1. What comes after modality?

This special issue investigates the ways in which modal meanings convert into intersubjective and discursive resources in different languages.Footnote ¹ The focus is on the meaning construction processes that lead from prototypical modal expressions (1a) to different types of discursive functions where the meaning of necessity or possibility does not surface (1b).

In (1a), should marks deontic obligation, whereas in (1b) it is part of a construction conveying mirative meaning – in other words, the speaker’s stance towards the unexpectedness of the interlocutor’s speech act (see Celle Reference Celle and Guentchéva2018).

From a grammaticalization perspective, in cases such as (1b), the modal meaning has undergone desemantization. This is a chain-like process in which instances of decreasing degrees of semanticity follow one another in a predetermined order. In the modal realm, lexical or otherwise semantically more concrete elements generate expressions of possibility and necessity until these eventually bleach into semantically less and less specific, abstract markers that have reached the latest stage in their evolution (e. g., Lehmann Reference Lehmann2015). Bybee et al. (Reference Bybee, Perkins and Pagliuca1994) established a cross-linguistic model for identifying the paths of development for various modal categories. What all these tracks have in common is the transition from agent-oriented source meanings through speaker-oriented and epistemic modalities to subordinate uses (cf. (1a) and (1b)).

Van der Auwera and Plungian (Reference van der Auwera and Plungian1998) combined these paths into a threefold map. At the initial stage, the premodal field gathers lexical source items. These pass into the modal field through changes in their grammatical profile, such as auxiliarization. For example, the Finnish modal verb pitää is considered to have evolved from the concrete premodal meanings ‘to grip’, ‘to stick’, and ‘to hold’ into a marker of (deontic) necessity and evidentiality (Laitinen Reference Laitinen1992, Reference Laitinen, Cheshire and Stein1997; see also Bybee and Pagliuca Reference Bybee, Pagliuca, Wolfgang U. and Fisiak1985 for a similar account concerning the etymology of the deontic to have (to) in English). In the process, the modal pitää “lost” its intransitive uses, as well as its full person agreement, and prototypical nominative-cased subject. Compare the intransitive construction, including third-person plural subject-verb agreement, in (2a) with (2b), which displays no agreement between the modal auxiliary pitää and the first-person plural genitive-cased pronominal form and where pitää is accompanied by a lexical verb in the infinitive form.

At the other end of the map, the modal field reaches the postmodal domain. Linguistic forms no longer encode participant-internal or -external possibilities and necessities but display various kinds of semantic-functional behaviours (e.g., concession, condition, complementation, optative, future-time, quotative, and consecution). This is also true of the aforementioned Finnish modal verb. Implicative pitää is used in exclamative and other types of affective utterances, as in (2c).

In addition to making manifest the relationships between modal subcategories (Badir and Polis Reference Badir and Polis2023), van der Auwera and Plungian’s (Reference van der Auwera and Plungian1998) model of semantic maps highlights how grammaticalization paths cross on the interface between modal and postmodal, as both possibility and necessity tracks may lead to certain postmodal meanings. This can be illustrated by the French translation of the Finnish exclamative utterance in example (2c): ‘(dire) qu’un tel endroit puisse exister!’, where the subjunctive form of the modal verb of possibility pouvoir shows a postmodal use comparable to that of the Finnish and English necessity-based verbs (see Peltola Reference Peltola, Hilpert, Cappelle and Depraetere2021).

The Romance future forms derive their roots from the modal-postmodal interface, as the Latin modal periphrasis INF + habere (cantare habeo ‘I can/must sing’) demodalized into a tense (in French, chanterai ‘I will sing’). Moreover, future tenses sometimes complete a full circle, offering in turn fertile ground for further modal and, thereafter, postmodal meanings. Example (3) shows how the French future tense, after having developed uses where deontic necessity surfaces, found its way to performative discourse functions (see also Celle Reference Celle and Pauchard2008):

The term postmodality illustrates the position of these elements beyond actual modal uses. However, there is a risk that the linearity implied by postmodality hides the fact that premodal, modal and postmodal items often coexist and interact. Another issue is how to deal with the metaphor of loss that features in the descriptions of the passage from modal to postmodal. When looking at utterances such as those exemplified in (2b), (2c), and (3), it is difficult to regard the modal items as void of meaning or lacking purpose in functional terms.

