1. On the place of interactional discourse in Cognitive Linguistics

Usage-based theories hold that the sole resource for language users’ linguistic systems is language use (Barlow & Kemmer, Reference Barlow and Kemmer2000; Langacker, Reference Langacker and Rudzka-Ostyn1988; Tomasello, Reference Tomasello1999, Reference Tomasello2003). Researchers working in the usage-based paradigm, which is often equated with cognitive-functional linguistics (e.g., Ibbotson, Reference Ibbotson2013, Tomasello, Reference Tomasello2003), seem to widely agree that the primary setting for language use is interaction, with spontaneous face-to-face interaction playing a primordial role (e.g., Bybee, Reference Bybee2010; Clark, Reference Clark1996; Geeraerts & Cuyckens, Reference Geeraerts, Cuyckens, Geeraerts and Cuyckens2007; Langacker, Reference Langacker2008; Oakley & Hougaard, Reference Oakley and Hougaard2008; Zlatev, Reference Zlatev, Dor, Knight and Lewis2014). It should, then, follow that usage-based models of language are not only compatible with evidence from communication research but also that they are intrinsically grounded in authentic, multi-party language use in all its diversity and complexities. This should be a logical consequence, as a usage-based understanding of language processing and human sense-making cannot be separated from the study of interaction. However, the overwhelming majority of the literature in Cognitive Linguistics (CL) does not deal with the analysis of dialogic data or with issues of interactional conceptualization. It is our firm belief that this is at odds with the interactional foundation of the usage-based hypothesis. Furthermore, we are convinced that an ‘interactional turn’ is not only essential to the credibility and further development of Cognitive Linguistics as a theory of language and cognition as such. Rather, CL-inspired perspectives on interactional language use may provide insights that other, non-cognitive approaches to discourse and interaction are bound to overlook. To that aim, this special issue brings together four contributions that involve the analysis of interactional discourse phenomena by drawing on tools and methods from the broad field of Cognitive Linguistics.

Despite the relative under-representation of interactional studies in CL, the papers in this special issue do not stand alone, isolated from the rest of the field. Rather, they line up with recent developments within CL and neighboring research areas that have adopted a dialogic view on cognition and language use as well as the emergence and the nature of grammatical systems. There is a growing consensus that the prototypically interactive nature of language use as a negotiation process between two or more participants needs to be treated as an inherent part of cognitive models of language (Brône & Oben, Reference Brône, Oben, Veale, Feyaerts and Forceville2013; Brône & Zima, Reference Brône, Zima, Giora and Du Bois2014; Chafe, Reference Chafe1994; Clark, Reference Clark1996; Deppermann, Reference Deppermann, Deppermann, Fiehler and Spranz-Fogasy2002, Reference Deppermann2007, Reference Deppermann2012; Du Bois, Reference Du Bois and Englebretson2007; Langacker, Reference Langacker2001, Reference Langacker2008, Reference Langacker2013; Levinson, Reference Levinson2006; Pickering & Garrod, Reference Pickering and Garrod2004; Verhagen, Reference Verhagen2005; Zima, Reference Zima2013).

Most prominently, this interactional perspective lies at the heart of Michael Tomasello’s (Reference Tomasello1998, Reference Tomasello1999, Reference Tomasello2003) work on shared intentionality and intersubjectivity, which has widely influenced core areas of Cognitive Linguistics. For instance, in Construction Grammar, there is growing awareness of the encoding of interactional meaning and function in the grammatical system (e.g., Boogaart, Colleman, & Rutten, Reference Boogaart, Colleman and Rutten2014; Fried & Östman, Reference Fried and Östman2005; Östman, Reference Östman, Östman and Fried2005; Verhagen, Reference Verhagen2005). A constantly growing body of work challenges the seemingly static concept of constructions as conventionalized pairings of form and meaning, arguing that, in interaction, constructions aren’t simply instantiated as “prepackaged wholes”, but rather they are “tools for the situated construction of context-sensitive turns: schemata which can be flexibly adapted to interactional contingencies” (Deppermann Reference Deppermann, Auer and Pfänder2011a, p. 121). Most notably in the German research context, ‘Interactional Construction Grammar’ has developed into a very vivid research area at the intersection of CL and interactional linguistics. Exemplary studies of this perspective are, among many others, Auer (Reference Auer2009), Auer and Pfänder (Reference Auer and Pfänder2011a), Brône and Zima (Reference Brône, Zima, Giora and Du Bois2014), Deppermann (Reference Deppermann2007, Reference Deppermann, Auer and Pfänder2011a, Reference Depermann, Lasch and Ziem2011b), Fischer (Reference Fischer2010, Reference Fischer, Lasch and Ziem2015, and this issue), Fox and Thompson (Reference Fox and Thompson2007), Günthner, Bücker, and Imo (Reference Günthner, Bücker and Imo2014), Günthner and Imo (Reference Günthner and Imo2006), and Imo (Reference Imo2007, Reference Imo2009).

