Title: Between understanding and misunderstanding


Author: Darinka Verdonik


Affiliation: University of Maribor, Faculty of Electrical Engineering and Computer Science, Smetanova 17, SI-2000 Maribor, Slovenia


Postal address: FERI, Darinka Verdonik, Smetanova 17, SI-2000 Maribor, Slovenia


Phone: +386 2 220 7 221





Bio-note :

Darinka Verdonik is researcher at the Centre for Language Technologies at the Faculty of Electrical Engineering and Computer Science, University of Maribor. Her research interests include discourse analysis and language resources. She received her Master’s degree in 2003 from the Faculty of Education, University of Maribor, the topic was misunderstandings in communication, and her PhD in 2006 from the Faculty of Arts, University of Ljubljana, the topic was pragmatic annotation of corpora for the needs of speech-to-speech translation technologies.

Keywords: misunderstanding, understanding, communication, conversation, coherence model



In this paper, we discuss borderline examples of (mis)understanding where it is not clear whether or not a misunderstanding has occurred, whether or not communication was successful, and where the participants do not try to negotiate an understanding, even though different interpretations are very likely to exist. By analyzing real data, we point out various types of such borderline examples of (mis)understanding, attempt to analyze their source and explain why they are a normal constitute of process of coming to an understanding. Using discourse comprehension theory, we define the level of propositional strategies, local coherence strategies, strategies for the use of knowledge, and interactional and pragmatic strategies as the main sources of reduced understandings. In spite of the fact that there is no complete understanding, and, consequently, some level of differences in understanding can be perceived as ‘normal’, we discuss other possible reasons why discourse participants do not explicitly negotiate an understanding.

1 Introduction


Misunderstanding (and communication failure) has been an important research topic over the last few decades (e.g., Zaefferer, 1977; Thomas, 1983;  Dascal, 1987; Schegloff, 1987; Weizman, Blum-Kulka, 1992; Weigand, 1999; Bazzanella, Damiano, 1999; Bosco et al., 2006). However, in recent years, it seems that interest in this subject has diminshed – we find very few studies on misunderstanding after 2000. Nonetheless, there are still many open questions; one of them is borderline or non-standard cases of (mis)understanding that we will deal with in this paper.


Most of the authors who have tried to shed mor e light on misunderstandings also present various classifications of misunderstandings. Zaefferer (1977) proposes a very structural taxonomy: in his view, misunderstandings can be misperceptions or misinterpretations, and each of these types of misunderstanding can affect the phonological, syntactic, semantic or situational level of interpretation, plus they can affect the illocutionary force, the propositional content, or both. Thomas (1983) introduces a pragmatics-based classification of communication failure: at level 1, she classifies failure to understand the proposition correctly, and at level 2, failure to understand the intended pragmatic force. The second level she calls pragmatic failure, and further claims that there are two types of it: pragmalinguistic failure and sociopragmatic failure (which – in the context of intercultural communication – refers to the social conditions placed on language use in different cultures). Dascal (1987) presents two modes of understanding (and consequentially two levels where misunderstanding can occur): comprehending and grasping, where the former means being able to understand the pragmatic level (determining the sentence meaning, the utterance meaning and the speaker’s meaning), and the latter means being able to detect what can and what can not be said in a given situation, i.e., being able to determine what rules (of social interaction) are to be followed. Another classification is presented by Weizman and Blum-Kulka (1992). They distinguish individual (I-level) and collective (we-level) misunderstandings, and each of these is further classified into three dimensions: I-level into aboutness (propositional content), point and mode; and we-level into aboutness (topic), purpose and mode. A more recent taxonomy (Bosco et al., 2006) builds on relevance theory and distinguishes failure of the expression act, failure of the actor’s meaning, and failure of communicative effect.


Along with the classifications, analyses of sources of misunderstandings (e.g. Schegloff, 1987; Weigand, 1999), ways of handling misunderstandings (e.g., Bazzanella, Damiano, 1999; Bosco et al., 2006), cross-cultural misunderstandings (e.g., Thomas, 1983) etc. have also received considerable attention. Weigand (1999) presents an important contribution to research on misunderstanding in her paper on the standard case of misunderstanding, in which she explicitly points out that the issue is not just understanding , but a process of coming to an understanding. However, she maintains a distance between understanding and misunderstanding as separate points in a model of communication ‘action games’. According to her findings, the standard case of misunderstanding is defined by the following features:

- understanding is partially or totally deviant from what the speaker intended to communicate,

- it is a cognitive phenomenon belonging to the interlocutor,

- the interlocutor who misunderstands is not aware of it,

- it is not a cognitive act, but represents an (in)ability of the hearer,

- misunderstanding is normally corrected in ongoing discourse.

Non-understanding, according to Weigand (1999) – as well as other researchers – is not the same as misunderstanding. In her late work, Weigand (2000; 2002a, b; 2004) developed a new model of communication called Dialogic Action Game, to which we refer more specifically in the next section.


Dascal (1999) points out many open questions about misunderstanding, among them the question of ‘non-standard’ cases of  misunderstanding and the logic of misunderstanding: “Is it a binary phenomenon or does it admit of gradation?” (Dascal, 1999: 756)


In this paper, we will not the discuss standard, typical cases of misunderstanding or their classification, since a lot of excellent work on this topic has already been done; we will concentrate instead on an area that is still an open issue, i.e., the borderline examples of (mis)understanding where it is not clear whether or not a misunderstanding has occurred, whether or not communication was successful, and where the participants do not try to negotiate an understanding, even though different interpretations are very likely to exist. By analyzing real data, we will point out various types of such borderline examples of (mis)understanding/reduced understanding and try to analyze their source. In the discussion we will try to point out some possible answers as to why they are a normal constitute of the process of coming to an understanding; however, it is beyond the scope of this paper to provide an analysis and a full explanation as to why these potential occurrences of misunderstanding are not dealt with at the conversational level.


The data we used for the analysis can be classified as persuasive dialogue, “where the aim of one party is to rationally persuade the other party to come to accept a proposition as true” (Walton, 2007: 89). This type of dialogue is of interest to pragmatic studies (eg., Kamalski et al., 2008; Dafouz-Milne, 2008) as well as to computational linguistics and artificial intelligence, especially to dialogue applications in autonomous agents and other computer communication systems (eg., Rahwan et al., 2006; Amgoud, Hameurlain, 2007; Walton, 2007). Methodologically, we will follow a theory (van Dijk, Kintsch, 1983) that builds on mental models, a concept from cognitive science (eg., Johnson-Laird, 1980). “Mental models are psychological representations of real, hypothetical, or imaginary situations.” (Johnson-Laird et al., 1998)  


2 Understanding


Many authors agree that the speaker’s and the hearer’s understanding of the same utterance and/or discourse can never be identical since each of them understands it from his or her own, personal point of view. Below I present different theories that support such a view.