Indeed, several studies have noted that the modal meaning does not fade away completely, but rather undergoes different transformations in its semantic structure. According to Celle (Reference Celle and Guentchéva2018), these forms combine different layers of modality, which renders their modal meaning more elusive, but not weaker. In Narrog’s (Reference Narrog2012) theory, an event-oriented modal element becomes more discourse-oriented (see also Narrog and Heine Reference Narrog and Heine2021 and Narrog, this issue). This happens through a gradual transition where the language form which initially encoded the potential or the necessity of an event taking place takes over intersubjective functions. It can now be employed to mark the speaker’s position (e.g., hedges), to construe the interlocutor’s perspective (e.g., concessive markers), or to organize interaction, textual structure or discourse (e.g., discourse markers). These accounts view the modal meanings that have undergone semantic change as more-than-modal rather than postmodal.

Following these approaches, the modal meaning does not disappear but is rather enriched with another expressive layer, as the conceptual structure on which the modal semantics leans is put to the service of interaction and discourse. This brings us to the question of the kinds of contexts where this reshaping typically occurs, in syntactic, semantic, and pragmatic terms. The evolution of the aspectual system may interfere in the life cycles of modals (Caudal 2018). it can also happen that, when a new type of setting for interaction emerges, the existing modal resources are reanalyzed in view of the new needs. The centuries-old epistemic uses of the Finnish construction [V₁ kuin ₁=kin], based on correlative verb reduplication, are revisited in contemporary online discussion group data when participants organize the interaction between discursive positions, sometimes creating long intertextual connections between posts within a discussion thread (Duvallon and Peltola Reference Duvallon and Peltola2025).

Studies in Construction Grammar have addressed the evolution of modal meanings in patterns where both grammaticalization and lexicalization processes come into play. This approach also allows us to deal with these developments within networks of constructions, beyond individual units (e.g., Cappelle and Depraetere Reference Cappelle, Depraetere, Cappelle and Depraetere2016, Hilpert Reference Hilpert, Cappelle and Depraetere2016, Hilpert et al. Reference Hilpert, Cappelle and Depraetere2021). In methodological terms, diachronic constructional maps can be drawn to show the evolutions on two different levels, as conceptual tendencies and as language-specific changes (Traugott Reference Traugott2016; for a discussion, see Georgakopoulos and Polis Reference Georgakopoulos and Polis2018). Combined with corpus studies, constructional approach can be employed to detect evolutions in the frequency of modal markers within patterns and in comparison with the general tendencies identified for the same marker; an otherwise declining item may increase its use in a given construction. This has happened with the English modal verbs may and might in concessive patterns (Leclercq Reference Leclercq2024).

2. Overview of this issue

The contributions in the present issue explore the semantic and discursive dynamics in the interface between modal, postmodal, and more-than-modal from different theoretical perspectives. Special attention is paid to the directionality of these movements and the relationships between different postmodal meanings, as well as the underlying modal meaning structure and its reanalysis.

Heiko Narrog’s theoretical account pursues the investigation of intersubjectivity in the evolution of modals, in dialogue with van der Auwera and Plungian’s (Reference van der Auwera and Plungian1998) model of semantic maps. Narrog argues that modal meaning is not lost but promoted from propositional level to a higher degree of discourse orientation. This change is considered to be unidirectional, and in Narrog’s model it is mapped onto a distinction between volitive and non-volitive modalities. Modal elements of both types can engage in the evolution towards more discourse-oriented categories. In the process, volitive can become non-volitive, and vice versa, but the overall direction is always the same: from event-oriented to discourse-oriented. Narrog illustrates this with examples of two modal verbs: the English can and the Romanian functional equivalent putea. Narrog furthermore elaborates on how the original modal meaning of possibility is the very basis for the emerging discourse functions. The elements thus remain modal.

Marco Favaro sheds light on the internal structure of the postmodal realm. The focus is on the evolution of the Italian adverb pure ‘also’ from additive semantics to discourse functions entailing concessive patterns, speech act marking, and illocutionary modification. Taking Narrog’s model as its main theoretical framework, the paper analyzes a rich array of data showing that although pure has not undergone the grammaticalization path of a modal element, its illocutionary functions are comparable to those identified for the items that have evolved from modal expressions. In doing so, the article highlights the crossings between the semantic maps of additivity and modality. Favaro furthermore argues in favour of giving illocutionary modification the status of a grammatical category.

