu ovlivněné věkem?

Během analýzy jsou ověřovány tyto dvě hypotézy: hypotéza H0 tvrdí, že konverzační techniky nejsou závislé na věku, hypotéza H1 tvrdí, že konverzační techniky jsou závislé na věku.

K tomuto účelu byly vybrány tři reprezentativní věkové skupiny a bylo sledováno jejich chování v chatovacích místnostech. Předmětem zkoumání jsou následující věkové skupiny: Mladiství (13-19 let), 30+ (30-59 let) a 60+ (přes 60 let). Vytvořený korpus obsahuje 2 382 Internetových zpráv a byl shromážděn mezi roky 2004 a 2005 od různých poskytovatelů chatovacích místností. Každá věková skupina zahrnuje 794 Internetových zpráv.

Konverzační analýza prezentovaná v této práci vychází ze dvou zdrojů: z výzkumu v oblasti mluvené konverzace a výzkumu z oblasti Internetového jazyka. První část výzkumu zahrnuje oslovení (Adler, 1978; Brown and Ford, 1964; Little and Gelles, 1975); sekvence zahájení konverzace (Schegloff, 1968; Schiffrin, 1977); sekvence ukončení konverzace (Schegloff and Sacks 1973; McLaughlin, 1974); konverzační strategie “turn-taking” (Sacks et al., 1974; Zimmerman, Boden, 1991; Schegloff, 2000). Druhá část potom vychází a navazuje na výzkum v oblasti konverzační analýzy chatu – “computer-mediated communication” (Herring, 1996; Werry, 1996; Rintel et al. 2001; Panyametheekul, Herring, 2003 a další).

Statistické srovnání výše uvedených věkových skupin odhaluje kvantitativně a kvalitativně rozdílnou distribuci jednotlivých prostředků a zároveň i jejich rozdílné

7 poměry. Data získaná analýzou tak také poukazují na rozdílné konverzační techniky a „rituály‟ uplatněné věkovými skupinami v jejich konverzaci.

Podrobná analýza a její výsledky se nacházejí v praktické části práce – kapitoly 4.1 Oslovení, 4.2 Sekvence zahájení konverzace, 4.3 Sekvence ukončení konverzace a 4.4 Konverzační strategie “turn-taking”. Celý korpus je k nahlédnutí v části Přílohy.

Největším přínosem této dizertace pro lingvistiku a konverzační analýzu jsou následující body:

1. Dizertace představuje poměrně nový a netradiční pohled na konverzační techniky synchronního Internetového chatování spojené s věkem.

2. Práce demonstruje různorodost konverzačních metod u jednotlivých věkových skupin s těmito stěžejními výsledky:

3. Oslovení - formy oslovení jsou úzce spjaté s úrovní a kvalitou komunikace v chatovacích místnostech. U Mladistvích se chatování podobá jednotlivým "výkřikům" bez jakékoliv koherence a omezeným počtem pokusů o standardní konverzaci. To znamená, že počet oslovení je u této skupiny nejmenší. Na druhé straně skupina 60+ rozvíjí sofistikovanější typ konverzace, které se podobají mluvené interakci; každý nově příchozí účastník je osloven a pozdraven. V této skupině je patrný určitý typ rituálu, který každý nový účastník chatování dodržuje.

4. Sekvence zahájení konverzace - analýza ukázala, mimo další skutečnosti, že pozdrav a "AJS výzva" zaujímají první místo s ohledem na četnost výskytu a efektivitu. Znamená to, že standardní sekvence zahájení konverzace se skládá z: AJS - pozdrav - otázka - kontaktní inzerát a představuje takzvaný "vstupní rituál" synchronního chatování na Internetu.

5. Sekvence ukončení konverzace - analýza ukázala tuto sekvenci ukončení konverzace "ALS" bez předchozí konverzace - "ALS" s předchozí konverzací - "continuity statements" - "pre-closing sentence + continuity statements+welfare concern". Výsledky analýzy ukazují, že čím jsou účastníci chatu starší, tím je důležitější a častější upevnit již navázaný kontakt v chatovací místnosti - "continuity statements", "well- wishing" nebo kreativní ukončení konverzace - zejména ve skupinách 30+ a 60+.

6. Konverzační strategie “turn-taking” - Strategie A - současný mluvčí si vybírá příštího - je dle tohoto výzkumu nejčastější strategií. To znamená, že se podobá okolnostem v rámci mluvené konverzace s omezeným počtem překrývajících se

8 promluv. Přesto analýza potvrzuje, že přenos dovedností je z prostředí projevu mluveného do prostředí synchronního chatování velmi problematické. Jedna ze specifických vlastností synchronního chatování na Internetu je možnost jeho uchování - předchozí promluvy a zprávy jsou viditelné a lze je znovu prohlédnout - "scroll-up affect". Tyto okolnosti mohou poskytnout účastníkům synchronního chatování možnost flexibility, která není možná při mluvené konverzaci.

7. Analýza potvrzuje hypotézu H1, že věk má středně silnou závislost na chatovacích konverzačních metodách u všech čtyřech prověřovaných denominací.

8. Kromě kvantitativní a kvalitativní metody analýzy práce představuje jiný typ statistického zkoumání, tzv. ² test nezávislosti, který prověřuje sílu nezávislosti daných proměnných.

Cílem této studie není pouze napomoci porozumění principům synchronního Internetového chatování, i když je to její hlavní záměr, ale také poskytnout cenné poznatky a postřehy ohledně pravidel, norem a rituálů, které řídí synchronní Internetové chatování jako takové. Analýza tedy může sloužit jako užitečná pomoc pro poskytovatele chatu, kteří neustále zlepšují interakční programy, a kde se věk stává jedním z hlavních jmenovatelů.

9 Table of Contents

Acknowledgments ……………………………………………………………….. 3

Summary…………………………………………………………………………. 4

Shrnutí……………………………………………………………………………. 7

List of tables ………….………………………………………………………….. 13 List of graphs…………..…………………………………………………………. 14 List of figures ………….…………………………………………………………. 15 Glossary of Terms..………………………………………………………………. 16 List of Abbreviations (acronyms) Used in Chatrooms ………………………….. 22 1 Introduction……………………………………………………………………. 25 1.1. Modern Trends in Communication and Linguistic Research……………. 25 1.2. Research Objectives – Specificity of This Research…………………….. 26 1.3. Computer-Mediated Communication……………………………………. 29 1.3.1. Asynchronous Groups……………………………………………. 29 1.3.2. Synchronous Groups – Internet Relay Chat ……………………... 29 1.3.3. Netiquette on Internet Relay Chat ……………………………….. 32 2 Theoretical Framework………...…………….………………………………. 35 2.1 Significance of Conversation Analysis and Sociolinguistics in This Work………………………………………………………………… 35 2.1.1 Conversation Analysis …………………………………………… 36 2.1.2 Sociolinguistics ………..…………………………………………. 37 2.2 Research on Conversation Analysis on Internet Relay Chat……………. 39 2.3 Computer-Mediated Versus Spoken or Written Interaction?…………… 42 2.4 Language of Internet Relay Chat…………………………………………45 3 Methodology…………………………………………………………………… 48 3.1 Research Methodology and Statistical Analysis…………..……………… 48 3.2 Data Collection ……………………………………………………………49 3.3 Corpus Description – Material Analysed………………………..……….. 52 3.4 Age Variable……………………………………………………………… 54 4 Conversation Analysis of Internet Relay Chat……………….……………… 56 4.1 Address Forms……………………………………………………………56 4.1.1 Theoretical Background…………………………………………….. 56 4.1.1.1 Address Forms in Face-to-Face Communication and IRC…… 58

10 4.1.2 Analysis……………………………………………………………. 61 4.1.2.1 Address Forms in Opening Sequences……………………….. 62 4.1.2.2 Address Forms in the Course of IRC Conversation…………. 69 4.1.2.3 Address Forms in Closing Sequences………………………… 76 4.1.3 Address Forms – Conclusion………..……………………………… 79 4.2 Opening Sequences………………….………………………………….. 90 4.2.1 Theoretical Background…………………………………………….. 90 4.2.1.1 Research on IRC Openings…….………………………………91 4.2.2 Analysis…………………………………………………………….. 94 4.2.2.1 Summons-Answer Sequence…………………………………. 94 4.2.2.2 Pre-Conversational Openings – Greetings……………………. 100 4.2.2.3 Question-Answer Sequence…………………………………... 104 4.2.3 Opening Sequences – Conclusion………………………………….. 108 4.3 Closing Sequences………………………………………………………. 113 4.3.1 Theoretical Background………...…………………………………... 113 4.3.2 Analysis……………...………………………………………………116 4.3.3 Closing Sequences – Conclusion…………………………..………. 124 4.4 Turn-allocation Strategies…………………………….….…………….. 126 4.4.1 Theoretical Background……...…………………………………….. 126 4.4.1.1 Turn-Taking in Face-to-Face Conversation………………….. 127 4.4.1.2 Turn-allocation on IRC ………………………………………. 131 4.4.2 Analysis………...……………………………………………………137 4.4.2.1 Turn-taking Strategies in the Three Age Groups…………….. 137 4.4.2.2 Statistical Analysis….………………………………………… 144 4.4.3 Turn-allocation Strategies – Conclusion…….……………………… 150 5 Final Considerations………………….……………………………………….. 153 5.1 Motivation, Objectives and Material Researched………………………… 153 5.2 Methodological Framework……………………………………………… 155 5.3 Results in the Context of the Research Question………………..……….. 155 5.3.1 Address forms..…………………………………..…………………. 156 5.3.2 Opening Sequences………………………………..……………….. 158 5.3.3 Closing Sequences…………………………………..……………… 160 5.3.4 Turn-allocation Strategies …………………………..……………… 162 5.4 Thesis Limitations………………………………………………………… 164

11 5.5 Concluding Remarks……………………………………………………… 164 Bibliography……………………………………………………………………… 166 Appendix A: A1. Teens………………………………………………………….. 174 Appendix B: A2. Teens………………………………………………………….. 180 Appendix C: A3. Teens………………………………………………………….. 186 Appendix D: B. 30+……………………………………………………………… 192 Appendix E: C. 60+...... 210 Appendix F: Author´s published articles in the context of this thesis...... ….. 227

12 List of tables

Table 1: Netiquette on IRC – “the do´s and the do not´s”

Table 2: Basic distinctions of virtual and real conversations (adapted from Quero, 2003)

Table 3: Teens: the form of address when entering the room – opening

Table 4: 30+ : the form of address when entering the room – opening

Table 5: Over 60: the form of address when entering the room – opening

Table 6: Strength of dependence – opening address forms

Table 7: Teens: the form of address within IRC conversation

Table 8: 30+: the form of address within IRC conversation

Table 9: 60+: the form of address within IRC conversation

Table 10: Strength of dependence – opening address forms

Table 11: Teens: the form of address when closing conversation

Table 12: 30+: the form of address when closing conversation

Table 13: 60+: form of address when closing conversation

Table 14: Total occurrence rate in address forms in the three stages of chatting and age

Table 15: Strength of dependence – address forms in the three stages of IRC

Table 16: The use of the pronominal address YOU in the three stages of chatting by the age categories (occurrences)

Table 17: Opening by means of AJS in the three age categories

Table 18: Occurrence rate of contact advertisements and their efficiency

Table 19: Greetings-returning greetings in the three age categories

Table 20: Relation between IRC greetings that lead to further discussion and greetings that were followed by no further discussion

Table 21: Types of questions in individual age categories

Table 22: Question-answer sequence on IRC

Table 23: The efficiency of openings in Teen groups

13 Table 24: The efficiency of openings in the group of 30+

Table 25: The efficiency of openings in the group of 60+

Table 26: The occurrence of closing sentences in the individual age groups

Table 27: The connection between age, type of responses and its effectiveness

Table 28: Per centage of use of turn-allocation strategies in the whole corpus regardless the age

Table 29: Turn-allocation strategies on IRC in relation with age

Table 30: Strength of dependence – Turn-taking – Strategy A

Table 31: Strength of dependence – Turn-taking – Strategy B

Table 32: Strength of dependence – Turn-taking – Strategy C

List of graphs

Graph 1: Relation between age and total per centage of occurrences of address when opening conversation on IRC.

Graph 2: Relation between age and total per centage of occurrences of addressing in the course of conversation.

Graph 3: Relation between age and total per centage of occurrence of address when leaving conversation.

Graph 4: Relation between age and the total per centage of occurrences of addressing on IRC

Graph 5: The influence of age on the frequency of use of any address forms in all three stages of chatting

Graph 6: Comparison between direct address by nickname and a general noun in all three stages of chatting

14 List of figures

Figure 1: Example of a chatroom screen

Figure 2: Schematic representation of turn-taking in an IRC sample (adapted from Herring, 1999)

15 Glossary of Terms adjacency pair – A sequence of two utterances by different speakers in conversation. The second is a response to the first, e.g. question-answer. addressing – It denotes a constant quality of speech, namely, the fact that any utterance is addressed to somebody, every utterance is „dialogic‟. It is the quality of turning to someone. Within this category, this thesis differentiates between the system of direct address and indirect address. access rituals – Sequences of speech acts in which parties display to one another whether and to what extent they are available for subsequent interaction. acronyms (IRC) – Words made from the initial letters of other words used extensively on IRC, i.e. LOL (laughing out loud). asynchronous chatting – The situation when the Internet allows people to engage in a multi-party conversation online in postponed time, for example: discussion groups, help lines, FAQs. automated joining signal – Written signal that announces that a chat participants has joined a conversation, for example: “T-ROCK has joined the room.” automated leaving signal – Written signal that announces that a chat participant has left a conversation, for example: “Grauwls left the room.” back-channel utterances – Brief arguments, repetitions, or mirror responses by a listener that are believed to occur primarily during pauses in the turn of the speaker who has the floor. channel – A site on the network, as on IRC, where online conversations are held in real time by a number of computer users. chatroom – A channel held in IRC servers on a particular network, each devoted to a particular topic where users are simultaneously in contact with each other.

16 chi-square test for independence – statistical method that studies the relationship between two variables through classifying multi-nominal count data on two scales or dimensions. This can be done by setting up a contingency table. closing – The act in which some rules or rituals are obeyed in order to depart from conversational exchange. coherence – The familiar and expected relationships in experience which we use to connect the meanings of utterances, even when those connections are not explicitly made. cohesion – The property of a conversation that its successive utterances can be seen to be about the same elements, usually evidenced through such devices as anaphora that are visible in the conversational text. computer-mediated communication (CMC) – It is in its nature near-real time conversation where writing is a primordial activity (an instrument) that registers voice as its graphic representation. Its synchronous form is called Internet Relay Chat (IRC). conditional relevance – A property of the functional organization of interaction that some speech acts appear to establish “slots” for the subsequent performance of other speech acts, such that failure to provide the called-for speech act may be marked, repaired, or sanctioned. contact advertisement – Type of advertisement on the Internet similar to a newspaper advertisement by which chat participants look for a partner, new friends or compatriots to chat with. context – The physical environment in which a word is used. cf. co-text. conversation – Relatively informal social interaction in which the roles of speaker and hearer are exchanged in a non-automatic fashion under the collaborative management of all parties. conversation analysis – The study of language use with reference to social and psychological factors that influence communication.

17 cooperative principle – A basic assumption in conversation that each participant will attempt to contribute appropriately, at the required time, to the current exchange of talk. co-text – The linguistic environment in which the word is used.

Cramer´s V index – statistical index measuring the strength of dependence between two variables. direct address –The way in which people address one another in speech or writing occurring between a speaker/writer and their immediate listener/reader. emoticon – (Emotional Icon) A sideways facial glyph used in e-mail or IRC to indicate an emotion or attitude, as to indicate intended humour, e.g. : - ). external legitimizer – Type of closing in the more formal conditions that invokes an external motive for termination of conversation, for example: “I have to buy bread and milk”. first pair part – The first utterance in an adjacency pair, e.g. „How are you?‟ See also second pair part. flaming – Originally, “flame” meant to carry forth in a passionate manner in the spirit of honourable debate. Recently, flaming has come to refer to any kind of derogatory comment no matter how witless or crude. floor – Mechanism in conversation analysis determining people‟s use of language in an extended, open conversational setting, in this case who holds the right to speak. honorific – Expression which marks that the addressee is of higher status. ideational function – The use of language as a means of giving structure to thought and experience. indirect address – Determines a system where speakers/writers refer to those who are either not present, or those who are present but do not participate in the interaction directly.

18 inference – The listener‟s use of additional knowledge to make sense of what is not explicit in an utterance. internal legitimizer - Type of closing in the more formal conditions that invokes an internal motive for termination of conversation, for example: “I think we have finished all the questions”. This type is non-existent on IRC.

Internet Relay Chat (IRC) – A multi-user synchronous communication facility that is available all over the world to people with access to the Internet network of computer systems. lag - An extreme slow-down of message exchange times caused by latencies in the Internet‟s packet-switched transmission-reception system. netiquette – The code of conduct, etiquette, on the Internet. nonterminality – The notion introduced by Schegloff when a completed summons- answer sequence cannot properly stand as the final exchange of a conversation. opening – The act in which some rules or rituals are obeyed in order to achieve coordinated entry into conversational exchange. overlap – More than one speaker talking at the same time. paralanguage – Complementary features of speech characterized by body language, i.e. gestures, mimics, laughter and other non-verbal communication. politeness – Showing awareness of another person‟s public self-image. pragmatics – The study of speaker meaning as distinct from word or sentence meaning. Pragmatics studies language as it is used in a social context, including its effects on the interlocutors. recipient design – A multitude of respects in which the talk by a party in a conversation is constructed or designed in ways which display an orientation and sensitivity to the particular other(s) who are the co-participants. ritual – Stereotyped behaviour; a formalised, predetermined set of symbolic actions generally performed in a particular environment at a regular, recurring interval.

19 second pair part – The second or response utterance in an adjacency pair, e.g. „Fine, thanks‟. See first pair part. server – A computer that delivers information and software to other computers linked by a network. sequence – A series of three or more speech acts, which constitutes a self-contained discourse unit with a coherent internal structure. In some sequences, each act in the sequence may be functionally dependent or conditionally relevant upon the act that precedes it. sociolinguistics – The study of language and linguistics behaviour as influenced by social and cultural factors. spamming – An inappropriate attempt to use a mailing list or other networked communication facilities by sending the same message to a large number of people who did not ask for it. synchronous chatting - The situation when the Internet allows people to engage in a multi-party conversation online in real time, for example: Internet Relay Chat. terminating rule – The rule used by members of the society to limit the number of repetitions of summonses. topic-initiating sequences – Sequences that serve to open sections of topical talk.

Transition Relevance Place (TRP) – A possible change of speaker point in an interaction. turn – A structural slot, within which a speaker has the right to one turn-constructional unit, renewable with the consent of the other parties. turn-allocation component – The notion of turn-taking system for conversation described by Sacks et al. (1974). Turn-allocation techniques are divided into two groups: a) those in which next turn is allocated by current speaker‟s selecting next speaker; and b) those in which a next turn is allocated by self-selection.

20 turn-constructional component – The notion of turn-taking system for conversation described by Sacks et al. (1974). It includes various unit-types with which a speaker may set out to construct a turn. Unit-types for English include sentential, clausal, phrasal, and lexical constructions. turn-taking – A prominent type of social organization, whose instances are implicated in a wide range of other activities. Turn-taking referred to in this work is used for the ordering of conversation. utterance – A spoken proposition: a unit of speech corresponding to a single sentence or independent clause. virtual reality – A computer simulation of a real or imaginary system that enables a user to perform operations on the simulated system and shows the effects in real time.

21 List of IRC Abbreviations (acronyms) Found in the Corpus

IRC abbreviations Translation Chatroom Occurrence rate Ali - g Character created by the A3, B 3+1 (4) comedia Sacha Baron Cohen any1 anyone A2 2 asl (alternatively age, sex, location (posed as a A1, B 1+1 (2) a/s/l) question) aussie person from Australia A1, A2, A3 5+7+1 (13) b, bye B, C 2+1 (3) buh(onomatopoic) bbl be back later A2, C 1+6 (7) booty call call somebody for sex, not B 6 offensive, comes from black English bots robots A1 2 brb be right back B, C 1+1 (2) b4 before A2 1 c see C 3 cst central standard time C 1 cya see you all A2 1 c2c camera to camera A3 1 dildo fake penis for masturbation, B 1 does not vibrate emt emergency medical team B 1 every1 everyone A2 2 ev 1 everyone C 1 f female A1, A2, B 12+5+3 (20) fag, faggot abusive word for a A1 1 homosexual male usually futrevu French phonetic transcription A3 1 of “fuck you” gf (alternatively g/f) girlfriend A3, C 2+16 (18) g2g got to go, gotta go A1 1 gm good morning B, C 1+64 (65)

22 hb hurry back C 5 hoe whore A3 1 hommie originally in black English B 1 “friend”, “pal”, short for “home boy” hon, hunn honey C 3 hubby husband B 1 id identity, I would A3, B, 1+4 (5) im instant message A1, A2, A3, B 12+11+3+1 (27) im I am A1, A2, A3, B, 10+11+9+20+1 C (51) jk just kidding A2 2 k ok A2, B 1+2 (3) lmao laughing my ass off A3, B 1+10 (11) lmfao laughing my fucking ass off B 5 lo hello C 1 lol laughing out loud A1, A2, A3, B, 6+17+5+30+55 C (113) lt ns long time no see C 1 m male A1, A2, A3, B, 4+4+4+3 (15) C mic microphone A3 3 msn messenger (website operator) A2, A3, B, C 1+3+3+4 (11) ni any A2 4 nic, nick nickname A3, B, C 1+2+6 (9) np no problem C 2 nvm never mind A2, B 2+1 (3) omg oh my god A3, B, C 1+9+2 (12) ozzie person from Australia A2 3 pic picture A1, A3, B, C 2+6+8+1 (17) pls please B 1 pm personal message A1, A2, A3, B 2+2+1+1 (6) ppl, ppls people A1, A2, B, C 1+3+1+1 (6) pro profile A1, A3 1+1 (2) sis sister A2, B, C 1+1+1 (3) some1 someone A2 1

23 sup what‟s up? A2, A3 1+2 (3) sux sucks A1 2 tc take care C 3 thx thanks A1 3 tks thanks C 2 ty thank you B, C 3+27 (30) tya thank you all B 2 u you A1, A2, A3, B, 14+12+16+7+14 C (63) ur your, you are A1, A2, A3, B 5+2+3+5 (15) urs yours A1 1 vic a code for Victoria, Australia A1 1 y why, you A1 3 ya you, yes A1, A2, B, C 2+4+8+7 (21) yall you all A3 1 yr year B 2 yw you are welcome C 2 yvw you are very welcome C 4 ywc you will see B 1 wb welcome back A2, A3, C 1+1+12 (14) webcam, cam web camera A1, A3 4+4 (8) wtf what the fuck A2, B 1+1 (2) 1 one (some1) A2, C 7+1 (8) 2 too, two, to A1, A3, C 3+3+2 (8) 4 for, four A1, C 1+1+1 (3)

24 1 Introduction

1.1 Modern Trends in Communication and Linguistic Research

The language of the Internet has recently attracted the attention of many linguists (Baron, 1984; Condon and Cech, 1996; Herring, 1996; Werry, 1996; Baron, 1998; Jones (ed.), 1998; Cairncross, 1999; Crystal 2001; Herring, 2001; Yus, 2001; Quero, 2003). It seems a natural tendency mainly thanks to the massive development of this media and its world wide usage. Various topics can be found dealing with virtual community life, politeness on the net, computer-mediated dialogue, humour in computer-mediated communication or framing and face in Internet exchanges.

In a modern science-fiction literature the topics of three-dimensional world created by a computer appear with high frequency. One of them, The Naked Sun written by Issac Asimov (1957), describes people of the Solaria planet who very seldom meet face-to-face. They distinguish between seeing and viewing. The following passage characterizes their life-style:

“In short, a Solarian takes pride in not meeting his neighbor. At the same time, his estate is so well run by robots and so self-sufficient that there is no reason for him to have to meet his neighbor. The desire not to do so led to the development of ever more perfect viewing equipment, and as the viewing equipment grew better there was less and less need ever to see one‟s neighbor. It was a reinforcing cycle, a kind of feed-back.” (Asimov 1957, p.100)

Even if this book was written thirty years ago, it seems that nowadays such concepts of the world directed by robots in which people are involved in near-real conversation at a distance and virtual reality do not appear unfamiliar.

In earlier times people had two ways of communicating with one another: either face-to-face (speech) or at a distance (writing). With only minor exceptions there was no other possibility. Then over the last century, other kinds of conversations have been created. The telephone itself made common something that had previously been rare, that is, a conversation with someone you cannot see.

In recent years information technology (IT) – computers and telecommunications – is also having the kind of revolutionary and restructuring impact. The rapid advances in the speed and capacity of computing devices together with the

25 pervasiveness of the Internet, digital storage, wireless and portable devices are causing major changes in the way we live and work. Namely e-mails, SMS and the Internet have become indispensable parts of our lives because they can convey the information written at a computer keyboard in near-real time. IT has introduced new communicative modalities that blend the spoken and written language, the so called computer-mediated communication (hereafter abbreviated as CMC), namely synchronous, Internet Relay Chat (hereafter abbreviated as IRC), and asynchronous Internet chatting. The former is of a major concern in this work.

Some of the common features of all the above-mentioned technological devices enable people to establish or maintain relationships and share information at a distance. Now, at the beginning of the 21st century, mainly the Internet is an example of the most used advanced technology that is accessible from any computer connected to the net.

Let alone the Internet‟s exceptional technical parameters, the fact that thousands of people all over the world can be connected to it at the same time and communicate, or chat as it is commonly called, beyond the physical barriers, must necessarily alter the way of communication and the organization of interactions.

The proliferation of virtual communities, and namely IRC, in recent years has resulted in the creation of not only new social spaces but also in new forms of interaction, identity formation and new means of expression. Thus, I believe that the most appropriate disciplines to study this remarkable communication modality appeared to be blending of conversation analysis and sociolinguistics.

1.2 Research Objectives – Specificity of This Research

The intention of this work is to evaluate, analyse and compare with the help of conversation analysis how chat participants use the language that is determined by the specific conditions of an Internet chatroom. The specific conditions are anonymity, exclusively text-based interaction, the absence of visual and audio cues and „decorporization‟ of the participants. The whole analysis is then based on the sociolinguistic variable of age as a main denominator. The age groups under investigation are as follows: 13-19 years old, 30-59 and over 60. The choice of these particular age groups is discussed in more detail in chapter 3.4.

26 My main objective is to verify the assumption that conversation techniques on IRC are affected by age of the participants. To reach this objective I focus on the following four phenomena of conversation analysis in the environment of IRC:

1. Address forms

2. Opening sequences

3. Closing sequences

4. Turn-allocation strategies.

Addressing as an act of turning one‟s attention to someone plays a pivotal role in achieving a successful interaction. In face-to-face conversation, there are various ways we can address other participants, using either verbal or non-verbal means, such as gaze, touch or head movement. On the other hand, the anonymous character of IRC does not account for these standard forms of address and chat participants must look for other ways to transfer the message to the right person. An investigation will be made and comparison provided of different methods used by the determined age groups.

Certain rules or rituals were observed taking place during conversation openings and closings. It is believed that through both opening and closing, a person gives the main impression and positive or negative signal to other participants. The main task is to observe which techniques members of the specified chat rooms use to open and close conversations on IRC with regard to which age category they belong to.

In the course of a standard interaction, the role of the speaker and listener frequently change. We can say that conversation is achieved through a series of turns. The question arises as to how and where these turns usually occur in the environment of IRC. A model applied to face-to-face conversation is used and various strategies are investigated and compared in individual age groups.

With a view to the above-mentioned facts, this empirical study of IRC language sets the aim to investigate the impact of age on the given conversation strategies on IRC and to answer the following research question:

To what extent are specified conversation techniques used on Internet Relay Chat affected by the variable of age?

I draw the following two hypotheses from the above research question:

27 My null hypothesis (Ho) claims that conversational techniques used on IRC are independent of age.

My alternative hypothesis (H1) is that the form of communication on a medium of IRC is dependent on age.

If H1 is confirmed, then my assumption is that older people will have a more conservative attitude to the IRC language use in all four specified phenomena of conversation, working with the language in the same way as if they were speaking to someone face-to-face. They will also show more respect, cooperation and politeness. On the other hand, I expect that the younger generation will favour an innovative approach, using mainly informal means of address and showing less respect for their peers. Further, they will cooperate less and be less polite to one another. The intermediate age group, that is mid-thirties, will have a tendency to direct their language use toward the younger generation still keeping some elements of the elderly.

The collected data are statistically analysed using quantitative, qualitative analyses and the method of chi-square test for independence and Cramer´s V. For further details see Chapter 3.1 Research Methodology and Statistical Analysis.

Furthermore, I inquire into the existence of norms or other markers that would indicate the development of a ritual in any of the four above-mentioned phenomena of conversation analysis in the IRC environment. (Firth, 1964; Schiffrin, 1977). In other words, I anticipate the possibility of being able to capture emergent conventional patterns of use within Internet chat behaviour.

Lastly, the results are expected to point to and enhance the understanding of conversational techniques and their operation in a newly emerged medium of IRC. The purpose of my selection of a “language-in-use” methodology is to discover the structuring principles behind chatroom language. Internet communication is a form of rapid conversation. It is rarely “frozen” for analysis, as it is when the chat is saved to examine.

I humbly believe that this unconventional view of IRC will at least partly contribute to the limited number of studies that integrate conversation analysis, sociolinguistic variables and the medium of Internet and that it will also encourage other surveys to be done in this relatively new field. To the best of my knowledge, this kind of research has never before been attempted.

28 1.3 Computer-Mediated Communication

This section attempts to introduce briefly two types of Internet multi-party conversations that allow people to communicate online: asynchronous groups, in postponed time, for example discussion groups, or synchronous groups, in near real time, for example IRC. The latter type is of interest in this work.

1.3.1 Asynchronous Groups

Communications vary in how synchronous they are. Spoken discourse is synchronous – speakers compose, deliver and hear messages with little time lag. Within the field of computer-mediated communication, electronic communication, bulletin boards, or discussion lists are asynchronous because messages are sent sequentially and can remain unread for a longer period of time (Greenfield, Subrahmanyam, 2003). This signifies that individual contributions are stored and distributed as they come in during a period of months or even years. The non-linear character of such interactions is highlighted. There is no given beginning point with topics being classified chronologically. The medium privileges personal and idiosyncratic features, which has some linguistic consequences (Crystal, 2000). Due to its character, it lacks the most elementary properties of conversation, such as turn-taking, floor-taking and adjacency pairing (Herring, 1999).

Even if asynchronous chats comprise considerable benefits, for example discussed in Crystal (2000), in the following chapter, I will proceed with the form of Internet interaction which is under investigation in this work, that is, synchronous chats.

1.3.2 Synchronous Groups – Internet Relay Chat

One of the largest synchronous chat systems, and that which constitutes the focus of the present study, is called IRC. It is a multi-user synchronous communication facility that is available all over the world to people with access to the 'Internet' network of computer systems. IRC, or chat rooms, are virtual “spaces” where people congregate for conversations and interactions. They are regarded a synchronous form of communication because participants communicate with others in the room in real time.

29 Still, as a consequence of composing a message in writing before sending it- hardware constraints when ordering the messages and server speeds, there is a short time lag between composition, input and appearance on the screen. Most chatrooms require participants to register and create a screen name or nickname. It is this nickname that is observable to others when a newcomer enters a chat room. Conversations take place via text that is visible to all participants (Herring, 1999). As people add lines, text continuously scrolls up. This fact represents input from all active participants and thus the screen shows parallel and multi-thread conversations simultaneously (Greenfield, Subrahmanyam, 2003).

