How to dispute with precision using Besѣda®

Content

Design

The goal for the speech genre “dispute” is to provide the tools for how to argue.

Issue

Requirements

On the one hand provide structure and on the other hand - ensure it is accessibile. If the rules are flexible than the coherence of the argument relies on the cooperativeness of the participants. If the rules are strict than the coherence is enforced. A balance needs to be achieved which accounts for the context.

Techincally, the implementation needs to be similar to and compatible with the implemenetation of the already existing genre “rapport”.

Limitations

By design, there is no method to evaluate the merits of the participants’ arguments. One reason is related to the technical feasability - it is computationally extremely complex to evaluate. Which means that it would need some sort of artificial intelligence (AI).

That is relevant for the second reason why arguments are not evaluated by the application on their merits - respecting the privacy of the participants means that their conversation cannot be sent externally from their device for processing by AI.

There is no reward for succeeding to convince the interlocutor nor punishment for submitting to the opposition. Any reward that does exist would be related to the performance when using the tools.

The argument may end with both parties continuing to hold different beliefs/stances/convictions - that is an acceptable outcome.

Deduction

What is the genre “dispute”? The author is breaking the tradition of this “making of” series of blog posts by presenting first a personal experience instead of academic research. Because the end goal is a very flexible and practical application, it should be possible to relate it to everyday experiences.

The author’s behaviour was the object of inquiry. The arguer introduced the central issue, took a Negative position in relation to it and made the initial constructive round stating a few arguments to strenghten the position. Then during his turn the author stated that he had already been thinking about and planned to change which aligns with the stance of the arguer. The topic concluded. Would you consider that to have been an argument? The popular understanding of an argument would lead you to conclude that it was not.

The author makes the following claims about the core attributes of the genre “dispute”:

Here’s why that simple experience proves those core attributes of the genre.

Orientation

Before the meeting, the conversationalist anticipates that the interlocutor is going to respond by denying and more generally speaking - opposing. For example, it has been proven by researchers that the response to complaining is often to admit to having done the action which is complained about but to characterise the action as reasonable to do or that there is a respectable justification why it was done that way and not the expected, conventional, “proper” way. The complainee is accountable but denies culpability. Accusation also seems to be responded to similarly.

Consequently, the conversationalist who anticipates objection prepares for the interaction according to the social conventions how to handle such situations - the conversation genre “dispute”. The conversationalist would not put the effort to prepare as much and would not provide such detailed explanations except if s/he is expecting opposition. Because it is more efficient to add to the conversation when necessary than it is to expand unilaterally the conversation according to some subjective considerations. Essentially this efficiency principle is what Paul Grice called the “Quantity maxim” and what Sacks and Schegloff called the “oversuppose and undertell” convention.

By preparing to provide explanations and justifications, the conversationalist is preparing to breach the Gricean maxim of Quantity. An intentional violation of the maxims creates an implicature - in this case that the speaker is committed to the stance.

The genre is used regarldess of whether during the meeting, the interlocutor does make the anticipated opposition or not.

Change

This goes to the core of language use - why do people externalise their feelings and thoughts vocally? It is useful to remind the reader of previous work by the author about epistemic and pragmatic actions: the epistemic action is to interpret the result of a change, without a predetermined expectation. It may be passive observation or it may be active trial-and-error. The pragmatic action is to make a specific change in the state of affairs with an expectation what the end result would be as well as monitoring that the end state matches the intended state.

People externalise their feelings and thoughts vocally to trigger some change - adjusting the state of the body (internal) and/or addressing a particular interlocutor (external). In the context of the genre “dispute” where there is anticipated/interactive opposition it is certain that the intended change is external because in preparation for the conversation the arguer would have had to have dealt with internal opposition and the consequent adjustment.

In everyday conversations often the intention of the arguers is to change the behaviour, feelings, thoughts of the interlocutor or her/his relation to other people. This is a very broad definition of argument. It is not exclusively logical, it is a competitive style of conversation with the fundamental goal to change.

The author is not claiming that the only goal of an argument is to change the interlocutor - the argument can be about evaluating the truthfulness or ethicality of an idea, planning, etc. But any such goal is in addition to the intention for change.