Yueh Hsin Kuo tackles the question of change in modal meaning within complex phrasal constructions. Operating in the framework of Construction Grammar (e.g., Goldberg Reference Goldberg1995, Traugott and Trousdale Reference Traugott and Trousdale2013, Trousdale Reference Trousdale2014), Kuo analyzes the semantics of the modal marker yào in Mandarin and its integration into syntactic patterns. While its volitional, deontic, epistemic, and conditional uses are well documented, the function of yào in complement clauses remains largely unknown. Analyzing the relationship between yào and the complement-taking predicate, Kuo’s article draws up a tripartite typology for complement patterns including yào. Their common origin is considered to be a complement construction with prohibitive matrix, where the semantics of yào is aligned with that of the prohibitive, and the meaning of yào can therefore become more procedural. From this starting point, yào spreads to other types of complement constructions. In this sense, the complement uses of yào resemble those of a subjunctive (e.g., English should, cf. (1b)). However, some of the more recent uses of yào in complement constructions have remained contentful, which suggests a more complex grammaticalization path.

Bastien Poreau’s article deals with the Russian deontic modal predicate nado when combined with the discourse particle zhe. The combination of the two – Nado zhe! ‘Well, well!’ – is used to convey affect in situations of surprise and incredulity. Poreau starts by comparing nado with a neighbouring deontic modal nužno, showing that the semantic difference between the two stems from the nature of evidence underlying the expression of necessity. The modal meaning of nado is founded on perceivable evidence in the described situation, whereas nužno activates the speaker’s subjective evaluation of the circumstances. The article then proceeds to analyze how the semantics of nado is activated in the presence of zhe. The study furthermore shows that the pattern takes on different functions depending on whether it is used with or without an infinitive complement.

Naizhen Xia’s article opens the second half of this issue, where the focus shifts to the variation in the uses of modal verbs expressing possibility. Xia compares two sentence types with the English may and a subject-verb inversion, optative (May he get well soon!) and interrogative (May I see your ticket, please?). Despite the shared word order feature, the two types of utterances are fundamentally different. Xia teases out these distinctions by looking at the choice of subject, different markers of subjectivity and intersubjectivity, and the types of event described by the lexical verb, as well as a syntactic particularity of the optative sentence with may, namely its capacity to take an adverb or an NP in the sentence-initial position (And much good may it do to them). The analysis is conducted within the Theory of enunciative operations (e.g., Culioli Reference Culioli1990, Gilbert Reference Gilbert2001). Xia argues that the optative may is not derived from the may expressing permission. The optative may has followed its own path, where the relationship between the speaker and the interlocutor, and the expectations regarding these positions, have been altered.

Marta Abrusan, Alda Mari, and Enzo Laurenti study the commissive readings of the English modal can in conditional sentences (I can wash the dishes, if you like). Their article demonstrates that the speech act function of can in offers of this type is founded on its semantic meaning of ability. Having an ability entails having a disposition to carry out the action described in the offer (e.g., washing the dishes). The ability meaning is, however, backgrounded in the offer uses of can, contrary to similar utterances with the construction to be able to. The authors furthermore identify conditions that must be met for an utterance to be understood as an offer. In commissive can, these include a desirability and an authority component. The outcome of the offer is preferable to the addressee, and the speaker considers themself to be in a position to make the subject referent carry out the action. Finally, the offer-can is compared to similar commissive uses of will, as well as a set of non-standard conditionals displaying, for example, a loose semantic link between the main clause and the if-clause.

Corinne Rossari, Cyrielle Montrichard, Claudia Ricci, and Linda Sanvido investigate the French modal verb pouvoir ‘can’ and adverb peut-être ‘maybe’ from a postmodal perspective. These are compared to other modal markers, such as the modal verb devoir ‘must/have to’ and the adverb probablement ‘probably’, which have not developed in the same direction. The authors first establish a semantic profile for the items studied. This allows them to predict the forms’ potential for use in different contexts in the data. Combining qualitative and quantitative methods, the article detects the most typical linguistic environments for each postmodal marker, showing that these are not shared with the other modal markers.

Daciana Vlad’s article closes the issue by comparing the aforementioned French markers pouvoir ‘can’ and peut-être ‘maybe’ with their equivalents in Romanian a putea ‘can’ and poate ‘maybe’. The analysis highlights the transformations in the semantics of these markers as they orient themselves more and more towards discourse functions. However, the initial modal meanings remain in the background, which, according to the author, makes the term more-than-modal more adequate for describing the later stage of the evolution than postmodal. The contrastive analysis between French and Romanian markers suggests that the Romanian modal verb a putea shows more variation in its syntactic behaviour than the French pouvoir.

Several of the contributions in this issue are based on papers presented at the conference Postmodality and the life cycles of modal expressions we organized together with Evgeniya Gorshkova-Lamy at the University of Caen Normandy on 2-3 June 2022. We would like to thank all participants at the conference, as well as the CRISCO lab for its support in organizing it.