Abstract
This article describes word meaning negotiation (WMN) in online discussion forum communication, a form of computer-mediated communication (CMC). WMN occurs when participants who are engaged in a discussion about a particular topic remark on a word choice of another participant, thus initiating a meta-linguistic sequence in which a particular word is openly questioned and the meaning of that word is up for negotiation. By closely studying the process of WMN and focusing on the practices of the participants engaged in it, this article empirically explores how situated meaning in concrete utterances is established by participants who are in disagreement about word meaning. Routine ways of entering into a WMN sequence are identified, as well as typical ways of contributing to the negotiation once the discussion has shifted focus from being on topic to being on word meaning. The article highlights the context-sensitivity of word meanings and illustrates how participants engaged in WMN orient toward both semantic properties of the words themselves as well as to relevant aspects of the conversational context.
Keywords
Introduction
In everyday communication, people do not usually reflect on the definitions or meanings of the words they use to produce utterances. On a moment to moment basis, people use language naturally as a resource that has been gradually adopted over time, from many instances of situated interaction in various language activities (Linell, 2009). People are familiar with the possible ways words can mean in the sense that they know how to use them in and across situations and contexts. In communication, interlocutors collaborate both in the production of utterances and in the interpretation of situated meaning (Clark, 1996; Linell, 2009). However, communicative partners typically do not go about establishing mutually agreed-upon situated meanings of each individual word used in utterances. Instead, interlocutors seem to collaborate with the purpose of achieving enough understanding for current purposes, which generally involves establishing sufficient mutual understanding to keep on communicating (Garfinkel, 1967).
Naturally, communication may sometimes run into trouble or even break down, at which point actions need to be taken to repair the communicative problem in order to restore enough mutual understanding so that a sufficient degree of intersubjectivity between interlocutors is maintained (Kitzinger, 2013; Schegloff, 2007). On occasion, particular words and the interpretation of these words are the causes of such communicative problems. When a specific word is identified as the root of a communicative problem, interlocutors may need to turn their attention to this particular word and its meaning, to collaborate with each other in a process of semantic coordination in which they negotiate their different takes on the meaning of the word (Larsson, 2015; Mills and Healey, 2008). This article focuses on such sequences in communication and analyzes how interlocutors coordinate with each other following a communicative problem caused by conflicting opinions about word meaning. This article aims to shed light on how processes of word meaning negotiation (WMN) are at work in instances in communication where participants are in disagreement about what a particular word can or should mean in a situated discussion context. The purpose of this article is thus to empirically explore processes of WMN, and to categorize the interlocutors’ coordinating efforts and contributions in the negotiation processes.
The interaction sequences analyzed in this article are taken from online discussion forums. This form of communication is chosen for this study for a number of reasons. Given that this form of computer-mediated communication (CMC) is asynchronous, participants have more time at their disposal to contemplate and reflect upon words and meanings than in spoken communication. They are under no pressure to give up the floor to someone else, or to hand over the turn to another participant at any particular point. In addition, participants share very little initial common ground with each other, since most of the discussions in online forums take place between formerly unacquainted participants who need to work out among themselves that they use words in similar ways when referring to people, actions and things in the discussion. Furthermore, since discussion forum communication lacks support for many of the non-verbal cues used to perform meta-communicative functions such as grounding and turn-taking in spoken communication, it is likely that more effort needs to be put into verbalizing processes of interpretation and understanding in CMC than in spoken communication, where gestures, body-language, positioning, gaze and prosody may be used to perform these meta-communicative functions (Clark and Brennan, 1991).
First, the theoretical background of the study will be outlined, followed by an introduction of the investigated phenomenon itself. In the section that follows, the methods and materials of the study are accounted for. Next, the results of the study are presented, mainly focusing on routine ways in which participants in discussion forum communication participate in and contribute to the process of WMN. In this section, several examples from the interaction data are presented and discussed in depth. Finally, this article presents a concluding discussion focusing on the significance of the results of the study.