Here are other everyday examples of arguments - a police officer requiring a citizen to show his identification documents (epistemic inquiry), a child who demands to be given attention (pragmatic protest), etc.

Research

An argument, according to Ian Hutchby, is a simple action-opposition pair. It is when the opposition gets treated as the action unit of a further action-opposition pair that the conversation turns into a dispute. Besѣda® is inspired by this technical definition of dispute which should not be confused with the legal term.

The ancient greek philosopher Aristotle (in his works “Topica” (Topics) and “De sophisticis elenchis” (On Sophistical Refutations)) expressed the concept of a conversation which has rules that regulate the logic implemented by the participants. Later, during the Middle Ages, there were similar dialogue games called Obligationes. For more than 2 millenia of human generations, arguing using logical thinking has been (re)mastered: for political participation, academic research, judicial trials, etc.

A debate is a more formal genre than argument - it is usually performed in front of an audience, in some cases there is a judge or multiple judges. With formal debates the goal of such debates is to persuade the audience (the meaning of the classic term by Aristotle - “rhetoric”).The participants have pre-allocated stances to which they are expected to stick. If a participant were to engage in genuine information seeking they would appear incompetent. Rather the standard procedure when there is some uncertainty, is for the participant to state their interpretation and continue without feedback.

Everyday arguments similarly often end with both parties holding the same stance they started in. However, with informal arguments there is more exchange between the sides and they are more flexible (often employ multiple subgenres).

The sophistication of the dispute can be discerned by the complexity of the arguments. Peter Suber distinguishes two domains of arguments according to whether the premises can be argued against. Mathematics, for example, has stipulated axioms which are not arguable. And where the premises are arguable - philosophy, ethics, politics, etc, the quality of the argumentation can be evaluated and classified into one of 4 stages: (stage 1) assertion without evidence, (stage 2) one-sided argument, (stage 3) two-sided argument and (stage 4) responsive argument. While this classification is primarily for teaching purposes it’s approach is neutral enough that it applies regardless of the validity, truth and soundness of the arguments.

Paul Lorenzen transformed the study of logic by focusing on the socially constructed reality of argumentation and introduced game semantics to the field of logic. He called it “dialogic logic”.

A dialogue game usually includes a set of communicative (classes of) acts and rules which determine the affordance of acts for the stage of the conversation (protocol) and possibly - when the conversation completes. Simon Wells and Christopher Reed provide a visual overview of the history of dialectic games in their definition of a domain specific language for such games.

According to the postmodern, relativistic standard of rationality:

an argument is correct if it is used at some stage in a dialogue to contribute to the overall goal of that type of dialogue by fulfilling the requirements for the kind of move an argument is supposed to be at that stage.

Scott Jacobs suggests the “zone of agreement” model of argument: the goal is to persuade the listener that the speaker believes certain statements and opinions. Interestingly, in this model the conclusion of the argument is the establishment of a common ground of the territory of knowledge. This is the opposite of a formal argument where estalishing the common ground is the beginning.

The consequence is that those arguments then become the boundaries of the zone. The end result is to allow the conflicting standpoints to coexist by aligning them to maximise each party’s satisfaction of their needs. This can be contrasted to the formal arguments where all contradictions must be resolved or accounted for.

Hutchby claims that any action is, in principle, open to being treated as arguable. This rule has significant implications: the focus spaces of the conversation genres can be layered on top of each other! This is also compatible with the concept of the communicative budget - changes to one genre affects all the other genres.

Subgenres

Here is the list of classes of genres which the designer identified:

For examples see the section below.

Deliberation is considered by many researchers to be a subgenre of argument. However because the deliverable (the tangible outcome from the conversation) is different from argument mapping the architect has decided to separate it as another genre - “plan”. Furthermore, in everyday situations, it is during the conversation that relevant circumstances are revealed or changed which forces the conversationalists to add, delete and modify options. Formal deliberation starts from fixed circumstances/principles.

Implementation

ContextRolesResourcesProgressionPositioning
InformalAsymmetric
(protagonist<->antagonist, examiner <-> examinee, etc)		one-way postures + specific tools3 stages
Confrontation -> Contestation -> Conviction		Two types:
initial/concluding stance (for, against, undecided)
tactical (support, against, undercut)

Progression

The designer considered the 4 stages provided by the leading logisticians Frans Van Eemeren and Rob Grootendorst - the “confrontation” stage, the “opening” stage, the “argumentation” stage, and the “concluding” stage. However, that and similar other definitions of the process seem to be more about the cognitive procedure than the competitive process of argumentation. The author presents the following abstract labels for the cognitive procedure: “Awareness”, “Interpretation”, “Meaning” and “Memorisation”.