Van Dijk and Kintch (1983) present a cognitive model of discourse comprehension, where subjective representations of the text and the situation are constructed. Their major assumption is that discourse processing “is a strategic process in which a mental representation is constructed of the discourse in memory, using both external and internal types of information, with the goal of interpreting (understanding) the discourse” (van Dijk, Kintsch, 1983: 6). According to them, understanding includes a construction of a mental representation of the text (the so-called textbase) and a representation of the situation (the so-called situation model). The textbase “is the semantic representation of the input discourse in episodic memory” (van Dijk, Kintsch, 1983: 11). The authors define the textbase in terms of propositions and relations among propositions. The situation model on the other hand “is the cognitive representation of the events, actions, persons, and in general the situation, a text is about” (Kintsch, van Dijk, 1983: 11-12). Such a distinction between situational and text representation corresponds to the tradition of distinguishing between ‘meaning’ and ‘reference’, known in philosophy and linguistics. The motivation for this is the acknowledgement that “to understand text we have to represent what it is about. If we are unable to imagine a situation in which certain individuals have the properties or relations indicated by the text, we fail to understand the text itself. If we do not understand the relations between the local facts and the global facts to which the text refers, we do not understand the text.” (van Dijk, Kintsch, 1983: 337) The construction of the textbase and situation model is supervised by a control system, which “is fed by specific general information about the type of situation, type of discourse, overall goals (of the reader/listener and of the writer/speaker), by the schematic superstructure and the macrostructures (gist, themes) of the text, or by plans in the case of production” (van Dijk, Kintsch, 1983: 12). Their model also includes knowledge, both episodic and more general and abstract, as well as other presupposed cognitive information, such as beliefs, opinions, and attitudes. Their central interest focuses on on-line processing strategies of comprehension. These strategies are: propositional strategies (i.e., strategic construction of propositions), local coherence strategies (i.e., the establishment of meaningful connections between successive sentences in a discourse), macrostrategies (which infer macropropositions from the sequence of propositions expressed by the text, and several levels of macropropositions form the macrostructure, i.e., the theme, the topic, or the gist), schematic strategies (used to activate a relevant superstructure/schematic structure, which is an overall form that organizes the macropropositions, e.g., narrative schema), production strategies (i.e., a component of discourse production, a semantic discourse plan), and other strategies, such as stylistic, rhetorical, processing of nonverbal information, conversational strategies etc., which the authors merely mention. The model of discourse comprehension was later supplemented by the context model (van Dijk, 2006; 2008; 2009). Just as discourse participants build a representation model of the text and the situation the text is about, they also build a cognitive model of the communicative event itself. To van Dijk, contexts are therefore “not ‘objective’ or ‘deterministic’ constraints of society or culture at all, but subjective participant interpretations, constructions or definitions of such aspects of the social environment” (van Dijk,  2006: 163).


Van Dijk and Kintsch’s (van Dijk, Kintsch, 1983) cognitive model theory of discourse comprehension presents an effective and flexible theoretical framework for analyses of discourse processes, and we find it an efficient tool for the explanation of the examples of (mis)understanding that we present in this paper.


However, other authors build from a similar general assumption that the processes of understanding and discourse contexts are subjective interpretations. We mention only those whose work has had impact on our analysis.


Just as van Dijk and Kintsch, Weigand (2000; 2002a; 2002b; 2004) also puts human beings in the center of discourse and claims that dialogue can not be described in terms of rules, but strategies. She uses the notion of ‘dialogic action games’ for cultural units, in which we behave and interact. But while van Dijk and Kintsch (1983) try to explain the cognitive processes of discourse comprehension, Weigand’s model of dialogic action game concentrates on social interaction. She defines ten fundamental premises of the dialogic action games (Weigand, 2000: 7), among them the premise that language can not be separated from human beings, that humans are oriented towards purposes and needs, that language use accepts misunderstandings, that not everything is (and can) be said explicitly, etc. Furthermore, she defines a set of basic methodological principles of how to behave in dialogic action: the action principle, i.e., “taking the communicative actions means pursuing specific dialogic purposes with specific dialogic means” (Weigand, 2000: 9), the dialogic principle proper, i.e., “rational basis which determines the internal interdependence of the initiative and the expected reactive action” (Weigand, 2000: 10), the coherence principle, i.e., “the joint attempt by the interlocutors to understand and to give sense to what is going on” (Weigand, 2000: 12), and a set of corollary principles: the principle of rationality, of suggestion or of convention, sequencing principles, rhetorical principles, and emotional principles. More explicitly than van Dijk and Kintsch (1983), Weigand (2000) points out that the fact that human beings are in the center of dialogic action inevitably implies differences in understanding, since each human being has a different cognitive background and different personal experiences. This view is very important for our work in this paper since it points that the phenomena which we analyze are not extraordinary or disturbing, but a normal part of communication.


In his discussion of context, Verschueren (1999) also points to the fact that discourse participants interpret discourse from different points of view. In his figure of context, the utterer and the interpreter (functional entities for real-world people) are presented as focal points (similarly, in Weigand’s model human beings are in the center of the dialogue). The utterer and the interpreter (as far as they are different human beings), according to Verschueren (1999: 75-7), inhabit different physical, social and mental worlds. “/T/here is overlap between those worlds, but even elements of common background from the overlapping areas may look different because the perspective always differs, at least slightly. Common ground, in other words, is almost never really common.” (Verschueren, 1999: 77) Following Verschueren’s ideas, the author of this paper carried out a series of studies on misunderstandings (Verdonik, 2003; Verdonik, 2005). A similar framework is presented in Ferbežar and Stabej (2008), who define comprehension as creation of meaning that depends on factors related to the comprehender, as well as on social and textual factor. They schematize comprehension as gradational overlapping of meaning created by the speaker and the comprehender. Vogel (2003) also focuses on the process of (mis)understanding in the classroom, explaining the process of interaction from the pragmatic and relevance theory point of view.