Words and meaning
The study presented in this article adopts a dialogical view of language and communication, recognizing that words are not containers of static, lexical meaning. Actual meaning can only ever exist in situated interaction, in which communicative partners collaborate to establish the situated meaning of utterances through joint interpretation (Bakhtin, 1986). Linell (2009) suggests that situated meanings and lexical meanings are in fact entities of different kinds, at different levels of abstraction (p. 327). Clearly, words and other language units possess abstract qualities which can be utilized in communication to enable production and interpretation of situated meaning. These qualities can be thought of as semantic potentialities, or meaning potentials, that are associated with words and that have been abstracted away from previous communicative situations. In dialogical theory, the meaning potential of a word is viewed as a semantic resource which, together with contextual factors, can help prompt situated meaning. Meaning potentials should thus be viewed as abstract properties of words that have accumulated over the years through repeated situated use. In this way, meaning potentials are open and dynamic and can change depending on how language users use them in different contexts over longer periods of time (Norén and Linell, 2007).
As meaning potentials can change over time, the situated meaning of an utterance is open to negotiation between participants in communication. When a speaker uses a word in an utterance, the meaning potential of that word is evoked in relation to contextual aspects, such as what the utterance is a response to, how other situated aspects are oriented to, how the utterance is addressed to the interlocutors and so on. Since communicators generally display their current state of understanding in their communicative contributions, what counts as a relevant interpretation of an utterance is dialogically determined as the participants take turns communicating with each other (Schegloff, 1992). Consequently, the main difference between meaning potentials and situated meanings is that meaning potentials are semantic properties of linguistic resources that have been abstracted away from repeated usage in various situations over time, whereas situated meanings are accomplished in dialogical interaction, where meaning potentials are used together with contextual aspects of the situation to establish a local meaning in situ (Linell, 2009, p. 341).
Defining WMN
In communication, participants do not always explicitly negotiate the meanings of the words used in conversation. If they did, they would make no progress on the topic being discussed. At certain points in conversation, however, participants do remark on the word choices of others, which may be an indication of a lack of intersubjectivity between the participants, that is, a sign that is insufficient understanding or disagreement between the participants (Linell, 1995). These kinds of issues of miscommunication must be dealt with in order for the conversation to proceed smoothly. Sometimes when a participant’s word choice is questioned by another participant, the conversation turns into a sequence that unfolds on a meta-linguistic level, where the meaning and use of that particular word are discussed in more detail. In such sequences, the communication turns from being ‘on topic’ to being ‘on language’. In the study presented in this article, sequences of this kind are called WMN sequences (WMNs) (Myrendal, 2015). These sequences are distinguished by participants using language to talk about language, and more precisely the meanings of words. The focus of the communication in WMN sequences is temporarily shifted to the meanings of words or the uses of words given in a particular context. Ludlow (2014) has described a similar phenomenon using the terms ’lexical warfare’ and ’word meaning litigation’ referring to instances in communication where interlocutors actively argue for a particular take on meaning to be acknowledged as ‘the correct one’. In this study, WMN is defined as
Myrendal (2015) proposes that WMN can be sorted into two main types, depending on the origin or cause of the WMN. The first type comprises WMNs that are caused by insufficient understanding of a particular word, that is, when one participant uses a word which is not understood by another participant, and the second participant needs to request clarification regarding the meaning of the word in order to restore enough understanding so that the discussion on topic can continue. The second type of WMN encompasses sequences that originate in disagreement between participants regarding the way a particular word is used in the discussion context. In this type of WMN, participants tend to disagree about the appropriateness of a particular word being applied to the current discussed situation, and the meaning negotiation that unfolds focuses on the meaning of the word and how it should be appropriately used in language in general and in the discussed situation in particular. This article mainly deals with the second type of WMN and focuses on how participants negotiate the meanings of words when there is disagreement about how a word can or should be used in interaction.
Myrendal (2015) found that both types of WMN typically start off in the same way, as a series of turns following a T-I-R-(RR) pattern, also found in negotiation of meaning between non-native speakers (Varonis and Gass, 1985). Here, T corresponds to the trigger causing the problem in the communication, I is the indicator pointing out the issue of miscommunication and R is the response usually addressing and repairing the issue. RR is an optional unit of the sequence – a reaction to the response – used to tie up the routine of repair before popping back to the main flow of the conversation on topic. Note that there needs to be a meta-linguistic shift that turns the focus of the conversation from being on topic to being on language in order for any conversation to turn into a WMN sequence (cf. the use-mention distinction (Quine, 1940: 23–25)). This shift is invited in the Indicator turn and then typically occurs in the Response turn. Once the shift has occurred, a WMN sequence has been identified, and any post that continues to build on the T-I-R-(RR) sequence is also considered part of the WMN sequence.
