Human mind, language, and action are nowadays understood and studied as thoroughly social and intersubjective phenomena in many fields of research; simultaneously, human sociality and intersubjectivity per se have become topics for multidisciplinary studies. At present, studies concerning human intersubjectivity have at least three foci. One approach focusses on intersubjectivity as a property of the human mind, drawing on research in philosophy, psychology, and cognitive and neuroscientific studies (e.g. Bråten Reference Bråten2006, Hari & Kujala Reference Hari and Kujala2009, Rochat, Passos-Ferreira & Salem Reference Rochat, Passos-Ferreira, Salem, Carassa, Morganti and Riva2009, Zahavi & Rochat Reference Zahavi and Rochat2015). Another focus is on language and intersubjectivity as cognitive phenomena (e.g. Verhagen Reference Verhagen2005, Reference Verhagen2008). The third line of research has its roots in sociology, where intersubjectivity is understood as an achievement of organized social interaction (e.g. Heritage Reference Heritage1984, Schegloff Reference Schegloff1992, Sidnell Reference Sidnell, Enfield, Kockelman and Sidnell2014). Within this approach, the focus is on intersubjectivity as situated in the social world rather than the human mind. Drawing on the methodology of conversation analysis, this approach is radically empirical, emphasizing the importance of recorded data from authentic interactions.

Regardless of a shared understanding of humans as thoroughly social beings, these three lines of research do not easily meet. While psychological studies have, to a large extent, neglected the role of language in the constitution of human intersubjectivity,Footnote ¹ cognitive linguistic studies of language and intersubjectivity are either purely theoretical or focus on language as a means for sharing experience at the cognitive level, neglecting the affective and action-oriented levels of human interaction. On the other hand, interactionally oriented empirical studies on language and intersubjectivity mainly neglect – some even reject – the cognitive dimension. The result is that there is only a handful of empirical studies that approach grammar as a socio-cognitive, profoundly interactional and affect-oriented phenomenon, and hardly any of these use authentic interaction as data.

This special issue wishes to bring language into the fore in the studies in intersubjectivity. Moreover, it aims at initiating a dialogue between interactionally and cognitively oriented studies on language and intersubjectivity. Interestingly, both of these lines of linguistic research have their roots in phenomenology, whose primary focus is on the structures of conscious experience. They draw, however, from differing lines of phenomenological studies. Whereas the approach to intersubjectivity in cognitive linguistics has its roots in Edmund Husserl's phenomenology (see Zahavi Reference Zahavi2003) and, in addition, on Maurice Merleau-Ponty's phenomenology of the body (e.g. Merleau-Ponty [1945] Reference Merleau-Ponty1962), conversation analysis and interactional linguistics have their roots in Alfred Schutz's phenomenology of the social world (Schutz Reference Schutz1967; also Schutz & Luckmann Reference Schutz, Luckmann, Zaner and Engelhardt1973). This is due to the originally different foci of these two approaches.

Cognitive linguists are, basically, interested in the relation of language, mind, and the world. In particular, they study meaning as residing in conceptualization (Langacker Reference Langacker2008), and as grounded in bodily experiences (e.g. Lakoff & Johnson Reference Lakoff and Johnson1980, Reference Lakoff and Johnson1999; Zlatev Reference Zlatev, Racine, Sinha and Itkonen2008, Reference Zlatev2013; Jääskeläinen 2014). The nature of conceptualization is, however, a topic of debate. Langacker (e.g. Reference Langacker1987, Reference Langacker2008) claims that conceptualization can be approached from the standpoint of cognitive processing; the papers in this volume (in particular, Zlatev & Blomberg and Möttönen), however, take the phenomenological standpoint. One of the main arguments for the latter is that language is by default a social and normative phenomenon. If it were reduced to mere cognitive processing of an individual brain, the intersubjective validation that is necessary for any meaningful communicative act would be lost (see also Möttönen Reference Möttönen2016).

