Go to Eric Miller's homepage

submitted by Eric Miller in May 1999 
for the course, Ways of Speaking
at the University of Pennsylvania. 
(This paper is approximately 20 pages.)  
Turn-Taking and Relevance
in Conversation







Perpetual Interaction

The Need for a Clear Agenda and Full Expression From All (in Purposeful Conversations)





This paper is about conversation, and especially about two of its essential elements: turn-taking and relevance.  I will ask (and begin to answer):  What is conversation?  What is turn-taking in conversation?  What is relevance in conversation?  I will elaborate on two points:  1) Even if generally only one person speaks at a time, all participants in a conversation are perpetually interacting with each other throughout the event.  2)  In purposeful conversations, there is a dual need for a clear agenda and for full expression from all participants. 


The English word, conversation, is made up of a combination of two Latin roots, 'con,' and 'vers.'  'Con' means: with, together.1   'Vers' means: to turn about in a given direction.2   Thus, to engage in conversation literally means, to turn about with others. 

Conversation can be said to contain two elements, the informational and the phatic.  The term, phatic--from the Greek word, to speak--was introduced into the English language by Bronislav Malinowski in 1923: "Phatic communion...is a type of speech in which ties of union are created by a mere exchange of words...  Words in phatic communion...fulfill a social function and that is their principal aim."3  In other words, "Phatic communication is used to establish social relationships rather than impart information."4 

Prior to, or at the beginning of, a conversation, participants often frame the event.  That is, they make clear to each other the intended nature of the conversation-to-be.  All conversations contain phatic communication; some conversations are also purposeful in that the participants have a defined goal, whether to impart information, formulate a plan, etc.  In a purposeful meeting, every conversational move ideally contributes to the overall goal as set by the participants. 

In order to begin a conversation, participants must form a relationship, and to do this they must in some sense be of the same order.  Any relationship must necessarily be based on partial equivalence.5  There is a need to establish a temporarily-shared reality among participants.6  Participants, to some degree, must agree upon a world-view, a cosmology.7   Common ground--a set of propositions which make up the contextual background for the utterances to follow--must be established.8 

Discourse-units can be of any length.  Small discourse-units are the building blocks of discourse.  As long as a discourse-unit is under construction, the listeners are discourse-unit recipients.9  While these discourse-units are being constructed (and spoken), ongoing negotiation can occur regarding their nature.10 

Discourse is spoken and written language, both monologic and dialogic (involving two or more participants).  Discourse analysis is concerned with how "structures and regularities (of language) are the product of more general social processes and norms of interaction through which people's interpersonal goals, selves, and relationships are negotiated, and out of which a sense of social order is created."11 

Conversation is a type of discourse: it is spoken dialogic discourse.  Thus, conversation analysis may be seen as a subfield of discourse analysis.12  Conversation analysis involves close examination of internal evidence within the (spoken) text.  One type of conversation analysis is conversational ethnomethodology:  "Ethnomethodologists are primarily concerned with the tacit rules which regulate the taking-up by speakers of the running topic, and hence the change-over from speaker to speaker."13 

The situation is the conceptual background of a discourse.14  Participants in a conversation describe and refer to scenes.  Together these scenes form the text world.  Participants must keep track of all the scenes developed in the text world, and of the relationships between those scenes.  This is the function of coherence.15 

The focus of attention constantly shifts in the course of a conversation.  Material that is freshly-introduced has prominence, at least temporarily.   Mentioning something, placing emphasis on it, or focusing on it, makes that thing the object of attention and brings it into the foreground.  If the new material is not responded to, it tends to recede.16 

Conversational sequences often feature a main point (issue) and an illustrating episode (event).  An issue is a generalization, an abstract statement of feeling or principle.  An event concerns a particular instance or example that has occurred in the past.  Together they form an issue-event.17

A landmark occasion in the study of conversation was H. Paul Grice's 1975 publication of his Principle of Cooperation, which was composed of four maxims--

    1)  Give the appropriate amount of information (Quantity).
    2)  Be truthful (Quality).
    3)  Give the appropriate type of information, i.e., be relevant (Relation).
    4)  Be clear (Manner).18 
Grice here is not very concerned with phatic communication.  He seems to be considering conversations in which there is a common purpose or set of purposes, or at least a mutually-accepted direction, conversations in which "participants have some common immediate aim."19 

Do any of the maxims generally have priority over the others?  Yes.  According to Grice, truthfulness is the most important: "Indeed, it might be felt that the importance of the first maxim of Quality is such that it should not be included in a scheme of the kind I am constructing; other maxims come into operation only on the assumption that this maxim of Quality is satisfied."20  Grice had little to say in the way of explanation about his third maxim, relevance.  Subsequently, scholars have begun to fill this vacuum, as we shall see shortly. 

Here is a general model for conversation proposed by Harvey Sacks and associates:

    1)  Speaker-change recurs, or at least occurs.
    2)  Overwhelmingly, one party speaks at a time.
    3)  Occurrences of more than one speaker at a time are common, but brief.
    4)  Transitions (from one turn to the next) with no gap and no overlap are common.  Together with transitions characterized by slight gap or slight overlap, they make up the vast majority of transitions.
    5)  Turn order is not fixed, but varies.
    6)  Turn size is not fixed, but varies.
    7)  Length of conversation is not fixed in advance.
    8)  What parties say is not specified in advance.
    9)  Relative distribution of turns is not specified in advance.
    10)  Number of parties can vary.
    11)  Talk can be continuous or discontinuous.
    12)  Turn-allocation techniques are used.  A current speaker may select a next speaker (as when he addresses a question to another party); or parties may self-select in starting to talk.
    13)  Various turn-constructional units are employed; e.g., turns can be as short as a single word.
    14)  Repair mechanisms exist for dealing with turn-taking errors and violations; e.g. if two parties find themselves talking at the same time, one of them will stop prematurely, thus repairing the trouble.21 
The point I would like to highlight is number 2, 'one party speaks at a time.'  Linguistic analysis of conversation is based on this idea of linear sequence.  As will be discussed below, however, there are in fact a great many things going on at the same time in a conversation: it is only on the oral verbal level that this view of conversation can possibly be maintained. 