Below, two excerpts from forum discussions will be used as examples illustrating the difference between sequences that make the meta-linguistic shift and thus contain posts that deliver meta-linguistic contributions (and therefore constitute WMN sequences), and sequences that do not contain posts that deliver meta-linguistic contributions (and thus do not count as WMN sequences but are considered repair sequences not particularly focusing on word meaning). All of the sequences in this study are of the former kind, that is, making the meta-linguistic shift in the third turn (the Response turn):
In the example in Excerpt 1, the situated meaning of the trigger word ‘sexism’ is questioned by P2, which can be viewed as an invitation to a meta-linguistic shift focusing on the situated meaning of that word. In turn 3, P1 responds to the meta-linguistic clarification request and explains the meaning of the trigger word. Since there is a meta-linguistic shift in the Response turn, and this post specifically addresses the clarification request as an issue about word meaning, this sequence counts as a WMN sequence.
The sequence in Excerpt 1 can be contrasted with another example, in which the clarification request in the second turn is not addressed as an invitation to negotiate word meaning. Therefore, in the example in Excerpt 2, there is no meta-linguistic shift, and as a consequence, the sequence does not constitute a WMN according to the criteria used in this study:
Here, in the Indicator turn, P2 issues a clarification request targeting the word ‘invandringsindustrin’ (immigration industry), which could have been interpreted as an invitation to clarify the meaning of this particular word. However, in the Response turn, P1 does not provide a meta-linguistic clarification targeting the trigger word’s meaning. Instead, P1 continues the main line of discussion, by raising questions on the topic of immigration directed at P2. Consequently, in Excerpt 2, there is no meta-linguistic shift in the response turn, and therefore this particular sequence does not constitute a WMN.
Methods and materials
The data used in the study presented in this article consists of 22 WMNs gathered from three large Swedish discussion forums, Familjeliv (http://www.familjeliv.se), Flashback (http://www.flashback.org) and Passagen Debatt (http://www.debatt.passagen.se). In total, the 22 interaction sequences comprise 631 turns produced by 357 participants. The data collection consists of over 44,000 words. In the excerpts presented in this article, the original Swedish discussions have been translated into English. The 22 WMNs in the data collection all have in common that they originate in disagreement about what a word can or should mean given a specific discussion context. This means that all sequences involve one turn where a particular word is used, and at least one turn in which another participant opposes the use of the word in some way. Following the initial use of the trigger word, and the indication that the word is somehow inappropriate to use, the discussion unfolds into a WMN sequence focusing on the meaning of the trigger word and how it should be appropriately used in language in general and in the discussed situation in particular.
As illustrated in Table 1, the WMNs display a great variation in length. The longest sequence is 85 turns long, whereas the shortest consists of only five turns. There is also a great difference in the number of participants active in each WMN sequence. In a few sequences, the number of participants is as low as 2. In the longer sequences, the number of participants actively contributing to the WMN is as high as 40.
Number of active participants and number of turns within each WMN sequence.
WMN: word meaning negotiation.
The study employs a qualitative method for interaction analysis influenced by conversation analysis (CA), which means that the general focus is on the routine ways in which participants organize their interactional activities (Hutchby and Wooffitt, 2008; Schegloff, 1999; Ten Have, 2007). More specifically, the focus of this study is on how participants manage to establish situated meaning by orienting to the underlying organization of text-in-interaction (asynchronous CMC), by orienting to individual and shared linguistic knowledge, by drawing upon aspects of the conversational context and by relating to viewpoints about word meaning as expressed by others in previous utterances. As interlocutors in conversation update their intersubjective understandings on a turn-by-turn basis (Heritage, 1984), the specific focus of the analysis is on how turns within each sequence relate to prior turns, which means that the analysis is focused on various features of sequentiality of the communication (Schegloff, 2007). By closely examining how participants build on the written contributions of others, the aim is to reveal how participants manage to accomplish orderly interaction in the negotiation of word meaning, particularly focusing on how semantic affordances of the negotiated words and aspects of the conversational contexts are oriented toward and utilized as resources in the negotiation of situated word meaning.
Results
This section will present the results of the study, first by describing how WMN sequences are initiated, then by analyzing how participants contribute to an ongoing WMN sequence in different ways.
Entering into a WMN sequence
In the 22 sequences, 2 main ways of initiating WMN are identified, either by a participant questioning the appropriateness of the trigger word by issuing a meta-linguistic clarification request, or by a participant opposing the use of the word by issuing a meta-linguistic objection.