3 Data and method


The general aim when collecting the data for our research on (mis)understanding was to obtain audio and video recordings of real life discourse in which misunderstanding or reduced understanding have occurred. At the time when this study was carried out, no speech corpus of Slovenian language was available, so we chose the TV as a source of data. Since the analysis was qualitative only, we attempted to collect a small number of conversations where a reduced level of common understanding or misunderstanding could be noticed immediately. The most appropriate for these purposes were interviews in the evening broadcast news show, in which the interviewer generally spoke with two interviewees who argued the opposite points of view about the topic of the debate – we expected that the opposition of opinions would encourage misunderstandings. The approximate length of these interviews was about 10 minutes, which was also a convenient discourse length for our purposes. However, we found that obvious misunderstandings were very rare. We recorded the interviews (more or less regularly) for a period of eight months, and selected four of them in which we immediately suspected that at some point a reduced understanding between the active discourse participants may have occurred. The topics of these four debates were: professionalization of the army, new Constitutional Court judges, the border between Slovenia and Croatia, and the change of the constitutional provision about the term of office of the president of Slovenia.

The specific characteristic of the data, which is very important for our analysis, is that the main and general aim of the discourse was to satisfy the expectations of the TV audience, that the interviewees are mostly opponents, and that discourse is strictly time-limited – all this may mean that it is not necessarily the aim of the discourse participants to come to a common understanding; and often there is not enough time for that.


In the first step of the analysis, we analyze each turn in every debate in qualitative terms, trying to answer the following questions: (1) did the discourse participants understand each other adequately, (2) is a specific utterance or turn too ambiguous/unclear/underspecified and therefore encourages different interpretations. In this way we extract the segments of discourse for further analysis.


In the second step of the analysis , we analyze these discourse segments using van Dijk and Kintsch’s (1983) discourse comprehension theory. We try to explain in what way the extracted discourse segments present reduced, partial understanding, and then explain how/on what level this happens. The second part of the analysis is presented in the following section.


4 Analysis of different types of reduced understanding


In our analysis, we focus only on the level of discourse comprehension where we notice a source of vagueness and ambiguity and which may therefore lead to reduced understanding. This is not an analysis of the entire process of comprehension in each of the analyzed discourse segments, which would be needed in order to understand the whole comprehension process in each example.


4.1 Propositional strategies


First, we focus on the propositional strategies. We will discuss two different examples of reduced understanding stimulated mostly on the level of propositional strategies.


4.1.1 Propositional ‘open ends’


W e label the first example propositional open ends. This refers to those elements of propositions that are more open to different interpretations than others; however, discourse participants do not pay special attention to their interpretation.


Example 1:

Context: B is a journalist, C is a member of parliament and K is the former president of the State Council. C has proposed that the following sentence in the Constitution should be changed [translation]: “The President of the Republic is elected for a period of five years, but may serve no more than two consecutive terms of office,” 1 by erasing the term ‘consecutive’. The reason for his proposal is that the current wording means that the same person can serve two consecutive terms of office, and while he or she may not run for a third consecutive term, he or she is entitled to run for the position of president again after another president has served a term. C disapproves of this possibility. On the other hand, K argues that there is no need to change the constitution the way C suggests.


K5:  tu mate v bistvu { deset držav je takšnih ki omogočajo večkratno izvolitev = fact 1}

K5: here you actually have { ten countries that allow multiple terms of office = fact 1}

B9: no

B9: well

K6: {s tem da mamo mi prepoved da zaporedoma ne more bit več kot dvakrat = fact 2}

K6: {considering that we have a restriction that no more than two consecutive terms of office may be served = fact 2}

B10: {gospod C vam takoj lahko navede deset držav ki tega ne omogočajo = fact 3}

B10: {mister C can of course list ten countries that do not allow this = fact 3}

K7: ne ne ne ne

K7: no no no no

C7: dvajset

C7: twenty

K8: kar navedite jih

K8: well list them

C8: dvajset

C8: twenty

K9: ne

K9: no


Van Dijk and Kintsch (1983) postulate facts as referents of propositions. In this fragment, three facts are discussed as annotated by brackets. The rest of the text are self-repairs (“here you actually have”), discourse markers (“well”), and a discussion about fact 3, contributing no new facts, but merely expressing an opinion and supplementing fact 3. Fact 3 can be analyzed 2 as:



PREDICATE – list, MOOD can

OBJECT (fact 1):

(SUBJECT countries, ATTRIBUTE1 ten


OBJECT terms of office, ATTRIBUTE2 multiple)


The discussion about fact 3 centres on two questions:

1. whether or not it is true that C can do what he claims he can (and therefore whether or not fact 1 is true)

2. what is the correct attribute 1 in fact 1


We will focus on the second question. It is very likely that attribute 1 is interpreted in different ways: it can be understood literally (‘ten/twenty’) or it can be understood in a more loose way (‘many/a lot/more than ten’...). Both interpretations are possible; therefore, if one participant (e.g., a member of the audience) interprets attribute 1 literally while another does not, common understanding is a bit reduced. Nonetheless, this does not have an important effect on the success of communication. The participants in the conversation did not pay special attention to this particular issue, for K and C it was just a way to convince the audience. What is more, if K, C and B had put more effort into clearing up whether they meant attribute 1 literally or not, it would have disturbed the process of argumentation which is in the center of their interest.


This suggests that propositions are (sometimes) necessarily open to different interpretations, so that discourse participants are able to have a pragmatically successful and effective conversation. It may be that some elements (not necessarily always the attributes) of propositions are less important for a common understanding than the others. It may also be that some other strategies of discourse comprehension, or even some other strategies (e.g., stylistic, rhetorical, processing of nonverbal information, conversational, context strategies etc.) are more important than (or may be superior to) propositional strategies.


4.1.2 Propositional vagueness


We label t he second example propositional vagueness. Every proposition contains a certain amount of implicit information. How much information is left implicit depends on the speaker’s estimation of the hearer’s knowledge, experiences, attitude towards what is talked about, familiarity with what has already been discussed... However, the speaker’s estimation may be improper and he or she may leave too much (or too little) information implicit. This may be done unintentionally, e.g., as a consequence of the speaker’s absentmindedness, low speaking abilities, or deliberately, e.g., in order to confuse the hearer. Too much implicit information can be seen as propositional vagueness and can encourage different interpretations. In the example from our data, the main reason for propositional vagueness is underspecified elements of proposition which would normally have to be filled up:


Example 2:

Context: The topic in the following example is the reform of the Slovenian army which involves switching from conscription to a professional military force. P argues that this reform should take place as soon as possible. B agrees that there should be a reform, but it should be slower and should follow the proposal that he developed himself.