In conversation, the norm of reciprocity--the idea that one will receive the equivalent of what one gives--may be said to operate as a social control mechanism.  People learn to reciprocate, in order to get what they want.  A conversation is a social arrangement.  To successfully conduct a conversation, participants must display a willingness and ability to collaborate.22  Successful conversation yields, for each participant, a degree of relational and interactional satisfaction.  Successful verbal communication requires one to be able to identify and satisfy the needs of one's fellow conversants.  To do this, one must be able to take the perspective of the other and adapt one's language to reflect that perspective.23 

A conversation is a type of social exchange: it consists of linked behaviors through which relational partners voluntarily provide one another with resources.24  As such, the conversational relationship is in some ways similar to relationships in which money or material goods are exchanged.  In both cases, the interpersonal behavior involves giving and/or receiving one or more symbolic resources; and in both cases, the interaction may be a means by which individuals may acquire supplies of resources needed to control their environment.25  Exchange partners of any sort are motivated to reciprocate when provided with a resource.  Unless each party to an exchange receives some amount of resource, it is unlikely that future interactions will take place.  Under the rule of reciprocity--an equity in the giving and receiving of positive and negative consequences--participants reinforce or punish each other to approximately the same degree that they feel they have been rewarded or punished.26 

The commodities in conversation, however, do not always consist of information that one wants from the other parties.  The valuable commodity may just be one's time and attention.  In this case, the speaker may be the consumer, not the producer, of the valued commodity.  One is sometimes told, 'Pay attention!'  If one must listen (or read!) when one does not desire to do so, one can feel that one is being robbed of one's time, or that one's time is being wasted: 

To take up time speaking in a small group is to exercise power over the other members for at least the duration of the time taken, regardless of the content....Within the small group, the time taken by a given member in a given session is practically a direct index of the amount of power he has attempted to exercise in that period.  27
Ideally, conversation involves each participant being interested in what the other has to say, each participant being a patient and empathetic listener, and each participant following Grice's Maxims.  In the real world, however, these conditions are often not present.  For example, one conversational participant may attempt to monopolize resources.28  Participants who feel slighted may react by withdrawing, resisting, and/or rebelling, for example, by being ironic.  Violations of reciprocity are sources of dissatisfaction within groups, and are a major cause of group disintegration.29 

Regardless of the nature of a conversation, participants must feel comfortable with each other on the social, personal, and feeling levels for there to be any chance for information to be imparted successfully.  That is, one needs to feel good about the speaker in order to be able to want to take in any of what he or she has to say. 

A conversation may end when 

1)  An external barrier is reached (for ex., people run out of time). 
2)  One or both participants have received enough goods. 
3)  One or both partners have exhausted their supply. 
4)  One or both partners is reluctant to give further.  30


Taking turns is one of the hardest lessons for children under five years to learn...  The young child cannot without much experience believe that 'his turn' really will come in due time.  All that he knows is that others 'have got it' and he hasn't...  Nor does he believe in the goodwill of the others who are enjoying their turns first...  Only the proven evenness of the justice of the controlling adult will make a transition possible from the impetuous assertion of 'I want it now!' to that trust in the future which makes 'taking turns' possible.  31
Turn-taking is a basic form of organization for conversation.32  As mentioned above, turn-taking is especially studied by conversational ethnomethodologists: these scholars study the methods by which groups conduct conversations.  The social organization of turn-taking distributes turns among parties.33  Groups develop turn-allocational systems and preferences.  "It has been the experience of numerous informants that they find themselves exercising the most exaggerated forms of a particular style when they are talking to others who share the style, and in contexts associated with the one in which they learned the style."34 

Transfers from the current-speaker to next-speaker occur at transition-places, or, competition-places.35  Two types of turn-allocational techniques are:  a) the current-speaker selects the next-speaker;  b) the next turn is allocated by self-selection.36  The current-speaker may select the next-speaker by foregrounding him or her.  This can be done in a number of ways, including, by looking at that person, or by asking that person a question.  Actually, if the current-speaker wishes to select the next-speaker, he or she must do so prior to the next transition-place, for an undesired potential speaker may self-select at that transition-place. 

A selected next-speaker has the right, and the obligation, to speak.  However, if the current-speaker does not select a next-speaker, he or she can come to a transition-place and set the stage for competition between aspiring next-speakers: the first one to speak becomes the next current-speaker.  If no one speaks, the current-speaker has the option of continuing.  Reaction-time latency is the time intervening between a current-speaker's completion and a next-speaker's start.  Initiative-time latency is the time intervening between a current-speaker's completion and the start of a follow-up utterance by the same speaker.37 

Deborah Tannen has shown that there is much overlapping and simultaneous talk among certain Jewish groups; Roger Abrahams has shown the same among certain African and African-American groups.38   Indeed, in some of these cultures, individuals perceive the failure to overlap as lack of interest, or dullness.  However, in mainstream Western culture, overlapping talk is generally seen as messy and unpleasant.  An aspiring next-speaker must time his or her utterance to come just after the completion of the current-speaker's utterance, but before the utterances of fellow aspiring next-speakers.  At these moments, instances of multiple-people-speaking-at-once are common--but they are usually very brief, as the norm is for the first speaker to continue and for the others to drop out.  The act of dropping out serves as a repair mechanism: that is, it fixes the situation that has momentarily 'gone awry.' 

Turn-taking systems can provide strong motivations for non-speakers to listen closely to the current-speaker: only by keeping track of upcoming transition-places can an aspiring next-speaker know when to speak; and there is always the possibility that one may be called upon by the current-speaker. 

Again, the turn-taking systems described above deal with single transitions at a time, between only two individuals (current-speaker and next-speaker).  In fact, a "turn-taking system functions to preserve one-party-talking-at-a-time while speaker change occurs."39  If there are two aspiring speakers, one will be left out; if there are three, two will be left out, etc.  The relationship between the current-speaker and possible next-speakers becomes a performance that other participants in the conversation observe. 

In addition to the full transfer of speaker, there are numerous other types of verbal turn-taking in conversation.  For example, there assent terms and adjacency pairs. 

Assent terms--such as, 'Yes,' 'Okay,' 'Uh-huh,' 'Right'--make it clear to the speaker that the listener has taken in and understood the previous message.  They also serve to establish the listener's ongoing availability, and they commit him or her to attend the speaker's next utterance. 40  Assent terms are among the few items that can be spoken while another is speaking that are generally not heard as an interruption-- although this depends on the speaker and the situation.41  One type of assent term is a ratifying repetition: here the listener repeats one of the last words spoken by the current-speaker. 