Meta-linguistic clarification requests
Excerpts 3 and 4 demonstrate the initiation of a WMN sequence issued by a meta-linguistic clarification request, in the turn contributed by P2. In Excerpt 3, the discussion is about a well-known Swedish blogger (Katrin Zytomierska) who posted a blog entry about her nanny running away from her during a family trip to the United States. In the blog post, the word ‘rymma’ (escape) was originally used to describe the nanny’s leaving the family without giving notice or telling anyone that she was planning to quit her job. At the start of the discussion about this blog post, the participants initially begin using the same word as the blogger, but early in the discussion meta-linguistic clarification requests are raised regarding the appropriateness of using the word ‘rymma’ applied to the current situation. The first post containing a meta-linguistic clarification request targeting the meaning of the trigger word is displayed in the turn provided by P3 in Excerpt 3. To show how turns are related in the interaction sequence, the alias for each participant (P1, P2, … Pn) is followed by the consecutive ID number assigned to each post by the discussion forum:
In Excerpt 3, P3 continues to build on P2’s post, which is manifested in the initial quote of that post. In P3’s post, corresponding to the Indicator turn, the appropriateness of the word ‘rymma’ is questioned by the meta-linguistic clarification request. This post is the initiation of a very long WMN sequence concerning the meaning of the word ‘rymma’, comprising in total 85 turns.
In Excerpt 4, the discussion is about whether or not co-sleeping with infants is beneficial to children. This discussion turns into a WMN sequence about the meaning of the word ‘onaturligt’ (unnatural), as one participant issues a meta-linguistic clarification request targeting the meaning of this trigger word, in relation to the discussed issue about co-sleeping:
In Excerpts 3 and 4, one participant questions the appropriateness of the trigger word by issuing a meta-linguistic clarification request targeting the situated meaning of the word in question (‘What do you mean by escape?’ and ‘How do you mean by unnatural?’). In the turns following this initiation, the discussions unfold into WMN revolving around the meaning of the trigger words.
Meta-linguistic objections
In Excerpts 5 and 6, the initiation of the WMN does not come as a meta-linguistic clarification request, but instead as a meta-linguistic objection explicitly opposing the use of the trigger word applied to the current discussion context. In Excerpt 5, the discussion is about a woman (P1) claiming that she is stalking another woman online. In the beginning of the discussion, P1 is vague about what the stalking actions entail, but as the discussion continues she explicitly states what she means by the word ‘stalking’:
Here, both P2 and P3 explicitly opposes the use of the word ‘stalking’ which launches a WMN targeting the meaning of this particular word. Similarly, in Excerpt 6, the discussion is about whether or not it is acceptable to flirt with a married person. In the first post of the thread, P1 asks other participants to give opinions about the topic in general, but shortly after, it becomes apparent that P1 has a very specific situation in mind, and wants to know if other participants find this particular flirting behavior acceptable:
In both Excerpts 5 and 6, a WMN is launched by participants explicitly opposing the use of a particular trigger word, as the word is perceived as inappropriate to use applied to the situation under discussion. In both examples, the participants explicitly contest that the circumstances can accurately be portrayed by the meaning of the trigger word. In each case, the participant opposing the use of the trigger word offers alternate words perceived as more accurately describing the discussed circumstances. In Excerpt 5, ‘stalking’ is opposed as an inappropriately chosen word, and is substituted with ‘following an open blog’. In Excerpt 6, ‘flirt’ is deemed inappropriate in the current situation, and the word ‘courtship’ is introduced in its place.
Meta-linguistic objections are not only used to launch negotiation sequences, making the shift from the discussion being ‘on topic’ to being ‘on language’, but they also contribute in other ways to the negotiation itself. This will be developed further in the next section.
Routine ways of negotiating word meaning
When a trigger word has been identified as problematic and an invitation to a WMN sequence has come in the shape of a meta-linguistic clarification request or a meta-linguistic objection in the Indicator turn, it is typically the third turn in the sequence that determines whether or not the sequence develops into a WMN. If the participant faced with a clarification request or an objection responds to the invitation by clarifying the situated meaning of the trigger word, the sequence is characterized as a WMN, as the third turn contains the necessary meta-linguistic shift. Consequently, in accordance with the so-called ‘next turn proof procedure’ (Sacks et al., 1974), it is the subsequent turns following the initiation to a WMN that dialogically determine whether or not the discussion participants turn their attention to word meaning in particular and enter into WMN. Once a participant accepts an initiation to go into WMN concerning a particular word, all of the posts following and building upon the initial posts of the WMN sequence are also considered as a part of that WMN sequence. This means that a WMN sequence typically starts off following the T-I-R-(RR) pattern, but then continues to develop into a larger and less regular structure.