Moreover, a growing number of studies also focus on the in vivo (Nerlich & Clarke, Reference Nerlich and Clarke2001, Reference Nerlich, Clarke, Nerlich, Todd, Vimala and Clarke2003) or online (Brône & Coulson, Reference Brône and Coulson2010; Coulson, Reference Coulson2000, Reference Coulson2006; Fauconnier & Turner, Reference Fauconnier and Turner2002; Langlotz, this issue; Zima, Reference Zima2013) character of meaning construction in language, and aim to gain further insights into the interaction of language structure and dynamically evolving discourse. In these studies, attention is drawn to the active process of meaning construction in longer stretches of discourse, which is also apparent in CL studies on discourse coherence, the dynamic construction of discourse representations, perspectivity and polyphony in different text types, and the nature of meaning-making in interaction as processes of joint construction and manipulation of mental spaces (a.o. Brône, Reference Brône2008, Reference Brône2010; Chilton, Reference Chilton2005; Coulson, Reference Coulson, Coulson and Lewandowska-Tomaszczyk2005; Dancygier, Reference Dancygier2005; Hougaard & Oakley, Reference Hougaard and Oakley2008; Langacker, Reference Langacker2001Reference Langacker2008, Reference Langacker2013; Pascual, Reference Pascual2014).

This nascent interest in issues of interactional meaning constitution is accompanied by the rapid evolution of cognitively inspired multimodality research. This particularly fruitful research area has enlarged the traditional focus of linguistics on verbal structures to include bodily means of expression, like gesture, posture, and gaze, into the scope of analysis. Drawing on insights from gesture studies, psycholinguistics, and Conversation Analysis, studies in this line have revisited core concepts of Cognitive Linguistics, such as conceptual metaphor (Cienki, Reference Cienki, Cienki and Müller2008; Cienki & Müller, Reference Cienki and Müller2008a, Reference Cienki, Müller and Gibbs2008b; Forceville & Urios-Aparisi, Reference Forceville and Urios-Aparisi2009; McNeill, Reference McNeill1992; Müller, Reference Müller2004), metonymy (Forceville, Reference Forceville, Caballero and Pinar2010; Mittelberg, Reference Mittelberg2006; Mittelberg & Waugh, Reference Mittelberg, Waugh, Müller, Cienki, Fricke, Ladewig, McNeill and Bressem2014), image schemas (Cienki, Reference Cienki2002, Reference Cienki and Pinar Sanz2013), and the notion of constructions as the basic units of language (Schoonjans, Reference Schoonjans2014; Reference Schoonjans, Sambre, Brône and FeyaertsSchoonjans, Sambre, Brône, & Feyaerts, in press; Steen & Turner, Reference Steen, Turner, Borkent, Dancygier and Hinnell2013; Reference Zima, Ruiz de Mendoza and Pérez SobrinoZima, in press). All these studies are intrinsically dialogic in orientation and put the actual, embodied speaker and his/her gestures, gazes, postures, facial expressions, etc. into the focus of attention, i.e., the data are no longer depersonalized corpus instances that are stripped of any reference to actual speakers. It also actively acknowledges that multimodal messages only acquire meanings in situated, dialogic, and necessarily embodied communication by the integration of input from verbal and kinetic modalities. This line of research is represented in this issue by Oben and Brône’s study at the intersection of Cognitive Linguistics, multimodality research, discourse studies, and psycholinguistics (for detailed descriptions of the papers in this special issue, see Section 3) and the theoretical plea for a multimodal reconsideration of the concepts of the ‘usage event’ and the ‘linguistic sign’ by Alan Cienki.