Moreover, as it has been suggested earlier, the users are spatially distant. It is a form of communication that is transmitted, received and responded to within a time frame that has formerly been only thought possible in spoken communication. IRC does not presuppose physical contact between users - either prior to or after communication via computer. It allows and encourages noncommittal communication between people who have never been, and most likely will never be, in physical contact.

However, users of IRC are not completely without knowledge of each other. Their anonymity is sometimes disclosed in various IRC live meetings where chatroom participants can meet face-to-face. They can also know each other from the chatroom where they meet for everyday communication and become so called “old friends”. The phenomenon of live meetings shows that people need to know who they are “chatting” with despite numerous advantages of chatroom anonymity. The following example gives evidence of such familiarity or even a kind of intimacy in opening phrases.

Example

1. Reen: julie!! 2. The Celtic Goddess: julieeeeeeeeeeeeee 3. JulieChats: Reen!!!! there you are 4. JulieChats: Goddess!!!!!!!!!!!!!! 5. JulieChats: Reen…open up 6. JulieChats: hehe 7. JulieChats: brb…need a private time with Reen 8. Reen: lol (30+, unpublished extract from the extended IRC corpus)

30 The technological, visual and social environments of chatrooms preclude face- to-face cues such as eye contact, gaze, body orientation, and gesture that enable speakers in a spoken conversation to interact, to know whom a speaker is addressing. Furthermore, the IRC conversations can be distinguished by the following features, some of which are more thoroughly discussed in this work:

1. several topics discussed in parallel,

2. overlapping conversation,

3. disrupted adjacency – juxtaposed turns do not refer to each other,

4. people involved in several interactions within the same chat room or more chat rooms,

5. relatively quick topic decay,

6. linear ordering of messages without any coherence.

Finally, I would like to emphasize that interaction on IRC is carried out with the knowledge that users are on a “rough equality”, a term utilized by Reid (1991). This equality is then intrinsic to IRC, it is one of its main advantages in terms of being anonymous and hidden behind the screen. Interaction on IRC involves a deconstruction of traditional assumptions about the dynamics of communication, and the construction of alternative systems. IRC is essentially a playground (Reid, 1991). Within its domain people are free to experiment with different forms of communication and self- representation. As David Porter (1996, pp.xi-xviii) observes:

„Participants adjust to the prevailing conditions of anonymity and to the potentially disconcerting experience of being reduced to a detached voice floating in an amorphous electronic void, they become adept as well at reconstituting the faceless words around them into bodies, histories, lives…Acts of creative reading…can and do stand in for physical presence in these online encounters.‟

To paraphrase F.R. Ankersmit (1989), users of IRC do not shape themselves according to or in conformity with the conventions of social contexts external to the medium, but learn to "play" their "cultural game" with them.

31 1.3.3 Netiquette on Internet Relay Chat

This section briefly discusses explicit linguistic guidelines to behaviour on IRC. It focuses on the most common rules and their violations (see Table 1 below).

In everyday conversation, we do not expect precise instructions about what we should say and how we should behave. There are no moderators who would tell us that we are off-topic or saying something inappropriate. In Internet chatting, however, explicit linguistic regulation is routine (Crystal, 2001, p. 72). The purpose of these guidelines is to maintain a certain level of stability, so that the channel is suitable for chatting. The netiquette is meant as a set of rules, or „suggestions‟ (http://www.msc.net.ph/irclang/html), which give the users a uniform and consistent way of dealing with different situations on IRC. As one IRC website states “Netiquette is mainly about using good manners, common courtesies known world-wide, and at the root of it all, treating others like we want to be treated”. (http://www.ircbeginner.com/ircinfo/etiquette.html).

Many sites provide advice which users are encouraged to read before they enter. The following account of channel etiquette and violations of these rules has been compiled from three different IRC providers (http://babel.dk/windowsnt/etiquette.html; http://www.msc.net.ph/irclang/html; http://www.ircbeginner.com/ircinfo/etiquette.html).

Table 1: Netiquette on IRC – “the do´s and the do not´s”

Channel etiquette Violations

Listen to a channel first, to get the feel of Try not to “double”, i.e. speak at the same what is and is not acceptable. time as someone else. Respect the culture of the group. Always Repetition is annoying. be friendly. Choose a nickname that is easy to type, No advertising in the channel is tolerable and which will not generate confusion – IRC is no place to advertise other with other participants. Nicknames must websites, businesses or personal be short and should not contain spaces. persuasions (religion).

32 Guard your privacy. Keep your personal No auto-greets acceptable. Do not greet information, such as your real name, every person in a channel personally. address, phone number, city or e-mail address, to yourself. Avoid using CAPITALS – it is like No editing of someone else‟s message shouting, people prefer soft tones. without permission.

Send maximum of four-line message at Flooding – sending too much information once. to the channel at once.

Spamming – sending of unwanted messages of excessive size. If you find a topic that offends you, do not Flaming - unacceptable conduct, or any join it. behaviour reducing the functionality of IRC may get the participant banned from that server. If you find yourself a channel that offends Harassing other users, extensive use of you, leave it. offensive language. Use only English in the channel provided Enticing others to violate the netiquette. the channel is international. Be nice, play nice, have fun Extensive use of colour when writing.

What is worth mentioning here as a conclusion is the point made by Crystal (2001, p. 73-74) that views such instructions and regulations of Internet language as obstructive containing a big amount of „subjective or institutional taste‟. He senses that there is

„a real risk that a biased account of Internet language will emerge, reflecting only the interests and background of the individual author, publication, or organization which produced it. If such accounts are then taken and promoted as guides to the Internet in general, an unhelpful prescriptivism can be the result, similar in its naivety, unreality, and oversimplification to that encountered in the grammar books of old.‟

33 In this respect, the work also tackles the issue of netiquette in the corpus and individual age groups and traces whether the regulations mentioned above are followed.

34 2 Theoretical Framework

In this chapter, I briefly review literature and theories relevant to conversation analysis and sociolinguistics. Further, I discuss the application of conversation analysis on computer-mediated conversation based on the work of different linguists. The main emphasis is placed on recent studies concerning the topics discussed in this work which include addressing, opening and closing sequences and turn-taking.

Next, I inquire into the problem whether IRC approximates more spoken or written communication. Finally, the language of IRC is presented with all its specific features differing from any other human interaction.

2.1 Significance of Conversation Analysis and Sociolinguistics in This Work

Conversation is more than the mere exchange of information. Shared assumptions, expectations, knowledge and behaviour are all brought to the conversational process. Disciplines that deal with studies of language in connection with society can be divided into three categories according to Trudgill (1978): those where the objectives are merely sociological (ethnomethodology), those working partly with linguistic and partly with sociological factors (conversation analysis and sociolinguistics) and finally those whose objectives are purely linguistic (secular linguistics, sociolinguistics as a way of doing linguistics with the focus on mechanisms of linguistic change). Although Trudgill himself draws on these three categories, he emphasizes that such classifications are quite arbitrary and not straightforward enough to apply fully in practise (Trudgill, 1978).

The concern of this study is represented by the second category which includes the above-mentioned studies of both linguistic and sociological character. In this section I would like to outline some socio-linguistic problems that are discussed and will become significant later on in this work. Therefore, I will very briefly discuss the theoretical background of conversation analysis and in the second part consider the significance of sociolinguistics with the emphasis on age.

When speaking of conversation analysis, it is appropriate to note that at the beginning of this work, a Glossary of Terms can be found in order to clarify the

35 meaning of new terms connected with IRC and also known terms of conversation analysis, the meanings of which might be modified in the context of this work.

2.1.1 Conversation Analysis

Conversation analysis that is derived from ethnomethodology, developed as a field of study in the 1960s through the intense cooperation of Harvey Sacks, Emanuel Schegloff and Gail Jefferson. As a linguistic discipline, studying language from the point of view of its usage, conversation analysis also seeks to describe the underlying social organization through which orderly and intelligible social interaction is possible (Goodwin, Heritage, 1990). Sentences and utterances are understood as forms of action situated in specific contexts and designed with specific attention given to these contexts. For conversation analysts, the starting point of their investigation and analysis of any utterance is, therefore, authentic conversation and its organization. Moreover language varies according to the social characteristics of the speakers (in this case, age) and also with regards to the social context in which they find themselves.

Many social factors can come into play in controlling which linguistic variety is to be used on a particular occasion. One of the key roles is played by the status and relationships of the speakers regarding which linguistic forms may be used. An example of this kind, discussed in this work, is that of address forms produced by altered degrees of status or intimacy and requiring different degrees of politeness and deference. (Brown and Ford, 1961; Little and Gelles, 1975; Adler, 1978, Trudgill, 1983).

Further, some researchers (Schegloff, 1968; Schegloff and Sacks, 1973; Schiffrin, 1977) studied in particular the structure of conversation. Schegloff (1968) suggested that it is actually the answerer of the telephone conversation, not the maker of the telephone call who should speak first, unless the ringing of the telephone itself is regarded as a summons, or a part of a summons-answer sequence. Schiffrin (1977) proposed that opening sequences consist of multi-level recognition displays that are all ritual signs of identity and help to maintain the structural differentiation of social organization. Similarly, closing sequences must be made to occur by coordinated activities of the conversationalists. Moreover, it appears that closings are preceded by pre-closings and a variety of strategies for closing conversations are developed.

36 Another group of scientists (Sacks et.al, 1974; Herring 1999; Schegloff, 2000) have pointed out that conversation can be regarded as a kind of speech exchange system which is organized in such a way as to ensure that at a given time only one speaker talks. It also guarantees that the speaker role is exchanged between individuals. There are further rules for turn-taking indicating that the current speaker selects the next speaker and that only if this is not done is it open to any speaker to self-select. In this respect, coherence of conversational discourse and the question of adjacency pairs have been of interest to many linguists (Sacks et al., 1974; Levinson, 1983; MacLaughlin, 1984; Herring, 1999; Mey, 2001 and so forth). Sacks et al. were the first to show that conversation does not happen at random, but follows a certain system. Thus, the notion of adjacency pairs came into play. The central characteristics rule implies that a current action (a “first pair part” such as a greeting or a question) requires the production of a reciprocal action (“second pair part”) at the first possible opportunity after the completion of the first (Goodwin and Heritage, 1990). The adjacency pairs pattern described by conversational analysts presents a procedure through which participants constrain one another, and hold one another accountable, to produce coherent and intelligible courses of action (Goodwin and Heritage, 1990).

Finally, I would like to emphasize the importance of conversation analysis for other disciplines. It is apparent that conversation analysis integrates the details of language structure and the social constitution of meaning and action within an analytic framework. Hence, it transcends the traditional disciplinary boundaries of social anthropology by providing a perspective within which language, culture and society can be analyzed not as separate subfields but as integrated elements (Goodwin and Heritage, 1990).

2.1.2 Sociolinguistics

The development of sociolinguistics as a modern science goes back to the late 1950s. Even before that some researchers noticed that a certain relation existed between society and language (Humboldt, Saussure, Meillet, Cohen). Bohuslav Havránek and other Prague school scholars (Mathesius, Mukařovský, Jakobson) are some of the most distinguished predecessors of modern sociolinguistics. In Úkoly spisovného jazyka a jeho kultura (1932) Havránek raises important questions about connections between

37 standard language and its users that forms a crucial part of any language culture. Among the most significant representatives of sociolinguistics are William Labov, who carried out his first sociolinguistic studies in the United States in 1960s, British scientists M.A.K. Halliday, Peter Trudgill, James and Lesley Milroy, Janet Holmes and in the Spanish environment H. López Morales, C. Silva-Corvalán, F. Moreno Fernández, B. Lavandera, M. Alvar and J.M. Lope Blanch.

Sociolinguistics can be broadly defined as that branch of linguistics which studies properties of language that must be explained in social terms. In other words, sociolinguistics studies verbal behaviour in terms of the social characteristics of speakers, their cultural background, and the ecological properties of the environment in which they interact (Černý, 1996). The properties of language mainly discussed in sociolinguistics are those of variation and language change. We can say that there is no change without variation. That means, when a variation appears in a language, it can, but does not have to, lead to a language change. We distinguish different types of variation – phonetico-phonological, grammatical, morpho-syntactical and lexical. The criteria for variation can be different: social stratification, class, gender, ethnic group and age.

Considering the works of Labov (1972), Milroy (1978) and Romaine (1978), for example, they appear to have one feature in common; they examine the variable of age and give quantitative evidence to demonstrate its significance for variation in language.

One of the main contributions of Labov (1972) is his theory of linguistic stratification and its dependence on social factors. He introduced statistical methods into linguistic research using random samples and collecting speech systematically in different contexts. According to him, it is casual speech which adequately revealed the irregularity of everyday synchronic and diachronic processes. His well-known study into rhoticity in New York City is a good example of the connection between age and status variables. The research results indicated that the younger upper middle class is the „reference group‟ for this new norm. Lower-middle-class speakers were using the younger speakers as their model of correct speech. Labov suggested through an innovative quantitative paradigm that hypercorrection serves to accelerate the process of change.

Milroy´s (1978) investigation of Belfast vernacular vowel sound /Λ/ in words like man revealed that there is a linguistic change in progress in the direction away

38 from the middle-class norm. Age, sex and style were considered. The results show that a strong sense of solidarity and group identity – even in terms of age – lead to both clarity of the linguistic norm of the group and a pressure to conform as an expression of individual identity.

Romaine (1978) investigated variation of the flap /r/ by studying the speech of twenty-four working class Edinburgh schoolchildren aged from six to ten years old. Even if the main stress is placed on gender differences and age-differentiation is not that striking, she found that ten-year olds make the greatest use of the flap (Downes, 1998) of all three age groups: 10 years old, 8 years old and 6 six years old, while the six-year-olds make the greatest use of r-lessness.

We have seen that if a linguistic change is taking place, it is reflected in the variable “age”. As each generation acquires its vernacular, the average score for that age group is further along the direction of the change than that of the immediately preceding generation. Each successive generation will have gone further than the last; so, in general, younger speakers will be more advanced, and older speakers less advanced with respect to the change.

Based on these facts, the analysis presented in this work is approached through the natural interplay of instruments of conversation analysis and a sociolinguistic framework applied to the “body” of the IRC communication.

2.2 Research on Conversation Analysis on Internet Relay Chat

Even if the majority of the material and theoretical background used in this work include studies and research in spoken conversation, other field studies and work that focus on conversation analysis as applied to the analysis of IRC could be found.

In the introduction, I have already mentioned some researchers who drew their attention to the questions of computer-mediated communication (Baron, 1984; Condon and Cech, 1996; Herring, 1996; Werry, 1996; Baron, 1998; Jones (ed.), 1998; Cairncross, 1999; Cherny, 1999; Herring, 1999; Parrish, 2000; Crystal, 2001; Herring, 2001; Rintel et al., 2001; Yus, 2001; Panyametheekul and Herring, 2003; Quero, 2003; Neuage, 2004). In the following outline, I look at the literature of those varying methods used to capture and analyse language in use, with special emphasis on

39 conversational analysis, paying attention to the texts and accounts of how IRC might be operating with respect to address forms, opening and closing sequences and turn-taking.

Even though address forms have not been of particular interest to many IRC studies until now, some minor research has been done in various fields mainly drawing from spoken discourse and their differences. The research focuses on new modes of coherence where vocative cues – addressing – are used to select the next speaker in a chatroom. These include conversational strategy adapted from oral discourse (Greenfield and Subrahmanyam, 2003); nicknames and the identity of their users (Bechar-Israeli, 1995). Finally, Werry (1996) describes sets of conventions that IRC users internalize in order to manage and follow the complex structure of conversational sequences that occur on IRC, namely addressivity that is crucial in this media.

Studies in openings and closings on IRC most notably derive from the works of Schegloff (1968) and Sacks and Schegloff (1973). Despite IRC´s popularity for both users and researchers, openings have received sparse attention up to now. The few accounts that exist are limited to the descriptive; for example Werry (1996) found evidence of the opening routine of users joining a public channel and collectively greeting the existing channel members but he does not further explain or investigate opening strategies. Rintel et al. (2001) make quite a thorough examination of openings in IRC interactions, concluding that turn-coordination in an opening phase is often ambiguous and has the potential to disrupt relationship development.

In their next survey of ambiguous non-responses on IRC, Rintel et al. (2003) admit that to initiate interaction on IRC can be complicated. Various strategies are utilized in order to begin conversation, among them collectively-addressed or individually addressed greetings. On the whole, they find that when users experience ambiguous non-responses they perform actions that attempt to clarify understanding as in face-to-face interaction, for example re-connecting, re-greeting or meta-lingual connection checking. These strategies bear a resemblance to „repair‟ strategies similar to Schegloff´s “recycled turn beginnings” (Schegloff, 2000).

Similarly, Herring (1999) in her study of interactional coherence on IRC finds that everyone who posted more than two messages – so called “multiple initiations” – was responded to. She further explains that overall frequency of multiple initiations may reflect the difficulty users have in determining other‟s intentions due to the paucity of feedback in IRC environments.

40 In other research, Ahti and Lähtevänoja (2004) raise the question of politeness in opening sequences in Finnish and Finland-Swedish chatrooms. They also inquire into the factors that influence gaining a response to an opening phrase. For the description of the data, they utilize the Sacksian notion of recipient-design where utterances are designed by the needs and expectations of the recipient. They conclude that the following three specific features have the biggest impact on whether an opening turn gains a response or not: individuality, clarity and minimalism.

As far as closings on IRC are concerned, not many studies could be found concerning this phenomenon. Cherny (1999) suggests that openings as well as closings are a part of “ritualized” or “conventional” (p. 202) actions and are loosely connected with Schegloff‟s and Sack‟s openings and closings. Into these categories she includes “hugging”, “waving” and so forth. Cherny concludes that given specific conversational contexts, not all disconnections are preceded by goodbyes or waves; long-idle users may just disconnect suddenly or may be “kicked off” by a firewall. However, she adds that when active people leave a room without any goodbye statement, it is generally seen as odd and against proper etiquette.

As regards turn-taking strategies on IRC, a relatively high number of works deal with the effect of this Internet medium on conversation structure, the length of individual utterances, adjacency pairing, back channels or overall coherence of the media. McKinlay et al. (1994) in their experiments investigated whether chat users attempt to reproduce the turn-taking behaviour observed in face-to-face communication. The results have shown that with a larger group, IRC participants tend to subdivide into smaller discussion groups whose membership changed spontaneously. However, they found and confirmed that the transfer of turn-taking skills from the face- to-face setting to the CMC condition seems problematic.

Accordingly, Herring (1999, 2001) stresses that the standard expectation that adjacent turns are related to one another does not count in IRC interaction due to a linear ordering of messages. Thus the problem of keeping track of topically related “threads” arises together with quick topic decay. Herring further investigates possibilities of turn coherence on IRC and suggests some techniques that facilitate the timing of turn allocation, such as typing a ”%” symbol at the end of a message indicating a person is not ready yet to give up the floor.

41 In a similar way Cherny (1999) describes a CMC interaction in which participants “raised their hands” and were called on by a moderator, as a means to facilitate discussion. Parrish (2000) finds in his study of turn-taking that on IRC there is a high frequency of stream overlap, which means new conversations appear and proceed with little confusion and participants often participate in more than one stream at a time. Greenfield et al. (2003) in their study describe how participants in an online teen chatroom adapt to the unique features of IRC environment to achieve conversational coherence. They discuss repetition, vocative cues (addressivity), visual cues and nicknames. They conclude that to be successful in chatting, a person must know the codified register and certain mechanisms used by chat participants to achieve coherence. Rintel et al. (2001) also note that inexperienced and non-regular users who appear not to know anyone on the channel may have difficulties both in initiating an interaction or developing further online relationship.

Finally, Panyametheekul and Herring (2003) base their analysis of IRC turn- taking and response patterns on face-to-face conversation strategies developed by Sacks et al. (1974). Their findings confirm that the most used strategy in the IRC media despite the absence of audio and visual cues is still the current-speaker-selects-the-next- speaker strategy. At the same time, they found that turn allocation in the chatroom is generally similar to that in face-to-face interaction: participants preferred to address one another rather than self-selecting to speak.

2.3 Computer-Mediated Versus Spoken or Written Interaction?

The difference between speech and writing, and the conditions under which they have been produced, have been given considerable attention in literature (Vachek, 1976; Baron, 1984; Biber, 1986; Halliday, 1989; December, 1993; Firbas, 1997; Baron, 1998; Crystal, 2001; Yus, 2001; Urbanová and Oakland, 2002; Quero, 2003; Urbanová, 2003). That distinction is based on a perception of temporal and spatial proximity in the case of spoken communication, and distance in the case of written communication.

Naomi Baron (1984) notes, for instance, that most analyses of linguistic interaction are based on the paradigm of two people speaking face-to-face. It is further assumed that alternative methods of communication - telephones and letters for example – supplement 'normal' face-to-face communication. An assumption existed

42 that physical contact is a necessary part of human communication. Moreover, Biber (1986) has suggested a more precise analysis of the spoken and written registers in terms of different dimensions: interactive text as opposed to edited text; abstract contents as opposed to contextualized contents; indirect style as opposed to direct style.

Until recently, physical contact was almost always a prerequisite for communication, with letters mainly being transmitted between people who had met in person. Even the telephone assumes physical contact (Schegloff, 1968). People have to use their voice and this factor, vocal quality, can reveal a great deal about communicators. Furthermore, people usually phone someone they already know, with the exception of businesses and the latest phoning for commercial purposes.

However, within the last few decades, the development of new types of communication have been observed which are more complex to examine only in terms of writing and speaking, creating a world that is based on text but exhibits characteristics of a primary oral culture.

This thesis deals with one type of such “hybrid” (Yus, 2001, p. 139) interaction, or so called „tertiary form of orality‟ (December, 1993), that is, computer- mediated communication (CMC). Internet Relay Chat (IRC) that is under investigation in the present work, is the most used kind of CMC. The rapid development of CMC since the 1970s shows that boundaries between spoken and written interaction are becoming blurred (Herring, 1996). CMC technologies transform thought and culture by engendering the creation of communities in which the participants can be involved in emotional and expressive communication (December, 1993).

A question linked to this problem is whether CMC is a variety of its own and whether it can be better compared to spoken or written interaction. While Kwang- Kyu Ko (1996) regards computer-mediated discourse as a written (typed) medium, Crystal (2001, p. 238), for example, characterizes CMC as neither „spoken writing‟ nor „written speech‟. He claims that the main differences with respect to face-to-face interaction are the lack of simultaneous feedback and overlaps, the lack of prosodic, paralinguistic and kinesic features (2001, p. 29-34). Alternatively, other scientists such as Yus (2001, p. 139) and Quero (2003, p. 15) argue that the language of chat is a type of anomaly and they call it “texto escrito oralizado” (spoken written text). Similarly, Greenfield and Subrahmanyam (2003, p. 722) view chat as an amalgam of spoken and

43 written language. Table 2 shows basic distinctions between a virtual and a real conversation adapted from Quero (2003, p.11).

Table 2: Basic distinctions of virtual and real conversations (adapted from Quero, 2003)

VIRTUAL CONVERSATION REAL CONVERSATION

1. textual (written) no textual (spoken)

2. less contextual and verbal extralinguistic information, features information (vocal or visual) 3. of weakened synchronic character of full synchronic character

4. screen-scroll factor no screen

5. slower pace of conversation standard pace of communication, depends on various factors (type of interaction, communicative disposition of participants, etc.) 6. limited turn taking turn taking

7. multiple simultaneous interactions less interactions

8. higher geographic dispersion no geographic dispersion

9. anonymous, use of nicknames real people, no anonymity

Hence, we may argue that all the above characteristics make this new and intriguing communication a hybrid connection of, on the one hand, stability and rigidness of written form, spontaneity and a temporary quality of speech on the other.

In the present work, I have compared CMC exclusively to spoken interaction for the reasons not only listed above, but also presented in the following account (CMC equals IRC in this respect):

1. CMC is in its nature near-real time conversation where writing is a primordial activity (an instrument) that registers voice as its graphic representation.

2. CMC possesses a lot of attributes of speech because it takes place in a „temporal immediateness‟ (Yus, 2001).

44 3. CMC as a text is not tangible or necessarily stable, but virtual that, unlike writing, accounts for spatial and temporal distance.

4. CMC represents a type of a tertiary orality (see above) that indicates that the discourse does not have to be based on a sound in order to be oral (December, 1993).

5. CMC resembles spoken interaction the most in that there is little elaboration of phrases, the flow of chat conversation is unpredictable (more improvisation), the argumentation is similar and it is not convenient to make long statements because they lose their interest and attention of other listeners (Yus, 2001).

To conclude, I have mainly drawn from the theoretical findings in a spoken interaction, however I support the opinion of Crystal that IRC is a specific type of communication that cannot be strictly contrasted to either a spoken or a written form. Some researchers even warn against such comparisons (Rintel et al., 2001) that they claim may be too restraining.

2.4 Language of Internet Relay Chat

At the beginning of the present work, there is a list of IRC abbreviations – acronyms – found in the corpus specified. The main intention of this list is to simplify further comprehension of the examples and the corpus. Furthermore, the reader can find the occurrence rate of each acronym and compare their preference in individual age groups. In addition, I have attempted to provide clearer evidence of the type of language that exists within chat rooms.

Taking into account limitations of audio and visual cues and other factors described in chapter 2.3, the endeavour of this section is to briefly describe the other cues, apart from abbreviations, that make the IRC language so specific.

Laver and Hutheson (1972, p. 12) in their study of face-to-face interaction maintain that behavioural means for communicating information create vocal, non- vocal and verbal, non-verbal divisions. By combining these elements inter se, we obtain the following categories: vocal – verbal (words in the oral language); vocal – non- verbal (intonation and paralanguage); non-vocal – verbal (written language); non-vocal

45 – non-verbal (kinesics). Further, they outline another division of information exchange in conversation, that is

(a) linguistic

(b) paralinguistic

(b.1.) non-linguistic, non verbal, vocal (tone of the voice)

(b.2.) non-linguistic, non verbal, non vocal (kinesics etc.)

(c.1.) vocal (voice qualities based on biological, psychological and cultural aspects)

(c.2.) non-vocal (language of clothes, colours)

According to this classification, then, the participants on IRC have great difficulty to textually reflect the above levels apart from level (a). In order to compensate for the absence of audio and visual signs, and based on the above classification, the following forms of IRC language have been observed:

For the level (b.2.), IRC users can count with some limited graphic icons, so called emoticons, for example [:-) ;-( ] Emoticons, or smileys, are used to imitate various facial expressions (Crystal, 2001; Yus, 2001). Their main role is seen in „suavizar‟ (Yus, 2001, p. 131), mitigate the meaning of an electronic text, or give it a contrary implication. Crystal (2001, p. 36) defines them as „potentially helpful but extremely crude‟ with very limited semantic role. He continues (p. 36-37) that „ those who get into the habit of routinely using smileys can also find themselves in the position of having their unmarked utterances misinterpreted precisely because they have no smiley attached to them‟. Usage guides warn against overuse. I find that the most frequent acronym in the corpus that bears the same function as smiley is a verbal expression LOL, in full “laughing out loud” (N=113) compare to the use of graphic smileys (N=5).

The level (b.1.) and voice qualities in (c.1.) tend to be manifested textually in a form of various punctuation marks (exclamation, question marks), repetition of letters, capitalization, italics, underlining, colours. For example: SpecialEd:

46 ohhhhhhhhhhhhhhh, KingKaosh: MmMMm, LIPGLOSS I LOVE ITTTT, screamz_05: biggggggg?!?.

The last level (c.2.) can be described by means of what was called by Yus (2001, p. 110) „acotación icónica‟, that is, a verbal description of a non-verbal message, for example „LINDSAY screams‟. Herring (2001, p. 633) calls it „contextualization cues‟ and adds that they are used to „expand dialogue into narrative performance‟. Later in the work, I also refer to it as “literary expressions”.

Crystal (2001, p. 225) views the language of synchronous chatgroups as bearing „clear signs of the emergence of a distinctive variety of language.‟ The question remains whether the limitations with which IRC has to cope do not affect the quality of a message. This work attempts to find an answer by analyzing standard conversational modules in the new Internet environments. Currently, we may view the arrival of the language of IRC enriching various communicative options available to us.

47 3 Methodology

The present chapter aims to outline and summarize the methodology I have used before engaging into the practical research. First, the choice of research methodology will be justified. Next, descriptions of data collection are discussed together with the statistical techniques used to analyze the data. Further, I strive to identify the corpus and finally, the explanation is given for the choice of the sociolinguistic variable of age.

3.1 Research Methodology and Statistical Analysis

In this work I analyze local conversational elements, cues and strategies for addressing conversation partners, opening and closing techniques, identifying relevant responses and turn-allocation strategies within the scope of three age groups.

I have set two main hypotheses – H0 claims that conversational techniques used on IRC are independent of age and H1 asserts that the form of communication on a medium of IRC is dependent on age. The aim of this work is to confirm or refute H0 or

H1 in the four analysed conversational techniques.

Most studies, conducted in the area of conversation analysis and IRC, have used quantitative and qualitative methods of analysis (Herring, 1996; Herring, 1999; Greenfield and Subrahmanyam, 2003; Panyametheekul and Herring, 2003; Rintel, Pittam, Mulholland, 2003; Ahti and Lähtevänoja, 2004). The aim of my work is to compare conversational techniques in the three age groups and answer the research question to what extent specified conversation techniques used on IRC are affected by age. The following statistical methods were proved to be the most functional for this work:

Quantitative: In the quantitative phase, focus groups were used to identify how frequent, or what percentage of occurrences is attributed to each phenomenon under investigation.

Qualitative: In the qualitative phase, focus groups were used again in order to explore what their methods of conversation and interactional relationships look like in accordance with age. Further, an analytical procedure of making comparisons was utilized.

48 Chi-square test for independence: Used in chapters 4.1 and 4.4 between two variables to verify the level of significance and then, the strength of their dependence. An independent variable is the age of participants and a dependent variable the categorical data expressed in absolute frequencies. When the level of significance is not verified ( >0.05), that is, the null hypothesis (Ho) of independence is true, it means that there is no relationship between variables. This method was used primarily to identify what influence age had on conversational techniques, if any, and to prove whether the division of Teens into sub-groups has any relevance at all or whether the data of the whole Teen group would be regarded as a single unit.

The Statgraphics programme for MS DOS was used as a statistical tool to determine this relationship. I utilized the chi-square test for independence on the significance level α =.05 and further, to measure the strength of dependence I used Cramer´s V. The standard interval of Cramer´s V is <0; 1>. The stronger the dependence, the closer the figure approximates 1.

Finally, it must be emphasized that I have applied this particular statistical method solely to „address forms‟ and „turn-taking‟ and not to „opening‟ and „closing,‟ because it is not possible to statistically test and verify the qualitative variables of the latter two.

3.2 Data collection

In order to conduct research in the field of IRC, a considerable database of Internet chatting taken from public chatrooms, designated to various age levels, had to be collected. Public chatrooms allow anyone to enter without registration and are an open conversational arena where what is said is clearly public. Even if chatrooms are mostly popular with teenagers, and according to one recent survey (Kaiser Family Foundation, 2001), at least 71% of teenagers participate in them, chatrooms for middle- aged and senior groups could be found without any difficulty.

Despite their relatively easy access, it must be emphasized that one of the biggest problems for collecting the kind of Internet database based on age is the aspect of anonymity. This means, on the one hand greater possibilities for chat users, but on the other hand raises the question: How can I be sure I am analysing the communication of adolescents and not the chat interaction of adults who are pretending to be under 19?

49 Or vice versa, how can I be sure I am entering a chatroom where only senior chat users over 60 are supposed to be interacting and not someone younger? On the whole, there is no possibility to verify who the writer of the text is, because online users are of a decontextualized origin.