5P-1:  to ne dr ži

5P-1: this is not true

5P-2: poglejte {analiza ki jo ima vlada in mislim da so jo pripravljali tudi zunanji strokovnjaki govori o tem da se bojna pripravljenost slovenske vojske ne bi zmanjšala in učinkovitost tudi če bi prepolovili profesionalne enote glede na sedanjo glede na sedanjo naborniško sestavo = fact 1}

5P-2: look {the analysis that the government has and I think that external experts were consulted as well shows that the military readiness of the Slovenian army would not be diminished and neither would be its efficiency even if the professional army units were half the size of the present conscript army = fact 1}

5P-3: to se pravi {vemo da imamo sedaj približno šejesetprocentno popolnjenost = fact 2}

5P-3: this means {we know that now we have approximately sixty percent fill up = fact 2}

5P-4: to se pravi {če bi samo polovico od tega = fact 3}

5P-4: this means that {if we could only half of this = fact 3}

5P-5: {to pa lahko naredimo res zelo hitro = fact 4} pospešeno bom reko da ne bom uporabljal izraza zelo hitro

5P-5: and {this we can really do very quickly = fact 4} or rather rapidly to avoid the expression very quickly


In this example, facts are not presented clearly. The most unclear are facts 2 and 3. In fact 2, the subject connected to the predicate ‘fill up’ is not explicitly expressed:







(PREDICATE fill up, ATTRIBUTE sixty percent

SUBJECT {not specified})



In fact 3, the predicate is underspecified because its attribute is not explicitly expressed:


PREDICATE could, ATTRIBUTE {not specified}

OBJECT only half of this


The se open positions in proposition structure are among the main reasons for propositional vagueness. In comprehension process, they can be filled up by different elements, and this can again lead to a reduced level of common understanding.


4.2 Local coherence strategies


For van Dijk and Kintsch (1983), coherence means mainly semantic coherence. Semantic coherence may be local or global. Local coherence refers to semantic relationships between the successive sentences of the discourse. Global coherence is formulated in terms of macropropositions, and it refers to generalized, central persons, objects, macroevents or macroactions, global goal, motivations as well as global time and place. Global coherence is a sort of a macrocontrol mechanism. Further, van Dijk and Kintsch (1983) distinguish two fundamental types of propositional coherence: conditional and functional. Conditional coherence is defined in terms of referential relations: propositions are coherent if they denote facts that are related (one fact is a possible, probable, or necessary condition for another fact). Functional coherence is a propositional link in which one proposition has some function with respect to another (it pertains to meaning relations, not to fact relations). The strategic establishment of coherence between sentences is based on cues, such as sentence topic, and cohesion cues, such as pronouns, paraphrase, identity, etc.


In the example from our data, when referring to macropropositions, i.e., when establishing global coherence, different interpretations are possible.


Example 3:

Context: The same as in example 2. Each turn is further split into utterances.



B10-1: ta sistem bi bil v tem da namesto splošne vojaške obveznosti ki pri nas traja od šestih do sedmih me scev bi vpeljali splošno državljansko obveznost ki bi bla tri do štirikrat krajša in bi jo opravljali samo in večinoma v poletnem času

B10-1: this system would be such that instead of the national service which lasts from six to seven months in this country we would introduce a general service programme that would be three to four times shorter and would be carried out only and mainly in summer time

B10-2: in bi znotraj te splošne državljanske obveznosti bi bilo vojaško in policijsko urjenje zgolj prostovol jno

B10-2: and within this general service programme military and police training would be only voluntary

B10-3: kar pomeni da potem bi odpadla potreba po tem da bi bile ženske izključene iz tega sistema

B10-3: this means that then there would be no need for women to be excluded from this system


P10-1: poglejte problem je v tem da že šestmesečni rok služenja vojaškega mislim vojaškega roka ne usposobi dovolj nabornikov

P10-1: look the problem is that even within the six months of military I mean military training the recruits don’t get skilled enough

P10-2: to se pravi da če bi bil ta siste() ta čas urjenja še krajši bi bilo še slabše

P10-2: this means that if this syst() this time of training is shorter it would be even worse

P10-3: je pa res da bi pa mogoče lahko to nekako vzeli kot dopolnitev našega predloga v smislu prostovoljne rezerve

P10-3: however it is true that maybe we could take this as some sort of complement to our proposal in the sense of a voluntary reserve force

P10-4: tu bi bile pa prav gotovo ženske tudi dobrodošle

P10-4: certainly women would be welcome as well here

P10-5: vendar (1) to tako enostavno ne bi moglo bit da bi (2) to imeli kot neka obveza

P10-5: but (1) this could not be so simple that we would have (2) this as some sort of obligation

P10-6: (3) to bi mogla bit prostovoljna rezerva

P10-6: (4) this should be a voluntary reserve force

P10-7: (4) to bi pa verjetno bla dopolnitev in to mi tudi predlagamo navsezadnje

P10-7: and (4) this would probably be a complement and that is what we suggest after all

P10-8: mi nis mo rekli ali za moške ali za ženske ampak na splošno

P10-8: we didn’t say for men or for women but in general


We will analyze only the pronouns which have been underlined (we cite the rest of the text in order to help the readers gain a better understanding of coherence relations in this discourse segment).


Pronoun (1) (‘this’) has a very general reference. It does not refer to anything specific in the previous utterances, but it establishes connections on the macrostructure level. It could be interpreted as a reference to the topic of the previous turn (B5), or as a reference to the whole process of professionalizing the army, i.e., to the general topic of the conversation.


P ronoun (2) (‘this’) may also refer to the topic of the previous turn (B5), i.e., ‘general service programme’, or it could, more specifically, refer to the ‘military and police training’, the sentence topic of B10-3.


P ronoun (3) (‘this’) may have the same references as pronoun (2), i.e., ‘general service programme’ (turn topic) or ‘military and police training’ (sentence topic). It is likely that pronoun (2) refers to the former, and pronoun (3) to the latter. There are some indications that references established using pronouns (1) to (3) go from more general (conversation topic) to less general (sentence topic).


Another vague reference is pronoun (4) (‘this’). It may refer to the previous utterances of the same turn; however, it could also be interpreted as reference to the previous topic turn.

The coherence cues in this discourse segment are therefore ambiguous and vague and can lead to different interpretations. Nevertheless, from a broader perspective, on a pragmatic level, this discourse segment is clearly understandable despite of some unclear references: P is claiming that what B proposes is not good for an independent army; however, B’s proposal could be used as a supplement to P’s own proposal. Again we have an example in which communication is not completely successful on each and every level of discourse comprehension, but in general it is successful and the participants do not feel the need to negotiate (mis)understanding.


4.3 Use of knowledge


Van Dijk and Kintsch (1983) use the notion ‘macrostructure’ to provide an abstract semantic description of the global content, and hence the global coherence of discourse. Macrostructures are defined by the so-called macrorules, which are deletion, generalization and construction. One of the central macrostructures is the discourse topic. In the next example from our data, we will try to show that active discourse participants share a common understanding of what the discourse topic is, but interpretations on what exactly it refers to vary. However, these differences are not caused by different macrostructure processes, but by different (use of) knowledge.