Nonetheless, despite these encouraging developments in Cognitive Linguistics, the observation remains that the study of authentic, interactional discourse data is not common practice in Cognitive Linguistic work. Most CL analyses rely on monological and to a large extent written data samples. As a consequence, Cognitive Linguistics has been criticized for being non-dialogic in orientation (cf. Fischer, Reference Fischer, Fischer and Stefanowitsch2006, and this issue; Langacker, Reference Langacker2008) and for suffering from a written language bias (Cienki, this issue; Linell, Reference Linell1998). The compatibility of CL theories and concepts with the nature and affordances of conversation is also put into question by more immediately dialogic research paradigms, most notably in Conversation Analysis that hardly ever incorporates insights from CL work (see, however, Depperman, Reference Deppermann2012, for an inside-CA perspective that argues for the need to take a more explicit cognitive perspective in research on talk-in-interaction). It seems, however, that both the under-representation of interactional data in CL and the negligence of CL work and ideas by CA and various strands of discourse studies are not so much due to obvious incompatibilities between Cognitive Linguistics and the nature of interactional communication. Rather, we think that the potential of Cognitive Linguistics in the domain of interactional discourse has simply been underexplored and largely remains to be uncovered.

This is particularly unfortunate as, to our mind, three basic tenets of Cognitive Linguistics strongly favor a CL approach to interactional meaning constitution:

• • the recontextualization approach;

• • the construal hypothesis;

• • the usage-based model.

Each of these tenets will be discussed in more detail in the following section.

2. On Cognitive Linguistics’ suitability for interaction research

As Geeraerts (Reference Geeraerts and Geeraerts2006, p. 25) argues, the recontextualization approach that is typical of usage-based approaches in general, and Cognitive Linguistic theories in particular, has put language use back in the focus of attention, from which it had disappeared in much of twentieth-century theoretical linguistics. While isolated, often invented, examples have long dominated as ‘empirical’ evidence in linguistic research, most notably in generative approaches, CL posits that the study of linguistic structure needs to be based on contextualized instances of authentic language use. As a consequence, especially over the past twenty years, the construction and skillful use of corpora has gained more and more importance, and today virtually all of Cognitive Linguistic research involves corpus studies or the use of other empirical methods (Janda, Reference Janda2015). In that vein, Stefanowitsch (Reference Stefanowitsch, Brdar, Gries and Žic Fuchs2011, p. 272) argues that “from this perspective, the linguistic corpus gains a central place in linguistic theory: it becomes a model of linguistic usage (both input and output) of an ‘average’ speaker”. By alluding to the ‘average’ speaker, he implicitly points to a fundamental principle of Cognitive Linguistics: the main interest is the relationship between cognition and the language system. It essentially deals with issues such as conventionalization, routinization, entrenchment, grammaticalization, lexicalization, etc. It is much less concerned with individual speakers or interactants, and as a matter of principle generalizes away from single, situated instances. This is a major difference with many other research paradigms that are much more concerned with the details of particular pieces of interactional data, most notably Conversation Analysis, and which in turn often shy away from generalizing over instances. This fundamental difference in focus and primary interest, among other factors mentioned, most likely contributes to the fact that interactionally minded researchers are by and large keeping their distance to CL and vice versa (for approaches arguing for fruitful synergies between CA and Cognitive Grammar, however, see Deppermann, Reference Deppermann2007, Reference Deppermann2012; Etelämäki & Visapää, Reference Etelämäki and Visapää2014; and this special issue’s papers by Langlotz and Fischer).