The interactive stages are 1. Confrontation -> 2. Contestation -> 3. Conviction

The conviction is the overall, final evaluation of the whole of the argumentation, the concluding stance. This word was selected because it reflects the unilateral action of evaluation - this is an informal context.

Resources

The initial focus of development is on the fundamental genre “dispute”.

Frans Van Eemeren and Rob Grootendorst use the same classification of speech actions which was made by Searle: assertives, directives, commissives, expressive and declaratives.

The resources in Besѣda® are the postures of the V.I.C.T.O.R.I.A. framework (the publication of which is pending). You can see some definitions in the analysis section.

Structure

As is the case for the genre “rapport”, the conversation of the genre “dispute” is visualised with the goals of providing more clarity and enabling better coordination. For example: whereas usually deixis (“this”, “here”, “then”) relies on implicit context, using visualisation makes it clear what is being referred; there is no need to restate one’s own or another person’s previous utterance to refer to it.

The visualisation was initially called information mapping and later - argument mapping.

The developer used an open source library to visualise the arguments. Construction transformations are “addition” and “permutation”. The transformation “deletion” is allowed only for relations/warrants (according to Toumlin).

Rules

One turn allows the participant to make a repair (ask the interlocutor to repeat or for a specific clarification) or reaction as well as one action.

The move has a set amount of time (10 seconds) which makes the participation fair. Besides this each subgenre modifies the basic rules. For example, in “quarrel” the turn is only one pragmatic action; in “teaching” the teacher role has additional moves per turn.

Commitment rules

Before the meeting, the conversationalist can create a visual diargram and write the substance of their arguments in their chosen language. This is the so called “dark” commitment store.

During the meeting, there are no rules about what the conversationalist can commit as the semantic content of the utternace is in no way analysed.

Structural rules

Stephen Toumlin provides an influential formula for arguments: three fundamental parts (claim, grouns, warrant) and three optional (backing, qualifier, rebuttal). In Besѣda®, the fundamental parts are included and some genres may require the optional parts.

The positioning can be one of “for”, “against” or “undercut” (it is about the relations/inferences/warrants between arguments - deny the relevance, doubt the importance and/or provide alternative relation).

Termination rules

Complying with the limitations, the main genre does not mandate termination rules - the participants decide themselves when to end. A subgenre may impose some rules for the termination.

In practice, it is usually only one of the parties that puts the conclusion into words, but if the other party does not accept this conclusion, no resolution has been achieved. This is not tracked.

Conversation Analysis

Considering the implementation details, it is appropriate to apply them analytically to some conversations.

Starting with two redacted excerpts from Excerpt from Jackson & Jacobs, 1980:

"Could you do X?" [genre "rapport"] posture=Ask for favour
"No." [genre "dispute", initial stance] posture=Repudiate -> subgenre "critique"
"Why?" [undercut] posture=Counter
"Just wait" [statement in support of "No."] posture=Instruct. "I'll be done in a second" [statement in support of "Just wait."] posture=Frame
# Stage 1 # Confrontation
1 S: "What are you doing?" [epistemic action] posture=Ask for justification -> subgenre "enquiry"
2 C: "Making coffee." [treats the question literally as in a cooperative conversation; fails to give a justification]
3 S: "No." [initial stance]
4 C: "Yes." [initial stance]

# Stage 2 # Contestation
5 S: "NO!" [counter-statement] posture=Complain
6 C: "YES!" [counter-statement] posture=Complain
7 S: "I'm warining you" [] posture=Threaten
8 C: "What?" [repair]
9 S: "It's four thirty AM..." [statement]

At line 7 the move is actually the threat “You’re going to regret this.” but it is formulated in a micro-political way. This example serves additionally as proof that we cannot rely for the speech action definitions on the language itself.

In September 2024, the author analysed a conversation which is relevant to this genre therefore a revisit is due. The analysis is largely sound and still holds true. However, because it is the competitive genre “dispute”, there is no first and second pair parts as with the collaborative genre “rapport”. The postures are also different.