In a similar vein, there is a debate on the relation of pre-linguistic bodily experiences to linguistic meaning and, in particular, on the relation of image-schemas to conceptual metaphors (see Zlatev Reference Zlatev, Gallagher and Schmicking2010, Jääskeläinen Reference Jääskeläinen2013, and their papers in this volume). If linguistic meanings were directly reducible to pre-linguistic bodily experiences, how could we explain their conventionality, and differences between metaphors across languages? Moreover, if linguistic meanings were directly reducible to individual bodily experiences, how could we maintain the crucial separation of Self from the Other, which is a prerequisite for intersubjectivity (Zahavi Reference Zahavi2003:114)? These problems can, however, be overcome, if conventional linguistic meaning is understood as sedimented upon pre-linguistic, embodied intersubjectivity (intercorporeality, Merleau-Ponty [1960] Reference Merleau-Ponty and McCleary1964:168),Footnote ² but not reducible to it (Zlatev & Blomberg, this volume).

Whereas debates on intersubjectivity between different approaches to cognitive linguistics exist, there is practically a silence between cognitive linguistics and interactional approaches to language (see, however, Etelämäki, Herlin, Jaakola & Visapää Reference Etelämäki, Herlin, Jaakola and Visapää2009, Etelämäki & Visapää Reference Etelämäki and Visapää2014, Zima & Brône Reference Zima and Brône2015). This has to do with the fact that the two paradigms have seemingly opposing views on cognitive notions such as intentionality: whereas human cognition, naturally, has an important role in cognitive linguistics, conversation analysts and conversation-analysis-influenced interactional linguists mostly take an agnostic stance towards cognition in the sense of cognitive events (Heritage Reference Heritage1984:54–61). The reason for this is that conversation analysis has its roots in the phenomenology of SchutzFootnote ³ who, instead of analyzing the structures of experience that provide for intersubjectivity, understood intersubjectivity as a practical problem in everyday encounters. We live in an intersubjective and a cultural world where, in everyday interactions, we routinely rely on the idealization of the interchangeability of the standpoints, and on the idealization of the congruency of the system of relevance:

In other words, as long as our common understanding of the situation seems to be sufficient enough for the encounter to go on, we rely on the belief that we are in an intersubjective understanding without problematizing what it is that actually happens in the other's mind. This has brought the organization of a social situation as a research object per se (Goffman Reference Goffman1964, Garfinkel Reference Garfinkel1967) along with the practices that members of a culture use in order to make sense of each other's actions and to engage in joint-projects (see, e.g. Sacks, Schegloff & Jefferson Reference Sacks, Schegloff and Jefferson1974, Sacks Reference Sacks1992, Schegloff Reference Schegloff2007, Sidnell & Stivers Reference Sidnell, Enfield, Kockelman and Sidnell2014).

It needs to be noted, however, that contrary to a common misconception, conversation analysts do not deny the existence of intentions. They only avoid making reference to individual intentions or other mental states in the analyses, unless these are displayed in the data. The participants of an interaction do not, themselves, have a direct access to each other's mental states; they only have access to the displays of intentions, emotions and the like.Footnote ⁴ Since the conversation-analytic endeavor is to disclose the ways in which the participants of an interaction make sense of each other's actions in a sequence of interaction, the method would lose its research object if turns-at-talk were explained by reference to invisible and inaccessible mental states (to the participants as well as to the analyst). Moreover, the product of an interaction is always something else than a total sum of the consciously intentional acts of the participants, thus interaction is not reducible to individual intentions of the participants.

Despite this apparent difference between cognitive and interactional linguistic approaches to intersubjectivity, they both share the understanding that linguistic intersubjectivity is based on normativity. There is normativity in linguistic structure as well as in language use (Itkonen Reference Itkonen, Haukioja, Helasvuo and Miestamo1997, Reference Itkonen2008), and there is normativity in everyday practices in social interaction (Heritage Reference Heritage1984:115–120) to the extent that they become conventionalized in linguistic structure (e.g. Couper-Kuhlen & Selting Reference Couper-Kuhlen and Selting2001, Hakulinen & Selting Reference Hakulinen and Selting2005, Thompson, Fox & Couper-Kuhlen Reference Thompson, Fox and Couper-Kuhlen2015). Moreover, both paradigms agree upon the fact that life-world in which language is included, which is an important notion in both Husserl's and Schutz's philosophy, is the intersubjectively validated social reality in which we live in. However, whereas cognitive linguists focus upon the more stable and conventionalized aspects of language and linguistic structure, conversation analysts and interactional linguists focus upon the means by which social interaction and linguistic structure are produced and intersubjectively validated in situ.