However, we believe that instead of regarding the CL interest in the language system as essentially incompatible with an interaction-focused perspective, a synergetic approach which combines insights from both fields will not only yield more encompassing insights in the cognitive underpinnings of interactional language use but, most notably, it will prevent CL from replicating long-established insights from Conversation Analysis and other discourse-oriented fields. In other words, doing interactionally oriented Cognitive Linguistics should neither be tantamount to ‘doing CL all over again’ nor to ‘doing CA / discourse linguistics all over again’. To be sure, by insisting on an interactional turn in CL, we do not wish to argue that the research done thus far in CL or every study that does not deal with interactional phenomena and data is irrelevant to the study of interaction from a CL perspective. On the contrary, the aim should be to gradually take well-established concepts from CL, to confront them with interactional discourse data, and to see how they react to this confrontation. That is, what needs to be seen is whether they do stand the proof or whether and how they need to be modulated/adapted. This approach, of course, may result in the observation that some CL concepts are not quite compatible with interactional language use. This should, however, not be seen as a threat to CL but as an avoidable step towards a plausible cognitive model of (interactional) language use.

Indeed, existing studies that have already applied theoretical concepts and methods from Cognitive Linguistics to the domain of interaction studies (and the papers in this special issues do just that; see also Hougaard and Oakley, Reference Hougaard and Oakley2008, for another interesting collection), in fact show that many concepts do hold in the interactional domain, although probably not un-modulated. Again, this is especially evident in work on Interactional Construction Grammar, where the concept of construction has proven to be generally well-suited to account for spoken language structures (much better than generative approaches), but which has also revealed the need to rethink some aspects of the concept (most notably its degree of fixedness and granularity; see especially Auer & Pfänder, Reference Auer, Pfänder, Auer and Pfänder2011b; Deppermann, Reference Depermann, Lasch and Ziem2011b; Fischer, this issue; Imo, Reference Imo2009).

Next to the recontextualization approach, another cornerstone of Cognitive Linguistics that makes it theoretically well suited for an extension to interactional discourse is the construal hypothesis. One reason is the fact that it regards meaning to be intrinsically speaker-oriented (again in the sense of focusing on an ‘average’ language user). It entails that meaning is not fixed and stable, but is rather construed by conceptualizers. Construal always involves a given perspective (spatial, epistemic, etc.) taken (i.e., ‘construed’) by given language users. In other words, meaning is grounded in the construal relationship between subjects of conceptualization (interactants) and objects of conceptualization (the things talked about or referred to). The construal hypothesis hence puts the language user – again, however, mostly as an abstract entity – at the very center of its account of meaning. The most explicitly dialogic account of construal has been proposed by Verhagen (Reference Verhagen2005, Reference Verhagen, Geeraerts and Cuyckens2007). Drawing on Langacker’s (Reference Langacker1987, Reference Langacker2001) classical viewing arrangement, Verhagen has argued for an extension of the subject of conceptualization (the ‘viewer’ in Langacker’s terms) as crucially involving two subjects – prototypically a speaker and a hearer. By means of language, speaker and hearer (or writer and reader) engage in cognitive coordination with respect to some construed object of conceptualization. In other words, for Verhagen (Reference Verhagen, Geeraerts and Cuyckens2007, p. 60) “the point of a linguistic utterance is, generally speaking, that the first conceptualizer invites the second to jointly attend to an object of conceptualization in some specific way […]; that is, both conceptualizers are involved in coordinating cognition by means of language, with one conceptualizer taking the initiative in each specific utterance”. The very essence of cognitive coordination relies in intersubjective attuning, i.e., while we speak we take into account others’ perspectives, their knowledge, attitudes, and beliefs. That is, we take into account and try to anticipate how others may understand our utterances.