Adjacency pairs are composed of any two types of utterance that are linked, either by logic or convention.  Given the first element of an adjacency pair, the second is expected; upon its occurrence it can be seen to be a second item to the first; upon its nonoccurrence it can be seen to be absent.42  For examples:  If a question is asked, it should be answered.  If someone in the audience is commanded, summoned, or invited by the speaker in any way, that person is expected to make an appropriate verbal response (unless it was understood to be a rhetorical question, that is, one that the speaker obviously desires no answer to).  If the first part of a saying or a proverb is recited by a speaker, it may be appropriate for listeners to finish the statement.  Deborah Tannen tells of conversationalists who routinely finish each other's sentences.43 

To take a turn in a conversation is always somewhat risky, as one's contribution may be ignored or rejected by one or by all.  There is always the possibility that no other party will wish to engage with one in a reciprocal relationship.  If this occurs, one may feel that one has lost face.  Face is an image of self, defined in terms of approved social attributes.  Protecting one's face is a very high priority for many people.44  This is perhaps one reason why many people dread taking turns in multi-participant, semi-public conversations: the fear of being embarrassed in front of a large group can be very strong. 


One dictionary defines relevance as, that which "has a bearing on, or connection with, the matter at hand."45  I find the metaphor, "at hand," to be useful and interesting.  According to this definition, that which is relevant is that which is within reach, that which can be handled, that which has direct practical application to one's present state and needs. 

Relevance, of course, is an absolutely relative term.  A focused interaction, such as a conversation, is an occasion when participants join together to sustain a single common focus of concern.  In any focused encounter, a particular 'definition of the situation' comes to be shared by the participants.  This serves to define what will be considered, for the time being, as relevant, as well as what is irrelevant.  In a purposeful conversation, a "remark is relevant if and only if it is related to the purpose of the conversational goal."46 

Likewise, as mentioned above, when people gather to begin a conversation, they create or refer to a (philosophical, ideological, cosmological, etc.) common ground, or set of common grounds.  Relevance is the relationship between the current proposition and the common ground(s).  A proposition's degree of relevance depends on its proximity to the center of the common ground(s).  If a proposition is accepted by fellow conversants--that is, if it is deemed relevant--this proposition is allowed to incrementally modify the common ground(s) in an appropriate manner, according to the consensus of the group.47 

Dan Sperber and Deirdre Wilson, authors of Relevance: Communication and Cognition, are two founders of relevance theory.  Their definition states:  "Relevance is a relation between the proposition expressed by an utterance, on the one hand, and the set of propositions in the hearer's accessible memory on the other."48  This definition refers to two aspects of the hearer's condition:  the set of propositions that he or she has previously internalized, and his or her access to those propositions (upon being presented with the new utterance).  Thus, an utterance is optimally relevant to the extent that: a) it is related to those already established propositions (maximum contextual effect), and b) it is easy for the hearer to realize the connections between the old and the new (minimum processing effort).49  The processing effort can be minimized by making the form and content of the utterance as simple and clear as possible.  In other words, if it requires too much cognitive effort of hearers to get one's point and make the connections (if one seems to be stretching it), one is deemed less competent at introducing new topics.50 

Relevance, like love, is in the eye of the beholder.  However, just as individuals can try to make themselves seem lovely, one can try to make one's utterances seem relevant.  In fact, one aspect of the art of conversation is the ability to make one's contributions appear to be inevitable and appropriate next steps.  There are numerous strategies for increasing the apparent relevancy of an utterance; these strategies all seem to involve linking the new to the old.  For instances:  One can say that the new is like the old, and leave it up to listeners to figure out the connections (procedural linkage).  One can show that the old issue is an instance of some larger issue, and then introduce the new topic as another instance of that larger issue (meta-issue linkage).  One can extend an issue in the old message (extension linkage).51  Three ways to extend an issue are: one may concur with it, disagree with it, or add to it. 

All variations of the relevancy rule seem to state that next speakers must respond to the conversational issue.52  A locally relevant remark responds to something said in the speaker's last sentence or two.  A globally relevant remark responds to an essential point made by the speaker.53  If a new utterance is deemed relevant, then the speaker has managed to reconcile his or her  meaning with the contextual background. 

In maintaining themselves as participants in a focused interaction, participants negotiate a common perspective of relevance for each other's actions.54  Above all, relevance plays a social regulating function: given what you have said or done, the constraints on what I say or do are social constraints. 

Changing the topic from the speaker's focus is a way of establishing one's own importance and right to define the situation.  By staying on a topic the other has introduced, one grants the other the freedom to define the nature of the interaction and therein indicates deference to the other.55  Successful communicators are able to simultaneously signal attentiveness to the other, and introduce new topics.  There need not be a conflict between these two goals--in fact it is possible to achieve both with a single interactional move.  Certainly, if one does a better job of signaling one's attentiveness to the present-speaker (and the common ground), one will increase one's chances of delivering one's new message in such a way that it will be well-received.56  A competent communicator is one who accomplishes his or her task goals while maintaining the face needs of the other.  Rhetorical sensitivity is seen as located between complete focus on self and complete focus on the other.57 

Each participant in a conversation is constantly engaged in topic management, which is one type of interaction management.58  Each participant must constantly be deciding if a given sequence of sentences and thoughts forms a connected text, or whether it is merely a random list.  Each must have the ability to discriminate between connected and unconnected structures of language.59  Linked discourse-units can be called reference-chains.  Numerous reference-chains run throughout, and perhaps compose, a discourse.60  Thus, conversational participants analyze the reference-chains and the discourse in front of them, even as they help to create these things. 

As mentioned above, the making of any contribution is risky, as it may be deemed irrelevant and be rejected--and by extension, the person who offered it may come to be considered irrelevant also.  The proposition and its producer may be deemed out of context, out of control, out of bounds, and out of line.  Additions to an issue are somewhat risky, and objections are most risky.  People may resist changing their frameworks, as they may have much psychological--and economic--investment in the system as it stands, in the status quo.  Individuals and groups tend to put a great deal of effort into framework maintenance.  If one raises what is considered to be an irrelevant topic, one which goes beyond the accepted framework or calls into question the validity of that framework, one may produce anger or disorientation in one's co-participants.  For new actions to be perceived and responded to as meaningful, the shared understanding concerning what is being talked about--i. e., the frame of interpretation all participants are applying to what each other says and does--must be changed first.61 

To summarize:  According to Grice, participants in a conversation actively assume each other's cooperation; it is assumed that contributions are intended to continue within the verbal and situational context.  "I expect a partner's contribution to be appropriate to immediate needs at each stage of the transaction."62  Conversation,  then, is a co-operative venture in which the participants seek to add to the commonly-accepted set of propositions by contributing further propositions which are relevant 
to it.63 

Perpetual Interaction

Bronislav Malinowski's definition of phatic communication is concerned only with communication on the oral verbal level: at least in this context, to him, as to many scholars of his day, the physical--including facial expression, gesture, and body placement and alignment--was irrelevant, beyond what was considered the text.  In reality, although only one conversational participant may speak at a time, each participant is perpetually feeling, thinking, responding negatively or positively, empathizing, etc.  It might be said that throughout the event, each participant is 'talking' to him- or herself.  Audible verbal expression is but the tip of the iceberg: a world of fervent activity is often occurring within participants.  It is through physical behavior that this internal activity is constantly, to some degree, expressed. 