A general solution cannot be offered, thus I took the age of participants to be in accordance with the age-oriented chatrooms, set the main hypotheses about the conversational practises influenced by age and later throughout the research had them confirmed by various statistical analyses. These results serve as sufficient evidence of the factual correlation between a chatroom and age, obviously with minor data divergence.

I gained access to the chatrooms through an account with an Internet provider under the nickname of sarku. The most useful websites for this purpose appeared to be yahoo.com, hotmail.com and msn.messenger. After that I appeared in the middle of various conversation flows. It must be emphasized that as long as there are chat participants in the chatroom, conversation is continuing. It can be 24 hours a day. As Neuage (2004, p. 72) stresses „Internet never sleeps […] making it difficult to say that there is a beginning or an end to any online communication. Simple conceptual structures will not transfer from CMC application to application, and are eroded by the very conditions of CMC technologies themselves: their boundarilessness and incessant interactivity‟. That is to say that at any given time, new chat participants are joining or leaving the chatroom and new topics are being raised or dropped.

Although my user name was still in the list on the right side of the screen (see similar example in Figure 1 below) room participants did not try to communicate with me except in one instance in the 60+ chatroom. I comment on this moment later in the work. Despite this one instance, I remained silent and did not interrupt the flow of conversation anyhow. In other words, I took the role of an observer rather than a participant. In one case, in the chat room 60+, I was following the conversation for as long as it took to solve the mystery about the nickname.

Figure 1: Example of a synchronous chatroom screen

My main aim was to capture the speech situation in the chatrooms as naturally as possible. The criterion for selecting a conversation corpus was to attain a sufficient number of messages for my analysis. My choice of such criterion draws on a number of studies on IRC that based their research on a number of messages in the corpus rather than number of words (Kwang-Kyu Ko, 1996; Greenfield and Subrahmanyam, 2003; Panyametheekul and Herring, 2003; Ahti and Lähtevänoja, 2004). Panyametheekul and Herring (2003), for example, collected 917 messages for their analysis, Greenfield and Subrahmanyam (2003) worked with 137 messages and similarly Kwang-Kyu Ko (1996) collected 94 messages. On the other hand, some researchers base their corpus on a time-scale. For example, Parrish (2000) views his 15 minute corpus of data as „sufficient‟ for studying Internet chat. He stresses that in conversation analysis a person must gather enough data to yield claims based more on structure, rather than content. Still, I believe number of messages as the main criterion is adequate for the type of analysis presented in this work.

51 I entered five different chatrooms designated for the following age groups: 13- 14; 15-17; 18-19; 30+, 60+, which were later divided into three major age levels, A. Teens 13 to 19 years old; B. over 30 years old and C. over 60 years old. At the end of each session, once I logged out, I printed out the log, or a conversation, which is enclosed in its entirety in the Appendix parts A-E.

At this point, it must be said that once I logged out of the chatroom, there was no possibility to return and capture the flow of the ongoing conversation from the latest place of departure. That is the reason why the authentic material I worked with (see Appendix) cannot be viewed from any website and does not exist in any form of official corpus version. Thus, my material is an example of an authentic, original Internet Relay Chat that occurred and was printed out in real time.

It is necessary to highlight, in this respect, that because of the difficulty involved in obtaining informed consent from chat participants, and in order to secure their anonymity, I modified all the nicknames.

3.3 Corpus Description – Material Analysed

I conducted my analysis on a corpus containing 2,382 IRC messages collected from February 2004 to April 2005 and taken randomly from the chat providers listed above. Each age group has been shortened to contain an equal number of messages, that is, 794 messages. The rationale for collecting a particular number of messages, rather than using the standard method of collecting a particular number of words, is discussed in chapter 3.2. With respect to this issue, it is adequate to state that the amount of messages collected is sufficient for this analysis, a point that has been confirmed following the examination of similar IRC studies. I mainly draw on Panyametheekul and Herring (2003) and their 917-message corpus that was sufficient for the analysis of turn-allocation strategies on IRC.

My data are in the form of a transcript of the conversations. It is necessary to note, though, that this is not a transcript in the usual sense; instead the transcript is the conversation. It means, contrary to the implications of the prefix trans-, this record has not been transferred from one medium to another. As it is a typed conversation, the record is only printed out. For this reason, it may be more appropriate, perhaps, to call it a printout rather than a transcript (Greenfield and Subrahmanyam, 2003; Parrish,

52 2000). It is also necessary to note that all conversations seen in the Appendix were printed out from the computer at the time they took place with the italicization, capitalization, punctuation, emoticons and the spelling. There was not made any correction in terms of grammar and spelling mistakes. The original typed conversations were in different colours and fonts, which I modified as these elements were not relevant to my study and they were of inferior quality for reading. For the purposes of this work I have transformed the original printout into a record using the Microsoft Word programme and numbered the lines to make the conversational threads easier to follow and clearer to present in Examples.

The age groups are as follows: Teens, ranging from 13-14, 15-17 and 18-19 years of age, 30+ and 60+. A more detailed explanation of age selection and age division is described below in chapter 3.4.

A: Teens (13-19 years of age)

I have collected three age sub-categories: A1. the age of 13-14 (283 messages), A2. 15-17 (241 messages) and A3. 18-19 (270 messages). The corpus contains altogether 794 messages collected over a period of one month from the www.yahoo.com web provider. The reason why this age difference ranging from 13 to 19 year-old chat participants was selected lies in the fact that it represents a valuable sample to be compared from the point of view of the maturity of participants and their language use. My hypothesis (H0) is that there is a dependence on individual age variables within a Teen group.

B. 30+ (over 30 years of age)

The corpus of age group 30+ contains 794 messages and was collected in July 2004 from the www.hotmail.com web provider. There is no further sub-division of this corpus.

C. 60+ (over 60 years of age)

The corpus of age group 60+ contains 794 messages and was collected in March 2005 from the www.msn.messenger web provider. There is no further sub-division of this corpus.

Last but not least, what should be pointed out is that the majority of the chat participants were American citizens. Other nationalities could be found as well, namely British, Philippinos and Australians. The character of the Internet medium does not

53 allow us to recognize people‟s nationalities via their mastery of the spoken language, accent or dialect. I could only presume their origin in the course of chatting when they exchanged information about where they were from with other chat participants. Nonetheless, for the purposes of our research, these were not crucial factors.

3.4 Age Variable

When revising various sociolinguistic studies, it becomes obvious that the age variable is one of the most relevant factors influencing language use next to gender, education or social status. Moreover, age has always been considered one of the crucial social variables in sociolinguistic research (Wolfram, 1969; Labov, 1972; Trudgill, 1978, 1983; Holmes, 1992, Milroy, 1992; Romaine, 1996; Valeš, 2002).

For the variable of age it is usual to differentiate three generations. Four generations are distinguished exceptionally in cases where the language use of minors is investigated. If only two generations are considered, this is because other objectives are pursued. In general, sociolinguists do not tend to collect the data of speakers younger than 15 years of age (Valeš, 2002). This research, then, works with three age groups, or generations, as is considered standard.

There is no doubt that a real problem arises in how to establish age limits of individual generations. The lack of unity among sociolinguists in this respect results in every author seeming to establish his or her own convenient limits. For example, Labov (1972) in his studies develops different age levels for different research objectives. Nevertheless, he usually works with age levels ranging from 15-30, 35-50 and 55-70. Serrano (1996), on the contrary, establishes the age groups from 20 to 34 for the first generation, 35 to 55 for the second and the third generation includes people who are 56 and older.

It is obvious that the age limits vary considerably. For this reason, in this research I have established age limits that appeared the most appropriate for my purposes. However, at the same time I was limited by the Internet chatrooms I employed. The chatrooms designated to different age groups were as follows: 13 to 14, 15 to 17, 18 to 19, over 30 and over 60. In these rooms, only chat participants of that certain age could enter. The system did not allow anyone outside the age limit to

54 participate in the conversation. There are certain potential risks given the anonymity of IRC and I discuss them in chapter 3.2.

The three age groups I selected for this work are – the first generation from 13 to 19, the second over 30 (30+) and the third over 60 (60+). I am aware of a possible vagueness of the terms “over 30 years old” and “over 60 years old”. Nevertheless, for my research of IRC, these were the only denominators I could draw from, given my interest in the relation between ages, a modern medium of communication and language use. Later, throughout my work, it becomes obvious that these three age groups, despite their vague boundaries, vary significantly and are sufficient for drawing broader conclusions concerning differences in their mastery of conversational techniques on IRC.

55 4 Conversation Analysis of Internet Relay Chat

The following practical part attempts to use methods of conversation and statistical analyses to describe the communicative strategies of IRC participants with regards to their age. Four separate chapters aim to cover fundamental topics of Conversation Analysis: forms of address, opening sequences, closing sequences and turn-taking strategies. Each section consists of a brief theoretical introduction, practical part and a conclusion attempting to answer the research question stated in Introduction and confirm or refute the hypotheses.

4.1 Address Forms

This chapter first outlines theoretical findings in addressing in face-to-face interaction and then in comparison with IRC. Further, it discusses the structure of analysis and focuses on stages of conversation in individual age groups and forms of address found there. In the end, final considerations are provided and conversational techniques of addressing are summarized from different points of view.

4.1.1 Theoretical Background

In most cultures, the way we address other people is very important. Addressing is a part of cultural awareness that also determines degrees of politeness (Trudgill, 1983, p. 102). Each address has its own social status in the society. In sociolinguistic terms, addressing is regarded as highly significant behaviour (Adler, 1978). According to Ammon (in Adler, 1978, p. 201) ´the function of address is to establish the external framework of conversation, within which reference is made to a particular subject. The partners in conversation have the possibility to orientate themselves socially because of the forms of address´.

The notion of addressing, or addressivity, as it was originally developed by the Russian linguist and semiologist M.M. Bakhtin, in standard means of communication, is basically the act of turning to somebody. Quirk et al. (1985) place an address in the category of vocatives together with a call and characterize it as expressing the speaker´s

56 relationship or attitude to the person or persons addressed. They further divide vocatives into the following categories:

a) Names: first name, last name, full name, with or without title, or a nickname

b) Standard appellatives: family relations; titles of respect: Sir, Ma‟am; markers of status

c) Occupational terms

d) Epithets: nouns or adjective phrase expressing an evaluation; favourable (dear, honey, beautiful) or unfavourable

e) General nouns often used in more specialized senses: girls, ladies and so forth

f) The personal pronoun you

Further, Quirk et al. emphasize that the hearer is addressed by name, not for clarity but out of courtesy and friendliness. This fact is observed further and comparison is made.

In addition, formal and informal, direct and indirect types of address can be identified (Brown and Ford, 1964; Little and Gelles, 1975; Adler, 1978). Present among formal address types are formal pronouns (vous, Sie, Vy - non-existent in Modern English), surnames, titles, honorifics and other forms. Informal address types are represented by first names, nicknames, diminutives, friendly expressions like guys, informal pronouns (tous, du, ty), terms of endearment such as hon, dear and so forth (Mey, 2001).

Brown and Ford (1964) have analyzed the norms of address in American English and the rules governing their use. The principal option of address seemed to be the choice between the use of the first name and the use of a title with the last name and the variant forms of address deriving from this structure, such as title without name, last name alone and multiple names including nicknames. The research demonstrated that the most common address forms are the first name and the title plus last name. When the intimacy dimension was taken into account, the first name address is the more intimate of the two patterns. If status is considered, then the person of higher status initiates each new step towards friendship. Brown and Ford also suggest that when

57 someone is uncertain about what form of address to use, this person avoids the use of any sort of personal name and „makes do with uncommitted omnibus you‟ (1964, p. 243).

In their survey Little and Gelles (1975) deal with forms of address as they relate to status transformation with particular attention to avoidance of any type of address. Personal distance (age) was found to be one of the most significant predictors for the use of formal forms of address and informal forms of address. Their findings support the hypothesis that among university graduates, as they progress through their degree program, there is a greater tendency to avoid using any form of address.

Adler (1978) very thoroughly discusses different forms of address in modern English. He emphasizes that the fact that Modern English has only one second person pronoun of address (you) makes it rather difficult for English speakers to recognize whether it is refers to „one person, more than one person, to males and females, even to animals‟ (1978, p. 177). Further, Adler highlights the social significance of addressing. Titles and formulas of greeting and so forth have the special task of helping to establish and define relationships. He adds „whenever social circumstances of any sort lead to the formation of a distinct group within the whole body of a society, the members of this group will tend to develop speech habits of their own‟ (1978, p. 169).

Wolfson (1983, pp. 61-86) surveys research that has been carried out in the area of address rules for a face-to-face conversation and discusses patterns that govern the way middle class Americans address one another. It was found that most dyads in this society follow a reciprocal pattern, addressing each other by mutual first names or by mutual titles and the last name depending largely on the social and occupational status or age. Where age and status do not coincide, status will usually prevail over age.

4.1.1.1 Address Forms in Face-to-Face Communication and IRC

While, as was mentioned above, the most used forms of address in face-to-face interaction are the first name and a title plus last name, face-to-face interaction does offer other methods of address. These can be the movement of a person‟s head or hand, eye contact, or just by turning to someone physically without saying their name. Initial naming is unnecessary in such circumstances, particularly once the talk has been established. In fact, it would be most unusual to hear: 58 Mary: John, are you going to rehearsal tonight? John: Mary, yes, I am. Mary: John, what time? John: Mary, about six. (Crystal, 2001, p.162) When closing conversation, we usually accompany the act of leaving with phrases like goodbye, take care, see you later and so forth without really having to address the person. Sometimes, it is not necessary to address people with words at all. Other means exist to express that you are saying goodbye to a particular person or to a group, for example eye contact or waving the hand.

On virtual IRC these physical acts are not plausible, and thus, other practices must be exploited. One example of importing a conversational strategy from oral discourse into the medium of IRC is the practice of including the name of an intended addressee in a turn, which is called addressivity by Werry (1996). He has provided us with the initial assumption that

“it has become entirely conventional for speakers to indicate the intended addressee by putting that person‟s name at the start of an utterance, followed by a colon: ... (3) frank: there` s a girl boot: where? where?

Such a high degree of addressivity is imperative on IRC, since the addressee‟s attention must be recaptured anew with each utterance. A contribution factor to the emergence of this convention may be that the role of the “listener” becomes more passive on IRC than in spoken dialogue”.

Greenfield and Subrahmanyam (2003) confirm Werry´s supposition and add that “often participants clarify whom they are addressing, with whom they are initiating a conversation, or to whom they are responding by prefacing their contribution with the name of the addressee.”

Stemming from the above statements and findings in other studies on IRC, it is believed that leaving out names will not be a very common phenomenon owing to the fact that participants cannot see one another, yet the message has to reach the right person. The lack of audio-visual cues is significantly impacting and altering the classic dominant model of address rules for a face-to-face conversation. In other words, direct

59 address by nickname, regardless the age, is supposed to be the most frequent phenomenon.

As mentioned above, another reason why addressing plays an important role on IRC is that the name, nickname and other expression of address are one of the few, if not the only, mutually known features participants of chat can rely on and draw from when chatting. The importance of the nickname for a chatroom community and further communication is illustrated in the following example.

Example

191. Snuffy has joined conversation. 223. Snuffy: hey there Pops 231. POPS: hi snuffyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyy 257. POPS: sorry snuffy I dont seem to know u 269. Fire: are you new Snuffy? or just a different nic? 278. Snuffy: little of both Fire n Ice 668. DeltaDawn: still, i don´t know who you are snuffy

(C. 60+)

Here, the chat participant called Snuffy who regularly visits the chatroom and probably knows the other chatters well, has changed his nickname so that nobody is able to identify him. In the course of chatting different participants try to guess his identity, which further becomes one of the main topics for discussion. The extent of that discussion, and its relevance, going from line 191 up to line 668 and further till the end of the chatting corpus can be observed.

The nickname may disclose something about a person‟s character or even reveal a gender. References to collective cultural, ethnic, and religious themes in nicknames might indicate that the individual belongs to a certain social group (Bechar-Israeli, 1995).

60 4.1.2 Analysis

Being exposed to a comparatively wide range of age-based chatting corpora over the past few months, I have noticed certain idiosyncrasies in conversational techniques related to address forms recurring across denominations, particularly in three phases of interaction. Stemming from this fact, the following three stages of chatting were investigated in each age group:

1. Address forms in opening sequences

2. Address forms in the course of IRC conversation

3. Address forms in closing sequences

Moreover, drawing from Quirk et al. (1985) and other research already done (Brown and Ford, 1964; Little and Gelles, 1975; Adler, 1978), the data have been further divided into the following address categories:

1. General nouns (direct address types, collective nouns – room, gang)

2. Nicknames - vocatives (direct address types)

3. Terms of endearment – epithets (adjectives)

4. Personal pronoun you

a. you without any address form

b. you and nickname, general noun

Concerning the pronominal address you, the reason for such division is that it appeared to be both the most common pronoun and the most problematic in terms of understanding the conversation flow on IRC. I suppose „you and a nickname‟ or a general noun would be more frequent in use than „you without any address‟. As Adler (1978) stresses, it is quite difficult to recognize to whom you refers to if it is not explicitly indicated. Besides, the specific features of the media under investigation must be considered.

61 4.1.2.1 Address Forms in Opening Sequences

The opening address is one of the most important phases of conversation, either face-to-face or virtual. For a successful conversation, it is important to address the person we wish to talk to in an appropriate way. The above-mentioned studies of addressing (Richards, Schmidt, 1983; Brown and Ford, 1964; Little and Gelles, 1975; Adler, 1978) showed that apart from standard address by title and name, initial no naming is also quite frequent in a face-to-face communication.

In the following research, it will be observed what forms of opening address are used on IRC and how they are used. It is believed that the anonymous character of IRC, on the one hand, limits the use of the standard forms of initial address and, on the other hand, can offer broader exploitation of opening address possibilities (Werry, 1996, Greenfield and Subrahmanyam, 2003). First, I discuss given age groups independently, and then more universal conclusion is made based on my hypotheses.

Age group Teens

Table 3 shows the relation between opening conversation on IRC and the use of address forms in the individual Teen groups.

First, I would like to consider the question of dividing Teen groups into sub- groups. I have found that the level of significance in this group is = 0.15. If the supposed level of significance is lower than the real value, the hypothesis H0 cannot be refuted, that means there was not proven any dependence of age variable and the variables of address forms given in the table. Thus, for a further comparison with other age groups, I will use the results of a Teen group as a single unit.

The figures in Table 3 indicate that the address by general nouns, such as “hi room”, “will someone please im me”, “good morning lady” occupies the leading position as regards its occurrence (N=37).

It is also notable that the frequency of nicknames is more than three times lower than the use of general and vague nouns such as someone, anyone, girls, and guys. It seems that in the Teen category the opening address by name and thus the whole opening ceremony, as we will see in chapters Openings (4.2) and Closings (4.3), tends

62 to be omitted and the opening address is more restricted to individual „shout outs‟ such as contact advertisements, for example blonde hair blue eyes 5´7 36d medium build tongue and eyebrow periced 20 from pa guys if u like what u hear im me (A2:244), looking for someone to chat with, 17/m/India here im me if u wanna chat (A2:154) , or looking for people from the same country hey ne 1 from aussie IM me (A2:6).

Table 3: Teens: the form of address when entering the room – opening

Forms of address A1: 13-14 A2:15-17 A3:18-19 Occurrence % Rate GENERAL NOUNS direct address, Example ( room, everybody, all, someone, guys, people, ladies, girls) 11 17 9 37 56% NICKNAMES direct address, Example (hey bina, wb butters, a/s/l lil, duracell, etc) 1 5 6 12 18% TERMS OF ENDEARMENT hunn (honey), hottie 0 0 1 1 1,5%

ADJECTIVES tiny, beautiful (damn tiny u jus gon leave me like dat, hi beautiful) 1 0 0 1 1,5% PERSON. PRONOUN you you without any address (How are you?, Which island are you on? Are ya f/male?) 6 2 5 13 20% you and nickname (are u a pimp Oscar? T- Rock you lookin for a wife…) 0 0 2 2 3% Total no.of occurences 19 24 23 66

The majority of these opening address forms, however, is lost in the number of other exchanges and is never responded to.

Concerning the pronominal address you, it can be inferred from Table 3 that „you without any address‟ (N=13) is considerably more frequent than the use of „you

63 and a nickname‟ (N=2). Given the specific environment of a chatroom, this fact is quite significant. It is supposed that participants want to make sure that their message reaches a particular person in the chatroom by adding the nickname as well. However, the results show the contrary. There is no doubt that such a lack of clarity on IRC leads to frequent misunderstandings or missing the message entirely like in the following “exchange”:

206. LIL ONE: KING NOTCH HALLA 208. King Notch: What it do ma 212. LIL ONE: WATS GOOD?? 238. King Notch: it aint nothin jus chillin!!! 244. King Notch: and u what u up 2 (A3. Teens) The participant called LIL ONE opened conversation with King Notch by greeting him and using direct address by nickname. Then, the exchange continued between King Notch and LIL ONE in lines 208 and 212. From lines 238 and 244 it is not apparent to whom King Notch speaks – the use of the pronominal address is vague and therefore, there is no reply to line 244. It is only presumed that the message was directed to LIL ONE as it appeared to be the only participant with whom King Notch was in contact. Further, the anonymity of a chatroom may lead to such unique types of opening address as using self-address in the expression “Happy B-day 2 me”, instead of a common phrase with the direction to a person “Happy birthday to you”: 240. Mz.Angie Makes Niggaz Scream Th: joined the room. 253. Mz.Angie Makes Niggaz Scream Th: Sup yall 258. Mz.Angie Makes Niggaz Scream Th: Happy B-Day 2 me 264. downsouthbabyboy: YO MZ ANGIE WHERES YOUR PIC 265. yungzeeky: I LL TELL U HAPPY BIRTHDAY IF U GOT A PIC 267. downsouthbabyboy: EXACTLY 268. Mz.Angie Makes Niggaz Scream Th: yea (A3. Teens) When the whole exchange from the time Mz.Angie Makes Niggaz Scream entered the chatroom is considered, there can be observed three functions this self- address fulfilled: 1. phatic – making the first contact, a type of opening different from the rest calling for being replied to; and 2. referential – giving private information,

64 whether true or not, about the person herself; 3. conative, Appell – the information was processed by other participants, commented upon and the first contact was made (Černý, 1996).

In contrast with majority of unsuccessful opening addresses in Teen chatrooms, this unique one was responded to and the first contact was established. Thus, I agree with Herring (1999) and Greenfield and Subrahmanyam (2003) that on IRC where there are tens of participants speaking at one time, a person with the most unusual and striking opening – in this case opening address, is more noticeable and has more chance of receiving a response from the others.

Age group 30+

Table 4 indicates the relationship between the opening stage of conversation and the use of address forms in the age group of 30+. It is worth noting that a direct address by nickname is the most used form of address (N=21) closely followed by direct address by general nouns (N=17).

Table 4: 30+ the form of address when entering the room – opening

Forms of Address Occurrence Rate % GENERAL NOUNS direct address (room, people, fox, all, everyone, ladies) 17 36% NICKNAMES direct address (hello fatal, hi lindsey...) 21 45%

TERMS OF ENDEARMENT 0 ADJECTIVES 0 PERSONAL PRONOUN you you without any address (how are you?, and you?…) 5 11% you + nickname (where are you from upfront? 4 8%

TOTAL 47

On the other hand, both pronominal address types turned out to be represented rather weakly (N=9). No type of endearment or adjectival form of address was noted.

65 Age group 60+

Table 5 shows the frequency of address types when entering the 60+ chatroom, a total of 244 occurrences. The figure deserving main attention is the number of direct forms of address by nickname (N=221), which constitutes more than 90 per cent of all address forms in the table.

Table 5: Over 60: the form of address when entering the room – opening

Forms of Address Occurrence Rate % GENERAL NOUNS direct address (morning room, hi gang, gmm everyone, gm sir) 4 1,5% NICKNAMES direct address (hi sixty, gm Weaver, Hi Delta, wb Chief...) 221 90,5% TERMS OF ENDEARMENT hon (hi donna, honnnn) 1 0,5% ADJECTIVES non-existent 0 PERSONAL PRONOUN you 0,5% you without any address (nice to meet u) 1

you + nickname (chief are u here, Dawn

how are you today?, thanks and you sady, 7% gm irish do i know you, …) 17

TOTAL 244

It is also notable that there are only four occurrences of direct address by general noun. It seems that senior participants highly prefer the use of nicknames when opening conversation, rather than addressing by vague and impersonal general nouns.

I have already mentioned the unique honorific expression sir used by the same participant in different places targeted to address different IRC users (In Appendix 60+):

77. donna: pops, gm sir

383. donna: Chief, when I walk into this room I have always found you to be a wonderful gentleman no way could you have did anything wrong sir

402. donna: ty chief very much bit no thanks needed sir just a fact

66 The way this honorific is used in the chatroom excludes any meaning of irony or sarcasm, on contrary shows high degree of formality, respect and politeness.

As the table indicates, the use of „you and a nickname‟ (N=17) highly outnumbers the use of „you without any address‟ (N=1). This ratio confirms the fact that the older generation favours clear and direct address types to make sure the message reaches a particular person.

Conclusion

Graph 1 shows the relation between age and a frequency of address forms when opening conversation on IRC.

68,5% 70,0%

60,0%

50,0%

40,0% Teens 30+ 30,0% 18,5% 60+ 20,0% 13,0%

10,0%

0,0% address forms when opening conversation

Graph 1: relation between age and total per centage of occurrences of address when opening conversation on IRC. Note: 18.5% (N=66), 13% (N=47), 68,5% (N=244)

The strength of dependence between age and the use of address forms in opening sequences was examined with the help of Cramer´s V index (C´s = 0.49). It indicates medium strength of dependence among the three age groups. The level of significance is 0.00 which refutes H0 and confirms H1, see Table 6.

67 Table 6: Strength of dependence – opening address forms

Summary Statistics for Contingency Tables

Value

Significance Level 0.00

Cramer´s V 0.49

Drawing from quantitative analysis, the older generation of 60+ utilizes the most opening address forms, thus making up for the absence of visual and vocal means of communication. Nevertheless, such a big difference in opening address, up by nearly 55 per cent compared to 30+ and by 50 per cent compared to Teens has not been expected. The abundance of direct address by nickname, representing 87 per cent of the total address types by nickname for all age groups, shows the tendency to make the communication in the otherwise chaotic IRC chatroom clearer and more understandable. This fact supports our initial assumption about a conservative, more polite and cooperative approach to IRC on the part of senior users.

The middle-age category of 30+ inclines more towards the younger generation, using an opening address the least of all three age groups. Therefore, it does not support my hypothesis H1. However, when figures of direct address by nickname are compared, 30+ shows the signs of a more cooperative attitude and the need to make the communication clearer (N=21) than in Teens (N=12).

The category of Teens with its low number of any type of opening addresses and furthermore the highest number of general or vague types of address – 64 per cent of all three age groups – does not support the initial assumption drawn from Werry (1996). It further confirms the fact that the younger generation has a tendency to use less traditional methods of opening address and to exploit the chatroom possibilities and its anonymity for reasons other than simply chatting with peers. This fact goes hand in hand with other features of conversation analysis such as opening and closing sequences and turn-taking.

The observed phenomenon of the pronominal address you is not of high significance in opening address, however an increase is expected in the following stage- address when turning to someone in the course of chatting.

68 4.1.2.2 Address Forms in the Course of IRC Conversation

On IRC, the physical acts replacing verbal address forms are not plausible, and therefore it was supposed for all three chatrooms that in further conversation direct address by a nickname will continue and will be the most frequent phenomenon. At the same time I assumed that the frequency of occurrence of address in the course of chatting would be slightly higher than in the opening address. This assumption is drawn from the fact that, owing to the anonymity of a chatroom and the lack of visual and audio means, chat participants have to address each other more to ensure successful transfer of their message. Another reason is that our corpus is not ´topic driven´ and people are addressing one another in the course of chatting rather than addressing a topic. Further, the synchronous chat structure is not very well organized or technically advanced as to show at any moment of conversation to whom a participant is talking (Crystal, 2001). This fact is convenient for our analysis because it approximates IRC with face-to-face conversation.

Age group Teens

Table 7 shows the frequency of address forms in individual Teen groups. Primarily, I would like to consider the question of dividing Teen groups into sub- groups. I have found that the level of significance in this group is = 0.02. If the supposed level of significance is higher than the real value, the hypothesis H0 must be refuted and H1 is confirmed, which proves that the use of address forms is dependent upon age. Thus, for a further comparison with other age groups, I will use the results of a Teen group not as a single unit, but as independent figures.

When the figures in Table 7 are analyzed, I find that there is a higher occurrence of address forms in the course of chatting than when opening conversation. Nevertheless, on the closer inspection, it was not reached by virtue of nicknames but by the pronominal address form „you without any address‟, (N=42), while the direct address by nickname remained nearly the same, (N=16).

69 Table 7: Teens: the form of address within IRC conversation

A1: 13- A2:15- A3:18- Occurrence % Forms of address 14 17 19 Rate GENERAL NOUNS

direct address (anyone, man, guys, damn fool) 5 4 2 11 14% NICKNAMES

direct address, Example (Samantha, Lil, Duracell, etc) 4 2 10 16 20% TERMS OF ENDEARMENT hunn (honey) 2 0 0 2 3% ADJECTIVES

silly, stupid, (Example: not u silly, ) 1 0 0 1 1% PERSON. PRONOUN you

you without any address (u see, that scares u) 12 12 18 42 53%

you + nickname (Kiwi, I already told you) 1 1 5 7 9% Total no.of occurrences 25 19 35 79

Here are examples of the pronoun „you without any address‟ and the pronoun „you + nickname‟ for comparison:

Examples:

69. burretsw: ? 70. burretsw: i cant seem to send 71. suhas7272: who r u 72. rm250002 joined the room 73. burretsw: can i receive 74. suhas7272: ya 75. suhas7272: who r u 76. burretsw: pleease

70 77. burretsw: send an im 78. suhas7272: nam,e 79. paU2x: im LOST!, (A1) and on the other hand

52. sharden87: I can see ya talking burretsw (A1) 43. JapanaMasturbation: Kiwi, I already told you you aren´t my best friend anymore (A3) 252. downsouthbabyboy: YO T ROCK YOU LOOKIN FOR A WIFE YO (A3) From the examples above, it is obvious that the use of „you without any address‟ is more vague and indeterminate. It is not as easy for the ´listener´ to understand to whom the message is directed as when the direct address by nickname is used. Especially, if the number of active participants at the moment of chatting is taken into consideration, the danger of misunderstandings and missing the message all together is very high.

On the whole, the preference for the pronominal address forms „you without any address‟ (53%) to the use of direct address by nicknames (20%) in the category of Teens can be noted, while the use of „you + nickname‟ is very low, a total of 8 occurrences (9%).

The result of the chi-square test for independence confirmed that the values in individual sub-groups are independent, and H1 can be confirmed. When I closely inspect the sub-groups, the address forms in category A3 vary significantly from the other two groups, mainly in the use of direct forms of nicknames (N=10) and in the use of „you + nickname‟ (N=5). The difference may be explained by a higher level of maturity of 18 and 19 year old IRC users compared to 13-14 and 15-17 year olds.

Age group 30+

Table 8 indicates the relation between the forms of address and conversation in the course of the Internet chatting in the age group of 30+.