The construal hypothesis, and must notably its intersubjective extension, brings Cognitive Linguistics closer to interactional theories of language use and communication than other linguistic paradigms. But without a doubt the road towards an interactionally plausible theory of cognitive coordination and meaning-making is still long and winding. Indeed, Verhagen’s version of the construal arrangement is being criticized as too static and too close to a meaning-transfer model from speaker to hearer (cf. also Cienki, this issue). This relates to the ongoing discussion as to where cognitive linguists believe that cognition and language reside. Most work in CL, and this is probably also the outside view on CL (Linell, Reference Linell2009), seems to attribute cognition and language to the individual. Work in dialogism, on the other hand, considers language to be essentially interpersonal and social. Cognition as individual thinking and language as a social (and cultural) phenomenon are often taken to be diametrically opposed, incommensurable perspectives (for a similar argument, see Clark, Reference Clark1996; Linell, Reference Linell2009, Reference Linell2014). However, one of the biggest, most vital challenges for any approach to language and communication is to bridge that gap, because it seems safe to say that language is best viewed as both an individual and inter-individual phenomenon.

Finally, the most important cornerstone of Cognitive Linguistics, which we argue provides a stable theoretical basis for an extension to interactional discourse, is the usage-based model. In contrast to traditional, top-down approaches that treat linguistic structures as largely determined by the language (rule) system and independent from language use, Cognitive Linguistics regards linguistic units to be inherently grounded and shaped by usage. This bottom-up spirit of CL inspires the usage-based thesis: speakers’ grammar arises through the abstraction of symbolic units from concrete, fully contextualized language use (Barlow & Kemmer, Reference Barlow and Kemmer2000; Bybee, Reference Bybee1995, Reference Bybee2010; Diessel, Reference Diessel and Aronoff2014; Geeraerts & Cuyckens, Reference Geeraerts, Cuyckens, Geeraerts and Cuyckens2007; Goldberg, Reference Goldberg1995, Reference Goldberg2006; Langacker, Reference Langacker1987, Reference Langacker and Rudzka-Ostyn1988, Reference Langacker2001; Taylor, Reference Taylor2002; Tomasello, Reference Tomasello1999, Reference Tomasello2003). By adopting the usage-based thesis as one of its basic tenets, Cognitive Linguistics has thus put discourse at the very heart of its theory:

A crucial role in the emergence of linguistic units is attributed to the processes of entrenchment and conventionalization. Whereas entrenchment refers to the storage of a unit in an individual speaker’s long-term memory, conventionalization concerns a unit’s status in the inter-individual grammar of a speech community. Both processes reduce the cognitive effort in language use, allowing for a unit’s “routinized nature of execution’ (Langacker, Reference Langacker2008, p. 17). Crucially, as CL argues against a clear distinction of linguistic and encyclopedic, experience-based knowledge, any usage aspect may get entrenched or conventionalized as part of a symbolic unit’s meaning. The strict distinction between semantics and pragmatics is discarded in favor of a holistic approach to linguistic meaning: “As units emerge from usage, becoming entrenched in the minds of speakers and conventional in a speech community, they retain as part of their value any feature which is constant across the events giving rise to them. This includes any facet of the context and the speaker-hearer interaction” (Langacker, Reference Langacker2009, p. 154). Usage events (as instances of languages use) hence are taken to “involve conceptualization, encompassing ‘the expression’s full contextual understanding’ – including ‘everything evoked as the basis for its apprehension’, as well as a means of expression. [This] includes the full phonetic detail of an utterance, as well as any other signals, such as gestures and body language” (Langacker Reference Langacker2008, pp. 457f.). In recent years, quite a few studies have adopted and developed further this multimodal definition of the linguistic unit and the usage event (Schoonjans, Reference Schoonjans2014; Reference Schoonjans, Sambre, Brône and FeyaertsSchoonjans, Sambre, Brône, & Feyaerts, in press; Steen & Turner, Reference Steen, Turner, Borkent, Dancygier and Hinnell2013; Reference Zima, Ruiz de Mendoza and Pérez SobrinoZima, in press). This special issue adds to this body of research the papers by Cienki and by Oben and Brône.

The usage-based model, as a fundamental tenet of CL, is thus intrinsically interactional (and multimodal) in orientation. However, as mentioned above, despite the interactional foundation and the non-restrictive approach to entrenchment and conventionalization, most CL studies do not explore the full potential of the usage-based thesis by basing the analysis on fully contextualized, dialogic discourse data. The papers in this special issue are thus welcome and much needed exceptions in offering CL-driven approaches to interactional discourse phenomena. The next section briefly presents the contributions to this special issue.