Example 4:

One of the topics of the evening broadcast news show was a discussion about the political orientation and world view of the Constitutional Court judges. The discussion also refers to other, previous discourses elsewhere, where the same topic was discussed.


In our data, the topic is first announced when the news headlines are announced, using the following words:


Example 4-1:

A: nazor ustavnih sodnikov na tehtnici

A: world view of Constitutional Court judges in the balance


After the announcement, another topic is discussed in the first two minutes of the programme. After that, the topic in question is introduced. The news presenter starts the introduction:


Example 4-2:

S0-1: zapleta pa se tudi pri izvolitvi novih ustavnih sodnikov

S0-1: and new problems have arisen in the election of new Constitutional Court judges

S0-2: eden izmed kandidatov Krešimir Puharič je namreč od svoje kandidature odstopil

S0-2: one of the candidates Krešimir Puharič has withdrawn his nomination

S0-3: zaradi kandidatove domnevne nestrpnosti do drugačnih pa so se odprla vprašanja etične primerljivosti kandidatov za us tavne sodnike

S0-3: because of the candidate’s alleged intolerance questions about an ethical comparison of the candidates for Constitutional Court judges have arisen

S0-4: polemiko o svetovnonazorski in politični usmerjenosti ustavnih sodnikov je znova sprožil nekdanji ustavni sodnik Matevž Krivic

S0-4: th is discussion on the world view and political orientation of Constitutional Court judges has once more been opened by former Constitutional Court judge Matevž Krivic



This discourse fragment introduces the topic; the phrase that explicitly expresses the topic (‘world view and political orientation of Constitutional Court judges’) has been underlined.


After the presenter’s introduction of the topic, a journalist’s report follows. Even though the topic was previously introduced as ‘world view and political orientation’, most of the propositions in the report (see the following fragments) are reduced to ‘political orientation’, as we can see from the main expressions (underlined below) that refer to the discourse topic (R stands for the journalist, M stands for the President):


Example 4-3:

R1-1: za ustavno sodišče v prejšnji sestavi naj bi veljalo da je bližje politični desnici

R1-1: the previous panel of the Constitutional Court was believed to be closer to the political right wing

R1-2: politična levica je to utemeljevala tudi na podlagi znamenite odločbe ...

R1-2: the political left wing supported this claim using the well-known provision ...

R2: predsednik države je očitke da politične uravnoteženosti ni upošteval zavrnil

R2: the President rejected the reproaches that he had ignored the political balance of power

M1-3: seveda nimam jaz nobene niti moralne niti formalne pravice niti možnosti da jih sprašujem po njihovi politični opredelitvi

M 1-3: of course I have no moral or formal right to ask them [i.e., candidates for Constitutional Court judges] about their political preferences


However, in her last turn, the journalist points out that not all interpretations share this view:


Example 4-4:

R3-1: nekdanji ustavni sodnik Matevž Krivic meni da ne gre le za vprašanje politične ampak tudi širše nazorske uravnoteženosti sestave ustavnega sodišča

R3-1: the former Constitutional Court judge Matevž Krivic claims that this is not merely a question of political orientation but also an issue concerning a broader world-view balance of the Constitutional Court


While we have a common understanding of what the topic is, we nevertheless observe at least two different interpretations of what this topic refers to:

1. it refers to the political orientation and the world view of the Constitutional Court judges (these are not interpreted as two parallel concepts),

2. it refers only to the political orientation of the Constitutional Court judges.


After the report , a studio interview follows. In the interview, the participants are the presenter (S) as the interviewer, and the president of the Slovenian Constitutional Court (W) and the former Constitutional Court judge already mentioned above (K – Krivic) as the interviewees. They discuss the same topic, ‘the world view and political orientation of Constitutional Court judges’. However, the interpretations on what exactly this refers to once again differ considerably:


1. Only political orientation is mentioned, most of all by S, but also to some extent by speaker K (we cite only the turn fragments where these concepts are explicitly discussed):


Example 4-5:

S2: gospod Krivic se tudi vi strinjate z mnenjem v prispevku ki smo ga slišali da je tudi sedanja sestava ustavnega sodišča politično neuravnotežena

S2: Mr Krivic do you also agree with the opinion that we have heard in the report that the present structure of the Constitutional Court is politically unbalanced


Example 4-6:

K4-1: mislim da je nujno zdaj bi bilo nujno da ta levo liberalni politični blok  ~LDS združena lista Desus ne da bi rekel se zavedajo svoje odgovornosti eee za bi rekel u() usodo ustavnega sodišča in njegov ugled vnaprej in da bi vnaprej zagotovili svojo podporo eee vsakemu resnemu eee dobremu desno usmerjenemu kandidatu

K4-1: I think that it is urgent now it would be urgent that this left liberal political wing ~Liberal Democrats United List of Social Democrats Democratic Party of Pensioners that so to say are aware of their responsibility em for so to say say the f() future of the Constitutional Court and its reputation and that they would offer their support em to any serious em good right wing candidate in advance


Example 4-7:

K4-3: ... da se resni in odlični kandidati enostavno včasih tudi bojijo kandidirat zaradi te politične neuravnovešenosti

K4: ... that serious and very good candidates are sometimes simply afraid to run for Constitutional Court judges because of this political imbalance


Example 4-8:

S8: ... ga spraševal tudi o njegovi politični eee opredelitvi

S8: ... asking him about his political um views


2. Political orientation and world view are mentioned as two concepts that are parallel and interconnected:


Example 4-9:

S5: zakaj ste vi prav zdaj v tem trenutku izpostavili to vprašanje politične in eee nazorske eee o() us() ustanovitve oziroma opredelitve ustavnega sodišča

S5: why did you choose right now to pose this question of political and em world view em o() us() of the Constitutional Court


Example 4-10:

S7: kako pa vi eee menite da bi se dalo doseči to politično in eee svetovnonazorsko uravnoteženost ustavnega sodišča

S7: what do you em think is necessary to achieve this political and em world-view balance of the Constitutional Court


Example 4-11:

K16: ... in zato je toliko bolj pomembna ta bi rekel nazorsko-politična uravnoteženost da se to nekako izravna

K16: ... and this is why this so to say world-view-political balance is so important so that this is somehow evened out


3. Only world view is mentioned or pointed out as more important/the only important reference for the topic:


Example 4-12:

W2-2: gotovo bi bilo idealno če bi lahko eee ustavno sodišče bilo dejansko svetovnonazorsko uravnoteženo mislim da se tukaj z magistrom Krivicem strinjava da ne gre za politično opredelitev

K3:   mhm

W2-3:   bolj za svetovnonazorsko

W2-2: certainly it would be ideal if the em Constitutional Court could really be world-view balanced I think that Mr Krivic and I agree that it is not so much about the political view

K3: mhm

W2-3: more about the world view


Example 4-13:

K8: ... ker mislim da je to nujna ustavna zahteva tako rekoč čeprav ni zapisana po tej nazorski uravnoteženosti ...