71 Table 8: 30+ the form of address within IRC conversation

% Form of Address Occurrence Rate GENERAL NOUNS

direct address (people, hommie, my hero, girl, damniot...) 10 8% NICKNAMES

direct address (you have a nice pic lindsay, you are calm now, Fhrun…) 55 47% TERMS OF ENDEARMENT 0 ADJECTIVES 0 PERSONAL PRONOUN you

you without any address (your still here why, im sure you look pretty, you can?, you two should get together…) 33 28%

you + nickname (omg james youre so smart, jen your already beautiful…) 20 17% TOTAL 118

The comparison of address forms reveals that the highest frequency of occurrence is direct address by nickname, the total of 47%, followed by pronominal address „you without any address‟, the total of 28%, while the direct address by general nouns reached the least frequency, 8%.

Age group 60+

The distribution of address forms in the course of chatting in the category of 60+, see Table 9, visibly indicates the effort of chat participants to be cooperative and clear by naming and addressing people in the chatroom.

72 Table 9: 60+: the form of address within IRC conversation

Forms of Address Occurrence Rate % GENERAL NOUNS direct address (sir, mon amis) 3 1%

NICKNAMES direct address (true dream, What is your age shimmering?, chief..you did not do a thing wrong, you are such a tease Snuffy…) 218 80%

TERMS OF ENDEARMENT non-existent 0 PERSONAL PRONOUN you

you without any address (you think?, you getting used to that smell?) 16 6%

you + nickname (darn Fire you are good, I am inclined to agree with you Donna…) 34 13% TOTAL 271

The most attention needs to be paid to figures of direct address by nickname (N=218) and pronominal address form „you + nickname‟ (N=34). More evidence of the older generation endeavouring to avoid any vague and unclear types of address can be seen when I compare both varieties of the personal pronoun you. The distribution of the more vague form of „you without any address‟ (6%) occurs half as often as the more clear form of „you + nickname‟ (13%).

It is notable that direct address by general nouns – hi room, mon amis – is the least utilized form of address with only 1 per cent of occurrence.

Conclusion

Graph 2 shows the relation between age and a frequency of address forms in the course of conversation on IRC.

73 58% 60%

50%

40%

Teens 30% 25% 30+ 17% 60+ 20%

10%

0% address forms in the course of conversation

Graph 2: Relation between age and total per centage of occurrences of addressing in the course of conversation. Note: 17% (N=79), 25% (N=118), 58% (N=271)

Cramer´s V index of address forms in the course of IRC conversation is C´s = 0.31, which indicates medium-weak strength of dependence among the three age groups. The level of significance is 0.00, which refutes H0 and confirms H1, see Table 10.

Table 10: Strength of dependence – opening address forms

Summary Statistics for Contingency Tables

Value

Significance Level 0.00

Cramer´s V 0.31

74 Graph 2 indicates a high occurrence rate of address forms in the age group 60+ (58%). Most of the address forms used in this age group are direct address by nickname (N=218) which serves to recapture the addressee´s attention anew with each utterance like in the following example. Every message contains a nickname of the person to which it is addressed:

Example

132. POPS : good to see u wineeeeeeeeeeeeee 133. retfirechief108 has JOINED the conversation. 134. donna : pops isn't it like old times to have wine back with us 135. Guest_Zoey122 has LEFT the conversation. 136. donna : wb Chief, 137. shimmering-wine : ty pops its great finding time to chat with friends 138. DeltaDawn: wb firechief 139. ROSENDOO has LEFT the conversation. 140. USUMBEACH has LEFT the conversation. 141. retfirechief108 : ty 142. retfirechief108 : dawn ty 143. shimmering-wine : ty donna gf 144. POPS : yes donna she should come in mereoften 145. donna : anytime wine g/f nice to have you back with us 146. retfirechief108 : hey mon hes back (C. 60+)

The second highest frequency of address forms occurs in age group 30+ (25%). The most used address form is direct address by nickname (N=55) as in case of 60+.

At 17%, Teens are the least likely to use address forms. The address form most used in this age group is „you without any address‟ (N=42); in comparison with „direct address by nickname‟ (N=16).

As expected, the pronominal address „you + nickname‟ in the groups specified has increased from 23 occurrences in opening address forms to 59 occurrences within conversation. This fact can be explained by a more established relationship among participants in the course of chatting where a sender more clearly indicates that a message is addressed to a particular addressee. Still, the occurrence rate of the

75 pronominal address form „you without any address‟ is 21 per cent higher than the occurrence rate of „you+ nickname‟.

4.1.2.3 Address Forms in Closing Sequences

The act of addressing is not only about how to “strike up” a conversation with someone, but also how to finish a conversation, how to “disappear” (Jones, 1998), to use IRC vocabulary. In this section, I focus on how participants of IRC address other participants when leaving the conversation.

Before making any detailed study about the way chat participants address one another when leaving the chatroom, it was supposed that this would be of lesser importance than the first two stages. The hypothesis was based on the fact that, given the anonymous character of a chatroom, it is easier to leave the room without addressing than to enter it. Moreover, I suppose more common address forms such as general nouns.

Age group Teens

The frequency of occurrence of closing address forms is quite low. With such low figures, there was no point in making statistical analysis of the chi-square test for independence and level of significance. As Table 11 shows, in Teen chatrooms, no direct address by nickname is used and only neutral expressions, N=2, have been noted.

Table 11: Teens: the form of address when closing conversation

Form of Address Occurrence Rate A1:13-14 A2: 15-17 A3: 18-19 TOTAL % GENERAL NOUNS PPL (people) 0 1 0 1 50% PERSON.PRONOUN you cya (see you all) 0 1 0 1 50% TOTAL 2 2

76 Age group 30+

Table 12 of the mid-age group of 30+ shows a higher use of direct address forms by nickname, N=6. The majority of address forms is represented by a direct address by a nickname (N=5).

Table 12: 30+: the form of address when closing conversation

Forms of address Occurrence rate % GENERAL NOUNS direct address (I am going to leave you bitches now) 1 17% NICKNAMES direct address (bye woogy, woogy woo) 5 83% TOTAL 6

Age group 60+

From Table 13 it is apparent that from a total of 35 occurrences of any address form, 28 (80%) are connected with direct address by nickname. This number represents the highest of the three age categories discussed.

Table 13: 60+: form of address when closing conversation

Forms of Address Occurrence rate %

GENERAL NOUNS direct address (god bless all, bye everyone...) 5 14% NICKNAMES direct address (by wine, take care shim, bye firefly) 28 80% PERSONAL PRONOUN you you + general noun (bye you all, see you all later) 2 6%

TOTAL 35

77 Conclusion

Graph 3 shows the relation between age and total per centage of occurrence of address forms when leaving conversation.

90% 81% 80% 70% 60% 50% Teens 40% 30+ 30% 60+ 14% 20% 5% 10% 0% address forms when closing conversation

Graph 3: relation between age and total per centage of occurrence of address when leaving conversation. Note: 5% (N=2), 14% (N=6), 81% (N=35)

When examining the strength of dependence between age and the use of address forms, I used Cramer´s V index to value the strength of dependence. However, the dependence cannot be measured because the marginal value vanishes, in other words the figures are too low and too diverse to be compared and statistically measured.

As supposed, the overall number of occurrences is very low compared to the previous stages. In this point, again the question of politeness and cooperation may arise. However, the lower number of address forms when leaving conversation can be caused by reasons other than simply violating Grice´s Principles. It can also be due to the somewhat loose and fragmented nature of computer mediated interaction. The so- called “noise” in the channel – the same that exists in face-to-face interaction – suggests that IRC is “noisy” concerning addressing when leaving a conversation as well (Herring, 1999). By this I mean that opening address on IRC is more visible thanks to the use of unusual forms of greeting that attract the attention of the chatroom participants, for example expressions like ´hello gang´, ´hi beautiful´, the use of unusual

78 nicknames such as ´TigerLily´, Dream Weaver´, writing in capital letters or in colours other than black.

On the other hand, closing a conversation is very simple and not that visible – a person just clicks on a “leave the conversation” button. It very often happened that the participants left and still received messages from the others like in this example:

Example: 243. Guest_patricia has joined the conversation. 593. Guest_patricia has left the conversation. 649. Firefly: Patricia gm sorry did not see you (C. 60+) The most usual address forms, apart from direct address by nickname are neutral expressions like ´bye all´ or ´take care everyone´. I have noted a rather peculiar address “I am going to leave you bitches now”, (B: 49), see example:

Example: 49. woogywoogywoo : I'm going to leave you bitches now 53. Im Trendy : bye woogy woogy woo 54. FatalisticHomeRun : bye bye 55. vοlkѕwagen racing : ciao 56. belle: bye woogy gooer 57. woogywoogywoo has left the conversation. (B. 30+) It is apparent that the address was not meant to offend anyone, rather, it was meant as a friendly address showing the sense of belonging to the chat community.

4.1.3 Address Forms – Conclusion

At the beginning of this work the two main hypotheses were set. Null hypothesis (H0) suggests that conversational techniques used on IRC are independent of age. The alternative hypothesis (H1) claims that the form of communication on the medium of IRC is dependent on age. Consequently, my research question is whether age has any influence on the communicative methods, in this case addressing, used on IRC at all.

79 When the distribution of address forms in the above analyzed stages of chatting for all three age categories is examined, Graph 4, a notable correlation between age and the frequency and the choice of address forms can be noted. The difference between the age group 60+ (63%) and the age group Teens (17%) in the use of address forms is the most striking.

63,0% 70,0%

60,0%

50,0%

40,0% Teens 30,0% 20% 30+ 17,0% 60+ 20,0%

10,0%

0,0% address forms in all three stages of conversation

Graph 4: relation between age and the total percentage of occurrences of addressing on IRC Note: 17 % (N=147), 20% (N=171), 63% (N=550)

Table 14 shows the distribution of address forms across age denominations. Nicknames are the most used address form in the corpus, however mainly thanks to the age group 60+. On the other hand, it is significant that Teens prefer general nouns and pronominal address without any name as a form of address.

It can be summarized that the younger the chat participants, the less they address their chatroom mates. Furthermore, the difference lies in the selection of address forms in each age category, see Table 14.

80 Table 14: Total occurrence rate in address forms in the three stages of chatting in relation with age Address forms Teens 30+ 60+ Occurrence Rate Total %

General nouns 49 28 12 89 10%

Nicknames 28 81 467 576 66%

Terms of 5 0 1 6 1% endearment Pronominal address you without 56 38 17 111 13% any address you + 9 24 53 86 10% nickname TOTAL 147 171 550 868 100%

When examining the strength of dependence between age and the use of address forms I used Cramer´s V index to value the strength of dependence. Supposing that the standard interval is <0; 1>, Cramer´s V index of address forms in the course of IRC conversation is C´s = 0.37, which indicates medium strength of dependence among the three age groups. Significance Level is 0.00 which completely refutes H0 and confirms

H1, see Table 15.

Table 15: Strength of dependence – address forms in the three stages of IRC

Summary Statistics for Contingency Tables

Value

Significance Level 0.00

Cramer´s V 0.37

The question emerges as to how we can account for such big differences and whether it has to do with politeness at all. The answer may lie in different expectations individual age categories draw from chatting and thus belong more to the field of sociology rather than linguistics, however, I would like to discuss this issue shortly now.

81 According to a sociological study dealing with Teens and chatting on the Internet (Jones, 1998), young people‟s aims may not be merely to chat with their chatroom mates, that is, to give and receive information, “most of the teenagers seem to be drawn to Internet chatrooms by the promise of fantasy and fun”. They may be looking for a ´cyber´ partner or Internet ´date´, adopting a new physical persona, escaping from boredom, talking about sex or just having fun without committing themselves to any relationship: “She was not interested in meeting the boys with whom she conversed, as this might undermine her attractive and aggressive on-line persona” (Jones, 1998). The absence of personal contact and the reasons given above result very often in such vague address forms – general nouns (33%) compare nicknames (19%) – and exchanges as found below in the corpus:

Example:

44. tikigaq.harpooner: so what does every1 do for fun in here 45. tikigaq.harpooner: ??? 47. klive3614: whats up ppls???????????????????? 48. xRx: y 50. tikigaq.harpooner: juss curious 52. freestyler616: Any1 wanna chat with 15m???? 53. Kaila Jo: hey room 54. Sexdevil: HYPERDINK IS A CUTTIE 55. klive3614: ni ozzies in here???????????????? (A2. Teens)

The sense of belonging to a chat community in the corpus was more an exception than the rule. I believe that these tendencies are reflected by the way and frequency young people address one another as well.

The mid-age category of 30+ on the one hand follows similar patterns as the Teens, indicating that people look for a partner and want to have fun without committing themselves to any relationship. The expressions like 129. Tweakdup1: This room is deader than hell, or 150. Tweakdup1: this is boring too, 332.watcher_of_souls: what are we talking about??? bring the room of 30+ closer to Teens. On the other hand, I observe more developed community awareness, chat participants know one another from previous encounters in the chatroom. Thus, address forms appear with more frequency – nicknames (47%). See example:

82 Example:

132. chief3212 : there's some life. hello fatal 133. pavinjohn11 has LEFT the conversation. 134. FatalisticHomeRun : hey chief 135. chief3212 : how are you today 136. malone21139 has LEFT the conversation. 137. jrbudman08 has LEFT the conversation. 138. FatalisticHomeRun : are you the same chief i was giving a hard time to a while back? 139. olivia354 has LEFT the conversation. 140. chief3212 : yeah, I'm over it though. big shoulders.... 141. manwidaplan45 has JOINED the conversation. 142. FatalisticHomeRun : good man (B. 30+)

The chatroom 60+ has the most developed sense of cooperation and belonging to a community. The abundance of address mainly in a form of nicknames (85%) in the category of 60+, I believe, serves participants‟ various purposes. First, due to numerous overlaps, all participants wish their utterance were conveyed just to a particular recipient. Second, senior chat users need to have better orientation in the text, thus helping themselves with the names. Third, senior chat users take the chat more seriously, discussing more specified topics such as housing, looking after grandchildren and so forth and thus they pay more attention at whom the message is directed. Fourth, senior chat users are more conservative in using individual types of address and choose more traditional types, ergo nicknames, see example:

Example:

132. POPS: good to see u wineeeeeeeeeeeeeee 133. retfirechief has joined the conversation 134. donna: pops isnt´t it like old times to have wine back with us 136. donna: wb Chief 137. shimmering-wine: ty pops its great finding time to chat with friends 138. DeltaDawn: wb firechief 141. retfirechief: ty

83 142. retfirechief: dawn ty 143. shimmering-wine: ty donna gf 144. POPS: yes donna she should come in mereoften 145. donna: anytime wine g/f nice to have you back with us 149. shimmering-wine: pops having a 8 year old to watch dosen´t give me much time 152. donna: wine g/f how is your granddaughter? (C. 60+)

Finally, it is necessary to consider the research of Brown and Ford (1964) and Wolfson (1983) that demonstrated that the principal option of address is the use of first name and the use of title and the last name. From the above results, it can be inferred that 60+ and 30+ approximate to a spoken conversation more than the Teen group, as their use of nicknames is the most frequent and thus bears a resemblance to the use of names to address someone in spoken interaction.

It can be said that the higher the formality, the higher the degree of politeness in address (Adler, 1978). In Internet chatrooms, however, the degree of formality diminishes due to its anonymous, impersonal and recreational character. As evidence, I encounter only three manifestations of formal address in the corpus specified (C. 60+, lines 77, 383, 402), namely the honorific “sir” in the category of 60+, that is, in the most conservative of all given categories. However, it does not mean IRC is less polite.

Although there are many other sociological features influencing every age category, I will return to the overall pragmatic classification of address forms through the whole spectrum of age categories.

The differentiation by age brought some notable comparisons and outcomes. The results of the analysis show clear and consistent stratification of address forms in the corpus specified. I will deal with the most significant ones in this section.

Apart from the total number of address forms and their influence on age, I focused on the relationship between age and the three stages of addressing; opening, in the course of chatting and closing. Graph 5 indicates that the three age groups show different tendencies.

The lowest figures appear in the category of Teens even if a higher number of opening address forms is noted than in 30+, a rise by 5.5 per cent. Then, there is only a

84 slight rise in “the address in the course of conversation” and then a very sharp fall to only two occurrences of “address forms in closings”.

68,5% 58,0% 81,0% 100,0% 90,0% 80,0% 70,0% 60,0% 50,0% 60+ 30+ 40,0% 30,0% 25% Teens 20,0% 13%

10,0% 18,5% 17,0% 14% 0,0% 5,0% opening in the closing course

Graph 5: The influence of age on the frequency of use of any address forms in all three stages of chatting

The mid-age category of 30+ illustrates a significant increase from „the opening address‟ to „address in the course of conversation‟, by 12 per cent; however, a big drop from 25 per cent to 14 per cent of occurrences of „closing address‟ can be noted. The category of 60+ has the highest numbers in all three stages of chatting, even if the occurrence decreases to a mere 35 occurrences in „closing address‟ forms. Generally, I find that the tendency to address on IRC is the highest during the course of chatting, while when closing, the numbers in the three age categories considerably decrease. It is comprehensible considering that even in spoken interaction the act of leaving is connected with less addressing than in other stages of conversation (Crystal, 2001).

General nouns form the third largest group of address forms in the corpus, 10 per cent. I placed these address forms under the heading of „direct address by general nouns‟. (see tables). Among the ones most used belong expressions like ppl, anyone, room, ladies, chicks, guys, everybody, gang.

85 Further, I find such address forms as the seemingly rude expression “bitches”, ´innovative´ expression “damniot” – connection of “damned” and “idiot”, flirty “my hero” both in 30+ category, formal honorific “sir” in 60+. It must be stressed that the highest occurrence of these general nouns can be found in the category of Teens, a total of 49 occurrences, followed by 30+ with 28 occurrences. The least, 12 occurrences, appear in the category of 60+.

The more frequent use of general address forms in the category of Teens is connected with a unique phenomenon of IRC opening phrases, which is what is called contact advertisement, often with sexual context, totaling 17 cases. Openings such as “hey…15/m…anyone wit [sic] pic wanna play before i get dressed…im me” (A1: 124) or “do anybody in her [sic] got mics and wanna get a freestyle session” (A4: 38) can be found. Usually these openings do not gain any response. Only 3 occurrences were observed in the 30+ and no contact advertisement attempts in the category of 60+.

The analysis of general noun address types proves my initial assumption that the older generation would use more names and nicknames than other categories; then the use of general pronouns for addressing ´anyone´ and in fact ´no-one´ in the group is less probable in their category.

I distinguish two types of direct address: 1. direct address by nicknames and 2. direct address by general nouns mentioned above. I believe that the use of „direct address by nicknames‟ on IRC signifies a clearer and a more open attitude to the hearer and the potential of ongoing discussion and further development of a relationship, which is one of the priorities on IRC. The comparison is made with a „direct address by a general noun‟ that is more vague and not so straightforward. However, Rintel et al. (2001) oppose by stating that: „…seeing a nick does no equate to knowing the other user…Indeed, given that nicks are almost the total embodiment of a user on IRC, their use without a token [a greeting] in the interaction-initial slot is potentially very threatening, strongly invoking the intimacy which may not exist.‟ Graph 6 indicates the use of direct address forms, either by nicknames or by a general noun, in the three age groups.

86 100% 90% 14% 80% 70% 31% 60% 81% 60+ 50% 30+ 40% Teens 30% 55% 20% 10% 14% 0% 5% direct address by direct address by nicknames a general noun

Graph 6: Comparison between direct address by nickname and a general noun in all three stages of chatting

We can see that in the three age categories the numbers of „direct address by a general noun‟ is a mirror image of the frequency of „direct address by nicknames‟. Teens show a less clear and open attitude to the other participants, thus confirming the tendencies in this chatroom mentioned above. In 30+ the difference between nicknames and a general address is two times greater; nevertheless, the development of relationships and the ongoing discussion is more established. Finally, in 60+ the frequency of direct address by nickname is nearly six times higher than address by general nouns. On the whole, it is assumed that the absence of audio and visual means of communication necessitates the use of direct address by nicknames in the IRC environment.

Using pronouns on IRC, mainly the pronominal address you is among the most common methods of addressing, see Table 14. The initial division of you into two categories – „you without any address‟ (13%) and „you + nickname‟ (10%) – enabled me to observe and evaluate each age category and the effort of its participants to address more clearly, thus follow the pragmatic Principle of Cooperation, namely the maxim of manner (be perspicuous). The initial assumption that „you +nickname‟ will occur with more frequency did not prove correct.

87 Table 16: The use of the pronominal address you in the three stages of chatting by the age categories (occurrence rate) Teens 30+ 60+

you without any address 56 38 17 you + nickname 9 24 53

Looking at Table 16, the descending trend of „you without any address‟ towards the category of 60+ is notable, and on the other hand the descending tendency of „you + nickname” towards the Teens. The category of 30+ is well balanced inclining more to the younger generation but at the same time showing awareness of the need for more clear expression. Compare the following examples: “yea u can see them pretty good” (A2. 216) or “who r u” (A1.75); “how r u today lidsay?” (B. 207).

By violating Grice´s principles of cooperation and its maxim of manner, the contributions become obscure and ambiguous. “Typing, not a natural behavior, imposes a strong pressure on the sender to be selective in what is said, especially if one is not a very fast or competent typist. And selectivity in expression must lead to all kinds of inclarity” (Crystal, 2001, pp.57-58). In this respect, when using “you” without any other reference, it is not clear to whom or about whom the participant is speaking and the text cohesion is disrupted (Adler, 1978; Herring, 1999).

Finally, the use of terms of endearment must be mentioned. With only one per cent of occurrence rate, it constitutes the least utilized address form in the corpus. The reason can be found in less developed sense of belonging to the community as such and the less emotional character of the medium.

Mey (2001, p. 271) says that addressing is regarded another angle of cooperation, “which is a part and parcel of any attempt at human contact, including conversation.” The group of Teens, in this respect, is the least cooperative group of the three.

After more detailed examination of the chatting samples, I can say that addressing is connected with the topic discussed in the chat room as well. In Teens, chatting is similar to individual ´shout outs´ most of the time with no coherence (Herring, 1999), while older generation of 60+ develops more sophisticated type of conversation, often similar to face-to-face conversation compensating for the weakened

88 link between sender and receiver – length of sentences (Crystal, 2001), more compound sentences, addressing and keeping to the topic. It appears that older generation of IRC users, that means 60+, developed a kind of ritualized method of approaching others - utilizing nicknames to convey the message to the hearer; in mathematic expression every 1.4th line contains a nickname. I have not noticed any kind of norm or ritualized tendencies in the other two age groups.

In this point, my results do not correspond with the above-mentioned statements of Werry (1996). On the contrary, as it could be seen, the research has proved two assumptions. First, I find that addressing on IRC appears in many forms and combinations. Despite its usefulness as a turn-taking device, it is not very conventional and is commonly avoided by younger generations, Teens and 30+. On the other hand, I did note its abundant use in the group of 60+. Second, the alternative hypothesis H1 has been confirmed and therefore, I claim that address forms used on Internet Relay Chat are affected by age.

89 4.2 Opening Sequences

This section first very briefly outlines the significance of openings for conversation and discusses some theories concerning IRC. Further, a closer investigation is made of different types of openings, as well as some specific techniques that differentiate IRC communication from face-to-face interaction. Openings could be considered a subgroup of adjacency pairs having the first pair parts and the second pair parts. The focus is on summons-answer sequences (SA), greeting-return of greeting (GG), question-answer (QA), and the notion of pre-sequences, such as the existence of pre-conversational openings. I mainly draw from Schegloff´s (1968) model of telephone openings, Schiffrin´s model of opening encounters (1977) and studies dealing with Internet Relay Chat openings (Rintel et al., 2001; Ahti and Lähtevänoja, 2004).

4.2.1 Theoretical Background

The research on openings deals with the method of how someone begins to create the conversational social relationship (Downes, 1998). Some scientists, namely Schiffrin (1977, p. 679), have proposed that opening sequences consist of three phases 1. Cognitive recognition (recognition of the other as a member of a class or category), 2. Identification displays (a smile, an eyebrow flash) and 3. Social recognition displays (greetings).

A number of scholars (Schegloff, 1968, Schiffrin, 1977) have investigated openings that are called topic initiating. Topic initiating openings, for example summons-answer sequences, are designed to shorten the social access process and gain a recipient‟s immediate attention.

It can be said that the structure of telephoning and IRC carries many similar attributes. In both methods of communication, visual contact is removed; however the speaker and listener – in case of IRC the sender and receiver – are logged onto the system simultaneously. Another common feature is spatial and temporal distancing that brings advantages and disadvantages; the most significant advantages are the importance of anonymity, freedom from external distraction, and revising what a person wants to say before transmitting the message. Disadvantages can be seen in the loss of communicative nuances, decreased feedback, and such phenomena as flaming and reduction of social bonding (Baron, 1984). All these above mentioned aspects lead

90 me to utilize, among other things, Schegloff´s analysis of telephone conversation openings for the description of IRC opening sequences.

Schegloff (1968) examined a corpus of telephone calls to a police dispatcher. The first rule that Schegloff formulated was the so-called distribution rule for first utterances. It is the one who picks up the ringing phone who uses it. Schegloff treats telephone openings as a special case of the summons-answer sequence. The ringing of the telephone, like an address-term, wave, or “Excuse me,” constitutes a summons, a first pair part of a two-part sequence.

Further, Schegloff introduces the term nonterminality of SA sequences, which means that a completed SA cannot properly stand as the final exchange of a conversation. Thus, exchange like (1) is well formed, while (2) is not:

(1) (Ring-Ring-Ring) “The Imperial Hotel”. “Yes, I‟d like to make reservations for 8:00”.

(2) (Ring-Ring-Ring) “The Imperial Hotel”. (silence)

Before proceeding to the situation on IRC, one final comment on Schegloff´s terminating rule. It states “…it is empirically observable that S´s [in telephone conversation] are not repeated without limitation, until an A is actually returned” (Schegloff, 1968, p.1085). There is, then, a question whether similar regulation used by members of the society to limit the number of repetitions of an S holds for IRC as well.

4.2.1.1 Research on IRC Openings

With a view to the above-mentioned facts, I will consider the situation on IRC. The fact is that there are many participants speaking at the same time, thus we can say its structure is multi-focus. Their utterances overlap and do not constitute a clearly organized text. It does not seem easy for the participants to enter and guarantee themselves direct attention from the others in this disordered form of conversation. It was observed that very few opening greetings, comments or questions on IRC gained any response at all (Ahti, Lähtevänoja, 2004; Rintel et al., 2001). In their study, only one-third of the chat participants gained a response to their greeting.

91 Further, no identification displays are possible, emoticons and other features like intensifiers – repetition of letters, the use of capitals, words underlined, and exclamation marks – are types of compensation for the absence of audio-visual contact. As mentioned above, Internet chat fulfils primarily a social function as a source of fun, relaxation, meeting new people, dating, cybersex and looking for a partner. This role of IRC must unavoidably be reflected in the opening phrases used.

When Schegloff regards the ringing of the telephone as the first pair part of a summons-answer sequence, the same applies to IRC and the automatic joining signal (AJS) or automated joining event (AJE) as it is called by Rintel et al. (2001). This signal shows who has entered the chat room and who has returned (for example, Yellowbird has joined the conversation; Abzzztalexie has returned). Its function is informative; to notify present chat participants that another member has come in. The AJS has several similarities to the telephone ring. As Rintel et al. (2001) say: „Both are produced as a result of a conscious action of one interactant…, both provide feedback about the attempted connection to another person…both signal only that interaction is possible not that it has begun‟. On the other hand, the ring sets up the condition of nonterminality while an answer to the broadcast of AJS is not conditionally relevant. That means that it does not have to necessarily complete the summons-answer sequence.

AJS can also be called identification displays (Schiffrin, 1977). By entering the chat room, a newcomer gives a signal to other chat participants that he or she is ready to chat. Further, it can be regarded an „attention-getting device‟, Schegloff (1968). It equals the situation when the ringing telephone indicates that there is a caller who wishes to speak to someone.

Another worth mentioning attention-getting device of a different nature is a contact advertisement (Ahti, Lähtevänoja, 2004), as the following example indicates:

244. kathleen_heim: blonde hair blue eyes 5´7 36d medium build tongue and eyebrow periced 20 from pa guys if u like what u hear im me

(A2. Teens)

Contact advertisements on IRC are similar to newspaper advertisements. They are a typical feature of IRC and are hard to imagine in face-to-face conversation or on the phone. It seems that these contact advertisements on IRC could be well regarded as a summons. Three types of IRC summonses were distinguished in the corpus:

92 1. Looking for people of the same nationality or from the same town or region;

2. On-line dating or soul mates, as it is sometimes called; they are similar to newspaper advertisements, “lonely hearts” columns;

3. Contact advertisements seeking a sexual relationship.

I turn now to considering the problem of social recognition displays on IRC. They represent, in Firth´s (1964, p.69) word a part of an opening “ritual” in face-to-face conversation and are sensed as pre-sequences to further conversation. This means that certain utterances are usually ´precursors´ to another utterance or a sequence of utterances. The identification of pre-conversational openings in the given corpus of data was a complex task. The problem was to identify what serves as a pre-opening and a recognition display on IRC. Finally, the theory of Richards J.C. and Schmidt R.W. (1990) was taken into consideration. It views greetings in a face-to-face conversation as invitations for further talk and therefore categorizes greetings within the group of pre- conversational openings. The analysis presented in this work will show whether greetings on IRC function in the same way.

Moreover, when a person enters a chat room, other participants might notice his or her nickname and address the newcomer first. It happened to me on several occasions that I wanted to enter the chat room unnoticed under the nickname sarku, however some participants addressed me first. The following extract from the category of 60+ may serve as an example:

58. sarku joined the conversation 59. DreamWeaver: hi sarku

Furthermore, it is quite paradoxical that I noticed this greeting only after carrying out an analysis of the collected text several days afterwards. Therefore, the greeting remained unintentionally unanswered. This event confirms my assumption that it is not a matter of impoliteness or a lack of cooperation not to receive initial greetings by other members of the Internet chatroom.

There is no doubt that in a real conversation, greetings can serve different purposes, for example as mere identification displays. Usually they are conditionally relevant. When the first pair part is produced, the next turn has an expected interpretation projected on it. Greetings are repeatable. If a speaker judges that a second pair part has not been achieved at all according to the rule, then the first act can be

93 reinstated. For example, if someone clearly did not hear a greeting, then the greeting can be repeated. IRC is specific in that it enables people to continue opening conversations as many times as they wish. It can be a greeting, a question or addition of a second pair part – which means joining the conversation flow directly. In contrary to face-to-face conversation, however, on IRC, it is not expected that if a newcomer greets the room, all participants will return the greeting. Such behaviour is technically impossible, as the vast number of greetings would flood the screen (Crystal, 2001).

Even if the QA opening sequence might be considered pre-conversational as well, it is placed into a separate category. Such division enables us to observe whether any further topics were provided by means of a greeting only or by means of initial questioning.

4.2.2 Analysis

Bearing in mind the theoretical findings, I now proceed to the more practical part. The following quantitative and qualitative analysis proposes an investigation of which rules and features hold for IRC openings in the corpus specified with the age being a prerequisite. The following groups of openings were investigated:

1. Summons-Answer sequence (automatic joining signal, contact advertisements)

2. Pre-conversational Openings (greeting-return of greeting – directed to all x directed to one person)

3. Question-Answer (opening questions)

It is necessary to point out that Teen group was analyzed as a single unit without any further sub-group structuring.

4.2.2.1 Summons-Answer Sequence

It may be observed that a summons appears in different types of conversations. Apart from the ringing of a telephone mentioned earlier, this attention-getting device includes 1. terms of address (“John?”, “waiter”), 2. courtesy phrases to get attention (“Excuse me?”), 3. physical devices such as a tap on the shoulder (Schegloff, 1968).