When a WMN has been launched, the participants engaged in it contribute in different ways to the negotiation of word meaning. In the data collection of this study, certain recurring and routine ways of contributing to a WMN sequence are identified. In their contributions to the WMN process, participants orient both toward aspects of the conversational context as well as aspects of meaning potential associated with the trigger words. In this study, several such routine ways of contributing to a WMN sequence are identified and categorized into five categories depending on if they mainly foreground aspects of the discussed situation (i.e. aspects of the context) or if they mainly foreground aspects of perceived meaning potentials associated with the word itself (beyond the particulars of the discussed situation).
Explicification
One recurrent way in which participants propose meaning to a word under negotiation is by introducing a definition-like component to the WMN, which is here called making an explicification. Explicification is a notion borrowed from Ludlow (2014) denoting the introduction of an explicit definitional component to a word meaning under negotiation. Explicifications may either orient toward aspects of the word’s semantic properties, that is, aspects of meaning potential, or toward aspects of the discussed situation. Here, the former type of explicification is called a generic explicification (Excerpt 7), and the latter is called a specific explicification (Excerpts 8 and 9).
Generic explicifications, which foreground aspects of meaning potential of the trigger word, sometimes come in the shape of attempted full definitions. For example, this occurs in the WMN of the trigger word ‘misshandel’ (abuse). In this discussion, the participants are debating whether or not piercing the ears of small children is morally acceptable or if it should be viewed as child abuse. The WMN sequence comprises 80 turns and takes place toward the end of this very long discussion thread (which in total comprises over 2000 turns). When the negotiation has been going on for a while, a few participants attempt to produce far-reaching definitions of the trigger word, using text from law paragraphs as a foundation for their definitions of the word:
In the last post in Excerpt 7, P2 uses paragraphs from the Swedish Penal Code to make a generic explicification of the word ‘misshandel’, defining what the word means by using an external source of information. Attempting to make an explicit definition of the trigger word by drawing upon an external source of information is one way of producing a generic explicification, but it is not the most common way of making generic explicifications found in the data of this study. Instead, the prototypical generic explicification comes in the form of an informal account attempting to explain what is perceived to in the meaning potential of the trigger word. One example of this is shown in P2’s first post in Excerpt 7 (#2085): ‘Child abuse is to inflict pain upon your child’. Here, the participant introduces a definitional component to the word under negotiation, suggesting that the meaning potential of the trigger word encompasses inflicting pain upon a child. This is considered an act of generic explicification, since the definitional component goes beyond the particulars of the discussed situation as it has nothing to do with ear piercing, but more generally concerns the conditions perceived as (at least) sufficient for calling a particular action ‘misshandel’.
In specific explicifications, aspects of the conversational context concerning the particulars of the discussed situation are foregrounded as the main part of the proposed definitional component and not the semantic properties of the trigger word itself (as in generic explicifications). Several instances of specific explicifications are found in the negotiation of the word ‘misshandel’ in the discussion about piercing the ears of young children. One example of this is shown in Excerpt 8:
In this case, the definitional component of the specific explicification is foregrounding the pain and physical change caused by the piercing of a child’s ears, which is picked up from the context of the discussed situation. By producing this specific explicification, the participant proposes that pain and physical changes are part of the meaning potential of the trigger word, but what is foregrounded are the specific circumstances of the discussed situation (‘to pierce the ears of young children’). A little later in the same discussion, another participant produces a similar specific explicification, which is shown in Excerpt 9:
As in Excerpt 8, the definitional component of the specific explicification focuses on the act of forcing someone else against their will, and even against their verbal protests. Similar to the previous explicification, these are contextual aspects picked up from the discussed situation (and from specific inserted YouTube clips of videos showing infants getting their ears pierced while screaming and crying), but again we learn something about what the participant in Excerpt 9 considers to be a part of the meaning potential of the trigger word.