3. The papers in this special issue

This special issue opens with a paper by Alan Cienki on ‘Spoken language usage events’. Cienki’s starting point is Langacker’s theory of Cognitive Grammar (1987, 1991, 2008) and his definition of the linguistic sign as a pairing of phonological form and semantic structure. In Cognitive Grammar – as in all usage-based models – signs (or constructions) are taken to emerge from usage events via “the progressive entrenchment of configurations [of semantic and phonological structures] that recur in a sufficient number of events to be established as cognitive routines” (Langacker, Reference Langacker2008, p. 220). Usage events are fully contextualized pieces of language use, involving a given conceptualization (semantic content) that is expressed by means of meaningful sounds and any other signals, such as gesture and body language (Langacker, Reference Langacker2008, p. 457). Cienki hence concludes that Cognitive Grammar posits that usage events are multimodal events and that the basic usage events of language are spoken language usage events. This, he observes, is, however, fundamentally at odds with the kind of data analyzed in Cognitive Grammar, i.e., constructed, monomodal examples that often take the form of full sentences. In his paper, Cienki hence proposes to take the claims of Cognitive Grammar seriously and to look at prototypical features of spoken language such as non-lexical sounds (like English uh-hm, hmm, mm), intonation, and, in more detail, manual gestures as being symbolic in nature to varying degrees. This multimodal perspective on symbolization, Cienki claims, has implications for the construal hypothesis (cf. supra Section 2). It is argued that gestures, intonation, and non-lexical sounds may provide vital information on a construal’s schematicity, and its degree of focus and prominence, as well as its perspective (character versus observer viewpoint). Cienki concludes from his observations and long experience with the analysis of multimodal language that whether a given pairing of form and meaning is symbolic in nature cannot always be answered with a straight yes or no. Rather, he argues for a more nuanced approach that allows for gradations of entrenchment. The application of the notion of a symbolic unit to spoken usage events, and its extension with non-verbal information, thus provides a novel insight into the flexibility and permeability of the borderlines of symbolic units as traditionally defined in Cognitive Linguistics.

Cognitive Grammar also is one of the theoretical cornerstones of the paper by Andreas Langlotz on ‘Local meaning-negotiation, activity types, and the current-discourse-space model’. Looking at conversations at a tourist information center in Switzerland, Langlotz observes a discourse-analytical dilemma: meaning in interaction is by definition locally managed and established between discourse participants. Yet activity types and speech genres impose restrictions on what can be said and how and what inferences are most likely to be speaker-intended. Hence, discourse structure and meaning are also conventional to some degree. For Langlotz, this entails that communication relies on “bridging the conventional with the situation-specific”, which “pares down to the cognitive challenge of performing the local planning of interactional moves against the background of … norm-abiding communicative conduct”. To resolve the inherent discrepancy between the online perspective that is typically associated with Conversation Analysis and the convention-oriented perspective on discourse as advocated in the Ethnography of Communication, Langlotz proposes a socio-cognitive model of discourse that combines two discourse models from the realm of Cognitive Linguistics / Cognitive Sciences: the ‘current discourse space-model’ as proposed by Langacker (Reference Langacker2001, Reference Langacker2008) and Lawrence Barsalou’s ‘situated conceptualization-model’ (Barsalou, Reference Barsalou, Cohen and Lefebvre2005). By performing a detailed, step-by-step analysis of two short extracts from his corpus of tourist encounters, he shows that successful communication within the tourist information activity frame means that both the tourist-information officer and the tourist need to gradually build up “a shared … conceptualization of the booking [a room]-category”. The trigger (or precondition) to arrive at such a shared simulation, Langlotz argues, is what he calls a “genre simulator”, which “mediates the discursive states in the interaction by simulating relevant conceptualizations for activity-specific meaning-coordination”. The model presented by Langlotz, and illustrated by means of real-life data, thus presents a useful tool for the analysis of meaning construction and negotiation in interaction, with co-participants who can flexibly and creatively manage shared conceptualizations and mutual (genre) knowledge.