The papers in this special issue are organized along a continuum which begins with empirical and interactional analyses of language, interaction, and empathy, continues to empirical analyses of mimetic schemas and the sedimentation of linguistic meaning, and ends with a meta-theoretical discussion of conventionalized intentionality. The papers can be, roughly, divided under three themes: (i) the construal of intersubjectivity in interaction, (ii) the sedimentation of linguistic meaning upon embodied intersubjectivity, and (iii) the linguistic architecture of intersubjectivity.

The first two papers approach intersubjectivity primarily as a social phenomenon that comes into being in-between the participants of an interaction, and analyze the linguistic means for creating and maintaining intersubjectivity in interaction. The paper by Per Linell & Jan Lindström discusses a particular Swedish construction, the x-å-x construction (e.g. flytta å flytta ‘move and move’). This construction is used in responsive turns; it picks up an element (x) from the previous speaker's turn, and the turn in which it occurs suggests a clarification for the situated meaning of x. By focusing on this particular construction, Linell & Lindström show the way in which the constant responsiveness of each turn in a conversational sequence takes part in the weave of intersubjective understanding. In this paper, intersubjectivity in interaction is understood as partial and dynamic (see also Linell Reference Linell2009); the mutual understanding between the participants of an interaction needs to be merely sufficient enough ‘for all practical purposes’ (e.g. Schutz Reference Schutz1953:8). There is, however, yet another aspect of the x-å-x construction, namely that its pragmatic meaning conveys that there are multiple understandings of x. The construction is, thus, itself a manifestation of interactional positions that are conventionalized in grammar (see Möttönen, this volume).

Whereas Linell & Lindström discuss the dynamically evolving intersubjective understanding in interaction, Ilona Herlin & Laura Visapää trace the roots of intersubjectivity by analyzing empathy in everyday interaction. By using one troubles-telling sequence as an example, they show the ways in which different dimensions of empathy are evoked and expressed in different phases of troubles-telling. They approach empathy as a dynamic process that includes three dimensions: emotional contagion, a cognitive dimension and an affective dimension. By combining theoretical insight from cognitive grammar with methodological insight from conversation analysis, Herlin & Visapää analyze the linguistic means for construing the degree of empathetic sharing in interaction. Moreover, they suggest that linguistic analysis may give concrete tools for analyzing and understanding how empathy takes place in real-life encounters.

The next two papers deal with embodied intersubjectivity and mimetic schemas (see Zlatev Reference Zlatev2008). Anni Jääskeläinen examines the use of two Finnish imitatives (ideophones that depict sound), humps and naps, in written texts. These two words are used primarily for portraying sounds: humps, portraying the sound of relatively soft falling, and naps (‘snap’), the sound of a sudden change, e.g. snapping. Jääskeläinen argues that the basic bodily actions that are the pre-linguistic basis for mimetic schemas are most often accompanied by sounds, and hearing is therefore a source-sense for mimetic schemas alongside vision and proprioception. She shows that these two words, humps in particular, are semantically synesthetic; together with sound, humps may also depict movement and other sensory impressions. Moreover, these two imitatives have more abstract, metaphoric usages for depicting emotions, particular kinds of temporal movement (slow or sudden), etc. Jääskeläinen suggests that the literal as well as the metaphoric uses of these words is based on intersubjectively shared mimetic schemas that render the sharing of profoundly individual experiences such as feelings.

Jordan Zlatev & Johan Blomberg focus on the relation between embodied intersubjectivity (intercorporeality) and linguistic meaning. The paper is a phenomenological as well as an empirical investigation on linguistic meaning, drawing on the philosophy of Husserl and Merlau-Ponty. Zlatev & Blomberg argue that linguistic meaning is intrinsically layered in human experience. It is sedimented upon but not reducible to pre-linguistic, experiential structures of embodied intersubjectivity. Whereas linguistic meaning is more abstract and stable than pre-linguistic, experiental structures of embodied intersubjectivity, the latter are more experientially rich, and latent but may be reactivated in specific contexts. Zlatev & Blomberg demonstrate this, first by reviewing evidence from ontogenetic semiotic development within the framework of the Mimesis Hierarchy model (see Zlatev Reference Zlatev2013) and then, by reviewing studies of non-actual motion (e.g. The road goes through the forest) in several languages. Their analysis shows that body and sociality are fused from the beginning and, by showing that pre-linguistic bodily experiences are not purely subjective, bridges the gap between mimetic schemas and normative, linguistic meaning.