Exemplification
Another recurrent way in which participants contribute to WMN is by exemplifying what the trigger word can mean, or usually means, in circumstances other than the particular discussed situation. In contrast to specific explicification, exemplification does not draw upon circumstances pertaining to the discussed situation, but instead introduces other (real or imagined) situations into the negotiation. By providing illustrative examples of what the word can refer to in other situations, and explaining why the word is suitably applied to these particular instances, participants display their understanding of the meaning potential of the negotiated word through acts of exemplification. The two excerpts below, both taken from the negotiation of the word ‘fullfet’ (full fat) illustrate the difference between exemplification (Excerpt 10) and explicification (Excerpt 11):
In Excerpt 11 (containing the explicification), the P2 contributes to the WMN by introducing a definitional component to the trigger word – ‘the fattest available option’. In Excerpt 10 (containing the exemplification), P1 is not explicitly saying anything about what is in the meaning potential of the trigger word, but is merely providing some examples of what the word appropriately refers to in the world. In order to draw conclusions about what is in the meaning potential of a trigger word operating from an instance of exemplification, some kind of abstraction process needs to be involved, which abstracts away from the concrete examples onto the abstract word level with the associated meaning potential. Judging from the enumerated dairy products, the act of exemplification in Excerpt 10 seems to point to precisely the same aspect of meaning potential explicified in Excerpt 11, namely that ‘full fat’ has to do with ‘the fattest available option’.
Meta-linguistic objection
As addressed in ‘Entering into a WMN sequence’, meta-linguistic objections regularly function as the catalyst for initiating a WMN sequence, but meta-linguistic objections are also found to serve other functions within the negotiation of word meaning. In most instances, meta-linguistic objections are raised toward the original use of the trigger word, indicating disagreement toward that word being used in the situated discussion. Typically, this kind of objection comes as a variation on the form ‘That is not X!’, and is regularly followed by an account explaining why the meta-linguistic objection is raised. Examples of this function of meta-linguistic objections were seen in Excerpts 5 and 6. In other cases, the meta-linguistic objection serves a delimitation function, and is used to address the issue of delimiting the meaning potential of the trigger word. In these cases, it functions to question or object to the suggested boundaries of the meaning potential of the negotiated word. One example of this is found in Excerpt 12 (a continuation of the negotiation shown in Excerpt 6), taken from the negotiation of the word ‘flirt’ (bold font added to highlight the addressed function of the meta-linguistic objection):
A similar example is found in the negotiation of ‘misshandel’ (abuse), as shown in Excerpt 13:
In both of the examples in Excerpts 12 and 13, the participants express that they perceive that there is something delimiting the meaning potential of a particular negotiated word, a border that can serve as a limit for deciding if a word can be applied to a certain situation or not.
Contrasting
Contrasting is performed in WMN when a participant positions the trigger word against another word. Acts of contrasting can perform a number of different functions in WMN. For example, contrasting can be used for delimitation purposes, that is, for drawing boundaries between the trigger word and another more or less closely related word. Contrasting can also function to highlight aspectual differences between the meaning potentials of the two contrasted words. A third function served by contrasting is to create a local opposite in the discussion, using the contrasting word as a juxtaposition of the trigger word.
In the negotiation of the word ‘flirt’, acts of contrasting are used to delimit the trigger word. An example of this is found in Excerpt 14, which is a continuation of the example presented in Excerpt 6 (bold font added to highlight contrast):
As illustrated in Excerpt 14, contrasting can be used to delimit the boundaries of a trigger word, by using another, closely related word as a contrast to the trigger word. These kinds of near contrasts can also be used to highlight an aspect of meaning potential shared by the contrasted words which display a significant variation which therefore marks out a difference in that particular aspect of meaning potential. Several examples of this form of contrasting are found in the WMN of the trigger word ‘super’ (boozing). This WMN takes place within a discussion about a woman who is denied wine in a restaurant, when the bartender notices she is breastfeeding her baby at the table. In the beginning of the discussion, the woman’s drinking behavior is characterized as ‘boozing’, but this choice of words is immediately questioned by several participants, which launches a WMN sequence negotiating the meaning of the trigger word ‘super’ (boozing). At one point in this WMN, the participants turn their attention to the alleged pace at which the woman was drinking her wine, to determine whether or not the drinking behavior can accurately be described using the trigger word. By using several contrasting words, the variation in pace is highlighted, which is exemplified in Excerpt 15 (bold font added to highlight contrast):
In all of the posts in Excerpt 15, the participants use words and expressions that display a variation with regards to the pace at which the woman was consuming her wine. Seemingly, the contrasting words are used to form arguments for or against using the trigger word ‘super’ (boozing), with the underlying assumption that this word also shares the same aspect of meaning potential, that is, that there needs to be a certain pace of the drinking activity for it to accurately be characterized as ‘boozing’. The participants who are arguing for describing the behavior as ‘boozing’ are using contrasting words that correspond to ‘drinking quickly’, such as ‘necking’ (halsa), or ‘chugging down the wine’ (bälga). On the other hand, the participants who are arguing against characterizing the drinking behavior as ‘boozing’ are opposing the use of both of these words, and are instead suggesting another word that corresponds to ‘drinking slowly’, namely the word ‘sipping’ (sippa). The difference in form (sipping-necking-chugging-boozing) here clearly marks a difference in meaning (Clark, 1993), focusing on that particular aspect of meaning potential which is shared between the words, but which displays a variation along a scale with regards to this particular aspect. What can be learned from all of these acts of contrasting is that there is an aspect of pace in the meaning potential of the trigger word ‘super’ (boozing), which is significant for determining if a behavior can be described using the trigger word, or not.