In ‘What you see is what you do: on the relationship between gaze and gesture in multimodal alignment’, Bert Oben and Geert Brône are concerned with a very pervasive, yet under-researched feature of communication: the inter-speaker alignment of gesture. Their account is an example of the aforementioned, particularly fruitful, research strand within Cognitive Linguistics that turns to the multimodality of communication and the linguistic system. Although it is epistemologically grounded in CL, the paper has a clearly multi-disciplinary focus. It integrates relevant concepts and methods from CL, gesture studies, and psycholinguistics. More specifically, Oben and Brône study the effect of eye-gaze on gestural alignment. The analyses presented are based on the InSight Interaction Corpus (Brône & Oben, Reference Brône and Oben2015), a corpus of recordings from face-to-face interactions between two participants who both wear head-mounted eye-trackers and perform a targeted collaborative task. Oben and Brône focus on one specific gesture type that occurs in the data, i.e., descriptive gestures that represent objects in an animation video that both participants have watched before retelling the story to each other. They specifically concentrate on the relation between gesture occurrences and the gaze behavior of the participants. In doing so, they address two issues that have previously been hinted at in the literature (Gullberg & Kita, Reference Gullberg and Kita2009): Are speakers who have focused their interlocutors’ hand gestures more inclined to copy these gestures? And does the focus on one’s own hand gestures lead others to reiterate that gesture? Without anticipating their results, their approach provides partly unexpected insights on how eye-gaze and gesture production are intertwined, and confirms that alignment in interaction indeed is a multi- and essentially cross-modal phenomenon.

Kerstin Fischer’s paper ‘Conversation, Construction Grammar, and cognition’ is yet another convincing example of how transdisciplinary approaches that cross the boundaries of interactional and cognitive paradigms can lead to a much deeper understanding of language and/in interaction than any monoparadigmatic account can do. Fischer’s focus is on the intersections between Conversation Analysis (CA) and Construction Grammar (CxG). Her aim is to show that both paradigms are actually more compatible than is often assumed and that a synergetic approach to constructions, combining CA insights and the grammatical perspective of Construction Grammar, gives a far more encompassing picture of the construction under scrutiny than a pure CA or pure CxG analysis does. To make her point, Fischer takes the example of the English interjection oh, studied in great detail by John Heritage from a CA perspective. Despite the fact that Heritage focuses on classical CA parameters such as the sequential contexts and social actions oh is used in, and hence explicitly leaves grammatical or cognitive factors – the CxG focus – out of the picture, Fischer argues that his description can be seen as a grammatical one, showing that oh is a stable form–meaning pairing in a Construction Grammar sense. One by one, Fischer contrasts Heritage findings and claims with central tenets of CxG and convincingly argues that there are many more similarities between the two approaches than other studies within the field of ‘Interactional Construction Grammar’ have previously suggested. She is also careful to elaborate the deeper insights a grammatical perspective can add to the CA analysis, and how CA-relevant conditions of use can and need to be integrated into the constructional description. Finally, Fischer addresses what is often seen as the red flag by conversation analysts but at the same time constitutes a core interest of Construction Grammar: the role of cognition. In doing so, Fischer’s paper is a carefully constructed plea to cross the boundaries of one’s own research field and to approach the inter-relations of cognition and interactional language use from more than one research angle.

Each in its own way, the papers in this special issue present innovative ideas on how Cognitive Linguistic concepts can fruitfully be employed to study aspects of interactional language use. In doing so, they cross the boundaries of their own discipline and search for commonalities with neighboring disciplines like gesture studies, psycholinguistics, and Conversation Analysis. We consider this transdisciplinary approach particularly well-suited for the development of cognitive approaches to interactional language use. By looking at other approaches and their findings in the domain of interactional language use, we may reduce the risk of reinventing the wheel, and instead can explore the full potential of an integrated, Interactional Cognitive Linguistics. The contributions of this special issue offer a glimpse at the possibilities for the use of Cognitive Linguistics to account for interactional conceptualization, suggesting uncountable ideas for future research.