Part of being a good conversationalist is being aware of and sensitive to other participants' simultaneous and perpetual non-verbal 'statements.'  "An aesthetic response is not an extra added attraction of communication, but its essence."64  "Involvement is created by the simultaneous forces of music (sound and rhythm), on the one hand, and meaning through mutual participating in sense-making, on the other.65 

In many cultures, there are artistic (storytelling, singing, dancing) and play activities in which members of the group move and make sounds together, whether in unison, in call-and-response, or in other fashions.  There may be a division between an individual and the rest of the group, or the group may be split into two halves or it may be divided in other ways.  Activities may be alternated and/or juxtaposed.  An example of the latter is: a participant on the left may lean rightwards, towards the center; while a participant on the right may lean leftward, also toward the center.  In such a case, the activities are co-ocurring, and participants are performing interlocking pieces of an ever-changing puzzle.  These types of physical activity occur on a more subtle basis during conversational communication. 

First of all, spatial positioning is available as an expressive resource for participants in a conversation: 

Participants in focused encounters typically enter into and maintain a distinct spatial and orientational arrangement.  By doing so, it seems participants can provide one another with evidence that they are prepared to sustain a common orientational perspective.  By arranging themselves into a particular spatial-orientational pattern, they thereby display to the other that they are governed by the same set of general considerations.  By cooperating with one another to sustain a given spatial-orientational arrangement, they can display a commonality of readiness.66
The physical positions that participants take in the course of the event can come to symbolize their social and intellectual experiences (and frames) in that event. 

The segment of space that an individual takes up in the process of an activity can be called his or her transactional-space.  When two or more individuals join to do something together, they come to have a joint-transactional-space.  The positions that these individuals take constitute their joint-transactional-space-formations.  For example, when people wait in a line together, they are in a joint-transactional-space-formation.  When co-participants orient their bodies so that each of them has easy, direct, and equal access to each other's transactional-space, and they engage in some sort of activity together, they can be said to be creating a joint-transactional-space-formation-system.67 

By establishing such a system of spatial and orientational relations, individuals create for themselves a context within which preferential access to the other's actions is established.  Furthermore, such a system of spatial and orientational relations provides a visually perceivable arrangement by which participants in a given focused encounter are delineated from those who are outsiders.  Indeed, it seems that the kind of arrangements that arise in the joint-transactional-space-formation provide a means of clearly demarcating the 'world' of the encounter from the rest of the 'world' around.68
Participants may change the joint-transactional-space-formation as a means of indicating desire of change in various aspects of the event, such as the topic, or the phase of the event.  One advantage of this type of communication is that one can be ambiguous: if one begins a move and no one else joins in, one can stop, and/or return to one's prior position, and pretend that nothing significant has occurred. 

If a participant leans backward, this might be a signal that he or she wishes for the conversation to end.  If a co-participant reacts by leaning forward (a compensatory move), the conversation must continue in its present phase.  However, if a co-participant follows one's lead and also leans backward (a complementary move), this could be a signal from the second participant that he or she also wants the conversation to end.  In this fashion, co-participants can move together into successive phases of the event, independent of verbal suggestions and commitments, which are much more easily identified with the individuals who utter them.69 

Through their joint-transactional-space-behavior, participants are able to keep each other continuously informed about how they feel and what they want.  Simply being responsive, in any way, to others' adjustments in spatial-orientational positioning sends a message of engagement.  All of these small moves serve as means of frame attunement.70 

Co-participants in a conversation often position themselves close to each other and orient their bodies at least partially toward one another.  In such cases, the co-participants' individual transactional-spaces overlap. 

Co-participants repeatedly focus their eyes on one another and, from time to time, their eyes meet.  The aiming of the eyes is perhaps the principal way in which the utterer in an utterance exchange system can indicate to whom his or her words and actions are addressed.  Likewise, the orienting of the eyes to the utterer is one of the principal ways in which a person can indicate that he or she is a recipient.  Indeed, one of the ways in which a recipient of an utterance may redirect the speaker to another is by looking away from the speaker and towards another member of the gathering.71
Discourse-unit recipients (to return to a more linguistically-oriented terminology) often adopt aspects of the speaker's physical behavior: gestures, headnods, and changes in facial expression often come to be patterned in a systematic relation with the speaker.  Entering into the current-speaker's rhythm can be part of an aspiring next-speaker's strategy: that is, doing so can encourage the current-speaker to foreground and even perhaps call upon, the individual who has entered that current-speaker's world.72 

My point here is that while only at most a few oral verbal expressions can be perceived simultaneously, an unlimited number of physical actions, whether voluntary or involuntary, can be perceived simultaneously, especially if these actions can be perceived in a single visual frame.  Sales and marketing people, including politicians, have developed a very practical application of this process:  In opinion-sampling groups, 

 ...potential voters are hooked up to hand-held meters and asked to respond to news reports, TV spots, and tapes of speeches to gauge what works and what doesn't.  The results are superimposed on the screen in real time, so you have an instant analysis of voter response.  When a shot of Hillary speaking was played, the line on the screen dropped like a downhill ski run.  "Oh, man," said Clinton, demonstrating both husbandly concern and his capacity for denial, "they don't like her hair."  73
The Need for a Clear Agenda and Full Expression From All (in Purposeful Conversations)
Citizens of mainstream Western society are still for the most part living in the age of mass media.  Despite the glimmerings of hope for true social dialogue and engagement offered by interactive telecommunication technology, the dominant communication paradigm of the day continues to be broadcast (for the electronic) and publishing (for the literary).  The broadcast/publishing paradigm calls for a silent, passive information and entertainment consumer.  For the operators of this system, feedback from the discourse-unit recipients is a low priority: the main type of feedback response that is encouraged is money payment for product. 