Endorsement
Participants regularly display which of the sides of the negotiation they affiliate with by taking a stance in the negotiation (Steensig and Drew, 2008). As discussed earlier in this article, issuing a meta-linguistic objection is one way of taking a stance in a WMN. Another way of taking a stance in an ongoing WMN is by endorsing the continued use of the trigger word after it has been made the trigger of a negotiation. Once a WMN has been initiated and a trigger word has been identified, the use of that word is being overtly questioned, and opting to use it again can therefore be interpreted as taking part in the meaning negotiation process. Passively endorsing the continued use of a trigger word, that is, continuing to use the word even after it has been made the trigger of a WMN sequence, occurs in most of the WMNs in the data of this study.
Interestingly, in a few cases, there are also instances of meta-linguistic endorsement in the WMNs. This occurs when a participant explicitly endorses the continued use of the trigger word by commenting on the word choice on a meta-linguistic level. One example of an act of meta-linguistic endorsement is found in the negotiation of ‘rymma’ (escape). In Excerpt 16, the participant explicitly supports another participant’s use of the trigger word, and thus endorses the continued use by commenting on the word choice on a meta-linguistic level (bold font added to highlight the meta-linguistic endorsement):
To summarize, this study has identified a range of routine ways of contributing to negotiation of word meaning in online discussions where interlocutors are in disagreement regarding the situated meaning of a particular word. The different ways of contributing to WMN have been classified and put into categories for the purpose of developing a better understanding of how negotiation of word meaning actually functions as a device in naturally occurring interaction for restoring intersubjectivity and achieving a shared understanding of meaning.
Concluding discussion
The results of this study show that WMN is a highly dialogical process in which many participants actively build on and relate to the contributions of other participants in order to calibrate and fine-tune situated meaning in online discussions. The results also indicate that although word meanings are open and up for negotiation, they are not endlessly open. Based on the empirical findings of this study, words cannot mean anything. On the contrary, there seem to be limits as to what a word can mean. These limits are regularly mentioned by the participants engaged in WMN. Participants explicitly orient toward perceived boundaries of negotiated words, using phrases such as ‘What counts as X?’, ‘Where do you draw the line of X?’. They also use other, contrasting words to indicate when a boundary appears to have been crossed and to show that they perceive that the limits of the meaning potential have been reached, and that another word is more appropriate to use when referring to the described circumstances. The ways of positioning words against each other through acts of contrasting can be seen as a kind of reification practice among participants engaged in WMN. To make abstract semantic properties more concrete and turn them into more substantial entities, participants engaged in WMN seem to develop various practices, in order to be able to address and negotiate what word meanings really are. Consequently, the development of these practices can be seen as a kind of reification of word meaning. In many of the WMNs in the data collection, the reification of word meaning appears to take the form of a spatial conceptualization. Several participants approach the notion of word meaning from a focal point which conceptualizes meaning as a kind of space with boundaries against the meaning potentials of other related words. Positioning the negotiated trigger word against the boundaries of the space is an important practice in negotiation process. As a result, it is relevant not only to negotiate what is inside the space, but also where the boundaries to other nearly related words should be drawn. On occasion, participants also address what lies outside of the space, that is, what falls outside the scope of the meaning potential of a particular word. In the data collection, there is ample evidence of reification using the spatial conceptualization of word meaning, and examples of this were presented in Excerpts 12 and 13 in ‘Routine ways of negotiating word meaning’. As shown in these examples, participants explicitly mention and negotiate perceived borders and boundaries of the trigger word’s meaning potential, and they use the word within to portray that they are representing the notion of word meaning as a space enclosed by some kind of borders.