94 When someone comes into a room, they usually greet first to follow the principle of politeness. It also depends on the kind of relationship, social status, age and cultural background. I have already specified that on IRC two kinds of summonses can be found: the automatic joining signal (AJS) and contact advertisements. Regarding AJS the main task was to find out how many times this signal gains a response from other chat participants, that is whether it draws the attention of others or whether newcomers open the conversation first. The idea was to specify the rules or rituals, should there be any. Thus, I divided this feature into AJS without any response, AJS that gained a response and AJS where a newcomer opened conversation first. In these situations, AJS can be interpreted as a first pair part and a potential answer as a second pair part of a conversation. Contact advertisements were analyzed in a similar way.

Referring to the research, Table 17 shows crucial differences in opening types within the three age categories. I find the most AJS summonses, a total of 159 occurrences, in the category of Teens. When the total number of messages – lines (N=794) is taken into account, this means that in every 5th line a new participant enters the chatroom. As it can be seen, though, 74 per cent (N=118) of the total are AJS without any response or further opening. This means people enter the chatroom and then leave without any contribution.

Table 17: Opening by means of AJS in the three age categories

OPENING - AJS TEENS 30+ 60+

AJS without any response or further 118 74 16

opening AJS that gained response 1 8 30

AJS where a newcomer opened 40 20 2 conversation first

TOTAL 159 102 48

In contrast, the 60+ category has as few as 48 AJS openings, however only 33 per cent of them (N=16) are AJS without any response. Further, I can see the disparity between AJS that gained response in Teen chatrooms, N=1, and in 60+ where the number exceeds the remaining two figures, N=30. When figures of AJS that gained response and AJS where a newcomer opened conversation first are compared in the chat room 60+, we can speak of a type of ritual, the sense of togetherness in this

95 chatroom. Whenever new people join the conversation, it is other chat participants who immediately react to the summons first, not the newcomers, thus forming a second pair part of a summons-answer sequence; see the following examples:

Examples: (1) 58. Sarku has joined the conversation. 59. DreamWeaver: hi sarku

(2) 191. Snuffy has joined the conversation. 193. shimmering-wine: hi snuffy 194. DeltaDawn: hi snuffy 195. Fire.n.Ice has joined the conversation. 196. Snuffy: hello wine 198. DeltaDawn: hi fire 199. shimmering-wine: hi fire 200. Snuffy: hi there delta dawn 201. Fire.n.Ice: Hello Delta Dawn

(3) 456. Black Widow has joined the conversation 458. Delta Dawn: hi bw 459. shimmering-wine: hi bw 460. Snuffy: hi bw 462. Fire.n.Ice: GM BW 463. Black Widow: hi fire (C. 60+) Furthermore, Table 17 shows that in the other two age categories, Teens and 30+, AJS summonses as an opening strategy seem quite ineffective; in Teens a total of 1 response was found and in 30+ a total of 8 responses.

In addition, from Table 17 it follows that the second largest group of newcomers, Teens, a total of 25 per cent (N=40), and 30+, a total of 17 per cent (N=20), after being unsuccessful in their AJS summons, turn to multiple summonses such as contact adverts, or to other opening strategies, see the example below. My findings correspond with Herring‟s (1999) when she says:

„…the vast majority (88%) of participants who were not responded to had posted only a single message, and the

96 remaining 12% of participants had posted only two messages. In contrast, everyone who posted more than two messages was responded to. Accordingly, some participants adopt a strategy of sending out multiple messages in an attempt to attract a response. The phenomenon of “spamming” – repeatedly posting the same message – is a noxious form of this practise.‟

Rintel et al (2001) also observe that Channel Entry Phase, as they call the whole process of entering the chatroom involves multiple opening progressions – a situation that does not occur in most FTF or telephone conversations. They furthermore state that the problem of IRC openings on public channels appears to be connected with the fact that ITC´s automated component (AJS) does not provide the same useful structural resource for openings that the ring does for telephone interactions.

With a view of the above mentioned facts, it can be claimed that the specificity of IRC is that it enables people to continue opening conversations as many times as they wish. It can be a greeting, a question or addition of a second pair part – which means joining the conversation flow directly. It is not unusual not to gain any response at all to any opening technique. Note the following example from the category of 30+:

Example:

759. Guest_ButterflyBack has joined the conversation. (AJS summons) 767. Guest_ButterflyBack: hi all (pre-conversational opening-greeting) 769. Guest_ButterflyBack: asl pls (question: age, sex, location) 773. Guest_ButterflyBack: 24/F/ Gadsden Alabama (contact advert) 785. Guest_ButterflyBack has left conversation. (closing) (B. 30+) It is obvious that the effort to join the conversation is not limitless. Schegloff´s (1968) terminating rule may be applied in this context. It is not possible to give a firm formulation of such a rule for the IRC context, except to note based on this research that openings are not strung out beyond four to six repetitions. The efficiency of other opening strategies is given later in this chapter.

A raw division of IRC contact advertisements was given above. Now, I will proceed to the more detailed characteristics:

1. Looking for people of the same nationality or from the same town or region; one observation of this phenomenon may be made about frequent repetitions of these

97 summonses. The following example represents more the rule than the exception on IRC:

Example: 7. Bluestar joined the room 11. Bluestar: Hey people 120. Bluestar: 15 f aus here 122. Bluestar: aussie 152. Bluestar: any one wanna chat to a 15 F aus im me 171. Bluestar left the room (A.2 Teens)

Example: 33. clive has joined the conversation 47.clive: whats up ppls????????????????? 55. clive: ni ozzies in here??????????? 108. clive: is there ni ozzie guys here?????? 154. clive: ni guys wanna talk to an aussie chick im me 175. clive: ni ozzie guys wanna chat im?????????????????? 204. clive left the room. (A.2 Teens) Note: in line 175 what seems to be a question is in fact a summons – contact advertisement

2. On-line dating or soul mates; it may be noted on IRC that after entering a chat room, the newcomer‟s first utterance does not necessarily have to be a greeting, take for example this utterance:

137. dog.with.out.eyes: MALE 36 CALIF, WITH 18 TO 25 YEAR OLD SINGLE FEMALES WITH WEBCAM AND OR PHOTOS PM ME PLEASE?

(A.1 Teens) The chat participant called “dog.with.out.eyes” issued a signal and is waiting for someone to contact him. He does not give any further opening or does not contact anyone specific. From our survey, it is clear that these types of summons are never repeated and also do not gain much response; see Table 18.

98 3. Contact advertisement in the form of seeking a sexual relationship; I have noted only eight occurrences of this kind, all in Teen chat rooms, like in the examples below:

242. HotGigolo: HEY GIRLS, IM SHOWING MY HOT DICK ON WEBCAM, WANNA SEE CLICK ON MY NICK NAME AND VIEW WEBCAM?? IM ME…

(A3. Teens)

144. hengjuanhao: ANY guy sex with their sis, IM me… (A2.Teens)

It must be stressed that on IRC any sexual remarks are heavily policed by other people in a room or by the supervisor. In my corpus, statements such as “that sux PORN? ew!” followed them (A1.13), or they were commented on in a more vulgar way like “wow these pathetic faggots who keeps buzzing me for cyber sex are pathetic” (A1. 118), on the other hand sometimes they were not commented on at all.

Table 18 below shows the occurrence rate of contact adverts in the three chatrooms. The idea was to evaluate the efficiency of a contact advertisement in the chat rooms specified. It is clear that a summons in the form of a contact advert does not work as an effective attention-getting device in any age category. A total of 86 per cent (N=57) contact adverts remain without any response.

Table 18: Occurrence rate of contact advertisements and their efficiency

99 CONTACT Teens 30+ 60+ ADVERTS Gained No Gained No Gained No response response response response response response Nationality, 5 30 1 5 0 1 friends Looking 0 12 1 1 0 0 for a partner

Sexual 1 8 1 0 0 0 offers

TOTAL 6 50 3 6 0 1

It follows from Table 18 that the chance of receiving any response in a Teen chat room is only 10 per cent even if the number of occurrences of contact adverts is quite high. The categories of 30+ and 60+, with their minimum frequency of contact adverts (N=10), indicate higher conversational maturity on IRC. This means that conversation does not involve individual “shout outs” but forms a more consistent unit. Table 18 also proves the fact that the Teen chat room is affirmed as a place for potential adult intruders. Hiding behind a nickname without any further identity displays, the risk of encountering such people in a teen chatroom is very high. In this respect, many chatrooms display their safety measures and warn chat participants of the danger of giving someone a real name or e-mail address.

4.2.2.2 Pre-conversational Openings – Greetings

As has been mentioned at the beginning of this chapter, certain utterances are usually ´precursors´ to other utterances or a sequence of utterances. In the case of IRC, I have stated that greetings may be regarded as pre-conversational openings, however there is no doubt that greetings on IRC are rather specific, as Crystal (2001, p. 154) suggests:

“You may greet everyone if you wish, by saying ´Hello everyone` or the like, but few of any of the other members will

100 reply…There is an automatic greeting facility, whereby the system immediately says ´Hi all`, or such like; however, many consider auto-greet to be poor chatgroup etiquette, because it removes the personal element which is a part of the medium. Some IRC help manners pages are quite firm on the point: ´Scripts that automatically greet people are considered rude and not welcomed.`”

It is said that in a real conversation to greet someone is considered polite but to respond to a greeting is an obligation. In the following survey, the main concern is how IRC greetings are utilized in individual age categories, whether greetings are returned, thus forming an adjacency sequence, and finally, whether greetings on IRC lead to a further discussion. This fact would confirm the initial hypothesis where greetings were regarded as precursors to further talk. It is also crucial to stress that according to a „netiquette‟ participants are advised not to greet every person in the channel individually.

For the purpose of a clearer analysis, responses to greetings were divided in two groups: greetings to all chat participants-response to a greeting – hi room, hello everyone; and greetings to one person-response to a greeting – hi dona, good morning Katie. The results of the survey represented by Table 19 below show that the variable of age influences behaviour in a chatroom very strongly.

Table 19: Greetings-returning greetings in the three age categories Teens 30+ 60+

No Gained No Gained No Gained response Response response response response response Greeting 13 0 6 5 0 4 to ALL Greeting 0 3 10 9 18 98 to ONE Total % 80% 20% 53% 47% 15% 85%

In the group of Teens with 16 opening greetings altogether, only 3 gained a response. On the contrary, participants in the category of 60+ are seen to open conversation by the means of a greeting in 120 cases out of which only 18 did not gain a response. As supposed, the number of greetings that did not gain any response (N=16) in the category of 30+ appears on the edge of Teen figures, however with its 49 opening greetings, out of which 33 gained a response, approximates the senior group. This fact

101 might confirm the above mentioned sociolinguistic theory that the middle-age group is influenced by both categories, dealing with the pressure from ´outside`. At the same time it proves my assumption that behaviour of the older generation on IRC approximates the conduct in a face-to-face conversation or even exceeds it, for example in the case of addressing discussed in chapter 4.1.

On a horizontal level, there is a significant difference between „greeting to all‟ and „greeting to one‟. Regardless of the age, more responses were gained when only one participant was addressed and greeted (N=110) than when the whole room was greeted (N=9). This finding is in accordance with the research made by Ahti, J., Lähtevänoja, H. (2004) where they examined 60 openings and finally stated that “a large amount of collectively addressed openings do not gain a response in a chat room, whereas individually addressed openings gain a response each time they are used.”

A specific category of greetings are those that were initiated by AJS, it means that the process of greeting was diverse, a person entered a chatroom and received a greeting by a present chat participant and afterwards responded to this greeting. It turned out that 30 such cases appeared in the group of 60+, unlike in other two age groups, Teens have one case and 30+ eight cases, see Example.

Example 422. skillsaw8 has joined the conversation 425. Friendly : hi...... Skill...... wb 427. DeltaDawn: hi skill 428. TigerLily7167 : gm skillsaw 429. slye_one : skill.. hello 434. skillsaw8: ty 439. skillsaw: hi gang

(C. 60+)

The ultimate concern is in viewing opening greetings as an invitation for further talk. The question is whether IRC greetings lead to further discussion and therefore can be regarded as pre-conversational openings. In a real conversation, greetings are considered either the lowest degree of conversation, that is, topic-initiating or just social identification displays. IRC with its absence of audio-visual means operates in other dimensions. It was observed that greetings on IRC in the majority of cases did not serve as an invitation for further talk but only as a phatic element. By phatic, I follow

102 the terminology of Malinowski (1972) where he defines phatic communion as establishing “bonds of personal union between people” and mainly fulfils a social function. Table 20 demonstrates this situation:

Table 20: Relation between IRC greetings that lead to further discussion and greetings that were followed by no further discussion Greetings Teens 30+ 60+ TOTAL Further 4 13 50 36% discussion No further 11 36 70 64% discussion

From the table it is clear that two-thirds of all opening greetings function only as social identification displays or as a means of phatic communication. It is mainly visible in the category of 60+ where out of the total of 48 newly joined participants only 42 per cent started to chat and lead a further talk, in the other two categories the total is as few as 27 per cent – compare the number of newly joined participants in those two remaining categories, in Teens 159 and in 30+ the total of 102.

The following exchange serves as an example of participants – in bold – who join the chatroom, exchange greetings or do not contribute to a conversation at all and after a few lines leave:

Example:

94. ROSENDOO has joined conversation. 95. USUMBEACH has joined conversation. 96. SweetSady: doing great here in NY with children 97. DeltaDawn: hi sumbeach 98. GuestLarry has joined conversation. 99. DeltaDawn: hi larry 100. USUMBEACH: hi delta 101. retfirechief: h b B W 102. TGUY has joined conversation. 103. shimmering-wine is away. 104. DeltaDawn: hi tguy 105.shimmering-wine: hi larry 106. teacup: hi tguy 107. GuestLarry: delta hi

103 108. GuestLarry: wine hi 109. TGUY: Hi Delta 110. TGUY: HI tea 115. TGUY has left conversation. 139. ROSENDOO has left conversation. 140. USUMBEACH has left conversation. 151. GuestLarry has left conversation. ( C. 60+)

The observation is then that 64% of all IRC greetings from the corpus do not serve as an invitation for further talk but only as a phatic element fulfilling a social function. This result is the key factor for further analyses of Cooperative and Politeness Principles on IRC. It appears that politeness on IRC functions on a different basis than in a real conversation. Therefore, I do not consider no response to a greeting or leaving without saying good-bye on IRC impolite but rather view this conduct as specific of cyberculture.

4.2.2.3 Question-Answer Sequence

The question-answer adjacency pair, in other words Personal Inquiry (McLaughlin, 1984), represents another opening sequence. It is a part of social recognition display together with greeting. In a spoken conversation the “ritual” questions such as “How are you?” or more informal “What‟s up?” are topic-initiating questions or are regarded „greeting substitutes‟ (Sacks, 1995, p. 554). I suppose few occurrences of these phatic questions. The reason lies in the fact that on IRC they are losing their original function of establishing relationship or initiating further conversation; especially, if they are not directed at one person but at all participants.

My assumption is that the lack of personal contact in IRC conversation and anonymity will result in a relatively high degree of straightforwardness reflected in the content and the form of the question. The lack of mutual and background knowledge can lead either to frequent occurrences of direct questions such as “Who are you?, What is your age?”, which would not be imaginable in a real conversation, or to a complete absence of opening questions. In the IRC research of Rintel et al. (2001) one of two types of question exchanges regularly follow: „how-are-you‟ and „where-are-you-from‟. They assume that “the function of „where-are-you-from‟ exchanges may be quite 104 directly relational – possibly as an uncertainty reduction strategy or as such exchange requires truth and detail, they may act to establish the first topic.”

Given the lack of audio-visual cues on IRC, questions that allow users to obtain information about the identity of a potential conversation partner are quite frequent. Greenfield, Subrahmanyam (2003) call them „slot-filler codes‟ and their variant on IRC is a conversation opener a/s/l (age, sex location). It might be noted that „age‟ and „sex‟ are the most universal markers of social roles. Greenfield, Subrahmanyam (2003) view it as a new type of adjacency pair – initiation-response pair – created for this communicative environment except that the two pair parts are rarely adjacent, as is the case of most of the adjacency pairs on IRC. I focused on the following types of opening questions:

1. The first contact questions; participants wish to know more information about age, sex, location (abbreviation asl) of other participants (Greenfield, Subrahmanyam, 2003).

2. Phatic questions; I supposed these initial questions to be “How are you?” and similar types creating an imagination of togetherness and the sense of belonging to a group (Rintel et al., 2001). 3. Establishment questions; more direct questions such as “Who are you?”, “Do I know you?”, “Where are you from?”, “What do you do?” with the help of which the first contact is established and further conversation can develop (Greenfield, Subrahmanyam, 2003).

Table 21 shows that the primary assumption proved correct in terms of few occurrences of phatic questions regardless the age.

Table 21: Types of questions in individual age categories and their occurrence rate Teens 30+ 60+ total total %

First contact questions 36 10 1 47 57% Phatic questions 5 12 9 26 32% Establishment questions 2 5 2 9 11%

Further, establishment questions contrary to our assumption do not play such an important role in establishing relationship on IRC in any age category. Finally, the first

105 contact questions appeared to be the most frequent, however this is thanks to the category of Teens that covers 77 per cent of the total. For comparison, in the group of 60+ the first contact questions are represented by only two per cent (N=1). The reason for such a low number lies in the fact that contact questions are used mainly in contact advertisements. I have already stated above that in the whole corpus of the 60+ category only one contact advertisement occurs.

In telephone openings Schegloff observed that in question-answer sequences, even after a considerable amount of time, the other speaks and gives a kind of an answer. This observation is not so obvious on IRC and depends on the type of address. From Table 22, it can be inferred that the question-answer sequence functions in 92 per cent of the cases where address by nickname is used.

Table 22: Question-answer sequence on IRC

Questions Teens 30+ 60+ Gained No Gained No Gained No answer answer answer answer answer answer Address by 4 0 9 0 9 2 nickname Vague 2 37 4 14 0 1 address

On the other hand, in instances where the address is vague (room, anybody), the answer occurs only in 10 per cent of the cases. The use of vague address (N=58) outnumbers the use of address by nickname (N=24).

The vague address types are usually neutral nouns or pronouns, the third person singular or plural such as anyone, guys, girls, ladies and mainly appear in the above mentioned contact advertisements. From the point of view of occurrences in the three age categories, contact advertisements appeared only in two of them, Teens and 30+. While in 30+ I encountered only 9, in Teens 56 occurrences were noted. In the group of 60+ only one was observed. The following example shows the efficiency of address by nickname.

Example:

677. Lunarwolfs has joined the conversation. 678. Lunarwolfs: hey people whats up (no answer – vague address)

106 684. Lunarwolfs: how is everyone doing today (no answer – vague address) 687. Lunarwolfs: hey Lindsay why the mud face for 689. Lindsay: cause its cute (answer – address by nick) 691. Lunarwolfs: okay 708. Lunarworfs: hey any ladies want to chat with a 24/m/florida who is a firefighter and emt pm me please (no answer – vague address)

(B. 30+)

Participant called Lunarwolfs opened the conversation by AJS, then greeting to all followed by an informal phatic question, his third attempt was the phatic question again. He did not gain any response to those attempts. In the end, the establishment question when he joined the conversation directly was successful because the specific question was directed at one particular person - Lindsay.

At this point, Schegloff´s terminating rule may be mentioned again. In his analysis, the terminating rule concerned summonses. The IRC data show that questions are repeated on IRC as well. The reason can be found in the simultaneous and overlapping character of IRC where it is easy to overlook a message. Chieftain had to repeat his question twice, once in line 168 and then again in line 177 before receiving a reply.

Example 167. FatalisticHomeRun: hey chief 168. Chieftain: how are you today 177. Chieftain: how are you today??? 180. FatalisticHomeRun: i´m having one of those days (B. 30+) The last phenomenon that is worth mentioning is the occurrence of impolite second pair parts on IRC, despite the fact that the frequency of such behaviour is not very high – I found only two examples. The anonymity of a chatroom may explain the occurrence of such exchanges as the following:

Example

511. Blueeys has joined the conversation. 514. Blueeys: hey room whats up 515. Blueeys: how is everyone

107 516. SpecialED: everyone is suck go away 541. Blueeys has left the conversation. (B. 30+)

Example

82. Oscar joined the room. 114. Oscar: 19 m ks pics and mic anyone want to voice or chat???? 122. Crucial Mami: are you a “pimp” oscar 129. Oscar: no im a sweetheart 133. Crucial Mami: or tryna be i should say 143. Oscar: oh whats wrong with talking to new people 144. Crucial Mami: there is nothing wrong 147. Ghengis.khan: new people suck (A3. Teens) Having introduced and discussed as many of the features of opening sequences on IRC as could be found, I turn now to a final outline of opening techniques on IRC.

4.2.3 Opening Sequences – Conclusion

I mentioned earlier that Internet chatting fulfils primarily a social function as a source of fun, relaxation, meeting new people, dating, cybersex and looking for a partner. This role of IRC must necessarily be reflected in opening phrases. My analysis concerning three age groups and their behaviour when opening conversation in a chat room revealed some notable facts that will be summarized mainly with a view to the research question.

First, I observed that Schegloff´s distribution rule and the notion of nonterminality of SA sequences do not function on IRC. It means that opening sequences can stand independently or in a changed order because of a high degree of disrupted adjacency, its incoherent and overlapping character. The data also show high level of initiation attempts in contrast with further discussion. The only rule that can be applied in IRC environment is Schegloff´s terminating rule. Various opening techniques of individual participants are repeatable, however not necessarily successful. The following “monologue” serves as an example:

Example:

108 759. Butterfly has joined the conversation. (AJS summons) 767. Butterfly: hi all (pre-conversational opening-greeting, vague address) 769. Butterfly: asl pls (first contact question, vague address) 773. Butterfly: 24/F/ Gadsden Alabama (contact advertisement) 785. Butterfly has left the conversation. (B. 30+) Further, it has been found that the least efficient type of opening is such opening that does not address a specific person. This refers to vague openings that do not carry any nickname, such as greetings or opening questions. Having said this, I claim that on IRC in order to be successful in the ensuing discussion, a person does not only have to differentiate herself or himself from the others by the use of capital letters, colour – despite the netiquette, or by the use of language, such as to choose the unusual style, but also has to address other participants individually. By adding a nickname, the sender emits a positive signal and shows the interest in a receiver of a message. I concur with an opinion of Quirk et al. (1985) who emphasize that the hearer is addressed by name not for clarity but out of courtesy and friendliness.

Finally, with respect to the research question it was found that age has an influence on the opening techniques on IRC. I investigated frequency and efficiency of opening strategies in individual age groups. Efficient openings are such openings that lead to a response or further conversation. The following results confirm the hypothesis

H1:

With regards to the conducted analysis, in the Teen group, AJS occupies the leading position as regards its occurrence, a total of 58 per cent (N=159). The second most frequent opening of the Teen group is a contact advertisement, namely looking for the people of the same nationality, from the same town or region, a total of 20 per cent. The third most common opening strategy seems to be questions, a total of 16 per cent, mainly the first contact questions (such as a/s/l (age, sex, location), constituting 84 per cent of all question types. The opening strategy with minimum occurrences in the Teen group is a greeting, 5.5 per cent. This fact verifies my above-mentioned hypothesis about different behavioural patterns of teenagers on IRC. It has been claimed that traditional opening strategies can be replaced by other more convenient opening techniques.

The analysis further focused on the most efficient opening strategies for the Teen group. It was revealed that contact advertisements (37.5%) and question-answer

109 sequences (37.5%) are the most efficient opening types. Greetings occupy the medium position with 19 per cent. On the other hand, the least efficient technique was summons (AJS) with only one response, 6 per cent. It may be noted that on the whole the percentage is very low. Table 23 shows the preferences of Teen groups:

Table 23: The efficiency of openings in Teen groups

Contact advertisements 37.5%

Questions 37.5%

Greetings 19%

AJS 6%

In contrast with teenagers, the mid-age group 30+ appears to use greetings the most, a total of 24.5 per cent. The second most common opening strategy are questions, 13.5 per cent, mainly phatic questions such as “How are you?”, constituting 45 per cent of all question types. On the other hand, the least utilized opening technique is observed to be contact advertisements with as few as 4.5 per cent of occurrence.

Regarding efficiency, analysis revealed that greetings are the most efficient of all opening strategies in the group of 30+, 58 per cent gained responses. It is followed by question-answer sequences which gained a mere 23 per cent of responses. AJS follows with 14 per cent. Contact advertisements are the least efficient openings as only 5 per cent gained a response. Table 24 describes the situation:

Table 24: The efficiency of openings in the group of 30+ Greetings 58%

Questions 23%

110 AJS 14%

Contact advertisements 5%

As the collected data indicates, the most used opening strategy in the 60+ group is a greeting with 66 per cent of occurrence. Summonses in a form of AJS hold the second position with 26.5 per cent. Questions, 6.6 per cent and contact advertisements, 0.5 per cent, appear to be the least used forms of opening.

As far as efficiency is concerned, the most efficient opening strategy for the group of 60+ is a greeting, 56 per cent, followed by AJS that gained 18 per cent of all responses. Questions with its 5 per cent occupy the last position together with contact advertisements, 0 per cent. Table 25 indicates the results:

Table 25: The efficiency of openings in the group of 60+

Greetings 56%

AJS 18%

Questions 5%

Contact advertisements 0%

When the whole corpus is taken into account, regardless the age, the following sequence of a standard, ritualized IRC opening sequence can be drawn from the analysis, regarding the frequency:

1. AJS (43%)

2. Greetings (31%)

3. Questions (15%)

4. Contact Advertisements (11%) and efficiency:

1. Greetings (62%)

2. AJS (18%)

3. Questions (16%)

4. Contact Advertisements (4%)

111

The only change I note are the first two positions where despite the fact that the most frequent is AJS, the most efficient are undoubtedly greetings. From a sociolinguistic point of view, the opening sequence may represent a stage in language development on IRC where each successive generation will have gone further than the last. As a consequence younger speakers will be more advanced, and older speakers less advanced with respect to the change. It means that the leading positions of contact advertisements and questions may represent the pattern which will be more advanced and will become a standard IRC opening in the near future, unlike more conservative greetings that approximate spoken interaction.

The above observations might suggest a way in which IRC systems could be designed to enhance more efficient and coherent openings. It would be satisfactory to introduce such a device that would enable the newcomer to examine the whole room in- depth in a user-friendly format and choose people they would like to chat with directly. On the other hand, present participants of the IRC interaction would receive a clear signal that another person entered the room, since the existing indicator - AJS is not sufficient, as it was observed.

112 4.3. Closing Sequences

This section first very briefly outlines the significance of closings for conversations and discusses the conditions on IRC. Further, a closer investigation is made of different types of closings, as well as some specific techniques that differentiate IRC communication from face-to-face interaction. In a real exchange, signals exist indicating that a conversation is ending. For example, Sacks and Schegloff (1974) point out that closings are preceded by possible pre-closings. I mainly draw from Schegloff´s and Sack´s (1973) and McLaughlin´s (1974) theories of closings in spoken conversation.

4.3.1 Theoretical Background

Closings like openings form a significant part of any written or spoken interaction. Its significance is quite aptly expressed by Laver (In McLaughlin, 1984, p.170) who claims that

“In the marginal phases of conversation, where the use of such linguistic routines is most dense, participants conduct their social negotiations about respective status and role partly by means of the choice of formulaic phrases, address-term, and type of phatic communication.”

Many studies have dealt with functions that are fulfilled by the particular behaviour carried out before taking leave from face-to-face conversation (Knapp et al, 1973, in McLaughlin, 1984) and at the same time closings in telephone conversations have proven to be of interest to several investigators (Schegloff and Sacks, 1973). Three functions in closing a conversation can be observed: (1) signalling that there is a movement towards a state of decreased access; (2) expressing appreciation for the encounter and a desire for future contact; (3) summarizing what the encounter has accomplished (McLaughlin, 1984, p. 176).

Among the verbal behaviours that appear to be typical of closing sections in informal spoken interaction were found to be reinforcement (short agreements such as “O.K.” and “Right”), buffing (short sociocentric sequences like “Uh” and “Well”), appreciation (statement of enjoyment of the preceding conversation), summary statements (summarizes what was said) and well-wishing. It is notable that actually

113 saying “Good bye” or “So long” which seem to be typical of closing was, out of the fourteen verbal behaviours examined, among the least likely to occur (McLaughlin, 1984, p. 176).

Several goals exist that this research of IRC closings wishes to pursue. The main goal is to delineate the strategies that participants of the individual chatrooms acquire in order to depart from a chat room conversation. The research stems from the above- mentioned closing features of spoken conversation, since I am not aware of any similar investigation on IRC. Mainly, the choice of formulaic phrases, types of phatic communication and the use of other closing remarks and responses are considered.

The initial assumption is that leaving a chat room is less visible and noticed than joining the conversation. When opening, AJS gives a primary signal to the present chat room participants that a new person has entered. Then, the person has a choice between talking or observing. In contrast, it is relatively easy to leave the chat room without being noticed. A person only clicks on the “leave the conversation” button and an automatic leaving signal (ALS) deletes his/her nickname from the screen. The aforementioned sociological study on teenagers and the Internet chatting reveals specific norms of behaviour that are performed by young chat participants. Fifteen- year-old Elizabeth discloses: “I never even told those guys I was getting off the Internet when I did. So I just kinda like, disappeared.” (Jones, 1998, p. 166). In other words, the anonymity of a chatroom enables its participants to leave without any closing remarks, like in the following example from my corpus:

Example 80. sk8erchick joined the room. 83. sk8erchick: hey room 84. sk8erchick: any one in here come from australia if so then im me 85. Kelliee (desiccatedberry) joined the room. 86. paU2x: no one... 87. paU2x: i guess... 88. Grauwls: yellowish fonts are scary 89. sk8erchick left the room. (A1. Teens) The participant sk8erchick made a very short visit to this chatroom. For a full comprehension of the above exchange, it must be pointed out that sk8erchick wrote in

114 yellow fonts. The first opening attempt – greeting to all – was disregarded by other participants, the second summons in a form of a contact advertisement brought a brief and rather brusque response from paU2x, which could be understood as an insult and made sk8erchick leave the room without any closing sentence.

On the other hand, I observe situations when a person closes by greeting and leaves the IRC conversation and after that participants say good-bye to him or her. The following example indicates such an event:

Example

623. shimmering-wine: see you all later 627. Fire: bye wine, take care 629. shimmering-wine has left the conversation 635. donna: bye wine g/f have a great days was nice to see you again 648. POPS: wine come back sooner (C. 60+) The chat room participant shimmering-wine entered the room among the first (first note in line 57) and was one of the most active and popular chatters. Before shimmering-wine left the room, she wrote the closing sentence but received only one response. After she left, two more responses arrived. According to “netiquette” (Crystal, 2001), it is quite common and not regarded impolite or offensive.

From the point of view of coherence of the text and face-to-face interaction, it is non-standard to observe that on IRC response to saying goodbye comes after the person leaves the room. The leaving individual logically does not wait until everyone says goodbye to him/her. The context of a chatroom gives this paradoxical situation new dimension. Considering the amount of chat participants present in the chat room, a person leaving would have to wait until all participants gave their responses. In reality, it is not technically feasible. This situation could be compared to the act of addressing or the act of greeting when opening conversation. The number of greetings and addresses would flood the computer screen. With a view to the above-mentioned facts, it can be said that such conversation incoherence is unique to IRC and forms a part of the freedom of Internet media.

Bearing in mind the above-mentioned findings, I have divided closings into the following groups for further analysis:

115 1. ALS without any previous contribution – includes people who left the room without any previous attempt to contribute to the conversation,

2. ALS without any closing sentence but with previous conversation – includes people who joined the conversation but left without any closing sentence,

3. Pre-closing sentence or an expression – includes people who joined the conversation and announced their departure from the room by a pre-closing phrase, such as „well‟, „OK‟, and so forth,

4. Closing sentence – includes people who joined the conversation and before leaving the room made a closing remark, such as „bye everyone‟, or used a continuity statement, such as „bbl‟ standing for „be back later‟.