This is the context in which mainstream interaction management experts call for 

 ...the establishment of a smooth and easy pattern of exchange, the meshing of one person's contributions with others.  This meshing is difficult, if not impossible, if the conversation consists of non-sequitors.  Failure to relate one's contribution to the ongoing discourse may cause awkward silence and other signs of disrupted flow.  Unless each conversationalist gives some attention to making his or her contribution relevant, smooth and continuous interaction is impossible.  74
I would like to take this opportunity to ask: Why is this superficial smoothness -- both in terms of relevance and turn-taking--presented as being so essential?  Why is so much emphasis being put on the need to regulate and maintain conversational coherence and propriety?  This concern, it seems to me, often functions simply as a coded type of support for the powers that be.  I suggest that, at least part of the time, people (including leaders) should exhibit more tolerance for diverse and simultaneous verbal and other expressions.  I believe that people should have more confidence in their inherent value, and in the human ability to sort things out.  Humans have a natural, organic organizing and prioritizing function and sense.  Things fall into place.  From my perspective, to achieve a healthy community, the most pressing need is for people to be able to put the issues--their thoughts and feelings--out in the open, get them on the table, out of the closet, and to be able to share these issues with the other members of the community.  Frequent and abrupt topic-switching may be a temporary result, but this can be a sign of openness and freedom, and all other things being equal, group members will settle down and coalesce after a short time.  Moreover, logical connections need not always be demanded: intuitive connections should be also be respected.  Of course, not everything can be considered at once, and not everything can be dealt with at once.  Once people have expressed their issues, the issues can be sorted and priorities can be adopted. Once contributions have been made, order will come. 

In the case of purposeful conversations, it is essential that the purpose be clearly stated at the outset, and that a general agenda be agreed upon.75   Participants need to ask and answer: What needs to be resolved at this meeting?  What can be resolved?  What action-related decisions need to be made?  Within this practical framework, then, people should be encouraged and allowed to truly speak their minds.  I submit that, by virtue of the fact that a single participant feels that a statement is relevant to a conversation, it is relevant to that conversation. 

This is not to say that I believe that ongoing disruption (from the group consensus point of view) should be be tolerated.  I have attended meetings at which individuals were so passionate (and frustrated and confused), that they could not stop speaking.  I am thinking of a particular planning meeting of an artistic group.  An individual was telling people that they were not doing enough, that they were not giving full commitment, that they were not following through. He spoke in generalizations, expressing disappointment and rage, and imploring and exhorting everyone else to act.  His insistence was intense and unquenchable.  Finally, it became clear to a number of us that this individual was preventing the other business of the meeting from being conducted--and people were there precisely to plan how to make contributions.  Finally, we had to intervene with a meta-message and say (paraphrase), "We sense that you are in a lot of pain.  We hear what you are saying and we agree.  We are trying to take action to resolve the situation.  Your excitement is now preventing that process."  People gathered around the excited individual and hugged him.  The individual then quieted down.  Sometimes a person just needs physical contact.  Physical contact be reassuring and can act as a sedative. 

Another, perhaps more difficult and delicate, obstacle to achieving full participation by participants in a conversation, is presented by the individual who, due to shyness, depression, fear, etc., does not speak and does not want to speak.  How can speech be elicited in such a case?  One method is to simply go around the group and ask each participant to take a turn.  If a person has nothing to say at the time, there is the option of setting aside perhaps thirty seconds, during which time nobody else may speak.  This empty space is reserved for that individual.  If that individual wishes to remain silent, the group experiences thirty seconds of that person's silence.  The quality of one person's silence will always be different from another's, as silence can express a wide variety of feelings and attitudes. 


This paper has begun to answer the questions, what is turn-taking, and what is relevance, in conversation?  When partners in a conversation take turns fairly and speak relevantly, the conversation is, on formal levels at least, a satisfying experience for those partners.  What constitutes this fairness and relevance is in part a matter of abstract rules, and in part is developed in the unique conventions and customs that develop in any relationship.


1   Comprehensive Etymological Dictionary of the English Language,Amsterdam: Elsevier, 1966, vol. 1, p. 326. 

2   Origins: A Short Etymological Dictionary of Modern English,London: Routledge & Kegan Paul, 1966, p. 770. 

3   Bronislav  Malinowski,"The Problem of Meaning in Primitive Languages," in The Meaning of Meaning:A Study of the Influence of Language Upon Thought and of The Science of Symbolism, C. K. Ogden  &  I. A. Richards, eds., London: K. Paul, Trench, Trubner & Co., 1923, p. 315. 

4   Oxford English Dictionary,2nd Ed., Oxford: Clarendon Press, 1989, vol. 11, p. 668. 

5   Paul Werth, Focus, Coherence, and Emphasis,London: Croom Helm, 1984, p. 58. 

6   Nancy J. Spekman, "Verbal Communication and Role-Taking: An Analysis of the Use of Deictics," in The First Delaware Symposium on Language Studies: Selected Papers,Robert J.  Di Pietro, William Frawley & Wedel Alfred, eds., Newark: U. of Delaware Press, 1983, p. 168. 

7   Erving Goffman, Frame Analysis,Boston: Northeastern U. Press, 1986 (1974), p. 28. 

8   Nancy J. Spekman, "Verbal Communication and Role-Taking: An Analysis of the Use of Deictics," in The First Delaware Symposium on Language Studies: Selected Papers,Robert J. Di Pietro, William Frawley & Wedel Alfred, eds., Newark: U. of Delaware Press, 1983, p. 168. 

9   Hanneke Houtkoop & Harrie Mazeland, "Turns and Discourse Units in Everyday Conversation," Pragmatics9 (1985), p. 601. 

10   Houtkoup, p. 605. 

11   Deborah Schiffrin, "Conversation Analysis," in Linguistics: The Cambridge Survey, IV: Language:The Socio-Cultural Context,Frederick J. Newmeyer, ed., Cambridge: Cambridge U. Press, 1988, p. 252. 

12   Schiffrin, 1988, p. 253. 

13  Paul Werth, "The Concept of 'Relevance' in Conversational Analysis," in Conversation and Discourse,Paul Werth, ed., NY: St. Martin's Press, 1981, p. 129. 

14   Werth, 1984, p. 45. 

15   Werth, 1984, p. 51. 

16   Werth, 1984, p. 7. 

17   Karen Tracy, "Staying on Topic: An Explication of Conversational Relevance," Discourse Processes,vol. 7 no. 4, Oct.-Dec. 1984b, p. 448. 

18   H. Paul Grice, "Logic and Conversation," in Syntax and Semantics, Volume 3: Speech Acts,Peter Cole and Jerry L. Morgan, eds., NY: Academic Press, 1975, p. 45. 