This study also indicates that negotiation of word meaning can be used as a tool to form arguments in polarized discussions and debates. Many of the WMNs in the data collection take place in discussions dealing with controversial topics, often concerning moral issues about what is considered right and wrong with regards to certain actions or behaviors. These topics in themselves typically divide the participants into two groups depending on their stance on topic, where one group is arguing for a certain position, and the other group arguing against it. In such polarized discussions, the key words used by the two opposing sides become very important as tools for gaining ground in the ongoing argumentation. This is reminiscent of what Danet referred to as ‘fitting of words to deeds’ (Danet, 1980). In her study on courtroom interaction from a trial where a physician was charged with manslaughter for performing an abortion, Danet found that the defense lawyer continuously referred to the aborted fetus as ‘the fetus’, whereas the prosecutor referred to it as ‘the baby’. Clearly, each party refrained from using a shared vocabulary when referring to the aborted fetus since both parties strategically attempted to convey different impressions of the committed act to the jury. If the defense lawyer instead had passively adopted the referring term of the opposing side, the defense would likely have lost ground in the overall argumentation, since they then would have accepted that the act of aborting a fetus entails killing a baby.
Similar to the defense lawyer and the prosecutor in Danet’s study, participants in the WMNs in this study are often at opposing sides of a polarized discussion on topic when a WMN is initiated. For example, in a discussion about whether or not it is acceptable to fool children into believing in Santa by actively not telling them the truth, the word ‘ljuga’ (lie) is negotiated, and the negotiation is carried out in parallel with the discussion on topic. The participants who are arguing against fooling children into believing in Santa opt to describe the action using the trigger word, since this word in itself carries negative undertones characterizing questionable or immoral behavior. Had the participants managed to get away with characterizing the action of fooling children as lying, they would likely have gained ground in the overall argumentation, since most people probably agree that lying to someone is wrong.
Likewise, in the discussion about piercing the ears of babies and young children, the participants who are arguing against ear piercings are opting to use words characterizing the action as ‘abuse’, making this word choice in itself a tool in the overall argumentation. If the participants arguing against ear piercings are allowed to describe these actions using the word ‘abuse’, surely they have already won the discussion on topic, since everyone agrees that abuse against children is wrong. The participants defending piercing the ears of children must therefore strike against the chosen trigger word, and argue against the action being described by this word; otherwise the discussion on topic is likely lost. These counter-arguments come in the shape of meta-linguistic objections protesting against characterizing the current action using the trigger word. Consistently, in the discussion about breastfeeding while consuming alcohol, the participants arguing against the woman’s drinking behavior opt to describe it using the trigger word ‘super’ (boozing), which in itself is a negatively loaded word characterizing a distasteful drinking behavior. The participants on the opposite side supporting the woman’s drinking, in spite of the fact that she was breastfeeding at the time, must oppose the use of the trigger word to be able to form counter-arguments in the discussion on topic. It seems reasonable to assume that if everyone agrees that the behavior can be called ‘boozing’, the discussion on topic has already concluded that the woman’s drinking in fact was immoral and generally bad. As a consequence, in these mentioned examples, the WMNs concerning the meaning of the trigger words and the discussions on topic are highly intertwined with and dependent on each other and the WMN phenomenon in itself functions as a tool for forming arguments in these discussions.
To conclude, this study has shown how WMN manifests itself in interaction as a tool for participants to compare and calibrate their individual understandings of the meanings of key words following disagreement concerning word meaning. The study has also shown that WMN can function as a mechanism for forming and framing arguments in polarized discussions. The classification into categories of routine ways of contributing to WMN outlined in this study marks the first step in developing an increased understanding of how WMN works as a distinct interactional phenomenon in naturally occurring communication. This study thus serves as a foundation of basic science, and hopefully as a catalyst for further research on the functions and purposes served by WMN in interaction. A possible extension of this work involves formalization of the categories identified in the study. Formalization could possibly make the theory of how word meanings are negotiated in naturally occurring interaction more precise, and would also enable future implementation of support for WMN in dialogue systems (Larsson, 2008).
Footnotes
Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