4.3.2 Analysis

The main concern of this section is to carry out quantitative and qualitative analysis and examine what type of closing strategies the age groups specified select and to what extent. From the initial observation, it is obvious that closings appear with less frequency than openings regardless of the age group.

Table 26 below shows the occurrence rate of closing sentences used in the three age groups. It is worth mentioning that most of chat participants, a total of 75 per cent, never contributed to the conversation; they only joined and then left the chat room. Teenage groups together with 30+ represent the majority of this total, 92 per cent. On the other hand, 60+ denotes quite a balanced relationship among the first two closing techniques (N= 19, N=20) and the last (N=15).

116 Table 26: The occurrence of closing sentences in the individual age groups

Teens 30+ 60+ %

ALS without any 127 81 19 75% previous contribution ALS without any 20 13 20 17% closing sentence but with previous conversation Pre-closing sentence or 0 1 2 1% expression Closing sentence 2 6 15 7%

In accordance with this finding, the question arises as to what purpose synchronous IRC serves. It seems that the anonymous nature of the media, the lack of commitment and an easy entry and departure from chat rooms enables people to „stand off‟, observe and then choose whether to join the conversation or leave. This phenomenon resembles a big social gathering going on at the same time in different rooms that are crowded by many people. Schegloff (2000, p. 3) describes such situations as „ the multiple conversations simultaneously in progress at parties, yielding the hubbub known as the “cocktail party phenomenon”. Newcomers have a possibility to look through the glass door, listen and then enter or leave as they wish repeating it as many times as they wish. The following example from the corpus shows such behaviour with four entries (AJS) and departures (ALS) with no other contribution from pavinjohn11.

Example

121. pavinjohn11 has joined the conversation. 135. pavinjohn11 left the conversation. 147. pavinjohn11 has joined the conversation 166. pavinjohn11 left the conversation. 479. pavinjohn11 has joined the conversation. 488. pavinjohn11 left the conversation. 543. pavinjohn11 has joined the conversation. 547. pavinjohn11 left the conversation. (B. 30+)

117 The second largest group of participants, a total of 17 per cent, does not make any closing statements despite their active participation in a chatroom. This fact is not unexpected given the overall incoherence of the media, the absence of audio and visual cues, its overlapping character, and the scrolling effect of the screen.

The following exchange of participants called Grauwls and paU2x can serve as an example of the ASL without any closing but with previous interaction. To maintain the authentic character of the Internet media, those participants that were directly or indirectly involved in the Grauwls and paU2x exchange remained in the discourse:

Example

3. Grauwls looks for divine inspiration 7. one_sexyguy4u joined the room. 8. one_sexyguy4u: porn movie 4 girls i'm me now 10. burretsw: will someone please im me 11. Grauwls: duh.. 12. burretsw: the computer is new and not sure if im works 13. paU2x: that sux! PORN? ew! 17. aussiebob181 has joined the room. 18. Grauwls: philippines.. according to the map, is made of lots of islands.. 19. one_sexyguy4u left the room. 20. paU2x: i love making ppl say "LOL" 21. Grauwls: which island are you on? 22. burretsw: someone im me and just write hello 23. hard36330 joined the room. 24. sharden87: lol 25. i_am_nakie000_000 left the room. 26. paU2x: im not in an islan! im in the water! 27. Grauwls: nevermind.. 28. burretsw joined the room. 30. aussiebob181 left the room. 31. Grauwls: um.. nice place to live in 32. paU2x: yeah! 34. Grauwls: lol 38. burretsw joined the room.

118 41. burretsw: someone please tell me if u seemy message 45. paU2x: i dont see any... 46. burretsw: computer is new and not sure if working 51. paU2x: my computer sux! 52. Grauwls: no we are not seeing any messeges.. 53. paU2x: yes? 54. sharden87: I can see ya talkin burretsw 55. burretsw: please im 56. Grauwls: we can? wow 61. burretsw: finally it works 63. paU2x: yehey! 66. paU2x: bubble party!!! 76. burretsw: pleease 77. burretsw: send an im 79. paU2x: im LOST! 80. sk8erchick joined the room. 83. sk8erchick: hey room 84. sk8erchick: any one in here come from australia if so then im me 85. Kelliee (desiccatedberry) joined the room. 86. paU2x: no one... 87. paU2x: i guess... 88. Grauwls: yellowish fonts are scary 89. sk8erchick left the room. 95. paU2x: that scares u... 100. Grauwls: yah.. 101. paU2x: thats... 102. paU2x: INCREDIBLE!!!!! 103. Grauwls: hahaha 104. im4uonly00 joined the room. 116. paU2x: g2g! add me up! if u skte and if ur punk! 118. Grauwls: wow these pathetic faggots who keeps buzzing me for cyber sex are pathetic.. 119. Grauwls: punk? 120. Grauwls: eww

119 124. Grauwls: punk looks shitty 125. Grauwls: whatever 127. Grauwls left the room. 129. paU2x left the room. (A1. Teens) The conversational thread between Grauwls and paU2x is going through various stages and topics. It was printed out just at the stage where they discuss their places of residence. Then another participant interrupts the flow and the topic is re-directed to help him settle a problem with the instant messages system, after that they start chatting about yellow print of sk8erchick (original colours of individual contributions were not maintained in the corpus) and seem to have a good time talking. The conversation terminated after paU2x issued a closing remark in line 116 („g2g‟ stands for „got to go‟) and an offer carrying elements of punk in it. Grauwls´ reaction was to reject that style and leave without saying goodbye or closing. Two lines later, paU2x left without any closing as well.

Table 26 further indicates that closing sentences represent only 7 per cent of all closing strategies, out of which 70 per cent could be found in 60+ chat rooms. Among the most frequent closing sentences on IRC unlike in the above-mentioned face-to-face conversation belong continuity statements like „brb‟ (be right back), „bbl‟ (be back later) or „bb‟ (be back) representing almost 50 per cent of all closing phrases. Their popularity might stem from the continual character of IRC where people come and leave as they please several times a day, as it was shown above in case of pavinjohn11. To „be back later‟ is understood as a marker of the actual absence in a chat room with a possibility of returning.

The other half of closing sentences ranges from a rather extraordinary phrase „I´m going to leave you bitches now‟ – B.30+ and a visual display „ „ – A4. Teens to ordinary „bye‟ – C. 60+, „cya‟ (see you all) – B. 30+, „tc‟ (take care) – C. 60+ to more formal „you all have a good day, God Bless and Keep the faith‟ – C. 60+ or more emotional „hugs‟ – C. 60+. From the perspective of individual age groups, it can be observed that the younger the participants the less formal and standard closings they use.

Pre-closings on IRC form the least frequent part of the closing sequence, a total of one per cent. The example below shows clearly that pre-closings coincide with the

120 category of reinforcement and buffing found in a spoken conversation, such as „well‟ and „ok‟. Knapp et al. (In McLaughlin, 1984, p. 177) also found that legitimizers varied according to whether they invoked an internal („I think we have finished all the questions‟) or an external motive for termination („I have got to go to the gym‟). In the case of KingKaosh, DreamWeaver and IrishHearty their pre-closings very closely resemble external legitimizers in spoken conversation.

Examples 777. KingKaosh: well jen and lindsay 778. KingKaosh: brb i gotta get ready (B. 30+)

87. DreamWeaver : well, i have to go get milk and juice will be back later (C. 60+) 709. IrishHearty: well, y'all have a good day ,, God Bless and Keep the faith Hugs around ! (C. 60+)

The above-mentioned analysis of closings on IRC leads now to the description of a standard closing sequence on IRC. Knapp et al. (1973) observed that a typical closing sequence in a spoken conversation in less formal circumstances was as follows: reinforcement, buffing, welfare concern („Take care of yourself‟) or continuation („See you tomorrow‟). IRC conversation shows rather different results. Apart from a very small percentage of closings in the whole corpus as such, with the majority of occurrences in the group of 60+, I can identify at least four types of closing sequences on IRC.

The most common sequence is a mere assurance of coming back or continuity statements (brb, bbl, bb) and after that leaving the room as the following two examples indicate:

Examples

173. SweetSady: BBL 177. SweetSady has left the conversation. (C. 60+)

230. Wave: brb with it 231. Wave is away. (B. 30+)

121 The popularity of continuity statements points to the continuous character of IRC functioning online 24 hours a day and thus providing a stable and easily accessible environment for further meetings on IRC. By stating „bbl‟, people view their communication as continuous and postponed rather than broken by saying goodbye.

The second most typical closing sequence appears to be a two-part sequence consisting of

1. a pre-closing, an explanation why the person has to leave or a continuity statement,

2. welfare concern or the hope to see everyone later. The following three examples indicate such situations:

0Example

237. duracell_plus: bbl ppl 239. duracell_plus: cya 240. duracell_plus left the room. (A2. Teens)

777. KingKaosh: well jen and lindsay 778. KingKaosh: brb I gotta get ready 779. SpecialED: bye jay 780. KingKaosh: I gotta put my face on 791. KingKaosh is away. (B. 30+)

709. IrishHearty: well, y´all have a good day “God Bless and Keep the faith” Hugs around! (C. 60+)

From the above examples and Table 27 below, it can be observed that the closing strategies emotionally gradate from a vague closing „see you all‟ in a Teen chat room to more open „bye everyone...take care‟ and „hugs‟ in the other two chatrooms. It is supposed then that age influences to a large extent behavioural patterns and norms in a chat room. The changes in language use are embedded in the context of age differences.

In contrast with the findings in a spoken conversation mentioned earlier (McLaughlin, 1984), „bye‟ on IRC is, together with a two-part sequence, the most used closing phrase in the corpus. It is also the most frequently used response to closings.

122 Example

683. donna: bye everyone... (C. 60+)

734. Fire: bye all... (C. 60+)

The least marked group of closings with only three occurrences consists of phrases and expressions typical of the cyber context. Placed in another context, they would be regarded as impolite, inappropriate or bizarre. The three examples below show their character:

Example

47. woogywoogywoo : I'm going to leave you bitches now (B: 30+)

501. skillsaw8 : off to see the wizard (C: 60+)

780. BlackWidow: got to reboot brb (C: 60+)

Concerning these creative types of “cyber-closings” the initial assumption was that the generation of Teens would be the most inventive group. As it appears from the analysis, creativity on IRC public chat rooms is not exclusively the domain of teenagers.

I conclude the investigation in closings with the efficiency of IRC closing sequences in the corpus. By efficiency, I mean whether the given closing statement received any response or not. Table 27 illustrates the connection between age, type of responses and its effectiveness. It is worth noting that in total 80 per cent of all closing remarks gained a response, however 67 per cent of these occur in the group of 60+.

123 Table 27: The connection between age, type of responses and its effectiveness

Teens 30+ 60+ Types of Gained No Gained No Gained No closings response response response response response response Continuity 0 1 1 0 6 3 statements Two-part 0 0 2 0 3 0 sequence Saying 1 0 2 1 3 0 good-bye Cyber- 0 0 1 0 2 0 closings Total 1 1 6 1 14 3

The result from Table 27 is significant when analyzing conversation and strategies that individual age groups utilize on IRC. The analysis has confirmed that the older the chat participants, the more important it is to establish a certain kind of relationship inside a chat room and reaffirm it with continuity statements, well-wishing or creative cyber closings. If there was no need for senders and receivers to act in this way, there would be no need for good-byes either, and the conversation could end with a vague pre-closing topic, termination section acknowledging with ALS or with no closing at all as in the majority of cases in the Teen group.

4.3.3 Closing Sequences – Conclusion

The above analysis has studied various strategies chat participants use to close conversation. The research has revealed that 75 per cent of chat participants do not utilize any closing at all. The most common “technique” appeared to be a simple click on the “leave the conversation” button regardless of age differences. The reason for such behaviour can be found not only in the incoherence of the IRC environment, but also in the lack of commitment and an atmosphere of distance indicating a lack of familiarity and intimacy, typical of virtual communication. Thus, my initial assumption that the frequency of closings would be much lower than conversation openings has been verified. However, it does not eliminate its significance in the overall structure of IRC.

When Tables 17 and 26 are compared, the similarity in the ratio of AJS and ALS in individual age groups is notable. In Teens it is 57:56, in 30+ it is 35:36 and in 60+ it is 7:8. Moreover, when placed aside, the ratio of AJS 57:35:7 and ALS 56:36:8

124 indicate that the highest fluctuation of newcomers is in Teens. This fact conditions the results of every other discussed conversation technique in this work in the sense that very frequently changing participants cannot develop any stronger relationships, or get involved in a more serious conversation. Consequently, Teens appear very vague and superficial in all four conversational methods discussed in this work.

Concerning the most frequent and efficient technique of closing, continuity statements have been identified. It can be observed that expressions like “be right back” or “be back later” creates an impression of togetherness and the atmosphere of sociability and a kind of security for current conversation partners. Continuity statements imply the continuous character of IRC, though break, at the same time, the maxim of quality in that they paradoxically never come back to the same conversation situation and encounter the same conversation partners.

It was said at the beginning that in a real exchange there are signals indicating a conversation is ending. It is obvious from Tables 26 and 27 that the majority of the conversation on IRC is not firmly embedded in certain rules, behavioural structures and linguistic routines like in spoken conversation. From the point of view of the sociolinguistic variable of age, teenage chatrooms must be differentiated from middle- age groups of 30+ and the older generation 60+ in the sense that the latter two seem to have developed a certain code of conduct accepted and followed by most chat participants which is, as of yet, missing in the former.

125 4.4 Turn-allocation Strategies

This chapter sets the aim of characterizing Internet Relay Chat turn-taking mechanisms with a view to the research questions posed in the introduction.

The analysis draws on models of Sacks et al., (1974) developed for a face-to- face interaction. The questions of interactional dynamics in the specified chatrooms and preferred turn-taking strategies are discussed. Statistical analyses including quantitative, qualitative and chi-square test for independence are made. At the same time, some approaches to analyzing turn-taking in computer-mediated conversations are considered and compared (McKinlay et al., 1994; Lunsford, 1996; Cherny 1999; Herring, 1999; Herring, 2001; Greenfield, P.M., Subrahmanyam, K., 2003; Panyametheekul, Herring, 2003).

4.4.1. Theoretical Background

One of the characteristics of conversation that distinguishes it from other forms of discourse, for example a lecture, is that during the course of interaction the roles of speaker and listener frequently change. Therefore, it can be said that conversation is accomplished through a series of turns. The basic unit for constructing a turn is said to be a single proposition or utterance. An utterance is each stretch of talk that can be interpreted as an independent clause and further including for example, terms of address, elliptical sentences, affirmations, tags and back-channel utterances (Sacks et al. 1974; McLaughlin, 1984). A turn may of course contain multiple utterances. It is further characterized by the fact that neither the length nor the order of turns and their distribution is predetermined (Sacks et al., 1974). Turns usually occur at certain points in conversation that are called transition relevant places (TRPs). The Sacks et al. model (1974), on which the present research draws, describes in more detail how TRPs are exploited by individual speakers.

The model of turn-allocation developed by Sacks, Schegloff and Jefferson (1974) posits three strategies for change of speaker turns in face-to-face conversation. The current speaker may use names, gaze or gestures to select the next speaker (strategy A). Speakers may select themselves (strategy B). In case no one self-selects, then the current speaker may continue speaking (strategy C). This model is

126 characterized as „locally managed, party-administered, interactionally controlled, and sensitive to recipient design‟ (Sacks et al., 1974, p. 696), where strategy A is preferred over B and strategy B over C.

In the case of turn-taking on IRC, specific features appear to govern its system. The lack of verbal cues and text-only communication are one of the reasons for disrupted adjacency, a situation where juxtaposed turns are not related to each other in meaning. Instead, they usually appear in a strict linear order, which results in incomplete or interleaved exchange sequences (Herring, 1999). All IRC participants are, in fact, free to self-select and turns are placed “democratically” in the order received by the system (Lunsford, 1996; Panyametheekul, Herring, 2003). Hence, especially in larger groups, there can be a possibility of competition (McKinlay et al., 1993; Cherny, 1999) for the next turn and in this case participants could show a tendency towards selecting strategies B and C.

However, when the variable of age is taken into account, it does not necessarily have to be so. From the above analysis of addressing, opening and closing techniques in individual chat rooms, it can be assumed that turn-taking in the chat room 60+ will be more based on strategy A: selecting the next speaker. This assumption draws, for instance, on the quantitative analysis of relation between age and percentage of occurrences of addressing on IRC (see Graph 4) that revealed vast differences in addressing strategies in the specified age groups. It has been observed that the group 60+ used address forms four times more than the group Teens. Accordingly, the analyzed opening sequences suggest similar results – greetings are the most utilized and efficient in groups 30+ and 60+. Hence, the affiliation of address forms and the use of greetings are closely connected with selecting the next speaker.

4.4.1.1 Turn-taking in Face-to-Face Conversation

Conversation analysts hold that face-to-face talk is „organized by use of machinery deployed in and adapted to local contingencies of interaction across an immense variety of social settings and participants‟ (Boden, Zimmerman, 1991, p.8). This implies that the truth value of any statement is indexical; it is assigned dynamically with respect to the context in which it occurs. Further, the indexical truth of a statement is constructed interactively through the ongoing creation of the situation

127 by its participants. Structure, then, is accomplished in and through the „moment-to- moment turn-taking procedures of everyday talk‟ (Zimmerman, Boden, 1991, p.17). Turn-taking, is then, essential to the creation of any interpersonal exchange.

Recipient design is „the most general principle which particularizes conversational interaction‟ (Sacks et al., 1974, p. 727). Through recipient design speakers create their turn with recipients in mind and all participants constantly track the trajectory of the talk to hear “their” turn. The organization of taking turns to talk is essential in any conversation, as well as in any other speech-exchange systems. Sacks et al. (1974, p. 700) conclude that „it appears to have an appropriate sort of general abstractness and local particularization potential‟, meaning that it still captures the most important properties of conversation without requiring any particular context. The following succinct list of essential features of turn-taking is based on Sacks et al. (1974, pp. 700-701) observations:

1. one speaker speaks at a time 2. number and order of speakers vary freely 3. turn size varies 4. turns are not allocated in advance but also vary 5. turn transition is frequent and quick 6. there are few gaps and few overlaps in turn transition

Zimmerman and Boden (1991) continue that in multi-party situations, conversation structure demonstrates a tendency towards a „breakdown‟ into two-party interaction. Sacks et al. (1974, p. 712) also claim that the model favours „smaller numbers of participants‟. A study on turn-taking in computer-mediated communication concerning larger groups conducted by McKinlay et al. (1993, p. 156) also revealed that subjects in the six member groups tended to subdivide into smaller, discussion groups whose membership changed of their own volition, see Chapter 2.2.

Further, to be coherent, conversations must follow implicit rules. One such rule is the notion of sequential relevance, implying that adjacent turns should relate in some way to what has gone before (Schegloff, Sacks, 1973). Consequently, turn-taking behaviour relies on expectancies generated by the semantic content of the preceding turn. Participants are able to follow a conversation flow and structure of turns with the help of adjacency pairs that organize and match utterances: greeting-greeting, question-

128 answer, invitation-acceptance/decline and so forth. According to Schegloff and Sacks (1973), the basic rule of adjacency pair operation is that when a speaker produces a recognizable first pair part, that speaker should stop talking and the conversational partner should produce a recognizable second pair part. Adjacency pairs thus provide for turn-taking mechanisms, and also prescribe the type of talking that the next speaker can do. In spoken conversation, if a speaker fails to provide the proper second pair part, it is usually noticed and commented on (Richards, Schmidt, 1983).

Another implicit rule related to the turn-taking system proposed by Sacks et al. (1974) is described in terms of two components: the turn-constructional component and turn-allocation component.

According to Sacks et al., the turn-constructional component includes a range of unit-types with which a speaker may create a turn, such as sentential, clausal, phrasal, and lexical. Phonology, intonation and so forth are not regarded as parallel unit-types.

The turn-allocation component in Sacks et al. view is provided in the following rules governing turn allocation „of a next turn to one party and coordinating transfer so as to minimize gap and overlap.‟ (Sacks et al., 1974, p. 704):

1. For any turn, at the initial transition-relevance place of an initial turn-constructional unit (TCU):

a. If the turn-so-far is so constructed as to involve the use of a „current speaker selects next‟ technique, then the party so selected has the right and is obliged to take the next turn to speak; no others have such a right or obligation, and transfer occurs at that place.

b. If the turn-so-far is so constructed as not to involve the use of a „current speaker selects next‟ technique, then self-selection for next speakership may, but need not, be instituted; first starter acquires rights to a turn, and transfer occurs at that place.

c. If the turn-so-far is so constructed as not to involve the use of a „current speaker selects next‟ technique, then the current speaker may, but need not, continue, unless another self-selects.

2. If, at the initial transition-relevance place of an initial turn-constructional unit (TCU), neither 1a nor 1b has operated, and, following the provisions of 1c, the current speaker has continued, then the rule-set a-c re-applies at the next transition-relevance place, and recursively at each next transition-relevance place, until transfer is effected.

129 Sacks et al. continue with more thorough observation of the next-speaker selection techniques which include the following:

1. the affiliation of address term to a first-pair part or some other device for achieving addressing, such as eye gaze.

2. next-speaker selection without addressing or any such technique, but with selection of only a particular speaker through questions directed to a particular speaker or through a relevant issue comprehensible to a particular or prior speaker.

3. the affiliation of the tag question as a generally available „exit technique‟ for a turn. The effectiveness of tag questions lies in the fact that they invoke rule 1a, making the start of a particular next speaker‟s turn relevant on their completion.

4. „starting first‟ technique, so called appositional beginnings, for example well, but, so, and. However, Sacks et al. do not regard these beginnings as turns but only „turn-entry devices with important turn-organizational uses‟ (Sacks et al., 1974, p. 720). However, speakers use these as pre-starts, the way that tag questions serve as post- completers, to get the floor.

A detailed account of the above speaker-selects-next-speaker techniques contrasts with no further description of the other two techniques. Therefore, for my further analysis, I use strategies adapted from Panyametheekul and Herring (2003) that are described below.

I should not neglect to point out one more feature of conversational cooperation, which is stressed by Cherny (1999): back channels. They include nods, body movements, interjections like “oh”, “aha”, “right” and serve to determine whether speaker intentions have been understood.

I have already mentioned that in order to converse smoothly, conversation participants must further coordinate transfer so as to minimize gap and overlap between adjacent turns (Sacks et al., 1974; Schegloff, 2000). Turns occur usually at certain well- defined points in conversation; such points are called „transition relevant places‟ (TRPs). In this context, the notion of predictability may be mentioned. Predictability has to do with what Sacks (1995, p.43) has called the “adjacency relationship” and Mey (2001, p. 140) described as „one‟s ability to foresee what‟s going to happen around the next bend in the conversational path, the next turn.‟ Under standard circumstances, it is indicated by a variety of prosodic and visual cues. These may include clearly

130 predictable turn-signals such as intonation patterns before a full stop, pausing, eye- contact and signalling gestures of hands or head. In a larger context, conversational closers, for example „OK?‟, „OK‟ serve as „pre-closings‟ (Schegloff and Sacks, 1973) or conversational starters such as „Excuse me‟ exhibit a similar predictable pattern (Mey, 2001, p. 141).

In the following chapter, some problems encountered by Sacks et al. (1974) are discussed and evaluated with a view to IRC.

4.4.1.2 Turn-allocation on IRC

While the previous section tackled the issue of turn-taking in spoken interaction, I would like to turn to the problem of turn-taking on IRC and how it is viewed by different researchers. Nevertheless, before examining turn-taking on IRC in more detail, I would like to specify what is further meant by a message, an utterance and a turn, as their explanation varies in the literature and has a different connotation on IRC.

I have already mentioned above that the basic unit for constructing a turn is a single utterance. I regard an utterance as a linguistic unit that might consist of several messages. A message, then, is a line of text transmitted electronically. A message may also be a single linguistic unit, containing the complete information. The turn in the IRC sense refers to all the information that senders intended to send as a whole unit, but were not able to because their flow was interrupted by received messages.

Herring (2001) claims that turn-taking rules and assumptions in spoken conversation described above do not hold on IRC. Among the properties of IRC that are often referred to as obstacles to interaction management are the following two: [1] the lack of simultaneous feedback caused by the absence of audio-visual cues and the non- existence of a message overlap; [2] disrupted adjacency of turns caused by the fact that messages posted on the screen appear in the order they were sent and are rarely adjacent.

Concerning the lack of simultaneous feedback, it is a consequence of the absence of non-verbal communication with conversational partners being spatially distant. In spoken conversation, as research has indicated (McLaughlin, 1984), simultaneous feedback plays an important role in timing turn-taking efficiently and maintaining conversational coherence (Werry, 1996, Herring, 1999; Greenfield,

131 Subrahmanyam, 2003). The audio-visual constraints refer to the situation on IRC when a person cannot determine who her or his conversational partner is and whether a particular utterance is directed to that person or not.

The second assumption stemming from a spoken interaction is that adjacent turns are related to each other (Sacks et al., 1974). This supposition, mentioned above, which provides basic interactional coherence in face-to-face settings, is repeatedly not fulfilled in chat rooms. This situation is caused by the fact that messages appear on the screen in linear order as accepted by the system and thus two logically adjacent messages can be separated by other unrelated turns (Werry, 1996; Greenfield and Subrahmanyam, 2003). Multiple conversational threads are interlaced with each other. Sometimes, it occurs, for example that the answer to the question appears before the question itself (Herring, 1999) or as we could see in the previous chapter on closing techniques, a response to saying good-bye appears after the person already left the room. The following sample of IRC followed by Figure 2 adopted from Herring (1999) demonstrates the disrupted nature of adjacent turns:

4. ashna: hello? 5. dave-g it was funny 6. how are u jatt 7. ssa all 8. kally you da woman! 9. ashna:do we know each other?. I`m ok how are you

Herring (1999) presents the anaphoric perspective of chat and patterns of interaction schematically in Figure 2. The messages lower in the diagram are responding “backwards” to previous messages.

In the example below, participants face the problem of how to keep track of who is talking to whom in synchronous chat. Every pair of logically related turns is disrupted by a message from another exchange.

132

Figure 2. Schematic representation of turn-taking in an IRC sample (adapted from Herring, 1999)

In a further example from my corpus two threads of conversation are simultaneously employed without any signs of misunderstanding or hesitation. Participant A keeps the track with B and participant C with D. It is worth noting that possible confusion and ambiguity is avoided by the use of address forms, topic development and phatic questions. There is no doubt which message belongs to which participant despite a disrupted adjacency. As regards turn-taking, the thread AB is as follows: ABAABAB, the second thread is: CDCDCDC, which very closely resembles a coordinated spoken interaction.

Example

200. FatalisticHomeRun : and i HATE these freakin people calling my office this morning A1 201. CHAZ4u2c2 : hi lidsey C1 202. LINDSAY : hey chaz D1 203. Chieftain : sounds like fun. I hate f B1 204. Chieftain : i hate phone calls too B2 205. Teacher_guy5 has joined the conversation. 206. FatalisticHomeRun : people! A2 207. CHAZ4u2c2 : how r u today lidsay? C2 208. FatalisticHomeRun : this guy just ticked me off A3 209. LINDSAY : Lindsay D2 210. LINDSAY : and im fine thanks and you? D3 211. Chieftain : how so B3 212. Guest_coxjames has left the conversation. 133 213. CHAZ4u2c2 : doing great. just a day off from work to have some work done at my house C3 214. LINDSAY : cool D4 215. FatalisticHomeRun : i'm over it now.... A4 216. sportster_857 has joined the conversation. 217. FatalisticHomeRun calms herself A5 218. CHAZ4u2c2 : I have contractors here C4 219. sportster_857 is away. 220. Chieftain : good B4 221. CHAZ4u2c2 : working to remodel my attic C5

(B. 30+) To conclude, it must be emphasized that a majority of IRC interactions does not follow such structure. Studies of IRC indicate that chat room users often participate in more than one stream of conversation at a time (Werry, 1996; Herring, 1999; Parrish, 2000; Greenfield, Subrahmanyam, 2003). It can be noted that among the most common strategies of how to avoid misunderstanding and incoherence is the use of nicknames, which means the vocative use of the intended addressee‟s name (Werry, 1996; Lunsford, 1996; Cherny, 1999; Greenfield, Subrahmanyam, 2003). By explicitly naming the intended addressee in each turn, chat participants compensate for the lack of non-verbal cues in the text-only medium (Werry, 1996; Lunsford, 1996). Together with this assumption, Cherny (1999) observes that frequent use of an address term (nickname) serves the function of eye gaze in face-to-face communication.

Other strategies that assist with maintaining coherence on IRC and thus better orientation in turn-taking are the following according to Werry (1996):

1. abbreviation: in order to keep up with the flow of conversation it is often necessary to respond quickly and this means that unless one can type very rapidly, messages must be kept short,

2. paralinguistic and prosodic cues: a complex set of orthographic strategies designed to compensate for the lack of intonation and paralinguistic cues that function to create the effect of voice, gesture and intonation through the use of capitalization, spelling and punctuation,

134 3. actions and gestures: IRC community employs a set of codes where words and visual images symbolize gesture qualities of face-to-face communication.

The analysis of turn allocation techniques on IRC are systematically described by Lunsford (1996) and Panyametheekul and Herring (2003). Lunsford compares the turn allocation model of Sacks et al. (1974) with turn-taking organization on IRC and concludes that turn-allocation on IRC is vitally different from that of spoken discourse. Lunsford (1996) found out that in the chat room, everyone has an equal opportunity to forward a message at any given time – the so called free-for-all condition (McKinlay et al., 1993). Based on this condition, she appoints three different turn allocation techniques for which I give examples from my corpus to illustrate the common appearance of such techniques on IRC. These techniques are in accordance with the strategies of Sacks et al. (1974):

1. speaker addresses individual participants by their screen name (nickname) – Strategy A in Sacks et al.

Example: 161. Northofborder107: What is your age SHIMMERING?? → 162. shimmering-wine: thanks for asking gf → 163. shimmering-wine: my age is 61 (C. 60+)

2. speaker addresses the whole group within a given room, the implication is that all present should respond – Strategy A in Sacks et al.

Example:

206. da_bling_queen: anyone in their school band hit 3 and say ur instrument → 219. slipknot_blast: im not in a school but im in a band i play drums (A1. Teens) 3. speaker elicits reactions from anyone who cares to respond, often by making a provocative statement – Strategy B in Sacks et al.

Example: 504. KingKaosh: Could we have a moment for children...who got raped and murdered or trapped in the system who never knew they father never

135 learned to dream...but was guided by drugdealaz killaz an crackpheins , for single mothers who are forced to play mom and dad → 508. SpecialED: nice depressing song jay → 517. LINDSAY: yeah makes me wanna go sleep or kill myself or something (B. 30+) Lunsford (1996) also notes that a chat message is mostly equivalent to a turn, except the messages that are too long to constitute a single message. Similarly, Herring (2001) points out that synchronous messages may contain less than a turn, when a sender has more to say than fits in a single message and continues his or her turn in an immediately following message. However, another participant might “jump in” and interrupt that flow. She continues that in order to retain the floor through an extended turn, some users invented „floor-holding‟ conventions, for example appending a special character to indicate that the turn is not yet completed.