19   Grice, p. 48. 

20   Grice, (no date), p. 27.  As cited in Stephen Neale, "Paul Grice and the Philosophy of Language," Linguistics and Philosophy15 (1992), p. 531. 

21   Harvey Sacks, Emanuel A. Schegloff & Gail Jefferson, "A Simplest Systematics for the Organization of Turn-Taking for Conversation," Language50 (1974), p. 701. 

22   John J. Gumperz, "Conversational Cooperation in Social Perspective," in Proceedings of the Sixteenth Annual Meeting of the Berkeley Linguistics Society, February 16-19, 1990: General Session and Parasession on the Legacy of Grice,Kira Hall, Jean-Pierre Koenig, Michael Meacham, Sondra Reinman & Laurel A. Sutton, eds., Berkeley: Berkeley Ling. Soc., 1990, p. 440. 

23   Spekman,  p.168. 

24   Michael E. Roloff & Douglas E. Campion, "Conversational Profit-seeking: Interaction as Social Exchange," in Sequence and Pattern in Communicative Behaviour,Richard L. Street & Joseph N. Cappella, eds., London: Edward Arnold, 1985, p. 163. 

25   Roloff,  p. 167. 

26   Roloff,  p. 161. 

27   Robert F. Bales, Personality and Interpersonal Behavior,NY: Holt, Rhinehart & Winston, 1970, p. 76.  As cited in Sacks, 1974, p. 711. 

28   Roloff, p. 162. 

29   Roloff, p. 170. 

30   Roloff, p. 162. 

31   Susan Isaacs, Social Development in Young Children,NY: Harcourt Brace, 1933, p. 223.  As cited in Sacks, 1974, p. 698. 

32   Sacks, 1974, p. 700. 

33   Sacks, 1974, p. 701. 

34   Tannen, 1983, p. 127. 

35   Werth, 1981, p. 129. 

36   Sacks, 1974, p. 703. 

37   Sacks, 1974, p. 716. 

38   Tannen, 1981, 1983.  Abrahams, 1974, 1983. 

39   Sacks, p. 729. 

40   Emanuel A. Schegloff, "Sequencing in Conversational Openings," American Anthropologist70 (1968), p. 1092. 

41   Schegloff, p. 1093. 

42   Schegloff, p. 1083. 

43   Deborah Tannen, "New York Jewish Conversational Style," International Journal of the Sociology of Language30 (1981), p. 137.  Tannen, 1983, p. 123. 

44   Erving Goffman, "On Face-Work: An Analysis of Ritual Elements in Social Interaction," in Interpersonal Dynamics: Essays and Readings on Human Interaction. Bennis, Warren G., ed.  Homewood, Ill.: Dorsey Press, 1968, p. 226. 

45   American Heritage Dictionary, 3rd Edition, Boston: Houghton Mifflin, 1992. 

46   R. Powers, "The Organization of Purposeful Dialogues," Linguistics17, p. 109. 

47   Werth, 1984, p. 55. 

48   Deirdre Wilson & Dan Sperber, "On Grice's Theory of Conversation," in Conversation and Discourse,Paul Werth, ed., NY: St. Martin's Press, 1981, p. 169. 

49   Dan Sperber & Deirdre Wilson, Relevance: Communication and Cognition,Cambridge, Mass.: Harvard U. Press, 1986. 

50   Neil Smith & Deirdre Wilson, "Introduction" (to a special issue on Relevance), Lingua87 (1992), p. 5. 

51   Karen Tracy, "The Effect of Multiple Goals on Conversational Relevance and Topic Shift," Communication Monographs,vol. 51, Sept. 1984a, p. 285. 

52   Tracy, 1984a, p. 282. 

53   Karen Tracy, "Staying on Topic: An Explication of Conversational Relevance," Discourse Processes,vol. 7 no. 4, Oct.-Dec. 1984b, p. 447. 

54   Adam Kendon, "The Negotiation of Context in Face-to-Face Interaction," in Rethinking Context: Language as an InteractivePhenomena, Alessandro Duranti & Charles Goodwin, eds., Cambridge: Cambridge U. Press, 1992, p. 327. 

55   Karen Tracy, "Conversational Coherence: A Cognitively-Grounded Rules Approach," in Sequence and Pattern in Communicative Behaviour, Richard L. Street & Joseph N. Cappella, eds., London: Edward Arnold, 1985, p. 31. 

56   Tracy, 1984b, p. 459. 

57   Tracy, 1984a, p. 274. 

58    Tracy, 1985, p. 31. 

59   Werth, 1984, p. 17. 

60   Werth, 1984, p. 24. 

61   Kendon, 1992, p. 327. 

62    Grice, p. 47. 

63   Werth, 1984, p. 153. 

64   Deborah Tannen, Talking Voices: Repetition, Dialogue, and Imagery in ConversationalDiscourse,Cambridge: Cambridge U. Press, 1989, p. 13. 

65   Tannen, 1989, p. 135. 

66   Kendon, 1992, p. 329.  See:  Kendon, 1973, 1977, 1985.  Goodwin, 1981.  Batchelor & Goethals, 1972. 

67   Kendon, 1992, p. 329. 

68   Kendon, 1992, p. 329. 

69   J. Lockard, D. Allen, B. Schiele & M. Wierner, "Human Postural Signals: Stance, Weight Shifts, and Social Distance as Intention Movements to Depart," Animal Behavior26 (1978), p. 224. 

70   Kendon, 1992, p. 331.  See:  Scheflen, 1964.  Scheflen & Ashcroft, 1976.  Kendon & Ferber, 1973.  Ciolek & Kendon, 1980.  McDermott, Gospinoff & Aron, 1978.  Streeck, 1984. 

71   Kendon, 1992, p. 332. 

72   G. C. Pelose, "The Functions of Behavioral Synchrony and Speech Rhythm in Conversation," Research on Language and Social Interaction20 (1987), p. 190. 

73   "Bill and Hillary Offstage,"  Newsweek(USA edition), March 15, 1999, p. 36. 

74   Tracy, 1985, p. 31. 

75   Roger K. Mosvick & Robert B. Nelson, We've Got to Start Meeting Like This: A Guide to Successful Business Meeting Management,Glenview, Ill.: Scott, Foresman and Co., 1987, p. 17. 


Abrahams, Roger.  "Black Talking on the Streets."  In Explorations in the Ethnography of Speaking,Richard Bauman and Joel Sherzer, eds.Cambridge: Cambridge U. Press, 1974.  pp. 240-262. 

Abrahams, Roger D.  Man-of-Words in the West Indies.Baltimore: Johns Hopkins University Press, 1983. 