The last paper is the most theoretical one: Like Zlatev & Blomberg, Tapani Möttönen draws on the philosophy of Husserl in his meta-theoretical discussion on linguistic construal in cognitive grammar (see e.g. Langacker Reference Langacker1987, Reference Langacker2008). Möttönen proposes that linguistic meaning, i.e. conceptualization, is best understood as acquired in multiple intersubjectively validated usage events and that it cannot be reduced to cognitive processing (see Langacker Reference Langacker2008:30). He argues that since linguistic meaning is social and, more importantly, normative (see Itkonen Reference Itkonen, Haukioja, Helasvuo and Miestamo1997, Reference Itkonen2008) by nature, and that since linguistic signs are used to convey meaning about extralinguistic, extramental reality, linguistic meaning cannot be grounded in individual cognitive phenomena. The key notions in Möttönen's paper are non-objective meaning, intersubjectivity, and intentionality. By non-objective meaning Möttönen refers to the fact that linguistic signs necessarily express an in-built perspective; any linguistically accomplished reference to an object is always a construal of that object, and any object that can be referred to linguistically has multiple linguistic construals. This means that linguistic meaning is intentionality that is conventionalized in linguistic structure.

Despite the seemingly different starting points of the papers collected in this issue, they share common themes. The papers by Linell & Lindström, Herlin & Visapää, and Möttönen all promote an understanding that linguistic structure is in itself a manifestation of intersubjectivity, because any linguistic object is necessarily perspectival and, moreover, any utterance necessarily positions the participants of an interaction in relation to each other (see also Verhagen Reference Verhagen2005, Reference Verhagen2008; Etelämäki & Visapää Reference Etelämäki and Visapää2014). Linguistic structure, thus, conveys the existence of the multitude of other subjects. The papers by Herlin & Visapää, Jääskeläinen, and Zlatev & Blomberg, all deal with embodied intersubjectivity, and its relation to linguistic meaning. Whereas Herlin & Visapää show how language and the body function for evoking and displaying dimensions of empathy in interaction, Jääskeläinen, as well as Zlatev & Blomberg, discuss the intersubjective basis for bodily experiences. They show how the embodied intersubjectivity can become conventionalized in linguistic structure for expressing more abstract meanings, these conventionalized expressions yet maintaining the possibility to evoke the experientially more rich content of their origins.

Finally, I would like highlight an apparent convergence of the radically empirical conversation analysis in the last, most theoretical and most cognitive paper of this issue (Möttönen). It emphasizes the social and normative nature of language, and the necessity of intersubjective validation of linguistic meaning. The paper follows Itkonen (Reference Itkonen, Haukioja, Helasvuo and Miestamo1997:55) in describing the structure of common knowledge:

1. 1. A knows x,

2. 2. A knows that B knows x,

3. 3. A knows that B knows that A knows x.

In conversation, intersubjectivity is provided by double-contextuality (Heritage Reference Heritage1984:242) and its interpretative corollary (Heritage Reference Heritage1984:254–260): every turn-at-talk provides an interpretation of the previous turn, and is interpreted with respect to is as follows:

1. 1. A does x,

2. 2. B's turn provides an interpretation of A's action, and is interpreted with respect to it,

3. 3. A's turn accepts B's interpretation by explicitly acknowledging it or merely continuing the conversation.

This means that if A does not initiate repair in the third turn, B's interpretation of A's initial action remains valid – regardless of A's original intention (see also Schegloff Reference Schegloff1992). In other words, as a method, conversation analysis focuses upon the intersubjective validation of shared knowledge in situ, and in this way, we can see where the thinking behind Möttönen's paper meets the ethnomethodological thinking originating in Schutz's philosophy. The cognitive approaches to language and intersubjectivity often deal with the more conventionalized aspects of language, and the interactional approaches examine the situated practices of intersubjectivity. Yet, the former presuppose intersubjective validation of linguistic conventions, and the latter also rely on normativity with respect to interactional practices. In order to understand human intersubjectivity with respect to language as a whole, we need both these approaches; and hopefully these approaches will find one another.