In addition to the above strategies, Cherny´s (1999) emphasis of the use of the third person present tense describing actions to stimulate other participants to take the conversation floor is significant. This phenomenon called in different literature either „nonverbal action displays‟ (Hudson, 1996, p. 465) or „symbolic enactment of physical actions‟ (Werry, 1996, pp. 59-60) or „conventional actions‟ (Cherny, 1999, p.202) is plausible only in text-based Internet media. It includes such examples as „X raises her hand‟ (Cherny, 1999) and a few instances selected from my corpus: 545. „FatalisticHomeRun powders her nose‟ (B. 30+) or 751. „Lindsay screams‟ (B. 30+) see Chapter 2.2. As Herring (2001, p. 623) concludes such „present tense actions are performative in nature and they count as acts solely by virtue of having been typed‟. The use and the effect of these non-verbal action displays in the above mentioned turn- taking strategies related to the age are discussed in the conclusion.

In contrast to the findings of Lunsford (1996) and Herring (1999) regarding the strategies of turn allocation, Panyametheekul and Herring (2003) observe that „it is not apparent that turn allocation behaviour in chat rooms is fundamentally different from that of face-to-face speakers in groups‟. They place an emphasis on the fact that addressing is a form of “current-speaker-selects-next”, the aforementioned strategy A, and Lunsford´s claim that any participant can “self-select” at any time is comparable to both Sacks et al.´s strategies B and C. Moreover, Panyametheekul and Herring (2003) stress that „the technical ability to take a turn must be distinguished from the social

136 appropriateness of doing so, both in face-to-face and in CMC [computer-mediated communication]. Social appropriateness is determined in part by speaker identities and roles.‟

It is obvious then, that even in such a specific audio-visual cue-lacking modality as IRC, the existence of similar strategies and rules as those found in spoken conversation would be present. That is what will be of major concern in the following section, discussing the relevance of age and synchronous IRC in the practises of turn- taking drawing on the model of Sacks et al. (1974) and the recent above-mentioned approaches to the issue.

4.4.2 Analysis

In the following section, methodology adopted to analyze turn-taking strategies on IRC is described. Further, quantitative and qualitative analysis is carried out with the help of the statistical analysis of chi-square test for independence to determine the relationship between turn-allocation strategies and age, namely the strength of dependence expressed by Cramer´s V index. In conclusion, I put forward several questions that arise with respect to a distinctive IRC character and previous research in this field.

4.4.2.1 Turn-taking Strategies in the Three Age Groups

Herring´s (1999) schematic representation of coherence in turn-taking was used for identifying turns and turn-responses in my corpus. For the analysis only those messages that do not include AJS (Automated Joining Signal) and ALS (Automated Leaving Signal) are utilized. AJS and ALS are regarded as non-turns as they do not carry any specific claim for the turn. On the other hand, discourse markers, interjections and any kind of emoticons such as „nods‟, giggles‟, „grins‟ „smileys‟ and „?‟ (questioning look) are treated as separate utterances.

Due to this fact, the present analysis is thus based on a total of 509 messages in groups A: Teens; 583 messages in group B: 30+; and 703 messages in group C: 60+. Teens are further divided into sub-groups to verify whether turn-taking strategies are influenced by age ranging from 13 to 19 years old.

137 By looking at the quantity of each sample, I am aware of the fact that the individual samples are not represented in a balanced way. I am convinced however, that the abundant examples of studied and analyzed turn-allocation strategies provided here are representative of authentic IRC language.

Further, three turn-allocation strategies adapted from the Sacks et al. (1974) model and Panyametheekul and Herring´s (2003) classification were used to position each turn in the structure of conversation. In addition, I have included my own factors for division and connected them with the ones of Sacks et al. (1974) and Panyametheekul and Herring (2003). This method, I believe, enriches the present findings and can facilitate and encourage other similar research.

The detailed division of each strategy is as follows:

Strategy A. The current speaker selects the next speaker by:

1. Affiliation of an address term, which means a nickname or a collective noun, term of endearment and other techniques described in the chapter on addressing; this technique is described by Sacks et al. (1974) as a major technique and is further discussed in the works of Lunsford (1996), Werry (1996), Herring (1999) and Panyametheekul and Herring (2003).

Examples:

122. Crucial Mami: are u a “pimp” oscar → 129. Oscar: no im a sweetheart (A3. Teens) 181. lilchickita: hi beutiful → 184. rolorooster: hey (A1. Teens) 2. No addressing but clearly directing the utterance to a particular person or a group; Sacks et al. (1974, p. 717) state that “ one variant of the use of first pair-part to select a next speaker will accomplish a next-speaker selection without addressing or any such technique, but will select only a particular other as next speaker”. This technique is more challenging from the point of view of attention-paying and coherence maintaining on IRC. Non-addressed initial questions, such as “How are you?” are also included in this section together with relevant nonverbal action displays, see examples below.

138 Examples:

780. KingKaosh: i gotta put my face on → 781. LINDSAY: where ya goin → 782. SpecialED: make it pretty 787. KingKaosh: is that possible? → 789. LINDSAY: if i help it is (B.30+) 718. KingKaosh: OMG 719. SpecialED: OMG OMG → 721. SpecialED jumps in place omg´ing → 722. LINDSAY: lmao (B. 30+)

3. Greeting when opening and closing; I have deliberately distinguished affiliation of an address term in the course of interaction and in openings and closings for one reason. I believe that opening and closing greetings on IRC are not really the current-speaker-self-selects-next-speaker technique in the real sense of the word and oscillate between this technique and speaker-self-selects strategies. Moreover, not all opening greetings and closings have an address term affiliated to it. Looking at Tables 20 and 27 above, it can be observed that the frequency of occurrence of opening and closing greetings compared to other techniques belonging to the strategy A is quite limited, especially in Teens and 30+ groups.

Examples:

68. DreamWeaver: hi sweet sady → 78. SweetSady: gm Weaver,Dawn how are you today? (C. 60+)

53. Kaila Jo: hey room (no response) (A2.Teens)

Strategy B. The next speaker self-selects by:

1. Interrupting a conversation in a relevant way; here a speaker self-selects by interrupting and responding to a conversation between other speakers or a group of speakers.

139 In the following example, the conversation is primarily proceeding between rolorooster and lil_chickita_13. The other participants, good_charlotte and screamz_05 attempt to join the conversation, however without any success.

Examples:

196. rolorooster: lil a/s/l 209. lilchickita: who´s? 226. rolorooster: urs chickita 234. lilchickita: 14/f/ca 240. rolorooster: im 14 m cali 241. lilchickita: awesome → 242. good_charlotte: im 13,f,ct 243. rolorooster: what part 245. lilchickita: south cali 246. rolorooster: of cali 251. rolorooster: im in la → 252. screamz_05: whos from cali (A1. Teens)

2. Changing the topic or initiating a new conversation; this strategy mainly includes contact advertisements mentioned above, action desriptions, pre-closing sequences and interrogatives, usually without a question mark.

Examples:

19. complicatedpunkhottie: hey if anyone hot guys wanna chat to aussie gurl 19 msn me at: [email protected] (A3. Teens)

709. IrishHearty: well, y´all have agood day,, God Bless and Keep the faith Hugs around ! → 710. DeltaDawn: bye irish → 713. LilyTiger: bye Irish. tc → 714. Snuffy: bye irish → 715. Fire: bye Irish…take care 716. IrishHearty has left the conversation. (C. 60+)

140 131. Chieftain: anything interesting in here today → 132. Tweakdup1: no (B. 30+)

Strategy C. The current speaker continues speaking:

1. Immediately; this strategy involves participants whose turns appear adjacent for different reasons. One reason is the technical ability of individual computer systems, the other reason is the above-mentioned division of one turn into more utterances as Cherny (1999) and Herring (2001) suggest. Apart from these aforementioned motives, the number of active participants on IRC might also play quite a significant role. I will mention this matter further in the analysis.

Examples:

47. abzzztalexie : I think so too....kids need the break

→ 48. abzzztalexie : But we have a lot of snow days....and they sure want that state funding

(C. 60+)

124. Tweakdup1 : go work → 125. Tweakdup1 : now → 126. Tweakdup1 : now → 127. Tweakdup1 : now → 128. Tweakdup1 : !!!!!!! → 129. Tweakdup1 : this room is deader than hell → 130. Tweakdup1 : this sucks (B. 30+)

2. After a pause; Panyametheekul, S., Herring, S. (2003) state in their research that participants „stop posting for a while‟. Usually on IRC, participants do not continue contributing to further conversation either for technical or personal reasons, meaning users can deliberately ignore one another, or stop chatting suddenly making use of the lack of visual cues. The use of external legitimizers is quite frequent, „brb‟, „bb‟.

The problem of time lag discussed in Herring (1999) and Rintel, E.S., Pittam, J., Mulholland, J. (2003) may come into play as well. Lag is an extreme slow-down of message exchange times caused by latencies in the Internet‟s packet-switched

141 transmission-reception system. It lasts from a few seconds to a stretch of several minutes. In their data, Rintel, E.S., Pittam, J., Mulholland, J. (2003) found that any lag over five seconds resulted in conversations losing coherence.

In the following example LilyTiger was silent for a while and some comments are made on this absence by Snuffy:

Example

732. Fire: bye all....BBL..... 733. Guest_Katie115 : LOL aint this livly 734. Fire is away. 735. SleepyJimmy: hello Tiger 736. Firehas LEFT the conversation. 737. Guest_Katie115 : b fire 738. Guest_Katie115 : wat happen I get in on the tailend of a good fight ?? 739. Snuffy : nope no fight in here 740. Guest_Katie115 : LOL oh dern 741. Snuffy : but looks like everyone took off 742. Guest_Katie115 : LOL yes it dose → 743. LilyTiger : still here. Just keeping quiet → 744. Snuffy : i thought you went to sleep 745. Guest_Katie115 : LOL wisper says still lotz of peeps still here 746. SleepyJimmy is away. → 747. LilyTiger : zzzzzzzzzzzz who me??? → 748. Snuffy : probably in other room 749. Guest_Katie115 : LOL → 750. LilyTiger : occasionally, I just like to watch things, Snuffy. (C. 60+)

Strategy D. Failures, shout-outs:

Last but not least, the category of failures was taken into account representing 4.5 per cent of the total occurrence of turn-allocation strategies (N=1713). The turns that were irrelevant, out of context or had no relation to other turns were regarded significant for the statistical analysis since they represent a newly evolving Internet communication, the aforementioned „free-for-all‟ condition and appear in the context of synchronous IRC. These turns might not be even called a strategy. Nevertheless, I

142 noted several cases that were meant to disrupt conversation, thus violating several rules of netiquette:, such as „repetition is annoying‟, „send maximum of four line message as once‟, spamming‟ or „flaming‟, see Example below.

Example:

14. hyperdink021: sexy devilis cuteeeeeeeeeeeeeeeeeeeeeeeeeeee 41. hyperdink021: sexy devil is hottttttttttttttttttttt 151. hyperdink021: nnnnnnnnnnnnnnnnnnoooooooooooooooooo oooooooooooooooooooooooooooooooooooooooooooooooooooo 169. hyperdink021: nnnnnnnnnnnnnnnnnnnnnnnnoooooooooooo oooooooooooooooooooooooooooooooooooooooooooooooooooo 231. hyperdink021: nnnnnnnnnnnnnnnnnnnnnnnnnnoooooooooo oooooooooooooooooooooooooooooooooooooooooooooooooooo oooooooooooooooooooooooooooooooooooooooooooooooooooo oooooooooooooooooooooooooooooooooooooooooooooooooooo oooooooooooooooooooooooooooooo → 233. butterfinger: hyperdink021, It would the best for all of us if you just killed yourself 43 people like you have done whats best for everyone…. (A2: Teens)

The above example indicates the “active participation” of hyperdink021 in the conversation. His intentional disrupting of conversation flow is finally commented on by butterfinger. Note the cyber metaphor “to kill yourself” meaning to leave the conversation. However, such reactions are scarce.

Before any further analysis is made, it is essential to highlight the fact that individual strategies may in some cases co-occur, that means Strategy A can simultaneously work with either Strategy B or C. Thus, when self-selecting or continuing to take a turn, a participant may select the next speaker as well. This is to stress, then, that one turn can serve multiple turn-allocation strategies. The analysis of turn-taking conducted in this work takes this fact into account.

143 4.4.2.2 Statistical analysis

This section sets the aim of supplying quantitative and qualitative analyses of turn-allocation strategies in the three-level corpus. Moreover, I verify the hypotheses by using the chi-square test for independence to determine whether turn-taking strategies depend on age.

Primarily, Table 28 presents results of the quantitative analysis of turn- allocation strategies in the whole corpus. From a mere look, it is worth noting that Strategy A, current speaker selects the next speaker, appeared to be the most used and dominant strategy, 75.5 per cent. It is then followed by Strategy C, current speaker continues, 11 per cent, and Strategy B, next speaker self-selects with an occurrence of 9 per cent. As expected Strategy D appears to be the least exploited.

Table 28: Percentage of use of turn-allocation strategies in the whole corpus regardless the age Turn-allocation strategies N Frequency in % Strategy A 1294 75.5% Strategy B 159 9% Strategy C 182 11% Strategy D 78 4.5% Total 1713

Table 28 indicates a very apparent tendency towards the methods utilized in spoken conversation. Consequently, the analysis has confirmed that Strategies B and C are not as popular as it would have been expected (Lunsford, 1996). The obtained results correspond to a certain extent with the findings of McKinlay et al. (1993) where the most successful CMC condition was „request-and-grant‟ and a turn was held until explicitly relinquished. This condition is the most similar to the circumstances in face- to-face conversation.

Furthermore, I agree with Panyametheekul and Herring (2003) who present the results of their investigation of turn-allocation strategies in Thai chat rooms. They also come to the conclusion that turn-allocation in these chat rooms are comparable to that in face-to-face conversation. This means that participants prefer to address one another than self-select to speak or continue speaking without waiting for others to contribute.

144 The other two figures are different, however. In their data, Strategy C is the least utilized of the three with only 7.6 per cent.

Nevertheless, my aim is not to hold spoken conversation as a „gold standard‟ (Rintel et al., 2001), but to show typical features and specificities of IRC turn-taking mechanisms as such and verify the strength of dependence on individual strategies and age, see Table 29.

At the beginning I set the aim of measuring and verifying the H0 and H1 regarding Teen sub-groups division. After including the data and carrying out the test with the 0.05 level of significance, I have found that the level of significance in Strategy A is = 0.2, in Strategy B = 0.1 and in Strategy C = 0.01. This means that

H0 is confirmed and it was proved that no dependence of age variable and turn-taking strategy A and B exist. For further analysis the Teen figures will be regarded as a single unit. In Strategy C, however, the H0 must be refuted and H1 is confirmed ( ≤0.05), which signifies that the data of individual sub-groups are regarded as separate units for further analysis.

As Table 29 indicates, the turn-taking strategies are affected by age. The occurrence rate of Strategy A is twice as high in the group 60+ (N=628) as in the group 30+ (N=384) signifying a higher community awareness, cooperation and text coherence. Attention should also be paid to Teens with their second highest occurrence rate of Strategy A2. It shows that unaddressed messages find their way to a particular addressee without disrupting the coherence. Moreover, it seems that young people have a more enhanced orientation in the overlapping message flow than the elder. Overall high numbers in Strategy A prove that IRC participants desire coordinated performance in conversation.

When examining the strength of dependence between age and Strategy A, Cramer´s V index of Strategy A in turn-taking is C´s = 0.40, which indicates medium strength of dependence among the three age groups. The level of significance is 0.00 which refutes H0 and confirms H1, see Table 30.

145 Table 30: Strength of dependence – Turn-taking – Strategy A

Summary Statistics for Contingency Tables

Value

Significance Level 0.00

Cramer´s V 0.40

Twice as low numbers in Strategy B1 (N=58) compared to B2 (N=111) point to the fact that IRC participants did not master the method of interrupting other participants in a relevant way, they rather change the topic or initiate a new conversation regardless of the topic already being discussed. On the other hand, it is one of the characteristic feature of any IRC interaction despite the „netiquette‟ that states: “Listen to a channel first, try not to “double”, i.e. speak at the same time as someone else” (Section 1.3.3).

Table 29: Turn-allocation strategies on IRC in relation to age

Age Strategy A1 A2 A3 Total 30+ 60+ TOTAL Teens A.speaker selects next speaker 1. by addressing 5 5 20 30 34 335 399 a particular person or a group 2. unaddressed 71 80 80 231 314 165 710 but clearly directed to a person or people 3.greeting 6 6 9 21 36 128 185 (opening and closing) SUBTOTAL 282 384 628 1294 B. speaker self- selects 1. interrupting 3 2 6 11 19 28 58 current conversation in a relevant way

146 2. changing the 26 28 17 71 21 19 111 topic or inititating new conversation SUBTOTAL 82 30 47 159 C.current speaker continues speaking 1. immediately 16 18 3 37 79 20 136 2. after a pause 4 2 5 11 27 8 46 SUBTOTAL 20 20 8 48 106 28 182

D.failures 18 8 17 57 15 6 78 (unclear turns)

This situation mainly applies to Teens (N=82) where a majority of such abrupt topic changes or initiations are contact advertisements, 68 per cent. The exception holds for pre-closing sentences and interrogatives. The example below shows an extract from a standard IRC interaction in Teens. It consists of the first contact questions, establishment questions, AJS and ALS, contact advertisements (three repetitions) and some disjointed messages – „shout outs‟. I note the absence of any adjacency pairs that construct a conversation.

Example

196. rolorooster: lil a/s/l (first contact question) 197. screamz_05: well iam scream! 198. jaizco: and u screamzzz 199. bentastic89 left the room. (ALS) 200. [email protected] joined the room. (AJS) 201. mandingo0369x joined the room. (AJS) 202. hot_boy0986 joined the room. (AJS) 203. justin_902102000 joined the room. (AJS) 204. zamboyz777: aussies (contact advertisement) 205. zamboyz777: any aussies (contact advertisement) 206. da_bling_queen: anyone in their school band hit 3 and say ur instrument 207. jaizco: ur name (establishment question) 208. wvfire99 joined the room. (AJS)

147 209. lilchickita: who's? 210. lil_ladybug_14f is away (Auto-Away) (ALS) 211. screamz_05: hatley 212. hot_boy0986: any girls want to cam to cam im me (contact advertisement) 213. zamboyz777: aussies female (contact advertisement)

(A1. Teens) When examining the strength of dependence between age and Strategy B, Cramer´s V index is C´s = 0.43, which indicates medium strength of dependence among the three age groups. Level of significance is 0.00 which refutes H0 and confirms H1, see Table 31.

Table 31: Strength of dependence – Turn-taking – Strategy B

Summary Statistics for Contingency Tables

Value

Significance Level 0.00

Cramer´s V 0.43

Despite the fact that data of Strategy C indicate that it is the second most used method of turn-allocation, they are also the most problematic to tackle. When examining the strength of dependence by Cramer´s V index, we find that C´s = 0.16.

This means a very low strength of dependence. Hence, H0 is confirmed and no dependence of age and Strategy C is observed, Table 32.

Table 32: Strength of dependence – Turn-taking – Strategy C

Summary Statistics for Contingency Tables

Value

Significance Level 0.13

Cramer´s V 0.16

148 However, as it was mentioned above, Teen sub-groups were proved to be dependent. I note this difference when examining the sub-group 18-19 with its lowest occurrence rate (N=8). It seems that this chatroom community is the most mature of the three, as it appeared to be in the forms of address, see Table 7 and the example below. It is worth noting that unlike in the Example (A1. Teens) above, the sentences are longer and more coherent. Most of them also begin with an initial address; participants know each other from a previous interaction, or even from personal contact. If they do not contain an address form, still it is clearly directed to a particular person (Sacks et al., 1974), line 50. The conversation is not that difficult to follow, see bold print. Messages usually equal one utterance with one exception in lines 40 and 42 where an utterance is divided into two messages. The adjacent turns are disrupted by AJS and ALS and other messages, however the coherence of the text is maintained.

Example

32. JapanaMasturbation: Wanna be my new best friend? 33. fuzz_buzz69 left the room. 34. ghengis_khan_the_pillager: bina, what about me? 35. luis_filipe_madeira_aka_figo: we already is 36. did_you_steal_my_bra left the room. 37. i_own_you joined the room. 38. snoopybgle23 joined the room. 39. ghengis_khan_the_pillager: bina has gone behind my back and got a new best friend. im shocked. 40. n1e43o: i wanna be japana's freind 41. ghengis_khan_the_pillager: what a friend-whore. 42. n1e43o: too 43. JapanaMasturbation: Kiwi, I already told you you're not my best friend anymore 44. luis_filipe_madeira_aka_figo: was that lissa the dawgin biartch??? 45. destroy_mee left the room. 46. ivegotthebluez2002: u got a pic complicated hottie?? 47. snoopybgle23: hello ladies 48. n1e43o: who ze futrevu is kiwi 49. luis_filipe_madeira_aka_figo: friend whore

149 50. ghengis_khan_the_pillager: you didn't have to get a new one right in front of me, dammit. 51. i_own_you: Wifey Biiiiiiiiiiiiiiiiiiiiiiina 52. ghengis_khan_the_pillager: tis i, the magnificent kiwi. 53. JapanaMasturbation: Wifey Jennnnnnnnnni (A3.Teens) Besides these specific features in Teens , certain differences can be monitored in other two groups. Group 30+ shows the highest occurrence rates in both strategies C1 (N=79) and C2 (N=27) compared to group 60+. The reason might lie in a low performance of any participants in the first part of a corpus where the number of active participants reached as low as 5 participants up to line 164. Thus, participants were mainly leading monologues, see Example below. The initial situation in the chatroom is commented on by FatalisticHomeRun and Tweakdup1:

Examples

105. FatalisticHomeRun : i don't want to work 106. FatalisticHomeRun : but no one is chatting

129. Tweakdup1 : this room is deader than hell 130. Tweakdup1 : this sucks (B. 30+)

To conclude, in my research, I notice that many initiations receive no response, see Chapter 4.2.2.1 Summons-Answer Sequence. Accordingly, Herring (1999) reports that in her study over a third of all participants (N=117) who posted messages received no response, which led to some of them sending their message more than once. Herring concludes: “Violations of sequential coherence are the rule rather than an exception in CMC” (Herring, 1999). I observe that Strategy C includes such cases as well.

4.4.3 Turn-allocation Strategies – Conclusion

The analysis shows how interactants coordinate conversation in the medium of IRC and whether their conversation techniques depend on age. The Sacks et al. (1974) turn-taking strategies form a part of the theoretical background together with my own

150 views and suggestions, such as more detailed distinction of individual turn-taking techniques.

I find that Strategy A is the most used technique regardless of the age group (75.5%). It has been proved that there is a medium dependence between the dependent variable of age and the independent variable of Strategy A, C´s V index = 0.40.

When closer examination is made, age group 60+ uses „selecting other speakers by address forms‟ the most (N=335), while the other two groups exploit the technique of „no address but clearly directing the message to another person‟ (30+ N=314; Teens N=231), see examples above. These results account for the above analysis of the address forms and opening sequences.

Strategies B and C, despite the previous assumption of Lunsford (1996), do not bear such importance in IRC turn-taking. Strategy C is the second most used strategy, even if the occurrence rate is very low (N=182). Further, it was proved that very weak dependence among age groups exists since C´s V index = 0.16. In contrast there was measured medium dependence among Teen sub-groups. The most notable is the group 18-19 where more forms of address are used, and hence text coherence is maintained.

I find that the immediate number of participants in the chatroom also plays a pivotal role in terms of using the „current speaker continues speaking‟ strategy. It seems that participants continue speaking only when there is a small number of them in the chatroom and there is no-one they could speak to, as in the example above. At the same time, it might be caused by disrupted adjacency when a participant continues speaking and for better and clearer conversation flow divides the turn into shorter messages.

Example

→ 153. tikigaq_harpooner: u dont really have to drill much tho 154. clive: ni guys wanna talk to an aussie chick im me → 155. tikigaq_harpooner: at least a foot

Strategy B is the least used with a mere 9 per cent of occurrence rate (N=159). The twice as high occurrence of „changing topic‟ or „initiating a new conversation‟ (N=111) insinuates incoherent entries to a chatroom and disregard to the conversation currently taking place. Even if it does not particularly violate any of the netiquette rules, a minor offence can be observed in terms of „respecting the group culture‟ or

151 „listening to the channel first before participating‟. Cramer´s V index is C´s = 0.43, which signifies medium dependence.

Apart from the use of turn-taking strategies, I note other significant features of IRC turn-allocation in my corpus. Primarily, no „floor-holding‟ conventions were observed in my corpus; however tendencies to wait until one person finishes what he/she wants to say are apparent in the group of 60+ and slightly in a mid-age group of 30+.

Compared to any spoken interaction, where the tendencies are to break down into two-party, or dyadic talk (Zimmerman, Boden, 1991), synchronous IRC does not usually break into dyadic exchanges. Rather, any conversation tends to maintain at least three members and more (McKinlay, 1994).

I confirm that using forms of address is the best way to avoid, or at least limit, misunderstanding and a lack of feedback. However, I observe that only the group 60+ retains this technique significantly.

The last phenomenon deserving attention is that of non-verbal action displays. I have found six occurrences in the whole corpus, limited to the group 30+. They include expressions like 274. SpecialED kicks the edge of the room; 545. FatalisticHomeRun powders her nose; 751. LINDSAY screams. These expressions, as it was already mentioned above, describe actions occurring in real life, as a kind of narrative commentary and usually appear in simple present tense. Their mere appearance in the corpus implies the necessity to overcome visual constraints and display the features of humanity in the Internet media, further, they might stimulate other participants to take the floor.

152 5 Final Considerations

In this section the motivation, objectives and material analyzed are summarized, the affect age has on conversational techniques on Internet Relay Chat is discussed and finally limitations of this work are presented.

5.1 Motivation, Objectives and Material Researched

Synchronous Internet Relay Chat with its specific blend of characteristics drawn from written and spoken interaction offers a wide range of possibilities for research in the field of conversation analysis.

My motivation for the analysis of the language of Internet Relay Chat related to age lies in the fact that in recent years I have noticed certain idiosyncrasies in the form of expression used in both teen and elderly chatgroups on the Internet. Even if it is obvious that Internet chatting is more connected with the younger generation, I intended to prove that the age range of participants using IRC is broader and plays a vital role in the modern medium of communication regarding conversational methods and practises. Moreover, the interplay of sociolinguistics and conversation analysis has appeared considerably challenging. Further, I am not aware of any study dealing with age as an influencing factor in the communication on the Internet despite the fact that age plays one of the key roles in any sociolinguistic research. The number of IRC studies that exist are usually limited to gender differences or broader descriptions.

As has been pointed out in the Introduction, the purpose of this work is to trace the conversational techniques inducing and facilitating Internet Relay Chat as well as to focus on how these techniques are exploited in the Internet chatting of different age groups. Besides, it attempts to answer the question of whether there are any differences in these techniques among individual age groups. Thus my research question is:

To what extent are specified conversation techniques used in Internet Relay chat affected by the variable of age?

In my research, I have drawn on pragmatic and sociolinguistic sources that have provided me with useful hints regarding conversational analysis of Internet Relay Chat. I focused on addressing, opening sequences, closing sequences and turn-allocation

153 strategies in order to cover the main elements influencing the process of any conversation. Furthermore, I inquired into the existence of markers that would indicate the development of a norm of behaviour or ritual, similar to the behavioural norms and rituals known from face-to-face conversation.

Throughout my work, I focus on confirming or refuting the following two hypotheses that were drawn from the above research question:

The null hypothesis (Ho) claimed that conversational techniques used on IRC are independent of age.

The alternative hypothesis (H1) was that the form of communication through the medium of IRC is dependent on age.

To confirm or refute these hypotheses, I conduct various statistical analyses further described in Chapter 3.1, Research methodology and statistical analysis.

If H1 was confirmed, older people would have a more conservative attitude to the IRC language use in all four specified phenomena of conversation, working with the language in the same way as if they were speaking to someone face-to-face. They would also show more respect, cooperation and politeness. On the other hand, the younger generation would favour an innovative approach, using mainly informal means of address and showing less respect for their peers. Further, they would cooperate less and be less polite to one another.

The intermediate age group – in their mid-thirties – would have a tendency to direct their language use towards the younger generation still keeping some elements employed by the elderly.

The transcripts of three major age groups have been chosen to make up my analysis corpus – a total of 2,382 IRC messages, each group 794 messages. The age groups are as follows – A. Teens 13 to 19 years old; B. over 30 years old (30+) and C. over 60 years old (60+). Concerning the Teen group, I maintained a division into three sub-groups in the chapter Addressing and Turn-Allocation Strategies in order to investigate possible differences and their dependence in the range between 13 and 19 year olds.

All the samples are tagged to the conversational techniques mentioned above. My main aim is to capture the speech situation in the chatrooms as naturally as possible. The criterion for selecting a conversation corpus was to gain an adequate

154 number of messages for my analysis. The choice of such a criterion draws on those studies on IRC that based their research on a number of messages in the corpus rather than a number of words (Kwang-Kyu Ko, 1996; Greenfield and Subrahmanyam, 2003; Panyametheekul and Herring, 2003; Ahti and Lähtevänoja, 2004). Panyametheekul and Herring (2003).

5.2 Methodological Framework

In the present work the objective of the analysis is to evaluate the results obtained from individual age group corpora. Special attention is paid to the quantitative distribution of the given conversational techniques and their comparison at the end of each chapter. The qualitative analysis enables clarification of the interactional relationship in the specified age groups and investigates how communication on IRC functions to instantiate and develop interpersonal relationships.

I also employ the statistical method of chi-square test for independence to verify the mutual dependence of conversational techniques and age. In the sub-groups of Teens my aim is to prove their dependence and thus confirm the alternative hypothesis

H1. I further use Cramer´s V to measure the strength of dependence. This method is used exclusively in the chapters on Address Forms and Turn-allocation Strategies. The quantitative and qualitative analyses are conducted in the chapters on Opening and Closing Sequences. Those tools help me to draw final conclusions about the influence of age and IRC interactive techniques and to confirm or refute the hypotheses and answer the research question.

5.3 Results in the Context of the Research Question

The objective of this work is to answer the research question: To what extent are specified conversation techniques used in Internet Relay chat affected by the variable of age?

I find that the perceived significance of age in the context of the four investigated and analysed conversation techniques on IRC has been proved without exception. The results of the analysis show that the older generation of IRC users approximates their conversational behaviour to a face-to-face interaction and in some

155 cases exceeds it, usually for the sake of better understanding. On the other hand, the young participants do not follow the standard concepts of spoken communication and develop their own techniques. The mid-age group fluctuates between these categories.

To answer the research question, I discuss how age affected each conversational technique analysed in this work. Each dimension is discussed in the order of its appearance in the work.

5.3.1 Address forms

The analysis conducted on forms of address on IRC revealed the following facts. I examined these forms of address: general nouns, nicknames, terms of endearment and pronominal address forms: „you and nickname‟ and „you without any address form‟. Further, I divided chatting into three basic stages: opening conversation, during the course of IRC interaction and closing conversation. This division enabled me to observe the type and frequency of address forms used by the age groups specified and make their comparison.

Based on the quantitative, qualitative and statistical analysis of chi-square test for independence, hypothesis H0 claiming that addressing is independent of age has been refuted. On the other hand, H1 stating that addressing depends on age has been confirmed at the level of significance α = 0.05. Cramer´s V index measuring the strength of dependence is then C´s V=0.37, signifying medium-weak dependence.

The Teen group is regarded as a single unit as there was not any dependence proved among the three sub-groups. The overall occurrence rate of address forms in this group is the lowest of the three age groups (17%). Nicknames, which are viewed as a basic form of address in face-to-face interaction, reached as low as 19 per cent. On the contrary general nouns reached 33 per cent. The reason is seen in different expectations teenagers have when chatting on IRC. Young people‟s aim may not be exclusively to chat with their peers but more significantly to look for fun, fantasy and partners without any commitment and more personal contact (Jones (ed.), 1998). Consequently, less cooperation has been noted.