American Heritage Dictionary,3rd Edition.  Boston: Houghton Mifflin, 1992. 

Bales, Robert F.  Personality and Interpersonal Behavior. NY: Holt, Rhinehart & Winston, 1970. 

Batchelor, J. P.  &  G. R. Goethals.  "Spatial Arrangements in Freely Formed Groups."  Sociometry35 (1972).  pp. 270-9. 

Ciolek, T. M.  &  Kendon, A.  "Environment and the Spatial Arrangement of Conversational Interaction."  Socilogical Inquiry50 (1980).  pp. 237-271. 

Comprehensive Etymological Dictionary of the English Language.  Amsterdam: Elsevier, 1966.  vol. 1, p. 326. 

Duranti, Alessandro.  "Ethnography of Speaking: Toward a Linguistics of the Praxis."  In Linguistics: The Cambridge Survey, IV: Language: The Socio-Cultural Context,Frederick J. Newmeyer, ed.  Cambridge: Cambridge U. Press, 1988.  pp. 210-228. 

Edmondson, Willis.  Spoken Discourse: A Model for Analysis. London: Longman, 1891. 

Goffman, Erving.  "On Face-Work: An Analysis of Ritual Elements in Social Interaction."    In Interpersonal Dynamics: Essays and Readings on Human Interaction.  Bennis, Warren G., ed.  Homewood, Ill.: Dorsey Press, 1968.  pp. 227-249. 

Goffman, Erving.  Relations in Public: Microstudies of the Public Order.  NY: Basic Books, 1971. 

Goffman, Erving.  Frame Analysis.  Boston: Northeastern U. Press, 1986.  (Originally published in 1974.) 

Goodwin, Charles.  Conversational Organization: Interaction between Speakers and Hearers. NY: Academic Press, 1981 

Gumperz, John J.  "Conversational Cooperation in Social Perspective."  In Proceedings of the Sixteenth Annual Meeting of the Berkeley Linguistics Society, February 16-19, 1990: General Session and Parasession on the Legacy of Grice.  Kira Hall, Jean-Pierre Koenig, Michael Meacham, Sondra Reinman & Laurel A. Sutton, eds.  Berkeley: Berkeley Ling. Soc., 1990.  pp. 429-41. 

Gumperz, John J.  "Contextualization and Understanding."  In Rethinking Context: Language as an Interactive Phenomena,Alessandro Duranti & Charles Goodwin, eds.  Cambridge: Cambridge U. Press, 1992.  pp. 229-52. 

Grice, H. Paul.  "Logic and Conversation."  In Syntax and Semantics, Volume 3: Speech Acts. Peter Cole and Jerry L. Morgan, eds.  NY: Academic Press, 1975.
pp. 41-58.

Grice, H. Paul.  "Further Notes on Logic and Conversation."  In Syntax and Semantics, Volume 9: Pragmatics. Peter Cole and Jerry L. Morgan, eds.  NY: Academic Press, 1978.  pp. 113-127. 

Holdcroft, David.  "Conversational Relevance."  In The Pragmatic Perspective: Selected Papers from the 1985 International Pragmatics Conference,Jef Verschueren & Marcella Bertuccelli-Papi, eds.  Amsterdam: Benjamins, 1987.  pp. 477-495. 

Houtkoop, Hanneke  &  Harrie Mazeland.  "Turns and Discourse Units in Everyday Conversation."  Pragmatics9 (1985).  pp. 595-619. 

Susan Isaacs.  Social Development in Young Children. NY: Harcourt Brace, 1933. 

Jakobson, Roman.  "Linguistics and Poetics."  In Style in Language,Thomas A. Sebeok, ed.  Cambridge, Mass.: MIT, 1960.  pp 350-377. 

Jakobson, Roman.  "Grammatical Parallelism and Its Rusian Facet." Language42 (1966).  pp. 399-429. 

Kendon, Adam.  "The Role of Visible Behavior in the Organization of Face-to-Face Interaction."  In Social Communication and Movement: Studies of interaction and Expression in Man and Chimpanzee,M. von Cranach & I. Vine, eds., London: Academic Press, 1973.  pp. 29-74. 

Kendon, Adam.  "Some Functions in a Face-to-Face Kissing Round." Semiotica15 (1976).  pp. 299-334. 

Kendon, Adam.  "The Behavioural Foundations for the Process of Frame Attunement in Face-to-face Interaction."  In Discovery Strategies in the Psychology of Action,G. P. Ginsburg, M. Brenner, & M. von Cranach, eds., London: Academic Press, 1985.  pp. 229-53. 

Kendon, Adam.  "The Negotiation of Context in Face-to-Face Interaction."  In Rethinking Context: Language as an Interactive Phenomena, Alessandro Duranti & Charles Goodwin, eds.  Cambridge: Cambridge U. Press, 1992.  pp. 323-34.

Kendon, Adam  &  A. Ferber.  "A Description of Some Human Greetings."  In Comparative Ecology and Behviour of Primates, R. P. Michael & J. H. Cook, eds., London: Academic Press, 1973. 

Kopperschmidt, Josef.  "An Analysis of Argumentation."  In Handbook of Discourse Analysis, Vol. 2: Dimensions of Discourse, Teun A. Van Dijk, ed.  London, Academic Press. 1985.  pp. 159-168. 

Laver, John.  "Communicative Functions of Phatic Communication."  In Organization of Behavior in Face-to-Face Interaction,Adam Kendon, Richard M. Harris & Mary Ritchie Key, eds., Paris: Mouton, 1975.  pp. 215-38. 

Lockard, J.  &  D. Allen, B. Schiele, and M. Wierner.  "Human Postural Signals: Stance, Weight Shifts, and Social Distance as Intention Movements to Depart."  Animal Behavior26 (1978).  pp. 219-24. 

Malinowski, Bronislav.  "The Problem of Meaning in Primitive Languages."  In The Meaning of Meaning: A Study of the Influence of Language Upon Thought and of The Science of Symbolism,C. K. Ogden and I. A. Richards, eds., London: K. Paul, Trench, Trubner & Co., 1923.  pp. 296-336. 

Mason, Jeffrey A.  "From Speech Acts to Conversation."  Journal of Literary Semantics,vol. 11, no. 2, Oct. 1982.  pp. 96-103. 

McDermott, R.  &  K. Gospinoff, J. Aron.  "Criteria for an Ethnographically Adequate Description of Concerted Activities." Semiotica24 (1978).  pp. 245-75. 

McDowell, John.  "The Poetic Rites of Conversation."  Journal of Folklore Research22 (1985).  pp. 113-32. 