K8: ... because I think that this is an urgent constitutional claim so to say even though world-view balance is not explicitly mentioned


Example 4-14:

K18: ... nekatere odločitve so take da izražanje nazorske pripadnosti ni defekt...

K18: ... some decisions are such that expressing one's own world view is not a defect ...


We believe that these differences in the interpretation of the topical concepts arise from the processes of the use of knowledge. Comprehension processes “continuously need information from long-term memory in order to construct or specify semantic and pragmatic representations of discourse” (van Dijk, Kintsch, 1983: 303). When the topic of  ‘world view and political orientation of Constitutional Court judges’ is introduced, discourse participants use their previous knowledge schemes of these concepts and adapt them to the present situational representation. Knowledge, as we know, is variable, socioculturally as well as personally, so we can presuppose that discourse participants have slightly different representations of these concepts when they enter the present discourse; however, at the same time knowledge is usually also sufficiently general to enable successful communication, so their representations also have much in common and the speakers presuppose that the hearers share some degree of common understanding of the topical concepts. In the discourse processes, the pre-established knowledge of these concepts is adapted to the present situational scheme. This is done through all levels of discourse comprehension; however, there are also two fragments where the speaker s more or less explicitly point to the preferred interpretation of these concepts (“ the former Constitutional Court judge Matevž Krivic claims that this is not merely a question of political orientation but also an issue concerning a broader world-view balance of the Constitutional Court ”, “ I think that Mr Krivic and I agree that it is not so much about the political view ”). Nonetheless, as we tried to prove above, there are probably slight differences in the interpretation of these concepts. We believe this is caused mainly by the manner in which these concepts were represented in the pre-established knowledge and how they where adapted to the present situational scheme by the speakers. Differences in representations of these concepts can be described as follows:

1. In the journalist’s report, the topic is reduced to ‘political orientation’ and it probably refers to whether a person supports the left or the right wing political parties.

2. For speaker S, the topical concept of ‘political orientation and world view’ probably refers to two interconnected, parallel facts, i.e., a person who is conservative, will support the right wing political parties, while a liberal person will support the left wing political parties.

3. For speaker W, the topic is reduced to ‘world view’, and for her political orientation and world view are probably not connected, i.e., a person who is conservative may not agree with the politics of conservative parties.

4. Speaker K, on the one hand, agrees that the topic refers only to ‘world view’ (example 4-12, example 4-4), but, on the other hand, he refers to ‘political orientation’ as well (examples 4-6 and 4-7). One possible interpretation is that speaker K is not very sincere, that he is trying to preserve his face by pointing to the world view imbalance of the Constitutional Court, while he is actually referring to the political orientation of the Constitutional Court. If this is so, he probably makes a distinction in his knowledge scheme between political orientation and world view, just as speaker W. However, it is also possible that he understands these concepts as interconnected and parallel, just as speaker S.


We must keep in mind that the above descriptions are only speculations, based on the text, and are not necessary correct – we do not have access to the actual interpretations. However, it is more or less certain that the interpretations of what exactly the discourse topic refers to are different, and common understanding is therefore reduced.


4.4 Interactive and pragmatic strategies


Our last example of reduced understanding will be explained on the level of interactive and pragmatic strategies. Van Dijk and Kintsch (1983) discuss pragmatic and interactive strategies within the context of production strategies, which in their view should be part of a complete discourse-processing model (p. 16). They see production of discourse as “the planning and execution of various actions” (p. 264). These “/a/ctions are represented cognitively as intentions , and their goals as purposes .” (p. 264) Speakers form plans for global (macro) speech act in order to change the situation in the desired direction. Local speech acts are, to some degree, determined by these global speech acts; however, they also depend on the previous speech acts and the reaction of the hearer to these.


In our example, we will see that there are various interpretations of the semi-global speech act of a particular discourse segment (semi-global, because we are only dealing with a discourse segment, so it is not the global speech act on the level of the whole discourse), and depending on how these semi-global speech acts are interpreted (and depending on the development of the propositional content, and the other speaker’s reactions), different interpretations of local speech acts are possible.


Example 5:

Context: The same as in example 1. B is a journalist, C is a member of parliament and K is the former president of the State Council. C has proposed that the following sentence in the Constitution should be changed [translation]: “The President of the Republic is elected for a period of five years, but may serve no more than two consecutive terms of office,” 3 by erasing the term ‘consecutive’. The reason for his proposal is that the current wording means that the same person can serve for two consecutive terms of office, and while he or she may not run for a third consecutive term, he or she is entitled to run for the position of president again after another president has served a term. C disapproves of this possibility.


B3-1: no nekteri so vam danes očitali paranojo kot da se v narekovajih bojite aktualnega predsednika

B3-1: we ll some have blamed you today that you’re being paranoid as if you were so to say afraid of the president

B3-2: je to res

B3-2: is this true

C3-1: glejte jaz se čudim da kar naprej nekteri vključujejo v ta pogovor aktualnega predsednika

C3-1: look I am surprised that some people keep bringing up the current president

C3-2: mislim da aktualni predsednik se pravzaprav poslavlja od te funkcije Barbara Brezigar še ni izvoljena ali kdorkoli drug in toliko politične modrosti pa že moramo met da takih rečeh la hko razpravljamo s tem da abstrahiramo trenutno aktualne osebe

C3-2: I think that the current president 's term is about to end Barbara Brezigar or any other candidate has not yet been elected and we have to have enough political wisdom to be able to discuss these things by abstracting the current office holders

B4: aktualne osebe torej niso vzrok za to vašo pobudo

B4: the current office holders are then not the motive for your proposal

C4: ne vem kdo je kdo to kar naprej vleče aktualnega predsednika v to razpravo

C4: I don’t know who keeps bringing up the current president into this discussion

B5: gospod C samo povzemamo to kar je bilo slišati v tem in prejšnjem prispevku

B5: Mr C we’re only repeating what we could here in this report and the previous one


On the semi-global level, the speech act performed by B may be interpreted as an objective question (typical of interviews), or it can be interpreted as a provocation and accusation (of C’s actions). The speech act, performed by C, can be interpreted in an even greater number of ways: it can be understood as a negative answer to the questions, or as surprise (caused by the content of the journalist’s question), or as an accusation (in response to B’s supposed accusation). On the interactive level, we may speculate whether either of the speakers is hiding his true intended actions: B may want to express a hidden provocation/accusation, or C may be feigning surprise in order to make his negative answer more convincing.