The age group 30+ on the one hand approximates Teens, in terms of looking for fun and relaxation from work as some examples indicate. Especially in the opening

156 address, the occurrence rate was lower than in Teens. (Compare 13% to 18.5%). On the other hand, I observe and the analysis confirms more developed community awareness, for example in a more frequent use of nicknames (47%).

The age group 60+ appears to be the most conservative of the three groups. The analysis discloses that senior IRC participants mainly utilize the methods of face-to- face interaction: the abundance of nicknames (85%) is significant mainly in the opening sequence. Furthermore, I believe the wealth of nicknames in the text helps elder participants with better orientation in the numerous overlapping messages, to convey their message to a particular addressee. The cooperative conduct and polite approach is a condition to being successful on IRC within this group.

As regards the occurrence rate of address forms in the three stages of IRC, the tendency is ascending slightly from opening address (N=357) to the address in the course of conversation (N=468) and further falling sharply to closing address with the rate of 43 occurrences. This situation is comprehensible with a view to the fact that even in a spoken interaction the act of leaving is connected less with addressing than in other stages of conversation, for example the opening.

I should also not neglect to mention the order of frequency of individual address forms on IRC regardless the age. Nicknames occupy the leading position with 66 per cent, then „you without address‟ with 13 per cent, general nouns and „you + nickname‟ both with 10 per cent and terms of endearment with merely one per cent.

Finally, I would like to point out that address forms are closely linked to the level and quality of communication on IRC. In Teens chatting resembles individual „shout outs‟ without any specific coherence and a very limited number of attempts for real conversation, therefore we cannot expect a high number of address forms. On the contrary, 60+ IRC participants develop a more sophisticated type of conversation that resembles spoken exchanges where nearly every participant entering the chatroom is addressed and greeted. In this conduct, I observe a kind of ritual or a norm of behaviour that every participant entering the chatroom should follow.

To conclude with a view to the research question, age affects the way IRC participants address one another.

157 5.3.2 Opening Sequences

The results of the analysis verify the hypothesis H1 in terms of the dependence of opening sequences on age.

Based on Schegloff´s (1968) analysis of telephone conversation openings, the opening sequences in individual age groups were further divided into:

1. summons-answer sequences (AJS, contact advertisement),

2. pre-conversational openings (greetings),

3. question-answer openings (opening questions).

The ritualized progression found in telephone openings (Schegloff, 1968) may be displayed in similar sequences in IRC openings, as we shall see further.

I analyzed the corpus from two different angles: frequency of the opening strategy used and efficiency of the strategy used. By efficiency I mean how successful, or efficient, the individual opening strategy was, that is whether it received any response or not.

In Teens the leading position is held by AJS (automated joining signal) with the occurrence rate of 58 per cent that indicates a high frequency of potential new-coming participants. The results of my analysis show however, that 74 per cent of this number leaves the chatroom without any attempt at further discussion. Notably, the least utilized opening strategy in the Teen group is a greeting with a mere 5.5 per cent. When efficiency of opening strategies is considered, then contact advertisements and questions, mainly first contact questions, are the most effective opening strategies (37.5%). This result is closely connected with the above-mentioned sociological consideration of why young people enter the chatroom.

In contrast to teenagers, the group 30+ favours greetings as its opening strategy the most, a total of 24.5 per cent. Consequently, from this figure, 67 per cent of greetings gained a response. The least utilized opening technique is a contact advertisement with an occurrence of only 4.5 per cent. Regarding effectiveness, greetings are the most efficient types of opening strategy with 58 per cent of responses gained.

As far as the 60+ group is concerned, the most utilized opening strategy is a greeting, 66 per cent. On the contrary, contact advertisements hold the lowest position

158 with occurrences of only 0.5 per cent. The most effective opening strategy is a greeting, 56 per cent, together with the group 30+, where the approximation to face-to-face interaction can be noted.

As to initiating interaction on IRC, I observed that it is a complex process. Multiple opening strategies were employed in order to gain a response, in most cases without any success. These strategies might bear a resemblance to „repair‟ strategies and concern mainly Teen groups.

Moreover, I found that in order to receive a response, an unconventional form of opening must be used; I am referring to the third person pronominal address „Happy B- day 2 me‟ in A3. Teens: 258. Besides this, successful openings were individually addressed openings. Therefore my results concur with Ahti and Lähtevänoja (2004) who found that a successful opening must be clear and bear signs of individuality, non- conformity.

When the whole corpus is taken into account, then greetings and AJS summons hold the first positions from the point of view of frequency and efficiency. This means that the standard opening strategy consisting of a range: AJS → greeting → question → contact advertisement represents the aforementioned „ritual of access‟ on IRC.

To conclude, contact advertisements hold the last position with respect to frequency (11%) so far. However, I dare say that the discovered variation in opening strategies in the Teen group – the preference of contact advertisements – might later lead to a change in conversational techniques on IRC. This fact is supported by etiquette rules found on various websites where greetings of every person in the room are regarded as a violation. Thus, more conservative greetings that approximate spoken interaction will be replaced by contact advertisement and questions that represent a pattern which will be more advanced and will become a standard IRC opening in the future.

159 5.3.3 Closing Sequences

The results of the analysis conducted can be summarized as follows. Hypothesis

H1 claiming that conversation techniques on IRC – in this case closing sequences – are subject to a dependence of age has been confirmed as I further show.

First, I delineate closing strategies that participants acquired when departing from chatrooms, see below. I partly draw from theories developed by Schegloff and Sacks (1973) and McLaughlin (1984) who studied closings in a spoken interaction.

My initial assumption that closings (N=306) will occur with less frequency than openings (N=633) has been proved. Low numbers of closings may confirm the Schegloff´s „cocktail party phenomenon‟ when participants come and leave as they wish without any serious commitment (Schegloff, 2000). Moreover, it is much easier to click on the „leave the conversation‟ button than to open conversation as we could see in Section 4.2 Opening sequences.

Closing sequences were divided into four groups:

1. ALS (automated leaving signal) without any previous contribution

2. ALS without any closing sentence but with previous conversation

3. Pre-closing sentence or an expression

4. Closing sentence

First, the rate of occurrence of the above closing techniques expressed in percentage and regardless of age, Table 26, disclosed that ALS without any previous contribution and ALS without any closing sentence but with previous conversation are the most frequent methods (92%).

Further, I found that the third most utilized closing is a continuity statement (17%), such as „bbl‟ or „brb‟. Its popularity might stem from the continuous character of IRC providing a stable and accessible environment for further meetings on IRC, despite the fact that participants never come back to the same conversation partners or never come back at all.

The least used closing sequence is a pre-closing sentence such as „well‟ or „ok‟ (1%). However, in connection with a continuity statement and welfare concern, it constitutes a larger group (2%). In this connection I noted a mark of emotional

160 gradation in groups 30+ and 60+. Closings contained words like „hugs‟, „take care‟ „have a good day‟ and other well-wishing phrases. There was no example found in the Teen group.

As regards the closing sequence from the most used to the least, I observe the following: ALS without any previous contribution → ALS with a previous contribution → continuity statements → pre-closing sentence+continuity statements+welfare concern.

However, when age is taken into account, the results show that the group 60+ follows the rules resembling a spoken interaction more than the other two groups. This means that the closing sequence in 60+ is: continuity statements → pre-closing → welfare concern → saying goodbye. Closing sequences in 60+ represent 65 per cent of the total occurrence rate.

Saying goodbye is also a frequent type of closings on IRC, in contrast with the findings in a spoken conversation (McLaughlin, 1984). The reason may lie in the fact that „bye‟ or even „b‟ is very a quick and short word to write. It is obvious that speed of communication and brevity are one of the most characteristic features of the synchronous Internet chatting.

As regards effectiveness of closing sequences, the analysis revealed that in total 80 per cent gained a response, out of which 67 per cent was in the group 60+. Due to the specific character of IRC, I noted that responses to a closing statement usually arrived after the person left the chatroom and not before, as it is expected in a spoken interaction.

To conclude, the results of my analysis suggest that the older the chat participants, the more important and frequent it is, after establishing a certain kind of relationship inside a chat room, to reaffirm it with continuity statements, well-wishing or creative cyber closings, group 30+ and 60+. Provided that there was no need for chat participants to act in this way, there would be no need for good-byes either, and the conversation could end with a vague pre-closing topic, termination section acknowledged with ALS or with no closing at all as in the majority of cases in the Teen group.

On the other hand, I do not agree with Cherny (1999) that when active people of any age leave a room without any goodbye statement, it is regarded odd and against proper etiquette. I examine many examples of this kind in my corpus. I view this

161 conduct as a combination of many factors – technical, spatial, temporal – that is specific of cyberculture. I conclude that principles of etiquette and politeness on IRC are based on rules other than those in face-to-face communication and I repeat that we should not hold spoken conversation as a „gold standard‟.

5.3.4 Turn-allocation Strategies

In my research on turn-allocation strategies, I wished to examine how participants of IRC coordinate their conversation and furthermore whether the choice of a specific strategy depends on their age. Along with the theories related both to spoken interaction (Sacks et al.,1974) and available studies on IRC (Murray, 1989; McKinlay et al., 1994; Lunsford, 1996; Cherny 1999; Herring, 1999; Herring, 2001; Greenfield, P.M., Subrahmanyam, K., 2003; Panyametheekul, Herring, 2003), I have conducted an analysis that has fully substantiated my hypothesis H1.

Let me summarize the most essential findings. The following strategies were analyzed in the environment of IRC:

Strategy A, that is, the current speaker selects the next speaker is divided into three parts: 1. affiliation of an addressed term; 2. no address form but clearly directed to a particular person and 3. greetings.

Strategy B, speaker self-selects, was further divided into two parts: 1. interrupting a conversation in a relevant way; 2. changing the topic or initiating a new conversation.

Strategy C, the current speaker continues speaking was divided into: 1. immediately speaking; 2. speaking after a pause.

Strategy D, failures and intentional disrupting conversation.

My findings based on quantitative and qualitative analyses point out that the most used strategy is Strategy A, 75.5 per cent regardless of age. In other words, this one bears the most resemblance to the circumstances of face-to-face conversation, with a limited number of overlapping contributions. Still, I claim that the transfer of turn- taking skills from a face-to-face setting to the IRC environment is problematic.

Contrary to the findings of Panyametheekul and Herring (2003), Strategy C is the second most successful protocol with an 11 per cent occurrence rate. I believe that

162 the prevailing occurrence of this strategy over Strategy B is connected with a disrupted adjacency and linear ordering of turns determined by the media and technology. The number of participants at every moment of the conversation also plays its role.

Strategy B with its occurrence of 9 per cent is the least used technique. On closer examination, we note that changing the topic or initiating conversation appears twice as much as relevant interruption of the conversation. The reason lies in the frequent use of contact advertisements and new topic initiations mainly in group Teens, 64% of the total.

The second part of my analysis was directed to the substantiation of hypotheses

H0 and H1 in the context of the research question. For a more precise verification, the statistical method chi-square test for independence was utilized. For Strategy A and B Cramer´s V index verifying the strength of dependence between variables showed medium strength of dependence. Therefore, I maintain that the age of participants influences the turn-taking techniques used. Strategy C appeared to be more problematic, in terms of C´s V index, the dependence is very low, C´sV=0.16. The subgroups of Teens showed medium dependence in contrast with the other two strategies where there was not proved any dependence.

I agree with Cherny (1999) that using address forms is the best way to limit misunderstanding on IRC. In this respect the group 60+ showed the most use of address forms (N=335), unlike the other two groups (Teens N=30; 30+ N=34). Furthermore, as a consequence of audio-visual constraints on IRC, participants developed non-verbal action displays which simulate back channels and describe their emotional state. The presence of non-verbal action displays on IRC supports the need for a closer social presence and more feedback.

To conclude, one of the most specific characteristics of IRC is its semi- permanence. This means that previous utterances, or turns, are available for direct review by conversational partners. This feature affects conversational practises in many ways. The prevalence of Strategy A may be the consequence of this effect. IRC, serving as a record of participants‟ input, may also help suppress some problems associated with the increase in group size. Thus, the circumstances described above provide users with a flexibility that is not possible in face-to-face conditions.

163 5.4 Thesis Limitations

I would like to stress the fact that my corpus and the analysis conducted are far from exhaustive. I am aware that a more detailed study must be done on politeness and cooperation, mainly in terms of determining strategies of face-threatening acts, negative and positive politeness.

Furthermore, the analysis of speech acts (Austin, Searle) could reveal some overlooked structures and conditions of IRC. For instance, the aforementioned non- verbal action displays may have characteristics of performatives.

To conclude, there is still no comprehensive knowledge of how interpersonal relationships are developed on IRC – how the relationships of users progress through mediated typography. Therefore, I believe that my findings concerning language of Internet Relay Chat can serve as a justifiable starting point for future investigations.

5.5 Concluding Remarks

The objective of this work was to examine how and to what extent age of participants influenced their conversational techniques on Internet Relay Chat. Four phenomena of conversational analysis were selected: forms of address, opening sequences, closing sequences and turn-allocation strategies. Two hypotheses were set at the beginning and verified throughout the work by the following types of statistical analyses – quantitative, qualitative and chi-square test for independence. Null hypothesis (H0) claimed that conversational techniques used on IRC are independent of age. The alternative hypothesis (H1) was that the form of communication on a medium of IRC is dependent on age.

The results showed that age has a medium-strong affect on the conversational practises of chatroom participants in all four categories. Further, the analysis revealed that senior chat participants show a tendency to introduce aspects of spoken interaction or even exceed it, while Teens establish novel techniques and conversation practices specific to the modern medium of Internet. In addition, the work confirmed or refuted theories found in other studies developed in this relatively new field of research.

164 Besides, correlations between individual conversational techniques were discovered. Hence, while address forms highly influenced the choice of the most successful turn-allocation strategy and the efficiency of opening techniques, there was a notable link between opening sequences and closings in terms of occurrence rate.

Last but not least, the study is expected to be an asset not only for understanding the principles governing synchronous interaction on IRC, but to serve as a valuable source for other works dealing with linguistics and the Internet.

165 Bibliography

Adler, M. K. (1978). Naming and Addressing. A sociolinguistic study. Hamburg: Helmut Buske Verlag.

Ahti, J., Lähtevänoja, H. (2004). Showing Politeness in Opening Sequences in Finnish and Finland-Swedish Chat Conversations. Proceedings of International Conference on Language, Politeness and Gender: The pragmatic roots. University of Helsinki. pp. 41-42.

Ankersmit, F. R. (1989). "Historiography and Postmodernism. History and Theory no.28 No. 2, p.151.

Asimov, I. (1957). The Naked Sun. New York: Bantam Books.

Atkinson, J.M., Heritage, J. (eds). (1984). Structures of Social Actions. Studies in Conversation Analysis. Cambridge: Campridge University Press.

Austin, J. L. (2000). Jak udělat něco se slovy. Praha: Filosofia AV ČR.

Baierova, P. (2003). The Impact of User Characteristics on the Perceived Importance of Web Quality Dimensions. Unpublished thesis. Victoria University of Wellington.

Baron, N. S. (1984). Computer-mediated communication as a force in language change. Visible Language, 32, pp. 35-53.

Baron, N. S. (1998). Letters by phone or speech by other means: the linguistics of email. Language and Communication, 18, pp.133-70.

Bechar-Israeli, H. (1995). From to : Nicknames, Play, and Identity on Internet Relay Chat. Journal of Computer-Mediated Communication 1 (2). Retreived in September 2004 from www.jcmc.indiana.edu/vol1/issue2/bechar.html.

Biber, D. (1986). Spoken and Written Textual Dimensions in English: Resolving Contradictory Findings. Language 62. pp. 384-414.

Boden, D., Zimmerman, D. (eds.), (1991). Talk and Social Structure. Cambridge: Polity Press.

Brown, G., Yule G. (1983). Discourse Analysis. Cambridge: CUP.

Brown, P., Levinson, S. C. (1987). Politeness: Some universals in language use. Cambridge: Cambridge University Press.

Brown, R. Ford, M. (1964). Address in American English. In Hymes, D. (ed.) (1964). Language in Culture and Society. New York: Harper&Row Publishers.

166 Cairncross, F. (1999). Konec vzdálenosti: Jak komunikační revoluce mění naše životy. Praha: Computer Press.

Cherny, L. (1999). Conversation and Community: Chat in a Virtual World. Stanford University: CSLI Publications.

Chomsky, N. (1966). Syntaktické struktury. Praha: Academia.

Coulthard, M. (1985). An Introduction to Discourse Analysis. London: Longman.

Condon, S. L., Cech, C.G. (1996). Discourse management strategies in face-to-face and computer-mediated decision making interactions. Electronic Journal of Communication, 13 (1). Retrieved on 15 April 2004, from www.cios.org/getfile/condon_v6n396.

Crystal, D., Davy, D. (1969). Investigating English Style. London: Longman.

Crystal, D. (2001). Language and the Internet. Cambridge: Cambridge University Press.

Čermák, F. (2001). Jazyk a jazykověda. Univerzita Karlova v Praze: Karolinum.

Černý, J. (1996). Dějiny lingvistiky. Olomouc: Votobia.

December, J. (2003). Characteristics of Oral Culture in Discourse on the Net. Paper presented at the twelfth annual Penn State Conference on Rhetoric and Composition, Retrieved on 15 November 2005 from http://www.december.com/john/papers/pscrc93.txt.

Downes, W. (1998). Language and Society. Cambridge: Cambridge University Press.

Escandell Vidal, V. (1996). Toward a cognitive approach to politeness. Language Sciences, 18, 3-4, pp.629-650.

Escandell Vidal, V. (1998). Politeness: A relevant issue for relevance theory. Revista Alicantina de Estudios Ingleses, 11, pp. 45-57.

Firth, J. R. (1964). On Sociological Linguistics. In Hymes, D. (ed.) (1964). Language in Culture and Society. New York: Harper&Row Publishers.

Goffman, E. (1955). On Face-Work: An Analysis of Ritual Elements in Social Interaction. In Laver, J. and Hutcheson, S. eds. (1972). Communication in Face to Face Interaction. Harmondsworth: Penguin Books.

Goodwin, C., Heritage, J. (1990). Conversation Analysis. Annual Review of Anthropology 19, pp. 283-307.

167 Greenfield, P. M., Subrahmanyam, K. (2003). Online discourse in a teen chatroom: New codes and new modes of coherence in a visual medium. Applied Developmental Psychology 24. pp. 713-738.

Grice, H. P. (1975). Logic and conversation. In Cole, P. and Morgan, J.L., eds. Syntax and semantics 3: Speech Acts, pp. 41-58. New York: Academic.

Gumperz, J. J. (1971). Language in Social Groups. Stanford: Stanford University Press.

Gumperz, J.J., Hymes, D. (eds), (1972). Directions in Sociolinguistics. New York: Holt, Rinehart and Winston, Inc.

Gumperz, J.J. (2001). Interactional Sociolinguistics: A Personal Perspective. In Schiffrin, D., Tannen, D., Hamilton, H.E. eds. (2001). The Handbook of Discourse Analysis. Oxford: Blackwell Publishing, pp. 215-228.

Halliday, M. A. K. (1989). Spoken and Written Language. Oxford: Oxford University Press.

Havránek, B., Weingart, M. (eds), (1932). Spisovná čeština a jazyková kultura. Praha: Melantrich.

Havránek, B. (1932). Úkoly spisovného jazyka a jeho kultura. In Havránek, B., Weingart, M. (eds) (1932). Spisovná čeština a jazyková kultura. Praha: Melantrich.

Herring, S. (1993). Gender and democracy in computer-mediated communication. Electronic Journal of Communication, 3 (2). Retrieved on 3 October 2005, from www.cios.org/getfile/herring_v1n393.

Herring, S. (1996). Computer-Mediated Discourse Analysis. Introduction. Electronic Journal of Communication, 6 (3). Retrieved on 15 April 2005, from www.cios.org/getfile/herring_v3n293.

Herring, S. (1999). Interactional coherence in CMC. Journal of Computer-Mediated Communication 4,4. Retrieved on 19 August 2004, from www.jcmc.indiana.edu/vol4/issue4/herring.html.

Herring, S. (2001). Computer Mediated Discourse. In Schiffrin, D., Tannen, D., Hamilton, H.E. eds. (2001). The Handbook of Discourse Analysis. Oxford: Blackwell Publishing, pp. 612-634.

Herring, S. (2004). Gender and politeness in Computer-Mediated Communication: A decade of new evidence. In Proceedings of the International Conference on Language, Politeness and Gender: The Pragmatic Roots. Helsinki: University of Helsinki Press, pp. 28-29.

Holmes, J. (1992). An Introduction to Sociolinguistics. London: Longman.

Hudson, R. A. (1975). The meaning of questions. Language 1 (1). pp. 1-31.

168 Hudson, R. A. (1978). Sociolinguistics. Cambridge: CUP.

Hymes, D. (ed.) (1964). Language in Culture and Society. New York: Harper&Row Publishers.

Jacobson, D. (1996). Contexts and Cues in Cyberspace: The Pragmatics of Naming in Text-Based Virtual Realities. Journal of Anthropological Research 52, pp. 461-479.

Jakobson, R. (1991). Lingvistická poetika. Bratislava: Tatran.

Jones, S. G. (ed.), (1998). Cybersociety 2.0-revisiting CMC and community. Thousand Oaks: SAGE Publication.

Kaiser Family Foundation. (2001). Generation rx.com: How young people use the Internet for health information. Retrieved on 30 September 2005, from www.kff.org/content/2001/20011211a/eibchartpack.pdf.

Kašičková, Š. (2004). Politeness Principles in Computer-Mediated Communication in the Czech Environment. In Proceedings of the International Conference on Language, Politeness and Gender: The Pragmatic Roots. Helsinki: University of Helsinki Press, pp. 103-106.

Kašičková, Š. (2005). Opening and Closing in Internet Relay Chat. In Discourse and interaction 1. Brno Seminar on Linguistic Studies in English: Proceedings. Brno: Masarykova Univerzita, s.52-62.

Kwang-Kyu Ko. (1996). Structural Characteristics of Computer-mediated Language: a Comparative Analysis of Interchange Discourse. Electronic Journal of Communication 6 (3). Retrieved in October 2005 from www.cios.org/www/ejc/v6n396.htm.

Labov, W. (1972). Sociolinguistic Patterns. Philadelphia: University of Pennsylvania Press.

Lakoff, R. (1973). The logic of politeness; or minding your p´s and q´s. In Papers from the Ninth Regional Meeting. Chicago: Chigaco Linguistic Society, pp. 292-305.

Lakoff, R. (1977). What you can do with words: Politeness, pragmatics and performative, In A.Rogers et al. Proceedings of the Texas Conference on Performatives, Presuppositions and Implicatures. Arlington, VA. Center for Applied Linguistics. pp. 79-106.

Lakoff, R. T. (2004). The politics of nice: From George W. Bush to Martha Stewart. In Proceedings of the International Conference on Language, Politeness and Gender: The Pragmatic Roots. Helsinki: University of Helsinki Press, pp. 30-31.

169 Laver, J. and Hutcheson, S. eds. (1972). Communication in Face to Face Interaction. Harmondsworth: Penguin Books.

Leech, G. (1983). Principles of Pragmatics. London: Longman.

Levinson, S. C. (1983). Pragmatics. Cambridge: Cambridge University Press.

Little, C. B., Gelles, R.J. (1975). The Social Psychological Implications of Form of Address. Sociometry 38 (4), pp. 573-586.

Lunsford, W. (1996). Turn-taking organization in Internet Relay Chat. Unpublished ms, University of Texas at Arlington.

Lyons, J. (1981). Language, Meaning and Context. London: Fontana Paperbacks.

Malinowski, B. (1923). „Phatic Communion‟ – excerpt from „The Problem of meaning in primitive languages‟. In Laver, J. and Hutcheson, S. (eds.) 1972. Communication in Face to Face Interaction. Harmondsworth: Penguin.

Mathesius, V. (1966). Řeč a sloh. Praha: Československý spisovatel.

Mathesius, V. (1975). A Functional Analysis of Present Day English on General Linguistic Basis. Praha: Academia.

Maynard, M. L. (1995). Interpretation and Identification of Gendered Selves: Analyzing Gender-Specific Addressivity in Japanese Advertising Text. Language and Communication, Vol.15, No.2, pp. 149-163.

McLaughlin, M. L. (1984). Conversation: How Talk Is Organized. London: Sage Publications.

McKinlay, A., Procter, R., Mastingt, O., Woodburn, R., Arno, J. (1994). Studies of turn-taking in computer mediated communications. Interacting with Computers 6 (2) pp. 151-171.

Mey, J. L. (2001). Pragmatics: An Introduction. Oxford: Blackwell Publishing.

Milroy, J. (1992). Linguistic Variation and Change. In Downes, W. (1998). Language and Society. Cambridge: Cambridge University Press.

Mohelská, L. (2006). Kapesní počítačový slovník anglicko-český. Brno: Computer Press.

Murray, D. (1989). "Turn-taking in Computer Conversations." In: Working with Language: A Multidisciplinary Consideration of Language Use in Work Contexts. New York: Hrsg. von Coleman Hywel.

Neuage, T. (2004). Conversational analysis of chatroom talk. Unpublished Ph.D. University of South Australia. Retrieved in November 2005 from www.neuage.org/All.htm.

170 Panyametheekul, S., Herring, S. (2003). Gender and Turn-Allocation in a Thai Chat Room. Journal of Computer Mediated Communication 9 (1). Retrieved on 5 October 2005 from http://jcmc.indiana.edu/vol19/issue1/panya_herring.html.

Parrish, R. (2000). Conversation Analysis of Internet Chat Rooms. Retrieved on 19 November 2005 from http://www.polisci.wisc.edu/~rdparrish/Research.html.

Porter, D. (ed.), (1996). Internet Culture. New York: Routledge.

Procter, P. (1996). Cambridge International Dictionary of English. Cambridge: Cambridge University Press.

Quero L. S. (2003). El Lenguaje de los ´Chats´: Aspectos Gramaticales. Granada: Port-Royal Ediciones.

Quirk, R., Geenbaum, S., Leech, G., Startvik, J. (1985). Comprehensive Grammar of the English Language. Longman.

Reid, E. (1991). Electropolis: Communication and Community On Internet Relay Chat. Retrieved on 6 November 2005, from http://www.alululei.com/electropolis.htm.

Richards, J. C., Schmidt, R.W. (1983). Language and Communication. London: Longman.

Richards, J. C., Schmidt, R.W. (1983). Conversational Analysis. In Richards, J.C., Schmidt, R.W. (1983). Language and Communication. London: Longman. pp. 117-154.

Rintel, E. S., Mulholland, J., Pittam, J. (2001). First things first: Internet Relay Chat Openings. Journal of Computer-Mediated Communication 6 (3). Retrieved in November 2005 from www.jcmc.indiana.edu/vol6/issue3/rintel.html.

Rintel, E. S., Pittam, J., Mulholland, J. (2003). Time will tell: Ambiguous non- responses on Internet Relay Chat. Electronic Journal of Communication, 13 (1). Retrieved on 30 August 2005, from www.cios.org/getfile/rintel_v13n1.

Rogers, A., Wall, B., Murphy, J. (eds.) (1977). Proceedings of the Texas Conference on Performatives, Presuppositions and Implicatures, Arlington, VA., Center for Applied Linguistics.

Romaine, S. (1996). El Lenguaje en la sociedad. Barcelona: Ariel.

Sacks, H. (1995). Lectures on Conversation. Volumes I and II. Oxford: Blackwell Publishers.

Sacks, H., Schegloff E. A., Jefferson G. (1974). A simplest systematics for the organization of turn-taking in conversation. Language 50 (4), pp. 696-735.

171 Schegloff, E. (1968). Sequencing in Conversational Openings. American Anthropologist 70, 1075-1095.

Schegloff, E. A. (2000). Overlapping talk and the organitazation of turn-taking for conversation. Language in Society 29, pp. 1-63.

Schegloff, E. A., Sacks, H. (1973). Opening up closings. Journal of the International Association for Semiotic Studies 8, pp. 289-327.

Serrano, M. J. (1996). Cambio sintáctico y prestigio linguistico. Madrid: Iberoamericana.

Silva-Corvalán, C. (1989). Sociolinguística: Teoría y Análysis. Madrid: Alhambra.

Sgall, P., Hajičová, E., Panevová J. (1986). The Meaning of the Sentence in Its Semantic and Pragmatic Aspects. Prague: Academia.

Schiffrin, D. (1977). Opening Encounters. American Sociological Review 42 (5), pp. 679-691.

Schiffrin, D., Tannen, D., Hamilton, H. E. (ed.). (2001). The Handbook of Discourse Analysis. Oxford: Blackwell Publishing.

Smith, N., Wilson, D. (1990). Modern Linguistics. London: Penguin Books.

Sperber D., Wilson D. (1986). Relevance. Communication and Cognition. Cambridge, Massachusetts: Harvard University Press.

Swan, M. (2000). Practical English Usage. Oxford: Oxford University Press.

Trudgill, P.(ed.). (1978). Sociolinguistic Patterns in British English. London: Edward Arnold.

Trudgill, P. (1983). Sociolingustics. London: Penguin Books.

Urbanová, L. (2003). On Expressing Meaning in English Conversation: Semantic Indeterminacy. Brno: Masarykova univerzita.

Urbanová, L., Oakland A. (2002). Úvod do anglické stylistiky. Brno: Barrister & Principal.

Vachek, J. (1959). On the functional hierarchy of spoken and written utterances. In Vachek, J. (1976). Selected Writings in English and General Linguistics. Prague: Academia.

Vachek, J. (1962). Lingvistická charakteristika současné angličtiny. Praha: Státní pedagogické nakladatelství, n.p.

Vachek, J., Firbas, J. (1962). Lingvistická charakteristika současné angličtiny. Praha: SPN.

172 Vachek, J. (1972). Z klasického období pražské školy 1925-1945. Praha: Academia.

Vachek, J. (1976). Selected Writings in English and General Linguistics. Prague: Academia.

Vachek, J. (1991). Chapters from Modern English Lexicology and Stylistics. Praha: SPN.

Valeš, M. (2002). La alternancia -ra/-se del imperfecto de subjuntivo en la ciudad de Granada. Unpublished Ph.D. thesis. Univerzita Palackého v Olomouci. Filozofická fakulta.

Valo, M. (2003). Workmates, friends or more? Perceived effects of computer- mediatedness on interpersonal relationships. Electronic Journal of Communication, 13 (1). Retrieved on 30 August 2005, www.cios.org/getfile/valo_v6n396.

Warschauer, M., and Healey, D. (1998). Computer and language learning: an overview. LanguageTeaching 31(2), pp. 57-71.

Werry, C. C. (1996). Linguistic and Interactional Features of Internet Relay Chat. In Herring (ed.) Computer-mediated communication: linguistic, social and cross-cultural perspectives. 47-63. Amsterdam: Benjamins.

Wolfram, W. (1969). A Sociolinguistic Description of Detroit Negro Speech. In Downes, W. (1998). Language and Society. Cambridge: Cambridge University Press.

Wolfson, N. (1983). Rules of speaking. In Richards, J.C., Schmidt, R.W. (1983). Language and Communication. London: Longman, pp. 61-87.

Yule, G. (2002). Pragmatics. Oxford: Oxford University Press.

Yus, F. (2001). Ciberpragmática: El uso del lenguaje en Internet. Barcelona: Ariel Lingüística.

Zimmerman, Donald H., Boden, D. (1991). Structure-in-action: an introduction. In Deirdre Boden & Donald H. Zimmerman (Eds.), Talk and Social Structure: Studies in Ethnomethodology and Conversation Analysis. Cambridge: Polity Press.

173