Moeschler, Jacques.  "Relevance and Conversation."  Lingua90 (1993).  pp. 149-71.

Mosvick, Roger K.  &  Robert B. Nelson.  We've Got to Start Meeting Like This: A Guide to Successful Business Meeting Management. Glenview, Ill.: Scott, Foresman and Co., 1987. 

Neale, Stephen.  "Paul Grice and the Philosophy of Language." Linguistics and Philosophy15 (1992).  pp. 509-59. 

"Bill and Hillary Offstage."   Newsweek(USA edition), March 15, 1999.  p. 36.

Origins: A Short Etymological Dictionary of Modern English.  London: Routledge & Kegan Paul, 1966.  p. 770. 

Oxford English Dictionary,2nd Ed., Oxford: Clarendon Press, 1989. 

Pelose, G. C.  "The Functions of Behavioral Synchrony and Speech Rhythm in Conversation."  Research on Language and Social Interaction20 (1987).  pp.  171-220. 

Powers, R.  "The Organization of Purposeful Dialogues."  Lingusitics17.  pp 107-51. 

Roloff, Michael E.  &  Douglas E. Campion.  "Conversational Profit-seeking: Interaction as Social Exchange."  In Sequence and Pattern in Communicative Behaviour,Richard L. Street & Joseph N. Cappella, eds., London: Edward Arnold, 1985.  pp. 161-89. 

Rundquist, Suellen.  "Indirectness in Conversation: Flouting Grice's Maxims at Dinner."  In Proceedings of the Sixteenth Annual Meeting of the Berkeley Linguistics Society, February 16-19, 1990: General Session and Parasession on the Legacy of Grice. Hall, Kira, Jean-Pierre Koenig, Michael Meacham, Sondra Reinman & Laurel A. Sutton, eds.  Berkeley: Berkeley Ling. Soc., 1990.  pp. 509-18. 

Sacks, Harvey, Emanuel A. Schegloff & Gail Jefferson.  "A Simplest Systematics for the Organization of Turn-Taking  for Conversation." Language50 (1974).  pp. 696-735. 

Scheflen, A. E.  "The Significance of Posture in  Communication Systems."  Psychiatry27 (1964).  pp. 315-31. 

Scheflen, A. E.  &  N. Ashcraft.  Human Territories: How We Behave in Space-Time. Englewood Cliffs, NJ: Prentice-Hall, 1976. 

Schegloff, Emanuel A.  "Sequencing in Conversational Openings." American Anthropologist70 (1968).  pp. 1075-1095. 

Schegloff, Emanuel A.  "Recycled Turn Beginnings: A Precise Repair Mechanism in Conversation's Turn-Taking Organisation."   In Talk and Social Organisation,Graham Button & John R. E. Lee, eds.  Philadelphia: Multilingual Matters, 1987.  pp. 70-85. 

Schegloff, Emanuel A.  "In Another Context."  In Rethinking Context: Language as an Interactive Phenomena,Alessandro Duranti & Charles Goodwin, eds.  Cambridge: Cambridge U. Press, 1992.  pp.191-227. 

Schiffrin, Deborah.  "Jewish Argument as Sociability."  Language in Society,vol. 13, no. 3, Sept. 1984.  pp. 311-335. 

Schiffrin, Deborah.  "Conversation Analysis."  In Linguistics: The Cambridge Survey, IV: Language: The Socio-Cultural Context,Frederick J. Newmeyer, ed.  Cambridge: Cambridge U. Press, 1988.  pp. 251-276. 

Schiffrin, Deborah.  "Everyday Argument: The Organization of Diversity in Talk."  In Handbook of Discourse Analysis, Vol. 3: Discourse and Dialogue,Teun A. Van Dijk, ed.  London, Academic Press, 1985.  pp. 35-46. 

Smith, Neil  &  Deirdre Wilson.  "Introduction" (to a special issue on Relevance).  Lingua 87 (1992).  pp. 1-10. 

Spekman, Nancy J.  "Verbal Communication and Role-Taking: An Analysis of the Use of Deictics."  In The First Delaware Symposium on Language Studies: Selected Papers, Robert J.  Di Pietro, William Frawley & Wedel Alfred, eds.  Newark: U. of Delaware Press, 1983.  pp. 168-86. 

Sperber, Dan  &  Deirdre Wilson.  Relevance: Communication and Cognition.  Cambridge, Mass.: Harvard U. Press, 1986. 

Streeck, J.  "Embodied Contexts, Transcontextuals, and the Timing of Spech Acts."  Pragmatics 8 (1984).  pp. 113-37. 

Tannen, Deborah.  "New York Jewish Conversational Style." International Journal of the Sociology of Language 30 (1981).  pp. 133-49. 

Tannen, Deborah.  "When Is an Overlap Not an Interruption? One Component of Conversational Style."  In The First Delaware Symposium on Language Studies: Selected Papers, Robert J.  Di Pietro, William Frawley & Wedel Alfred, eds.  Newark: U. of Delaware Press, 1983.  pp. 119-129. 

Tannen, Deborah.  Talking Voices: Repetition, Dialogue, and Imagery in Conversational Discourse.  Cambridge: Cambridge U. Press, 1989. 

Tannen, Deborah.  "Remarks on Discourse and Power."  In Power through Discourse, Kedar Leah, ed., Norwood, NJ: Ablex, 1987.  pp. 3-10. 

Tracy, Karen.  "The Effect of Multiple Goals on Conversational Relevance and Topic Shift."  Communication Monographs, vol. 51, Sept. 1984a.  274-287. 

Tracy, Karen.  "Staying on Topic: An Explication of Conversational Relevance."  Discourse Processes, vol. 7 no. 4, Oct.-Dec. 1984b, pp. 447-464. 

Tracy, Karen.  "Conversational Coherence: A Cognitively-Grounded Rules Approach."  In Sequence and Pattern in Communicative Behaviour, Richard L. Street & Joseph N. Cappella, eds., London: Edward Arnold, 1985.  pp. 30-49. 

Werth, Paul.  Focus, Coherence, and Emphasis.  London: Croom Helm, 1984. 

Werth, Paul.  "The Concept of 'Relevance' in Conversational Analysis."  In Conversation and Discourse, Paul Werth, ed., NY: St. Martin's Press, 1981.  pp. 129-54. 

Wilson, Deirdre  &  Dan Sperber.  "On Grice's Theory of Conversation."  In Conversation and Discourse, Paul Werth, ed., NY: St. Martin's Press, 1981.  pp. 155-178.