1. Parent: Tell me about your day. [stance] posture=Assign -> subgenre "examination"
2. Parent: What did you learn? [supporting statement] posture=Discriminate
3. Child: We went on an adventure in the park.= [statement] posture=Discriminate
4. Parent: Yiea:h, an'- an'- tell me about it. [supporting statement] posture=Assign
(1.5)
5. Child: Uh:m_ .h It was fu:n?, [statement] posture=Discriminate
(0.5)
6. Parent: No. [counter statement] posture=Repudiate
(.)
7. Parent: You're gonna h(h)afta do uh lot better than that. [] posture=Rectify

Finally, an interesting intervention of a smoker, provided in the book “Nonviolent Communication A Language of Life ”:

Al: Burt, I know we've talked about this a dozen times [genre "rapport", pre expansion] posture=Concede
    but listen [topic transition] posture=Counter
    I'm scared your damned cigarettes are going to kill you! [genre "dispute"] posture=Frame -> subgenre "persuasion"
    You’re my best friend, [supporting statement] posture=Discriminate
    and I want you around for as long as I can have you. [supporting statement] posture=Demand
    Please don’t think I’m judging you. I'm not [counter statement] posture=Threaten
    I'm just really worried. [supporting statement] posture=Discriminate

The beginning is of particular interest because it includes a transition from the genre “rapport” to the genre “dispute”.

User manual

Before the meeting

You can navigate to “Prepare strategically”: there is a new interface that you can use to prepare your assertions, statements, explanations, etc and save them. Saving the map stores it locally on your device, not on the Besѣda® server.

Please note that this map is not going to be seen by your interlocutor - it is private to you. If you chose to, you can copy and paste the contents during the dispute so that the interlocutor sees the content.

Similar to the previous version of Besѣda® - for your orientation you can add topic labels that you intend to initiate and save them. During the meeting, you can use the labels to set the names for the focus spaces.

During the meeting

Select the dispute genre (if the interlocutor selected it your application is updated automatically).

How to progress

There are three stages: 1. Confrontation -> 2. Contestation -> 3. Conviction. You can interact with the image below.

You decide when to proceed to the next stage by clicking on the corresponding element from the image.

Depending on the subgenre you may skip stages. The mechanism to select the subgenre is simply by using a specific posture as the first move of the Confrontation stage. If you did not make the first move and you prefer not to participate in the selected subgenre you can create a new focus space (by clicking the button ”+ New Focus Space +”) or even a new dispute genre entirely (by clicking the button ”+ New Genre +”).

How to take a position

Before you can say your message, you need to select a position. If you initiated the topic the position is automatically selected and you can proceed to using a resource.

If you did not initate the topic, you need to select one of the existing statements (represented graphically as a rectangle) - three options appear (currently only two, the third one is in the roadmap):

Selecting one is going to remove the other two and bring to your attention (by scrolling the view) the resources interface.

How to use resources

Finally, you can select what your speech intention is. After you select it your microphone is switched on and the other side cannot interrupt you. Be mindful that the microphone is switched on for 10 seconds (this is probably going to change or be more flexible in the future - you can request such a feature!).

How exacly to select a resource is not explained in this guide because the V.I.C.T.O.R.I.A. framework on which it is based is not yet public. You are going to find out when you purchase access to the application. Good luck!

When you are listening, you are prompted by Besѣda® if what is being said by the interlocutor is clear (you can ask for clarification, paraphrase or directly ask for their justification).

When the interlocutor completes their move (limited to 10 seconds), Besѣda® provides some relevant suggestions how to respond - they are important because they contain 2 postures which you can use to make a concession. Normally, while in the genre “dispute” you do not have any other way to concede besides this suggestion. No public image of this prompt.

Roadmap

Some of the planned features which are going to be part of the application are:

To see the roadmap check out https://roadmap.andonovi.net. You are welcome to request features.

Document history

Read Next

How to combine genres in Besѣda®

This new feature enables mixing and matching speech genres for the greatest flexibility.

What are genres in Besѣda®?

Introducing conversation genres - how they are defined and implemented. This higher order structure allows participants to engage easily in more complex conversations.