According to this semi-global speech act analysis, different interpretations of speech acts on the most local level are possible, for instance:

B3-1: introduction

B3-2: (objective) question / provocation-accusation

C3-1: (negative) answer / surprise / accusation

C3-2: explanation / explanation / argumentation

B4: (objective) question / provocation-accusation

C4: (negative) answer / surprise / accusation

B5: explanation


It is important for our purpose that it is very likely that B and C have different interpretations of each other's speech acts. B may want to ask an objective question, but C interprets it as an accusation, and B, noticing this from C’s acts, explains in B5 that this is not his intention. Or C may only want to give a negative answer to B’s questions, and B interprets this as an accusation, and therefore explains his acts in B5 in order to manage the misunderstanding. We can also speculate about how truthful one or the other speaker is in his acts and how they interpret each other's sincerity. One way or another, the explanation in B5 reveals that it is very likely that each discourse participant's the understanding of the other participant's speech acts was (partially) different; however, the discourse participants do not find it important enough (in view of the global goals and plans of the whole discourse) to pay any further attention to this possible lack of common understanding.

5 Discussion


In this paper we have tried to prove that in real-life communication we can find different types of reduced understanding (or ‘non-standard’ cases of misunderstanding), where, for instance, some components of propositions are open to different interpretations (examples 1 and 2), coherence cues are vague (example 3), topic concepts are used in such a way that they build slightly different situation representations, and speech acts and interactional strategies are open to different interpretations. In spite of this, discourse participants did not explicitly show that they have recognized reduced understanding (or misunderstanding) and they did not explicitly try to handle it. How can we explain this?


At the beginning of this paper, we have pointed out that there is no such thing as perfect, complete understanding, since discourse participants always understand discourse from their own points of view. This fact is especially noticeable in examples 4 and 5 above. However, we chose the examples above to show a higher level of differences in understanding than usual. Therefore, we will briefly discuss why the participants in our examples did not recognize and/or explicitly try to handle reduced understanding or misunderstanding, and we will point out some possible answers that could provide a starting point for further research. But it is not the goal of this section nor this paper to give a full answer to this question. In order to do so, a larger variety of data would be needed, including different types of dialogue. Here we analyze only a very specific format of TV discourse that could be classified as persuasive dialogue – a larger, representative set of dialogue types, such as listed for example by Walton (1992) would need to be added to our data: information-seeking dialogue, negotiation dialogue, inquiry dialogue, and eristic dialogue. Furthermore, we believe casual conversation would be important data for the analysis as well. By analyzing such a broad set of real life data we could make the above set of different types of reduced understanding more complete. Then we should make a further step in the analysis, in which we should consider all the available contextual information, and perhaps also use interviews or similar techniques to come to a more reliable explanation as to why these potential occurrences of misunderstanding are not dealt with at the conversational level.


However, we can interpret the following on the basis of the analysis presented in this paper:


We have to consider the fact that discourse participants do not have as much time for comprehension as we do in our careful step-by-step analysis, so it is possible that they did not detect reduced understanding in the examples above as we did. As van Dijk and Kintsch’s model of discourse comprehension explains it, comprehension is complexity oriented, building from the most local to the most global, but continuously looking back to the local. From this perspective, we could expect that as soon the hearer is able to construct the textbase and situation model on a more global level, he or she does not lose time to process details on a more local level anymore. In this way, we could explain some of our examples at least partially:


Example 1: no matter how we interpret the particular elements of proposition, on a more global level, propositions of the discourse segment analyzed are unambiguous, i.e., ‘K claims

that there are many countries that allow the president to serve multiple (more than two) terms of office, while C claims that the same number of countries or even more countries do not allow this’.


Example 2: on a more global level, the propositions of the discourse segment analyzed can be unambiguously understood as: ‘what B claims is not true, what the party that speaker P belongs to proposes is a good proposal for the professionalization of the Slovenian army’.


Example 3: as before, on a more global level, the discourse segment (of the speaker P) can be unambiguously understood as: ‘P is claiming that what B proposes is not a good system for the state army, however, B’s proposal could be used as a complement to the professional army, but not in the form of obligation, but as a voluntary reserve, where women could participate as well’.


For examples 4 and 5 , such explanations are not necessarily satisfying. In example 4, the understanding probably differs considerably from participant to participant on a very global level (i.e., reference of the discourse topic), while on a more local level, discourse segments seem unambiguous. We believe that this may be due to the overall goals and interests of the speakers: as we already pointed out at the end of our analysis (4.3), speaker K may be insincere, trying to preserve his face by pointing to a world-view imbalance of the Constitutional Court, while he actually wants to encourage a change in the political structure of the Constitutional Court. If this is so, vague and mixed use of the notions ‘political’ and ‘world-view orientation’ corresponds to his goals. Moreover, as we mentioned in section 3, the specific characteristics of the data analyzed (short evening news interviews) mean that it is not necessarily the aim of the discourse participants to come to a common understanding. Furthermore, we must also consider the possibility that it is not expected and common behaviour in this genre to explicitly point out and handle misunderstandings, since this might not be interesting for the audience, therefore reduced understandings are handled in a more or less indirect way – as in segment 4-12 of the example 4.


In a similar way, we could explain example 5: utterance B5 may have been intended to manage a case of reduced understanding, but the specifics of the genre do not allow more time to be spent on this, therefore the journalist quickly continues, leaving the half-managed reduced understanding behind (without waiting for the addressee’s reaction) and addressing another discourse participant with a new question.



6 Conclusion


In this paper, we have analyzed the borderline examples of (mis)understanding in which it is not clear anymore whether or not a misunderstanding has occurred, whether or not communication was successful, and where the participants do not try to negotiate an understanding, even though differences in understanding are probably bigger than usual. By analyzing real data, we pointed out four different types of such borderline examples of (mis)understanding/reduced understanding. We analyzed these examples following van Dijk and Kintch’s (1983) theory of strategies of discourse comprehension. This theoretical model explains discourse comprehension as strategic processes, the goal of which is the construction of a textbase and situation model. The various components of this model are propositional strategies, local coherence strategies, macrostrategies, schematic strategies, production strategies, strategies for the use of knowledge and some other, non-major types of strategies (stylistic, nonverbal, etc.). In our examples, we defined the level of propositional strategies (in examples 1 and 2), local coherence strategies (in example 3), strategies for the use of knowledge (in example 4), and interactional and pragmatic strategies (within the production strategies) (in example 5) as the main sources of reduced understanding.

In the last part of the paper, we briefly discussed the possible answers to the question why the speakers failed to handle (non-standard) misunderstanding or reduced understanding in our examples. We think the possible answers – in addition to the fact that there is no complete understanding, and consequently, some level of differences in understanding can be perceived as ‘normal’ – include the following: it may be that discourse participants do not lose time  processing details on a more local level as soon they are able to construct a textbase and situation model on a more global level; vague and ambiguous comprehension may correspond to the overall goals of discourse participants; or explicit and detailed managing of a possible misunderstanding may not be appropriate and expected behaviour in the genre analyzed here. However, all these are hypotheses that present a starting point for further research.




Amgoud, Leila, Hameurlain, Nabil, 2007. An argumentation-based approach for dialog move selection. Lecture Notes in Computer Science, n. 4766, p. 128-141, Springer.


Bazzanella, Carla, Damiano, Rossana, 1999. The interactional handling of misunderstanding in everyday conversations. Journal of Pragmatics 31, 817-836.


Bosco, Francesca M., Bucciarelli, Monica, Bara, Bruno G., 2006. Recognition and repair of communicative failures: A developmental perspective. Journal of Pragmatics 38, 1398-1429.


Dafouz-Milne, Emma, 2008. The pragmatic role of textual and interpersonal metadiscourse markers in the construction and attainment of persuasion: A cross-linguistic study of newspaper discourse. Journal of Pragmatics 40, 95-113.


Dascal (ed.), 1999a. Journal of Pragmatics: Misunderstanding 31 (6).


Dascal, Marcelo, 1999b. Introduction: Some questions about misunderstanding. Journal of Pragmatics 31, 753–762.


Dascal, Marcelo and Isidoro Berenstein, 1987. Two modes of understanding: Comprehending and grasping. Language & Communication 7 (2), 139–151.


van Dijk, Teun A., 2007. Discourse, context and cognition. Discourse Studies 8 (1), 159-177.


van Dijk, Teun A., 2008. Discourse and Context: A Sociocognitive Approach. Cambridge University Press.


van Dijk, Teun A., 2009. Society and Discourse: How Social Contexts Influence Text and talk. Cambridge University Press.


van Dijk, Teun A., Kintsch, Walter, 1983. Strategies of Discourse Comprehension. Academic Press, New York, London.


Ferbežar, Ina, Stabej, Marko, 2008. Razumeti razumevanje. Jezik in slovstvo 53, 15-31.


Johnson-Laird, P.N., 1980. Mental models in Cognitive Science. Cognitive Science 4, 71-115.


Johnson-Laird, P.N., Girotto, Vittorio, Legrenzi, Paolo, 1998. Mental models: a gentle guide for outsiders. Available on: (24 th Aug. 2009)


Kamalski, Judith, Lentz, Leo, Sanders, Ted, Zwaan, Rolf A., 2008. The forewarning effect of coherence markers in persuasive discourse: Evidence from Persuasion and Processing. Discourse Processes 45, 545-579.


Pasquier, Philippe, Rahwan, Iyad, Dignum, Frank, Sonenberg, Liz, 2007. Argumentation and persuasion in the cognitive coherence theory. Computational Models of Argument: Frontiers in Artificial Intelligence and Applications 144, p. 233. IOS press.



Schegloff, Emanuel A., 1987. Some sources of misunderstanding in talk-in-interaction. Linguistics 25, 201-218.


Thomas, Jenny, 1983. Cross-cultural pragmatic failure. Applied linguistics 4 (2), 91–112.


Verdonik, Darinka, 2003. Pragmatični vidik nesporazumov v komunikaciji, master thesis. Faculty of Education, Maribor.


Verdonik, Darinka, 2005. Nesporazumi v komunikaciji. Jezik in slovstvo 50, 51-63.


Vershueren, Jef, 1999. Understanding Pragmatics . Arnold, London.


Vogel, Jerica, 2004. Razvijanje razumevanja in vrednotenja govorjenega besedila. Jezik in slovstvo 49, 37-56.


Walton, Douglas, 1992. Types of dialogue, dialectical shifts and fallacies. In: van Eemeren, F.H., Grootendorst, R., Blair, J.A., Willard, C.A ( Eds.), Argumentation Illuminated. SICSAT, Amsterdam, pp. 133-147.


Walton, Douglas, 2007. Dialog Theory for Critical Argumentation. John Benjamins, Amsterdam/Philadelphia.


Weigand, Edda, 1999. Misunderstanding: The standard case. Journal of Pragmatics 31, 763–785.


Weigand, Edda, 2000. The Dialogic Action Game. In: Coulthard, M., et al. (Eds.), Dialogue Analysis VII. Working with dialogue. Niemeyer, Tübingen, pp. 1-18.


Weigand, Edda, 2002a. The Language Myth and Linguistics Humanised. In: Harris, R. (Ed.), The Language Myth in Western Culture. Curzon, Richmond/Surrey, pp. 55-83. Weigand, Edda, 2002b. Constitutive Features of Human Dialogic Interaction: Mirror neurons and what they tell us about human abilites. In: Stamenov, M., and Gallese, V. (Eds.), Mirror Neurons and the Evolution of Brain and Language. Benjamins, Amsterdam/Philadelphia, pp. 229-248.


Weigand, Edda, 2004. Emotions: The simple and the complex. In: Weigand, E. (Ed.), Emotions in Dialogic Interaction. Advances in the complex. Benjamins, Amsterdam/Philadelphia, pp. 3-22.


Weizman, Elda, S. Blum-Kulka, 1992. Ordinary Misunderstanding. In: Stamenov, M. (Ed.), Current Advances in Semantic Theory. Maxim Stamenov Institute of the Bulgarian Language, Sofia, pp. 417–432.


Weizman, Elda, 1999. Building True Understanding via Apparent Miscommunication: A Case Study. Journal of Pragmatics 31, 837–856.


Zaefferer, Dietmar, 1977. Understanding misunderstanding: a proposal for an explanation of reading choices. Journal of Pragmatics 1, 329-346.



1  Original text: “Predsednik republike je izvoljen za dobo petih let, vendar največ dvakrat zaporedoma.” (Constitution of the Republic of Slovenia)

2   The analyses in this and the following example follow the tradition of syntactic parsing of Slovenian sentences. Even though they are shown on the English translations, they are primarily done for the original, Slovenian utterances.

3  Original text: “Predsednik republike je izvoljen za dobo petih let, vendar največ dvakrat zaporedoma.” (Constitution of the Republic of Slovenia)